Min-Maxing Your Steam Launch

Steam tests all games like Steam gives all games a relatively fair shot.
So, can I convert as many people as quickly as possible to reinforce the signals to the algo that like, hey, this is something that you want to drive more and more and more and more to. Do you want to know the controversial thing?
This is a conversation with Cory Darby, an indie game business strategist who reached out to me with some interesting data. I was indeed intrigued by the data, but also by Cory's personality because together with a business partner, he developed a successful incremental game with just 400 hours of time investment. That is wild. 35,000 copies sold and it's over $100,000 it's made. Might be controversial because I've had other people say like, "Well, is that sleazy? Is that the seto of the consumer?" May not want to hear it. I think a lot of the industry is going to shift from this purely passion creativity and more and more to this capitalist businessdriven. I guess I am the evil corporate capitalist who uh sees the divide and I'm like I am going

to be on the right side of the halves.
The interview I recorded with him is controversial. It's going to spark some discussions and it's going to haunt me for weeks to come. I hope you enjoy. I've been playing around with some of the data across the other indies that come to me typically since uh the game I had done, Click and Conquer, was very successful. And one day I had someone reach out and say, "I don't know why my game failed." And I didn't know because I'd looked through everything and it was like 1 hour before the call. I was like, "I wonder what their first 72 hours kind of look like." And I had some comparison of other games I can compare to, but also this Cliff and Conquer game and it was pretty shocking. It looked much more um equal footing compared to the other games I would say. Like there's pools. There's like a left pool on the on your axis. There's a right pool. There's a spot where I'd say like 90% of indies land in this middle.
Let us know what exactly are you describing on the graph. What are the axis?
Yeah, exactly. So the axis would be like

uh sales and visits. So how many people are viewing the store page versus how many purchases they got. So like the bottom left on my side at least is um like you know uh low visits uh low sales versus a top right would be high visits high sales.
Basically the interesting thing you noticed is that across multiple games that you have compared now the visits in the first couple of days are always relatively similar. Right.
Yeah. So it it's something like they'll get between 30 to 60,000 visits and typically half of that's driven by discovery queue.
Um now it's a small sampling like the games I publicly put on this graph are like 10 games. Behind the scenes for people who didn't want to have their data exposed it's another like 30 and it still holds true.
How do you have access to so many data points? Is this all just devs reaching out to you about their games or
Yeah. So, word of mouth and I do like a 30-minute free consultation calls. So,

that gets a ton of people coming in and then usually a part of that is, hey, I'd like to use the data um without the name. Are you fine with that? Yeah, but obviously some people aren't. So, those aren't the ones on any of the graphs I produce. Got it.
But I personally know where they fall. Uh so, it's all up here. Fuse are always kind of similarish, but where they differ is in how well they convert to sales, right? That seems to be the differentiating factor in the first couple of days.
Absolutely. Like this is not uh let's compare everything to Click and Conquer. It was just a single data point.
But if we do the comparison like we're talking in some cases 100 times the amount of sales. Uh even though you would say like the visits for click and conquer might be like 20% higher. So, we expect like I don't know 20 to 40% higher on the sales, but the sales are like 90 95% uh higher. Uh we're talking like 35 sales versus 3,000 sales. And

for context, this is eliminating everything else like Click and Conquer for those who don't know.
That game didn't have any YouTube or social coverage for first three days. And that's particularly why I looked at 72 hours. uh it was an unpopular upcoming uh or uh popular recently released. It had 2,800 wish list when it launched. Uh like it it didn't have anything unique about it. It just converted very well. So Steam what you're or I guess the theory we would basically derive from your observations would be Steam tests all games like Steam gives all games a relatively fair shot and views in the first couple of days are often very similarish for for indie games and then based on how your game performs uh your games either going to go bigger bigger and get more visibility from Steam or Steam's just going to be like nah there's there's better games to show and then your your visibility obviously dies down right So the observation is not that everybody gets the same amount of

visibility forever, but it's more like Steam gives every game a relatively fair shot.
Yeah. So there there's a couple caveats like in the bottom left. Uh if you can't basically secure 10 reviews in 24 hours, you're just going to be way less visits than everyone else.
Yep. because obviously discovery Q, you'll still get some traffic from discovery Q, but once you hit that like 10 review threshold, it just um spikes way more. And then the very top right of that graph would be things like if you hit popular up counting, your your visits are just going to be disproportional to everyone else as well. U but we get this little bucket in the middle here and that's where pretty much everyone falls in. if you're got your 10 reviews in your first 24 hours and you weren't like popular up and coming plus or minus about 20%. Now, obviously that sample uh no one in there went viral.
So, so again that's why they're they're bucketed but yeah they're it seems like

Valve does a good job of what I call the seed audience. Like I have a background in uh software for ad tech company, a pretty major player in the space and we do very similar just like on YouTube you're going to get a seed audience and how well does that seed audience convert for like the click-through ratio the watch time and that's going to determine how much more you're going to get. I guess the thing is like why do they convert so much better?
Yeah, that would be that would be the the follow-up question, right? Cuz being like oh the the games that convert better, they get more visibility from Steams. If you think about it, it's it's almost kind of a little bit of a common sense thing. Um, if you think about you're like, "Duh, yeah, that's that's how the algorithm is supposed to work, right? If if I were to write an algorithm like this would probably look the same. I would like test the game and if it does well, I would give it even more visibility. That's kind of makes sense. But it's still firstly it's still nice to see that that's how it works. That that's actually how it works. That it's actually like relatively fairish,

right? Once you hit that minimum threshold and yeah but the secondly the question is what do we what do we do with this now like how do we make sure that the game converts well why do you think your game click and conquer converted so well
not one individual thing I would say a combination so uh I I I think there's three camps I see it's like can you market the game well okay and then on uh so I guess this is a how how to market game camp
and then there's uh the Jonas camp just make good game camp
and and I think I'm somewhere in the other camp of both important but how do we convey this on a page cuz the thing we're talking about here is you could have a not great game and this doesn't matter in this equation. You could have a great game and still doesn't matter necessarily in this equation because they haven't played your game. It's only what you can convey to the the player watching. It's the perception of is this good game and good value is the other thing I should say. So, it's like maximize the store page, the trailer as

good as possible. Uh, the pricing should be a good cost versus value um for the consumer. Not what you think it's worth. Like, oh, I put a,000 hours in, it's it's definitely worth $15. It's like, if I put this in front of the consumer, do they feel like it's worth $15? And if it's no, then they're unlikely to convert. Um, so Click and Conquer, like I had gone and looked at all the similar games and we priced even below what we thought we could get away with. Like we did a 40% discount at launch, which maybe that's uh controversial in itself there. Um, we did maximum time, maximum discount. So minus 40% off for two weeks. Um,
some of that is like psychological like trying to convey to the consumer it's a great deal. just like retail 40% off. Um, looks like a good deal. You're like, "Oo, wow. Purchase now." Uh, I don't want people clicking the wish list and coming and buying it later. I would much prefer the immediate. Um, my whole

mantra in this is like shortest amount of time, biggest impact. So can I convert as many people as quickly as possible to reinforce the signals to the algo that like hey this is something that you want to drive more and more and more and more more to because you know obviously after 72 hours you want to know what your long tail is on your uh sales and things like that is
interesting cuz that's the same thing told me what one her of her biggest regrets with the 1.0 O la la la la la la la la la la la la la la la la la la la la la la la la la la la la la la la la la la la la la la la launch of the wandering village was which was was like we had so many wish lists we should have launched with a way higher discount right so it's interesting that you kind of share share that idea
and we had baked it in like honestly you can work back like okay just base price your game with the 40% discount already factored in so you're not pulling money off the table for no reason like if you intended to price the game for I don't know $5 and you're like, "Oh, but 40%

squid really hurt. I'm not gonna make like anything." Well, take whatever number you want, $5, add in the 40% discount, and start your base price that uh there's no disadvantage on Steam to decreasing the price. There's no cool down period. You can decrease the price of game at any point with no cool down. If you try to increase the price, you take a 30-day penalty where you can no longer do discounts for 30 days. But your your game for example, the base price is only uh four bucks, something like that. So it's already very cheap, right?
Do you want to know the controversial thing?
Did you decrease the price after the discount or what do you do?
Exactly. Yes. Um so we did it where the price is higher to factor in the 40% discount.
After that's done, we decreased the price uh because we're only going to be doing 20% discounting. The people originally bought the game with 40% off, they still got a better deal than the new base price minus 20%. But I was aiming for this retail psychological oh

my goodness, the game's off 40%.
And uh that is why the game price is there. You can see on SteamB uh Steam DB. I've never hidden this. We've never hidden this.
Uh it might be controversial because I've had other people say like, "Oh, is that sleazy? Is that deceptive to the consumer?" But the way I've justified it is like the original buyer still got a better price than um the current discount and um everyone else is going to discount the game eventually. Like eventually the price is going to be lower across all of our games than the original group. But
yeah, to be honest, I'm kind of in the blame the rules, not the player camps. I'm like if if Steam allows it, then you you shouldn't be surprised if people do it. Usually Steam if if they find like these exploits, they usually fix them, right? If if that's something that they don't want, they'll they'll fix it.
Yeah,
we'll see if they do that or not.
Well, I I might be the reason why, you know, if everyone's doing it, it's not a problem if one person's doing it.
Um but and the reason to why we did it

was I knew the market wouldn't bear that base price. like they would pay whatever it was minus 40%, it was in the sweet spot for these incremental games and where other incremental games fell. But if I left it at that, it it was way too high for
the amount of other popular games and the quality of those games and what we'd be asking for, it would just be unreasonable. The consumer would just not justify. It's also a very short game. It's only two hours of play time. So, you know, you have to have a game that's price proportional to what they're getting out of it.
I think Landfall did that with one of their games where the game was entirely free for the first day, right? They did like a 100% discount. Have you heard of that story? I
I have not.
Yeah. Unfortunately, I don't have too many details, but I think they had similar thinking where where it was just like, let's get as many people as possible in in the first day. We'll basically give the game away for completely free. and they will create so

much buffs that we'll still get a very very good tail after that. And in that case it also worked out very very well. And I think Steam has since uh since already fixed that exploit cuz other devs were like oh we want to do that as well. And Steam was already like no this is not not a thing that's going to happen regularly on this platform. So this has now this is not a not a thing anymore. But
still interesting to see.
Mhm. you're always looking for the tactics, right? And and a lot of what we're talking about here is like this isn't going to make or break a lot of indie games. This is more like, oh, can I minmax to get more out of it?
Uh I I mean, I guess if you overly price your game, and I have case studies on my side that show this, like you can essentially kill your game. uh a game that normally would have did decent uh but you have the price like two times what consumers are willing to pay uh with like a 40% refund rate and it just kills out the game like less people are

going to buy uh the signal to the algo is okay less purchases probably not good game even if the game is fun um it's just you're indicating the signals to the algo one of the many signals would be hey there's not a lot of purchases it probably means the game is not great or the Steam page is not converting yada yada
right so the the price being too high can definitely kill your game but like these special tactics of is it like 20% or 40% discount or whatever it's it's not going to kill your game most likely right
see I I don't know because I I can't go back for click and conquer and like well would it have done as well with only a 10% discount
I think it would have I think honestly I think it would have done okay because I can see that your store page is very well optimized and Like my personal gut feeling would be that it would have done okay, but you're right. We can obviously we can't know for sure.
And I've had the argument, well, why don't we AB test it on other games,
but you you can't because there's so

many other variables. The only true AB would be it Valve itself being able to do this because you would have to get two near identical consumers of purchase power. both show them different pricing and different like uh yeah exact same pages different pricing and then have a big enough sample size to actually see hey does it impact or not. Uh, a different game is all new, different variables, different trailer, different tags, different audience, different meta, too. Like, you can't just say, "Oh, 10% on this game, 40% on this game. Let's see." The the only way to truly do that AB would be at the platform level. And and Valve doesn't give us those tools. So,
you think we're slowly, seem like slowly turning into like the mobile market where games are becoming cheaper and cheaper like we can see these incremental games very often. they go like for for a couple of bucks, right? And the the the trend seems to be going downwards. So games are getting cheaper, discounts are getting higher. Do you have any idea where where that is

headed or do you think it's going to stabilize or what what's your take? uh particularly for the incremental games. I just think the time you get out of those games uh of like pure novelty, pure enjoyment is much more lower and people see the market as easy money cash win. So there are more games entering that market. Uh and you get doesn't help that there's success story like click and conquer and people see that and they're like you built a game in 400 hours and made how much I'm going to do that. M
so you know consumer has more choice and there's more variables and now it's like do I want to spend $9 for this incremental when I could buy two $4 incremental games. So I think that genre is currently kind of getting pulled down. Um I also think it's fair to the consumer. Like if you're only going to get two to three hours of play time, you kind of
it would take a lot at least for me and I played a lot of these to be convinced to pay like $9 for the game. It would

have to be like astronomically great on something like think of it as a radar. Like artwork would have to be phenomenal or the mechanic would have to be insanely new or the content I'm getting out of the game and not feel like it's just like the same repetitive grind in and incremental would have to also be good for me to justify that. Uh right now a lot of these games are like average round out rand uh average round it out and I know it's two to max maybe four hours of play time and it's like okay you're going to be priced at least in US dollars probably something between 3 to6 US then I don't think the global market is heading down. Um I I I don't know do do you think there's just a global trend of price decreasing? I mean, a little bit would be normal, right? Cuz games are getting easier and easier to make, and if something gets easier to make, then typically typically it does get a bit cheaper over time. So, I think, yeah, sure, we'll see a little

bit of that. But then, of course, there's also the fact that you're also constantly competing with the entire back catalog of Steam where there are tons of really, really good games on Steam and a lot of them are often times on like 80 90% discounts. And arguably that sometimes does make it a bit more difficult for for new games to break through. But I think it's still it's still totally possible. You just got to come into the market with some some new ideas and some stuff that has hasn't been done before. Then I think it's actually uh still still kind of fine.
I think there maybe even a counterpoint to how far you can pull down the prices and discount. Uh the example I could give is
and this person's h has on their YouTube channel. Um, I I I worked with Aramis, who has created a game called Chess Survivor, and it's been around, I think, three three and a half years. And he took the standard advice of like just discount your price over time, yada yada. And and there's truth to that, but he had gone to the point, I think, of

the extreme where the game was so price low that it was actually hurting sales cuz like when I went in and looked, it's like, you know, you're converting better on sales and wish lists and making more money at a higher price point than where you are now. And I actually can't definitively say this, but I think consumers are looking at like whatever minus 80% discount and going dead game low quality. Uh like it's so cheap that like why would I buy it? Um because we bumped the price back up, you know, from let's say - 80% to - 60% and it's selling more even when you factor out festivals and content because it's a three-year-old game. There's not really a lot of new content on it. Um, and it's like look, I don't know if it's just perception, like this people are willing to pay and they feel like they get a better value because the price is a little bit higher, uh, versus a game that's like, you know, a dollar, you're like, ah, it's it's dead game. It's probably not great quality. So,
yeah, it's it's I think it's pricing is very much a psychological thing, right? So, if it's like it's like 95% off, then

you're then you're like, why why not playing this anymore? you know, just buying it, you're not going to get any updates, no patches, like it's unmaintained. At least could be a perception of it.
But besides price, what what else can you do to make your page convert better? What do you think?
I think it's underrated and and I worked with a game where I spotted this is the tags. Everyone talks about like put the proper tags to your game. Discoverability, you want people to be able to find your game based off the tags. Do you want them to be on store page and recognize the type of game from tags? Yes. All valid, all true. Please optimize. Uh what I think misses is people don't pick the tags that work for Steam, they pick the tags they think match their game. And what I mean is like when you are in the tag builder, tag simulator in the portal, whatever they call it, um you should be looking at the games that are similar like and if those aren't matching like as close

as possible that you could say like these are games that are like my players exactly
then it's not going to do well um because I have a case study where a game launched great trailer uh a good publisher uh amaz subjectively amazing store page from my perspective good gifs good description I couldn't find anything. Even the tags make sense for the type of game. I can't say the game name, but what I'll say is it's actually similar to the Islander game that you made.
Uh, so it's like this little cute cozy builder ass game. And they're more like this game was Factorio and Rim World. And I'm like I I from my perspective, I don't think there's a lot of overlap on those types of players. for going from pretty hardcore like you're destroying people and your little villagers dying or full automation to hey uh I got like an island and it's minmaxing building placement and making like a design of what I want it to look like.
Yeah. And what I saw was, hey, they got

the same amount of traffic of visits uh from Discovery Q. And their sales were very, very low, like you know, 100th what I would expect. And I think, like I said, that seed audience from Discovery Q, we don't know. It's a black box. But if I go back to adtech world where I would be building this, I'd say, well, you need some sort of starter audience. isn't the ideal starter audience the ones that you're saying your game is similar to. Um because you could do random, but again you could get tons of false positives, but what random? Like the first 100 players that log on to Steam, there's a disproportional amount of players on Steam that are going to be like say FPS because of CSGO. So that's not going to work well with these games. Ideally, we want people that we think would convert. And the way that you know Valve can get that is okay, what tags are placed on a game? You said it's similar to these games. I'm going to try and take that audience and show them your game. Oh, it didn't do well. Well,

yeah, it didn't do well. It's World and Factorio and you got like a cute cozy island building game and uh I don't think that audience is going to work. So,
right. So, seat audience means the the audience that the algorithm tests out your game with basically, right? And you want uh you want the seed audience to be as suitable for your game as possible so that the algorithm gets better signals from from that audience. That's the the idea you're saying.
Yeah. You get um let's say you know both of us I don't know uh we we like to maybe um and now we're talking about this like
cutesy cozy game that isn't ours and we and it's recommended because it says oh this game's similar. we look at them, we go, "No way. This is not a game I'd ever play." Um, so we don't convert. So those first like, let's say, 72 hours look not great. Your sales are lower. So indie is like, "No tears. My game is not going to do well."
Um, and then what also hurts is you're

signaling to the algo that like, "Hey, this is either not a good game or the store page isn't doing well." Like, it doesn't care why. It's just like it's not converting to sales or wish list.
And what it does is just get pull back traffic. It's just going to claw back, claw back, claw back. Um, because it's saying, "Hey, I just gave you the most ideal audience you told me and it's not converting. I don't care why. I'm just going I just know I should send that traffic somewhere else on a game that does convert." So like your you'll have some initial spike and then immediately this drop cuz some wish list will convert obviously that you have and these are the more ideal audience. Some discovery overlap will oh you play Factorio. Ah, you also happen to play cozy games. Um, but then you'll get this giant dip down and you'll pretty much never recover like outside of like sales and festivals. I don't want to say dead game, but dead game.
You you like my theory is essentially just make a great game, right? And what what a very important part of my theory is is what I call appeal. And I think

this idea of appeal is very similar to what what you're talking about. It's like when you come on a store page and you check out a couple of seconds from the trailer and you check out the screenshots and you're just it just the game instantly connects with you and you're like I want to play that. What do you think? What creates this kind of immediate connection cuz every everybody's talking about it. I also like Filina when she was like how does she pick games she publishes she's also talking about this I need to feel this immediate connection with the game. How do you create this immediate connection with a player with with someone who sees your game? What do you think?
I would say I am probably the least game dev person that's been on this show. So, I consider myself not very skillful in this
and I let the other smart people do it. So, I am a person who says what is already winning in the market. I go play that game and say, "Ah, that was smart.

That mechanic is good. that game loop is super game juiced and then I go I'm going to build game like that but ABC differently with XY Z twist. Um so I don't want to be the one doing the market validation and try and find that that that appeal. I want someone else to find the appeal and let the market say yes it's an appealing game and it checks off the box and I just want to use that as like a north guiding star if that makes sense. I I'm not a very creative individual.
Got it. While people like Renee are like, "We're making 20 different prototypes and seeing which one does the best." You're like, "You make the prototypes and I just cleverly pick the things that I think worked well."
Yeah. Because I think it takes a certain type of individual to have that mindset to come up with good mechanics and novelty, particularly novelty, something that we haven't seen. And, you know, I think I'm a productive person and I can spot it. I think I I'm good at spotting it once I play it and see it. I think a

lot of people are. But to come up with a concept in my head or purely in my head or even prototyping like I I know you've talked about prototyping with like gray boxes, minimal art, but I really really struggle with this. Like if I take Click and Conquer and we strip out the assets and strip out the music and go back to like the core game mechanic, I don't think I would say it's a fun game. Like I would really have a hard time envisioning how fun it is. But once I add in all the things like I can get there uh Click and Conquer is inspired by a game called Dixium and then that's how we create that game. So you know the next game that me and Sakas are working on is based off of another game. like we are very much a like you said, let the really good creative people go do it and then we just find something that we didn't like or could enhance and we're like, okay, we're going to be this vertical slice of that game but with this different thing.
So yeah, the thing with the gray box prototypes is that almost anything can

become fun when you polish it enough and when you, you know, when you play test it and iterate it a bit. So, I'm I'm slowly converting to the camp where I'm like making making the game fun isn't maybe even necessarily the biggest problem. It's like obviously for for newcomers and for like beginners, making a fun game is still quite a challenge, but for somebody who's been making games for a couple of years and knows how to polish and knows how to play test and iterate, making a fun game isn't really the main problem. the the main problem is really how do you make something appealing that connects immediately and that players immediately want to play when they see it. And a lot of that is, as you said, also about context and and theming and so on, right?
So you for your game you chose a like a semi-realistic war theme. Is there a specific reason behind it or where do you go with that? Uh it was way different than Digium

and um we had an asset pack that fit the theme. Uh me and Sakus are both not artists.
Uh and we had a time constraint like when you're really The original plan was only 300 hours for everything.
Like we're talking Steam page trailer development, QA, testing, balancing. It's not a lot of time and neither of us are strong in art. So it was like, well, what it assets do we have to save time? Oh, this one. It's got little soldiers and tanks. Oh, it's got a bomb. Okay, we're we're we're doing war themed. You know, there was a pro uh I guess no pro, but con to this was people thought we were like tapped into the news and making some controversial political statement and it's like neither of us even watch news. We didn't even know what was going on. We were just making a game around an asset pack. Um but yeah, people did not like that it was a bombing theme war game. So
yeah, I mean to to be fair, you know, I can can kind of understand that I'm not not the the biggest uh fan of realistic

war games, but I do have I mean I do have the intelligence to differentiate between game and reality. So I think in in that regard, it's less of a problem for me. But I I can see why it's maybe not some people's favorite theme. But then on the other side, for some people, they really like that kind of stuff. So fair enough.
There was no statement in the game or story, too. Like we we don't even say that you're the the good or bad people. It's actually up to the player to determine whatever they want. Like um you know and you're only bombing um soldiers and tanks. Like there are no civilians in the game. It's you know it's up to the player to imagine whatever story they want.
Yeah. I mean and also you know these these these war games have been around for a for a long time. There are tons and tons of war games
and as far as pick something now like different looking back now. Sure. Yeah, but
would you
Yeah, because in the middle of development, I don't remember where, maybe it was like level design book or something and I remember reading that uh

Space Invaders, it used to be a similar game when it first got prototyped. I I think it was you either bombed cities or bombed uh villages or something like this or like you shot villages or something like this or you shot people, I think it was. And um they said that they thought that would be like over the line and people wouldn't resonate with the game and that's why you shoot these like alien looking shaped
ships that don't connect to anything. And I read that and I thought we should probably take like you know a guiding principle from that. But we were already 80% into the game and it's like do we like do we go back and rip it out? And we didn't even think the game would be that popular. like we didn't think it would be this as as successful as it was. So, it was like let's just ship it. We're already invested. We're already too far. Uh maybe was sunken sunken ship fallacy, but we we just committed and doubled down.
So, in terms of game play, click and conquer is pretty straightforward or

what how exactly does it play?
It's only one button. Literally left click there. There's no keyboard shenanigans at all. There's no WD. There's no shift alt control. You literally left click.
Uh you're given a map with clouds on it and you drop the bomb and you either hit the target or you didn't hit the target. Kind of like a mind sweeper.
And the bombs you upgrade they get a reveal radius so you can see more of the map to actually and you upgrade these. There's a prestige. Those are really all the the game um you know behind the scenes. And what I think I want to cling on one of the things you said about this peel and we're talking about making a good game and like peel I always associate to like the perception.
It's it's like you can
you can see something that you desire. I don't know a car. Wow, it looks so great. You get the car, you get in the car, you drive it and you realize car not good. And um the appeal for me is

like the store. what is the perception or what can you convey to the end user that looks and tells a story and and kind of actually a trailer you can tune it to be like quicker pace than your actual game and then people are interested like I'm not saying deceive the user but I'm saying you can present all the highlights uh then you get in the actual game and you know it is is the mechanic fun yes make good game as you were saying the other piece that we really tuned in on click and conquer was Is the pacing correct? Is there novelty? Like I had done the analysis. I'd gone to like games like Batcher and literally Excel sheeted out like how many times am I clicking? When am I clicking? What's the time between each reward? Um you know, and got it down to basically like, okay, every 30 seconds the player has to be presented with a choice that they're making some decision. They need to be doing an action at least every 5 seconds. And there needs to be like a core new introduced mechanic or twist to

a mechanic every six minutes. And like if you want to build a further game for Click and Conquer, we knew like every 60 80 minutes you have to introduce something that's entirely new. Maybe socially break the contract of how the players used to playing the game. Um but if at least for us that's kind of what we based around. So
okay. And this is based on the analysis of different games. So you kind of checked out how are other successful games doing it and then you derived some rules of thumb from those essentially that and just like playing it through oursel always tracking exactly where um like I got a foot pedal. So I record myself playing the game whenever I put down my foot
it marks on like OBS recording a disengaged moment
and then I go and look exactly where does that happen? Where was the last reward? Where was the last interaction? How long was that? Why do I feel this way? Okay, it takes us 9 minutes to get

here. It's a slump in the game. It feels like a grind. Okay, we got to tighten it more. Um, and just repeating, repeating, repeating like, yes, the other games gave us a idea of where we should fall and we went through that, but then we just playing our own game going, is there any spot or getting other people play any spot where you felt like it just kind of got slow or where would you close out on the the game? uh right around third level of XY Z I'm not getting new stuff okay we just need to decrease the time when you get that new stuff um it was very maybe like data driven like I think I don't know if many people are doing it this way and they're just like I they're more creative and they build a game of how they want to build but I'm all about like the Excel sheets and like where are people dropping off where are they disengaged like where do they click less where do they take longer to put their like mouse on so the action they're supposed to do. Um because that could be a sign of boredom. So yeah, maybe it's different. I don't know. Jonas, are you ever building games this way?

The This is definitely one method of play testing that I haven't seen before. I'm certainly intrigued. Like you record yourself playing the game and you have a pedal that you press when you notice that you're disengaged.
Yes.
And then you analyze like your own play test video and are like, uh,
it goes even deeper than this. Okay, so I wrote a tool so that it cuts out all the dead the spots that the pedal isn't pressed down so that I can get these minimum feedback clips because when I hold down the pedal uh I usually talk like I feel really bored. I don't like doing this. I didn't have anything in the last five minutes yada yada
and we end up with this super edit down. So the VOD might be an hour of me playing the game, but we only get seven minutes of actually clip data. And then I can send that over to whoever I'm working with. So it's a timesaver, too. Like I don't want them watching an hour of gameplay. Um I want them watching just the direct feedback and then they

have like the the full VOD but also spots of only where. Um and then I also take that I transcribe the audio and then I run it through chat GPT to give a summary of the feedback too. Like I'm telling you, man, I'm building tools of tools for this.
Okay. I I'm Yeah, I'm intrigued because uh I in when it comes to some things, I'm I'm starting to um like experiment with similar approaches. For example, uh this podcast is also very very much automatically edited. not not like fully by AI but just like simple Python automations and just automatically cutting between speakers editing out silence and then I only go in for the
for the for the refinement final cut is essentially
I think there's a lot of lot of power in automations especially but like especially if you mix it with like these new possibilities that AI gives you now right
absolutely because maybe this is also unique or I just don't know like I will

ship seven more games probably let's say 4 to 7 by June and the reason is I do collaborations with other people
so I come in okay 80% of the project you 20% of me and that 20% will be like balancing pacing uh testing um maybe there's some development or UI stuff and the moment you start working with more than like two people you start realizing common problems and you start realizing okay I can't scale to do this across six people. I just need to build the tools to like simplify. Uh even the example there where I cut an hour down to six minutes of direct feedback. That means the people I'm working with, they don't spend an hour watching a VOD to get like feedback from me through the play test. They're spending six minutes exactly what they need. It's also helpful for me to have the data so when we're talking, they're like, "Didn't you say something about that level in the past? It's all transcribed so I can just search through

everything." Ah, yeah, here it is. Here's the VOD. Where exactly did I struggle when I said I didn't understand the UI? Um, like it creates this archive of progression that we can go back and look.
Uh, I built a generic skill tree editor for all these incremental games that I'm working with so that I can go in my editor, build out the tree how I want, and have no concept of the game, and then pass along the JSON, and they just load in the JSON, and there's their entire skill tree. And that means they can build a game and I can build the skill tree. And it's interesting because like it separates out like you just build your game and I just build out the pacing and like the balancement of it. Um this has been interesting.
But again,
how do you how do you start so many projects with different developers? Do they reach out to you or do you start these projects yourself?
I would say sock house and I uh because we had done in our game, it was obvious we should continue this. we work together. But all the other ones are just either they've seen me or it's word

of mouth or they reached out to like sockos about click and conquer or um like it was after being on that podcast but also mentioning him like hey I'm really interested in the data of your games can we review the games and then just talking back and forth and you know finding a sweet spot where for him he was like I just want to build a game and what that meant to him was like the the polish and the mechanic and like features and he's like I don't want to do all this super tuning balance and you know cost economy and making sure the player's always engaged and I'm like I love that stuff that stuff is just like tuning so we partnered and that's when I realized okay I need to build an editor tool so we can separate out responsibilities it's like I have my own GDAU application that's just tree building and he's got his game and I export JSON I'm like load in this and just build the tree from that and it's uh very convenient actually. It makes

like I don't have to worry about the project and how it's evolving. He doesn't have to worry about any of the tree stuff. Um yeah.
Okay. So working are you working on a bunch of different uh incremental games basically or are there different types of games you're working on right now? I'd say most are incremental, but there is another that's a rog like like Bellatro um game.
And then we've with one of them we've talked about doing a simulator game because they're convinced that like that genre is just you know overpowered. Like in their mind they're like if you build simulation game like these uh grocery store simulators, it's just going to do well. Um I don't know. We'll see how it goes. Uh I I think that is a a trending genre still. Um, but they they they range. They range. And it's interesting working with people like different personalities, different work types, uh, different ideas, different approaches. Uh, I kind of feel a bit maybe spoiled is the right word because it's easier

for me to spot like not trends maybe, but like commonalities. Like once I've worked with more than two people, I started realizing like, oh, a lot of people spend too much time on whatever, like let's say QA, like the quality assurance part. And that was obvious to me
in your own bubble. If you ask them, they're like, uh, it's it's not that crazy amount of time. It's not that bad. U, but when you start working with more and more people, you realize, okay, there's a better way. There's got to be a tool or we could streamline this. Or why is this person able to do their their testing in like onetenth the time compared to other people? What are you doing? Oh, they have like a makeshift tool that they're using. Um, even I started building a tool. So, um, every time we play the game or sorry, build the game, uh, it plays a basic version of the game to ensure like all the UI elements actually work on clicking, um, that you can actually interact with the tree and we didn't introduce any regressions or any bugs. So, um, I don't know. I I think this is stuff like once

you start working with more than one person, you're just realizing like tools are the way to go.
Yeah. And you often times think tools are not worth it for just just one game, but like even for for individual games, like whenever I made some tools for my games, I've never really regretted that. I was always like, "Yeah, that was a good decision." And I have often regretted not making tools earlier. Like best example for this is, for example, the level editor for Will Snail. I made that after I made the main game and I was like, "Oh, damn. This would have been so useful for making the game."
Absolutely.
Yeah.
The the skill tree editor at this point has already probably saved me about 70 hours. It took me maybe 25 30 hours to build and just in this one game,
it's already probably, if I guessed, 70 hours save time. Now,
I I think the issue is people will fall in love with their tool and they'll build the tool like a game, like, "Oh, I I'll add more features. I'll make it

more polished. Oh, it could be more streamlined." Like, my editor tool doesn't even have a UI to select like different painting modes. It's just keyboard keys that I've memorized. Why? Because I'm the only person who has to interact with it. It doesn't need to be refined like a game.
Uh even though I know it' probably better to have a UI on it with like, hey, this is a race mode. This is painting mode. This is move mode. Uh for the the tiles. No, no. It's like q wer t i fns. Like those are the keys. And what do they do? Doesn't matter. Only I need to know.
Yep. Yeah. The the level editor for my new game Kobaku is kind of the same. It's like the I the entire keyboard does like every button does something and I'm like I'll never have to document this. I'll I'll never build a visual editor for this. But like for myself, it's super nice cuz I can build levels super quickly just by by using all of the buttons on my keyboard basically. Yeah, I see that. Switching topics a bit,
what do you in general think about these

trend waves that we see in in indie games? Like the three main waves I can kind of observe at the moment are incremental games. Everybody's like, "Oh, they seem to be easy to make and they seem to be profitable. Let's make incremental games." The other one is friends slop. Horrible name for a genre, but it's like the apparently what we settled on like games that you play with your friends make like for funny moments and so on. Pe like peak. Exactly. And as you mentioned also these shop simulator games. I think that's probably another one of the the waves we're currently in. Like what how do you think about these trends? I I assume you like write them while they're there or like what's your strategy? Do you try to catch them early? Do you try to uh write them? What do you think about those? So, I talked about this in another podcast, but like we did pick Click and Conquer because I got in all the games uh from uh scraping Steam,

plotted it out and kind of realized, okay, once a game peaks uh total games released uh by peak is what I mean, you got about two years after that to maintain the success of having a thousand plus reviewed game. It's not impossible, of course not. there's still a thousand plus review games even 10 years after it's peaked on you know top number of games released for it. It just seems to become harder which you know my instinct is that makes sense like consumers have determined what has won. Um you know once you say like uh incremental game people list off key games they will recommend um so it just becomes a little bit harder to get in there. They already have a lot of choice selection. Those games are discounted probably more. So now it's even more attractive. Uh for me I we picked sorry me and sock house picked click and conquer because what we spotted was incrementals were seemed to be on an uprise when we looked at it. They didn't

they hadn't peaked yet and we thought okay we will start now and we got two years and we think we could do maybe two three games in a time and ride this wave.
Can you actually see these waves in the data? Like have you can you actually just I don't know plot out all of the incremental games and see like were curve going up or how did you do that?
Yeah, that's what we did. So, it's harder to see the like the up and coming, right? Like you'll have flat and like you can't really tell if that's just a one-year little bump or is a good dip.
But we would look for like okay one to two years kind of of like more than like 25 to 50% growth on the number of release. So let's say it was 10 all of a sudden and now it's 15 the next year all of a sudden it's 30. It's like okay we probably have started on some sort of like momentum. And when you think of it from development, it kind of makes sense. It's like you have a game that

comes out that's either novel, insane artwork, a new mechanic, and you know that that game comes out and then you have people are like, I want to build a game like that, but it's going to take some time. Um, so let's say I don't know 8 months to two years. Um, so then you see the second bump and then people start hearing success stories. Oh, they they made a game like this game and I want to be on that. Okay, eventually you get saturation and then you dip down. there's a time window like there's maybe one thought leader type game that's like sparks a new trend and then that starts starts this waves of not copycats but deaths who are inspired by this and were like oh this looks easy enough I could make something like this or or even justification that there's a player base like I had never looked at these uh fun social games until all of a sudden like I played repo with friends and then when peak came out it was like oh this is so obvious of course like this is just like repo but not scary and more like we have

an objective of like like a visual objective to do and it's more casual, laidback and funny and like it just
and repo wasn't even the the first kicking off one of that. I can't even remember what it was called but the
there there was Phasmophobia. There was also
we're both blanking here.
I know exactly which game you mean like probably just have to look in similar games for repo. Lethal company. You probably mean lethal.
Exactly. Yes. Lethal company.
So, so like as you can see there even a progression on those games, right? Like in terms of release and like building upon like in my mind I see those games as fairly similar even though most players would say they're very different games. Yes. But like the core playing together mechanic and working together and how to work together and having a team objective and these micro objectives like Yeah. So when I plot this out like I I just see them. You know, there are games that I think could do well in that that I haven't seen a lot of momentum. So, we haven't gone down that path. Like, I think trains,

horses, and things with dwarfs, when I look at the data, they haven't really had this peak. They haven't quite got enough momentum.
Uh, but recently, I've noticed a lot more train themed games either in the incremental space or even in building space. And I'm like, "Yeah, because no one's fulfilling that need, that experience." And you know, um, it's not that shocking that we'll probably see more once there's a successful thing, um, that people go, "Oh, trains sell. Oh, there's people on Steam that like trains and like themed things around trains and they go, "Okay, what can I do with trains?"
It's not just it's just not genres. It's just like all all kinds of themes. People are like seeing what D does well and then can can all kinds of mini waves essentially.
What I should have said when I plotted it out is I plot it out by tags.
Quick question, how do you get that data? Where do you get that data from?
So I used to have like a machine learning site that had an older version,

but now I just use I think it's called GG Insights. Yeah, I use GG Insight. So, they have a paid I think it's like $50 a year and every month they do the full dump in a JSON or CSV format. Um,
all Steam games.
Exactly. For all Steam games. And they're they're just doing the scraping probably behind the scene and I just don't want to maintain a a scraper and my time is just worth more than that. So, it's like here's $50 lifetime access. Okay. No, this is a no-brainer for me. Bro, the first purchase I'm going to make after ending this interview, like that's a actually that's a really good tip.
Yeah, I think that sounds totally worth it.
Yeah, like there there are other spots that you can get the data publicly for anyone who's listening and don't want to pay the 50 bucks,
but uh when you start manipulating and looking across all of Steam, it's like is it up to date? I don't want a thing that got released one year ago that someone used for machine learning

project and did do the scraping. I don't want a GitHub project that's like h it only works if you're doing it and running this and only in certain things. Sometimes it breaks and good luck. I just pay the 50 bucks and it's up to um that individual to maintain it. And so far they've done a great job. And actually I corresponded with them on email before with a bug and they fixed it. So, um
I I I don't want to say I'm recommending it, but I can say it works.
I mean, the the thing is, of course, people are noticing these kind of trends anyway, like with the data, without sort of, right? So, it's like But obviously, it's very nice to to have the data and maybe you can maybe that way you can spot the curves a bit earlier than everybody else.
Uh yeah, I think so. And I know you had uh Gavin from Choo Choo Charles on here and he made a comment about concepts and you know you don't want to be working on something when everyone's talking about it. And I think if it's obvious that it's a trend like I'm not saying it's

too late. I'm saying like you know maybe
maybe you want to find it before it's obvious to everyone like oh of course these incremental games are doing well. Yeah. because now you've seen 30 of them out and you're like, "Oh, another success story. Another success story."
Um,
yeah, sorry, I keep interrupting here with with little thoughts. And once the trends are really really obvious, uh, bigger players get into it as well, right? For example, once people realize that survivor likes start doing really well, then all of a sudden you have like a bigger player like Deep Rock Galactic also making a survivor like game and now like the market's already like very very saturated. Now you're if you like if you make a survivors like now you're going to have a very difficult time.
Not to say it's not possible but like from my perspective you know coming from like a corporate environment in enterprise software it's like why like it's it's not impossible but why take on the additional risk? Like you want to be early enough that you're not number one typically like you

can be but being number one doesn't not mean that you get the most success. We've seen this time over time. Think about products you buy. You know, Apple was not first in almost every category it's ever done. Uh they didn't build the first laptop. They just built a very well-defined and refined laptop. Uh the, you know, iPod, same thing. The iPhone, same thing. So, but you don't want to be so late that now you're just drowned out by competitors because now you have to break through the noise. It's that just sweet spot, you know?
Right. So, I mean, the later you are to the wave, the more creative you need to be to find your spot in that, right? It's not impossible. But if you're kind of late to the wave, then you probably need a more unique twist or you need to like out compete everybody in quality. It's just going to be a bit harder, but it's obviously it's always possible.
People people once again say in the comments uh like that that they don't like all of this trend chasing and you should just make what you want to make.
Yeah, you can. But I I I don't know if

it's laziness is the right word, but like I don't want to have to fight an uphill battle. I would rather be walking down the hill um as a metaphor. So I want to pick an easier path. Uh you can do that. Build what you want. Um if that's what's going to drive you to finish the game and that is the motivation, well, it's better to finish the game than never finish the game. So okay, go that path. Um, if that's not a factor of your motivation and you're just like, I don't care what I'm building. I'm like, well, why would you pick the thing that's probably going to be harder because now you have to come up with something unique because there's so many of them.
Maybe you just look in a different space or combination of games. Like I started looking at incremental skill tree kind of progression and going, where would that work in a game that's not seen it before? like, huh, what happens if I looked at like this slot machine gamblyes blocker games, but more skill tree based and more like you're going to eventually win? Um, and less about just

like was RNG gods on your side like can can I take this incremental like I will hold your hand and eventually you'll be guaranteed to win because you have unlocked everything which makes it like pretty much guaranteed you're going to win. I don't know if that will work, but I I haven't seen many people do it. So, I'm looking for blending across.
I mean, genre mixing is something that uh people often warn about, and that is a bit like that. I also think is a bit dangerous. Um, exactly. Because it uh can confuse both the algorithm as well as the audience a little bit. They're like, is this for me? Is it like only half for me? And if it's only half for me, then why don't I just buy something that's like entirely for me instead? I was going to say I think it's also who we're talking to. Like when it's me and you, it's like
okay, you have enough successful hits and I have a successful hit that I could go make a bet on something that maybe is not going to work out and it's not going to be fundamentally going to like ruin

my life or I'm not going to be able to make rent because of it. But like if you're just starting out and you're like betting the firm, betting the house, saying like this is the bet, like okay, maybe you don't want to go against conventional wisdom and be like I'm going to take X genre and Y genre. It's never been done before cuz it just increases the yes the reward but also the risk. And and for me when like Sockhouse and I started, it was always about this investment strategy mentality which is like we never want our loss function, how much we could lose to ever be zero.
Mhm. But that caps what your potential gains are going to be. So we don't want a one where our gains could be 100 times if it means we could go down to zero. We would much rather go for like the three times and have a 20% chance or a 20% loss, right? Like we we never truly go down to zero. Uh because it just gives us another attempt, another attempt, another attempt versus you do the big bet where you're like, "Yeah, but the payout could be another whatever Stardew Valley." Yeah, but your loss could also

be a complete zero and you could be spending two, three years of your life, burn all your savings and have a subjectively flop of a game or a game that does not allow you financially to sustain this and go, "Well, that was fun, but I guess I have to go get a job and this is not going to happen again." Uh, versus, you know, you've got a, you know, quite a few hits on your side that I'm sure you could take a bet uh for eight months on a game and it would not like end your career. Let's say it that way.
Sure. You made click and conquer in 400 hours is I think what you said earlier. Is that
400 hours per person or across both of you or how do you measure that?
I'd say 90% of it sock house cuz he did all the development even though I'm a software developer at that time. It was like he wanted to build this game. He had just come off another game. I was still working a corporate job. So I'm like and I've been doing software for like yeah plus 15 years. And I I was just tired. I just did not want to write code. I wanted to do something else. So

I was all like, "Oh, this business side of things, the strategy, Steam, let's go do that. Level design and pacing and bouncing, QA, different." Um, so 90% SOS, like 10% me. I'd say there's another, let's say, 50 100 hours that was like non game development. It was like me just tinkering around like what's trending, what's not trending, what's what could we get away with doing, does this idea look like it makes sense, does this idea look like it makes sense, how long would it take us to build that game? Like uh 400 hours was like once we had set on we are building this game. Um that included the the trailer, the Steam page, the game development, reaching out to the u content creators, like everything. Then Stephan or sock house, he had a full-time job. Like we we were both if you want to say part-time on this. Like no one was like
15 20 hours a week. Uh when we had time. We started January of 2025

and the game released June 2025. So you know what's that? Six five six months of calendar time. And then hourly uh because we tracked our hours because I had this theory and we want to prove if it was even viable. So we tracked out all our hours and uh yeah we came in around just shy of 400. So if we divide like 400 by 8 or something then we would arrive at something like if like if it were full work days then would have been something like 50 50 days along the lines. So would have been 2 months or something like that for a full-time developer.
Yeah. I and I think that's why the space is like trending or becoming hot. It's not necessarily because like ooh crazy amounts of money. It's more like okay it's not a huge amount of time to get feedback and you know if the worst thing is I spend two months and then have to go back and whether either get a job or freelance or whatever your scenario is.

It's not as bad as, okay, it's going to take one, two years of hunkering down 40 hours every week and I have to burn through my life savings to do it.
Um, you know, it depends on where you live, but like some people their severance will be almost two months if they lost their job and they're like, "Okay, I'm going to bet that and we we'll see how it works." So, I'm not recommending that. I'm just saying examples.
Mhm. Um, but do you think this is the case for all genres or is this specifically for this genre? Do do you think it's it's always an advantage to make games in a relatively short amount of time so that maybe once you cuz if you if you follow this this theory that you want to ride the wave,
then maybe you can ride the wave more easily if you make your game quickly or how do you think about that? I don't think there's a disadvantage of being able to do it quicker if you're not sacrificing the quality and polish and all that. I don't think it's about the wave as so much more like I think people

should try and get a complete cycle through. Like even if you've done game jams, you've done five game jams, you never truly have gone through the process from A to Z and you don't realize certain things. Like I I've had plenty of people finally take the leap, say, "Okay, I'm next game and doing full release and then realize like, wow, Steam polish or player polish portion of it is like 30 to 50% of the time." Um, at least from my perspective, that's typically where it lands and depends on what you define as polish, but like let's say the balancing, the pacing, all the animations, all the refinement of the UI polish. And you think, "Oh, it's not going to take me that long. this is going to take me two months. I've done jams before. And then all of a sudden, you're sitting there and you realize it takes four months because uh half the time you've never actually truly accounted for at the end. Uh things like Steve achievements, have you ever implemented that? No, because you were doing jam games. Okay.
Uh I think shorter allows people to

actually go through the full cycle and then you can come back for another one quicker with all the lessons too. Like, oh, I launched my game but didn't understand how important the store page is. Yeah. But if you spend two years building a game and the first time like you get to complete that cycle is two years and the next time is even one year. We're talking three years. Three years just to have gone and reiterated the first full loop. And for me the risk tolerance is just through the roof for that. Like I I just I can't justify it in my model like how I look at is it too riskier not to do this? Like the the 400 hours like you said two months let's round up three and a half months. Okay. and you factor out holidays and time off and brainstorming, you're probably going to be able to squeeze out almost just three games a year. Um, you know, now, can you do it in every genre? No, I do not think it's possible in every genre. Like, unless you really can find like a slim down vertical slice of each genre. Like, I don't know of a openw world MMO

RPG that you're going to be able to fully build in 400 hours. Like, uh, I've never seen it. If someone figures out how to do it, great. Maybe AI will get us there. Right.
Yeah.
But incrementals will eventually of course be saturated I assume for a while at least. Right. So at that point what should you do if you want to make a short project? Should you like find another trend to serve that sort of in a in a similar scale or or do you just go back to making whatever then or what do you do then? Personally, I think I'm looking still at the trends. And like I said, even though let's say incremental is on a down trend. I I don't think it is yet. I don't think it's peaked. It's I still see each month it's breaking alltime highs. But let's say it has. Okay. What can I do? Like I said, adjacent to this, like can I find something that's somewhat similar but different? Maybe there's a revamp on like I said I gave the example of this more like rogue like experience but I

want to bring more of the incremental I hold your hand and you're going to progress and go up and up and up
and I think something like that I could attain in this 400 500 hour range. Me and Sakas had actually done a game jam called Poker Tower and it was like a very very basic elatra if we call it. there wasn't all the crazy Joker combination rule sets, but we did that in about 100 110 hours. So, I thought, okay, that seems viable. We could probably do something in there and try and find combinations of what we like in building these incremental games and bring it over. Uh, like you said, there there's a risk of mixing and we'll see how the audience resonates.
But, um, the risk is also proportional to the time and how successful the game needs to be is also tied to the amount of time. So my biggest risk there is like we're talking three four months and you know if these other games do very well then it's not as big of a risk because I can fall back on how the other

games do you know uh everyone's risk is going to be different and their situation is different. You know you're you're brand new in this and you don't have another game to fall back on either the sales or uh you quit your job and you're doing this full-time then you're trying to look for the thing that's most likely to win. or at least that would be my advice is to minimize risk and look for the thing likely to win.
You can I guess you can always have a look at what's doing well and try to do some spin on that.
But are there specific games where you're like maybe it's not a good such a good idea to do that strategy or do you think this is just something that's like always a good starting point? Other successful games
trend falling I I I don't think there's like like I said a negative. People might not enjoy it because it doesn't align to what they want to build and fair enough. Uh then the advice there is like I said if that's going to be the motivating factor to finishing your game then so be it. Do what you want to do. Uh if it's a it's a puzzle platformer

type of game which you know wisdom says do not do u because they just don't resonate well with the audience. They're hard to make content on yada yada. There's so many of them. But if that's what's going to get you to complete it so be it. Um because completing is better than not building a game at all. uh if you're trying to make this like sustainable career and you're willing to forego like just passion only driven like I have to build this puzzle game because that's what drives me instead you say I'm open to building a game I want to build games and that's the driver and yeah I think looking at what's already doing well is a good spot also you can immediately see like in the reviews what people are saying they wish they had that's typically where I'm looking too like I I grab all the reviews of the game I think I want to do pass it through like sentiment analysis and chat GPT to get an idea of like what's everyone complaining about and what did people really like and
oh yeah that's also probably a good use case for for AI I just just feed in all of the reviews. Yeah, because like I said, I the market has spoken that

they're willing to put their dollars towards that and if I can find a twist or something that's, you know, filling a need that's not being met by that game. I I can give a an actual real life example.
Uh because I'm not doing this game with anyone and if someone wants to do it, they can do it. Uh there is a game called Moonlighter. It's this RPG style game. You go into dungeons, you collect loot, you sell loot in said shop, you upgrade said shop in town and they have a Moonlighter V2 coming out or it is out and I hate it. I hate it. It's It's 3D. I don't know why it's 3D. I read the reviews. People are like, "Why is the game in 3D? They changed the entire shop mechanic. People are like, "The old way was nice. Why didn't they just add more content to the game? Why do they introduce this new thing?" I'm like, "You have people saying that there's a want." You know what? What is great is people telling you their pain points who bought a game like the original, don't like the new one. There's this gap

probably somewhere here that I'm not saying it's guaranteed success. I'm just saying that sounds better than just I think I'm going to make a game. What What idea? I don't know. I'm going to come up with something.
Right. So find finding a community with existing pain points is also probably a very good starting point. You're right. That's sort of similar to what phenomena told me cuz she was like she she had this theory that you should go into a genre that has a lot of reviews but not so not so positive reviews. So you you want to go into a genre that has uh a lot of mixed reviews or a lot of very mostly positive reviews or something like that. You never want to go into a genre that has a lot of very positive reviews on their games. I think that logically makes sense. There's a lot of like opportunity.
I wonder if that's some analysis we could run on that on that Steam data to to find the genres that have the most reviews but the worst ratings.
Absolutely. Um like I'm thinking of the

columns there
and it's definitely doable and you know the only data that's I don't think really exposed well is like the reviews over time. you see how many reviews there are. Like you won't be able to see like, oh, the negative reviews used to be X, now they're Y. Each week it changed. Um, because maybe it got review bombed. Maybe it's after a content update, the game was actually all positive, but then people didn't like one component being changed. Um,
also reviews tend to trend downwards a bit after launch because, you know, the the less interested players come in, the people who buy it during a sale without reading the description come in and they're disappointed more often. What would you describe yourself as? Are are you a marketing person? Are you a games strategy person? What are you?
I don't know, Donis. I don't know. I'm in this weird spot right now. I think I come from software

background, went into game dev, but have always been drawn to the like operational business side. You know, I I ran a startup, so I've worn all hats. I guess I'm playing the own my own incremental game. I like numbers go burr. I I like minaxing.
That's what I like. I like I like being the person who can just minax regardless of what it is. If it's in the code, making the game like feel better. uh tweeting the animations and just tweaking the core mechanics, okay, so be it. If it's there diving into the, you know, whatever Steamworks portal and trying to be like, why aren't we using this functionality? Why aren't we doing this? Like, oh, this is not even really documented anywhere. Um, I don't know. I'm trying to figure that out is I guess what I'm saying is I'm enjoying my time and I have fun and
that's what matters at least.
Sure. cuz I mean the the feeling I sort of got from talking to you is that you're kind of a marketing person, but

what's what threw me off is that you uh still care a lot about the experience of how the game feels so much so to the point that you have this elaborate workflow for for testing games and automating feedback and so on. So, it's kind of but I I'm I think I'm actually kind of on on board with that because the way marketing works in my opinion
Yeah.
is mostly by making a great game. So, so to be like if you were like I'm a marketing person and then you still care about all of these gameplay related things. In my opinion, that still totally checks out. So, but I don't know. We don't need a description. We just as long as it's fun, it's fun. That's great to hear.
Yeah. And some of it's just a necessity, right? Like uh like the the the foot pedal, as much as you
are are kind of laughing at it. It It's very efficient. Like it makes my life so easy. Like I The problem is some games bind different functionality on the keyboard and mouse. So I can have the

foot pedal be like um a special function key
um that no one's going to be listening to. It's not going to accidentally close the game or open something. And uh then I can be playing the game with keyboard and mouse and I don't play any racing simulator so I never need my my feet.
Um so I'm just playing the game and I just hold down my foot and I just talk out loud and um it works for me. I mean I don't know.
You said no one else is really doing it. So I guess it just works for me.
I haven't heard that that one before. No, no, no. as like a professional minmaxer. What is your let's just call you that a pro professional minmaxer?
That will be the title right there. Professional minmaxer.
Um what are your thoughts on the future of game development in particular looking at what's coming with AI and so on. So some of this will be biased from like a more software development perspective than purely just game design

um or game development I should say. I don't think either industry are prepared whatsoever um to any level of the extent this will change. I can already even see it in my own workflows. um like the entire skill tree editor, I'd say 80% of it is me running multiple agents building that
and um in that case it didn't truly matter because I think where this where AI has a lot of struggle is like making this cohesive structurally sound thing that other people can work in very easily. It's kind of like once you go down this AI bucket, it's like it works for the AI and if the AI is dealing with it, no problem. But once you want to start involving yourself, it's kind of like there's friction. Now, obviously, someone's going to be in the comments, but you just don't have a good enough prompt. You're not running enough agents, you're not doing this new paradigm of like where you're acting as a peer reviewer, and the and then the

agents are proposing ideas. Okay. Yes. But, you know, that tool probably would have taken me 400 hours because it was dealing with things I had never done. It wasn't that I was incapable, but like uh I'm typically using GDAU and I don't spend a lot of time with like custom draw functions to draw like dynamic grids and things like this.
Um or panning and zooming a lot on but still structurally keeping everything exactly where it should be relative to where I was and like Yeah, I can open up the docs and read docs. Yeah, sure. But it would have taken me way more time versus like I know kind of what I want and I know enough software development that's like, okay, I have a good idea of like, hey, we're going to structure things this way. This is kind of what I'm looking for. What do you think? And it's like, oh yeah, GDAU can do blah blah blah. You just need to override the draw method and do these and don't worry about canvas sizing because I'm using this method. And it's like, okay, cool, great. Like, that's exactly what I want. And it worked especially well for that because I don't care how sloppy it is.

It's it's a tool for me. I'm not working with a team on it. Y
uh but it say at least 100 hours at least. The more these improve like the more impact they will have. Even recently with like chat GPT5 and deepseek I I've been able to get these to produce like even UIs for me which is where used to struggle. Anything visually it would just have no concept of being able to do at all. It like throw on some buttons and that's it. But now I can pretty much specify like I want um like a minimalist UI that looks stylus and you should use like color theory to match out things and all of a sudden I look and I'm like oo that's like 80%.
And the other 20% is me going in and tuning and the time difference is like astronomical. I think that's what people don't realize is like even if I have to spend I don't know we'll say 100 hours. So it produces 100 hours of work I'd have to normally do and I have to spend 15 20 hours refining it or tweaking it

and the prompts and okay I netted out 80 hours and and like to me that that's just maybe insane is not the right word but it is just light years ahead of what I thought it would be able to do.
Mhm. Um
yeah I think especially for for tools development it's so powerful right because this in the past this calculation of oh does it really makes sense to write a tool for this you know sometimes it's just faster to do the manual labor than to write the tool this was often times the problem in the past
but now I think this equation has shifted so much with um with AI because creating the tools is so so much faster and more efficient it's like almost always the correct decision to just create the tool now and to use the tool.
You talked about using it for the video. Well, sorry, you didn't talk about using AI, but um you talked about having a tool that did the video editing.
Well, that that tool was written by AI, of course. Yeah. Yeah, exactly. Like
I wasn't sure, but
Mhm. like written by II is not exactly

the correct way to put it because it's a very honestly it's a very very big code code base a very big Python code base and it's like catch could not have written that alone but also I could not have written that alone so it's like uh a human machine collaboration that would would have not been possible with just one of the parties at the moment I have a feeling we're obviously shifting more and more towards it's going to be able to do more and more alone for example I used to have this uh personal rule of thumb of like you can have the AI write 300 lines of code and it will do a very low uh mistake rate usually that was like my rule of thumb and then if it's more than 300 lines then I need to find a way to split it into multiple files or section things off and now with GPT5 I'm I'm like actually it's more like a thousand lines now it's like crazy for for programming you can still very much feel the improvements even So GPT5 had a bit of a rough launch of course, but for

coding nowadays it's absolutely insane sometimes what it spits out for me. It's like holy hell. Sometimes I'm very impressed. Yeah. Oh yeah. Uh the whole QA uh tooling is AI like with me just guiding being like okay I wrote it I wrote an OBS plugin that hijacks subtitles.
So when I press a foot pedal it actually injects a subtitle for the time constraint of me holding the foot pedal. Then I wrote a tool to in ffmpeg to interface with that to pull out the subtitles, clip out where they are. It allows me to add a bunch of params like pre and post time from where I held the footprint. Um, so that maybe like I released it too soon so I can add an additional second. And like this would have taken me I don't know at least a day or two to red docs, figure out where everything is, does it work? and instead it was like I don't know two three hours and I was like cool I got this whole tool and like now it does exactly what I needed to do.
What what does this mean if we

extrapolate this into the future a bit? You you said like the the industry is not ready like what what do you mean?
I don't know what will happen. I just see even the usage of everyday people is nowhere near what is it capable of? example I can give you. My my wife, she's learning GDO because she wants to build like this little cute cat game thing she has an idea for. And the way she works is she's very like once she has a checklist, she's like zero motivation required. She's like a once she has a structure there's she's just disciplined to it. Like getting a structure is very hard for her to like not get like analysis paralysis. I like you know what's great for that? I'm like, "Go get ChachiBT to just make you the checklist." And to her, like, it didn't even dawn on her. I'm like, "You have this assistant that will basically do whatever you want with very little friction for like 80% correctness." And like, I know it's not the best example,

but like people just don't even realize where it goes in their day-to-day and like how much it can be used for. product decisions I don't even do myself anymore. Um, I literally just give the URLs and like compare X and Y and Z, tell me which is the pros and cons and do price comparison and then I get like super short summary. Something that would take me normally an hour is like, you know, 5 minutes, not even. I I I uh interact with so many people that's hard to keep up to date. So, I actually wrote a Twitch monitor. So, now like I get a summary of every VOD from a Twitch stream, what was said. So, I'm up to date with each of these content creators and streamers like uh and it took me less than an hour uh of having chatbt like okay uh download the VOD oh there's open source GitHub project for that that it's aware of great uses that here's the command to run okay it slice up the video run open whisper translation or transcribe on it produce a summary like normally I would have just never been up

to date with those people and they probably would have died off on my follow list but now the fact that I can get a summary of what they're doing and kind of keep, you know, up to date. Even though that sounds very scary cuz like people should hear that and also have a moment of like surveillance and privacy concern. And I'm not one who's a big like privacy advocate, but I do think the path forward is like worrisome to some extent.
I mean, on a on a social level, there's like a it's one once again like a very different story, right? Because I mean I I'll have to make a video talking about this at some point as well because there's like the the example just the one example that I like to give for example is we'll have AI that will over the duration of this podcast just analyze the reflections of my eyeball and will create a perfect perfect 3D replica from of my entire room including very sharp resolution of everything that's in my room. We'll probably have that. Who knows maybe secret services have that kind of stuff already right?

It's like to totally possible and like all of these these weird things that you even didn't even think would be possible they they'll suddenly become possible and will be very very strange for privacy in particular. So I think yeah privacy I don't I don't even know if it's going to be possible honestly
the the barrier there used to be a barrier of effort. Yeah.
So even if the data was all collected, it was like even though you don't want all the data to be collected, it was like okay, who's actually going to parse through and have all the context and like okay, unless the government has a reason to suspect something, that's probably when. But now it's like the effort is zero. like it it's literally okay. You go grab an open whispered model and you can transcribe all data for all audio that's ever been recorded and saved and get a complete summary and easily search through the entire thing. Um that is where it changes is the level of effort is no longer there.

Um so
uh on the on the good side is it reduces the effort for us to build tools. bad side. You know, everything you've ever had recorded somewhere is now easily
digestible for anyone to ask anything about you, people. You know, that's the worst part is um even if you have the greatest privacy, you're as weak as your weakest link. Like let's say I am very secure and very minimalistic. I mean, I'm not on a podcast, let's be real. But let's say I I was um then my wife communicates things or images of me elsewhere. Um then you know I'm not saying don't do anything. I'm saying like you realize there's only a limit to what you can actually do because like if if I open up Slack and I write something about someone like oh
Jonas and Mia just had this podcast he said blah blah blah enter
even if you were very aware of your privacy like now I've contributed something.
Yeah sure. I don't want to be on the

tinfoil hat podcast.
No, I I just just very very small tangent because we're we're not really here to discuss the social issues even though they're very very interesting. Um I'm more more interested in extrapolating um what's going to what what this going to mean for gamedev, right? Cuz it's like what what is the next step after you know we we have AI that does really good coding and obviously it's already very useful for that. A lot of people could probably get a lot of benefit out of learning how that works correctly and how to use that in their workflow. But what what do you think? What what are the likely things that are going to happen from here? Not not that anybody can tell for sure. Let's just say one thing that's very likely is that these things are going to get better, even better at coding. Another thing that I can imagine that's going to happen at some point in the next couple of years that would be very very high impact is if AI um could become smart enough to operate a computer. So perform clicks just use all

of the software that we use basically. I think that's probably one of the next things that's going to happen. And then the next step up of that would probably even be that it could also for example play an entire video game. And I think that's if that happens, that's the the breaking point where where we know that we're probably going to be out of a job soon. As soon as I can play video games, it's probably not too far from from making full-on video games without any humans in the loop. And maybe I'm I'm the one with the tinfoil head now. But do you sort of agree with with with that or do you see something different? Like where where do you see what where do you think things going?
I see it like that. Uh I mean I'm actively actually getting you know the tools I'm writing to play a game.
Um so I I don't think it's I think each game is going to be different. The complexity is going to depend on the game but like you know okay these incremental games where it's just like mostly clicking. It's like
okay I can task the AI to use image recognition for like hey by the way you

want to click start button on each round. You want to click these asteroids. They look like this. Um, and then collect and spend your points on whatever skill tree looks this color, which mean indicates it's unlocked. Um, yeah, I mean, I'm not using it to necessarily just play a game for the sake of playing the game. I I want to build a QA tool um to make sure like when we produce something, the game still works.
I think it will increase, and history kind of indicates this, the inequality gap. I think the the people who have done well on Steam will do even more like greater income, greater results because they will have access to the best tools. They'll have access to um spending the money for GBT Pro, whatever it is, $200 a month, which will give them an advantage ahead of everyone in terms of time.
Um even if that becomes eventually like the free tier. Yeah. But if you have a a model that is like let's say two times better than the one everyone else is using on $20 a month or free um you're

able to produce even more content and or hopefully produce more content and of higher quality hopefully or should be and grab more of the market and make more money and just now you can afford to pay somebody whose only job is to minmax your AI models which should perform better. uh you just I think GameDevil will see a widening of the gap
and already it's pretty wide on like how where people land as indies versus like you know top tier games but I think it will get way worse.
Um
Mhm. I think for a lot of artists listening to this what we're describing here sounds like very nightmarish probably. So the the imagination that the gamedev job will essentially become about um buying AI compute and managing that AI compute more efficiently than your competitors. Right? That's sort of the the thing you're describing here is that

gamedev will develop might develop in this direction a bit.
Now on the optimistic front I would say this you make it more accessible to everyone. People who did not know development could potentially make a game now that actually in their head is is a good game and on like implementation is actually a good game, but they do not have the skill set or time or were willing to invest it to do it.
You could get games that are extremely more tailored to players. You could get more accessibility like everyone's games are now always working for color blindness because it like you don't do anything. You're just like, well, we just import that module that got produced by the AI and now it's accessible game. You could have AI allow games that maybe didn't have something like accessibility now become accessible because something's able to recognize and explain through what different things are on the screen for somebody who's impaired vision.
Like there's I go back history is a great example. We go back to the internet, right? There's good and there's bad. And what we hope is that the good nets out more than the bad.

And yes, the internet was super disruptive and it changed people's jobs. It changed the way we work. you know, we see depression on the rise and social media anxiety and things like that, but like you I don't think it's fair. I think it'd be hard stance for someone to tell me like the internet is purely this evil bad thing and it was not worth having because it's connected us more than ever. uh you know the technology improvements like you know just the transportation of communication even this wouldn't be possible like me and you normally very unlikely would ever have this conversation you're all the way in I think Germany I'm all the way in Canada I'd have to like get on a plane fly we'd have to figure a coordination of time and when's available and all this and it's not all gloom and doom I think it's just easier and at least for me it's easier to spot the gloom and doom I'm sure there are many people who see the optimistic light of like it's going to make this more accessible for many people that have

ideas that are great locked in their head
that but at the same time what you say is that it's likely also going to be a bit of a winner wins more mechanic right because if you have capital you can potentially afford more compute
that's what we see in the business market all the time right industry starts there's let's say 50 competitors in that space and eventually you get down to like this three uh leer the second tier and first tier winner.
So,
so do you think because gamedev is kind of such a newish discipline, we aren't quite there yet where they're just like two or three big companies or or is or are these arts disciplines like by nature maybe a bit more fragmented?
I think what I can say and this is controversial hot take. Sure,
there is a reason to why I entered gamedev when I entered gamedev. Uh I mean I guess I am the evil corporate capitalist who uh sees the divide and

I'm like I am going to be on the right side of the halves and I think now is a the time to not be you know left behind
and I I think sadly while we may not want to hear it I think a lot of the industry is going to shift from this purely passion creativity and more and more to this capitalist businessdriven
businesses are going to see more opportunity move in more and drive out a lot of solo indies is what I think because they will have the money.
What we see at the moment is sort of that bigger companies like AAA seems to be struggling a bit, right? Do you do you think they're sort of not really like they they they almost seem like they're not really sure what's going on? Is it just like this current generation of AAA studios are going to is going to get washed away and somebody's new news is going to raise rise to the top or are they just eventually going to be like here are all of our new AI games and everybody's going to be like yay because at the moment I can't really see that

happening because I I don't know they honestly the their way of making games it seems a bit too out of touch.
Yeah. Uh I so I worked in an industry for a company that was number one in the space. Uh, I'll let I'll let listen listers use their imagination. It sounds like Cornhub and it's in the not safer work space. Um, so I work for the parent company that runs all that sorts of stuff and they were number one in that industry by a large margin
and I can confidently say that businesses like these AAA studios, you get comfortable and lazy and you know when you have something to lose, you don't think typically of innovation. You think of defensive maneuvers and that's kills out a lot of a lot of innovation, a lot of change.
You have execs at companies that um got brought in because they're experience and they have done it multiple times with that experience and they will typically do the same playbook over and

over. The that works in a stable market and a predictable market. If you have a fragment or unpredictable or unstable change, massive innovation, uh, pricing disruption, those playbooks no longer work. And, you know, at that level, um, typically it's not your problem because you have a nice option, stock option plan, and you just need to make it to the exit. And why would you be the person to try and rock the boat and do something innovative or reckless or could be seen reckless and risk that? you're like look um you know we're going to do the same thing we've done before and uh we will produce still gains and yeah someone who doesn't have that experience and maybe even some naivveness will come in and be like no I'm going to build a company that's purely around AI or purely whatever or this disruptive thing and I'm not going to approach things like that I'm not going to have big budgets and I'm going to take you know imagine you took the

same budget as a AAA game but instead of doing it for that. You just said, "We're just going to go for the Steam market. We're gonna run tons and tons of these smaller teams, invest in AI, so they're even more productive, and we're just going to take that whole market."
Um,
you know, is it worth it? I don't know. Like, uh, we'd have to look at how much money they could make out of Steam versus how much on AAA on all platforms and all that. Um, could they have teams to work that efficient? Could they reduce their communication overhead? Because as you increase these teams, you're increasing the communication and management
which I think is going to be disrupted with AI because like I said when I used to work at a corporate environment and back in January I was starting to build tools that would summarize my meetings, summarize all my emails coming in. I was able to take like 10 times the amount of email influx as anyone because I wasn't reading every email. The the agents were summarizing 90% of the email for me. So like people are just like, "Man, this guy is a beast." Yes, I was a beast, but

I was also just smart and using tools to become like probably five to 10 times more efficient than any other worker. I don't know, Jonas. We'll see. I I I I think someone will come and disrupt that whole space. I don't think it would be the AAA's. The only downside that market shows me is they also have the beauty of not needing to. If someone else comes, the the the Jonas and Corey up and cominging people find a way. They show up and they go, "You know what? We have though capital
and here's a giant check. We buy you and now we get you and we didn't have to be first and we didn't have to do the innovation. We just have to have the checkbook to know when you become a risk." And most people will sell. It's very very rare that you have a founder who is either fortunate enough financially that they don't want to sell or is
naive yet or egotistical or just out there enough to just say
there's no amount of money. There's no amount of money. You couldn't buy it.

You you would have to kill me to take it.
Especially cuz if you if you don't sell often times the alternative is just okay, we'll just give the check to somebody else and take you down that way. Basically, it's more like um
do you want to be part of it or not is more like the question you have at this point, right?
Yep.
Yeah. Very interesting times ahead for game development and and for for everything really. Yeah.
Could I could I ask you what
what are you changing and what are you doing with all this stuff or are you doing it?
Um I don't know. I'm obviously I'm or not obviously but I'm I'm playing it uh a bit defensively because I have noticed that for sure the timing is not entirely right because of also people hate AI. They really really hate it. Especially artists and especially like people who who like art and often times also players who play indie games fall into

this category, right? Indie game enjoyers are often often people who who enjoy art and supporting the work of individuals and so on. So I think I'm extremely I I would be extremely careful about like using stuff like AI generated art, AI generated sounds. I would be extremely extremely careful about that. Um like I don't honest I have to be honest I personally I don't have anything against it in in principle but I'm um listening it to it to it because I I am willing to follow what the people want right and at the moment the message is very clear people don't want an AI generated uh art and sounds and stuff. So, as long as people don't want it, I'm not going to do it. Very simple. But for code, it's a it's different, right? For code, it's I don't know why it's different for code, but like for code, people are a lot more accepting and definitely going to use it for that if I can, right? And there are also, as you mentioned, a lot of workflow and tool

things that that become possible now. And I find them very interesting. I pay pay very close attention to the opportunities there. where I'm constantly thinking about tools I could build that support me. Also, I think weird things like writing your own game engine or your own framework for games, it becomes more and more attractive the more time passes because having a custom ecosystem, custom tools for your game can be very useful for a lot of different circumstances. It can be very tailored to how you like to work and so on. But at the moment, like writing on own engine for your game is it's not worth it. But I I can see it becoming more and more worth it as it gets easier and easier. It's like you just tell your group of AI agents, hey, I need I need an engine that supports this and that to make my game and then wait a week or something and they they write a custom engine for you. I I can see all of that. And uh as mentioned, I think the the main breaking point for when will AI be

able to make full video games. I think the the main bottleneck for that is AI being able to play test the games.
Uh-oh. So, I'm the I'm the evil one making that move forward. Uh-oh.
It'll happen for some games, it'll happen sooner than for others, right? But as soon as soon as AI can play test the games that it makes, the loop closes, right? It's like it can feed itself with the feedback and then it can play test and then it can improve and it can play test and then the loop is closed. But as long as it can't play test the games in a way that a human would play test, um, we're safe. And I think that'll luckily the good news is I think that'll be another while. I think it'll take quite a bit more time till we have AI that is universal universal enough to play test any game that you put in front of it. It can already give good feedback. Like people like I did a video about this in the past and people were like, "Yeah, but AI can't know what a human would do or what a human would

think." I'm like, "Yeah, of course it can." What it can't do is probably play the game. But it like if you give it a series of screenshots, it can perfectly tell what's going on. It can perfectly say, "Oh my god, this like the user was stuck here for a long time. You should probably do something about this." Like that part is already working, right? That's not the problem.
Problem is also
Yeah. Oh, I was going to say it's also an incredible sounding board. like I present information to it and have it like find counterpoints, counterargument, different perspectives, like things that I would normally have like confirmation bias to and and you know not be aware of
you know also expert I'd say like 80% of expertise in something uh enough for me like color theory like uh one of the games I was working on um has these asteroids that are are different colors and I never felt like the game popped out that much. And the person I was working with just went to chatd and was like, "Okay, these two colors I definitely want to keep. I like these this background and this foreground

color and these other colors I have use color theory and tell me what they probably should be." And it came back with, you know, much more brighter saturated colors and we put them in, we're like, "Yeah, that does look better." Like, holy, I mean, maybe it's placebo like because you think, "Oh, it's more uh it's better than your information. and it made a choice and you're now um have this placebo thinking it's just going to be better. But so far we showed other people and they say the same thing like oh the game experience just feels richer. It's like all we did was just change the colors and I guess color theory and
looking back at that like you can ask why and it's like oh the colors I chose are naturally brighter like fruits that the na like human eyes are drawn to and you're like oh once you hear it it makes sense like ah I would never thought of that myself even though it seems so duh. Um, but you know, I don't have a four-year bachelor's in I don't even know what would be related to that, but uh, color theory, whatever.
So, being able to just have a sounding

board and and you know, you're still the gatekeeper,
right? Still still checking and confirming, but, you know, not having to go in and play around with RBGs and be like, "Okay, I got one, I think." But, does that work with everything else?
Yeah. Controversial topic. I don't know. What do we say to to come come the people who are mad at us now? Like like what a soulless way to make games. What do you think? What is what for you is the the soul or the purpose in making art and games?
I'm just going to make them angrier. I am the soulless corporate capitalist that is like AI.
I see. But um I I mean to be fair it you're you're following your kind of fun, right? You are
you do what you find enjoyable and that that kind of minmaxing is what you what you find enjoyable. That that is the art for you. You could say in a way maybe.
Exactly. Professional minmaxer. I think you nailed it.

Okay. I see. Well, I mean what what I would add maybe is as you can see with these podcasts, there are a lot of different ways to make games and lot lots of different approaches. And what's always important is that you find your own way and what what works for you, right?
Yeah.
Would you plus one that
uh yeah, I will plus one that I will click on it. Uh you know, add more.
Awesome. Awesome. Yeah. Um thank you so much for for taking the time. I really appreciate it. got very got very interesting there in in the end as