A Little Reset - Channel Update

[@aarthificial]

ATTENTION: I've already shipped one game with Unity. I have an entire video about this: kzbin.info/www/bejne/sILJXnhnl7OirZY If you're new to this channel, please watch the video above before commenting, I beg you. As a side note: this has little to do with the recent licensing debacle (though it did make it a bit easier to announce this news) My dissatisfaction with Unity as a tool has been slowly growing throughout the development of Astortion. And working on Foundations only solidified these feelings. While I considered pushing through it, I think it would negatively affect the quality of the game and risk burning out in general. I hope you see where I'm coming from.

[@Trombi01]

Even if it does not have anything to do with the debacle, I am glad you moved away from Unity. I think brilliant devs doing so gives confidence for the rest of us that there are other legimate possibilities out there. The whole Unity thing was really horrid, a show of blatant corporate greed. If they would not see demonstrable ammount of people leaving, they might think they can try again. Even if not exactly the reason, you still contribute to the push back, which I appreciate. This new approach will surely have it's own struggles, so I just hope you will find enjoyment in the solving the problems coming your way. I at very least am intrested in seeing how this new approach will turn out.

[@MikeDesaro]

Love the decision. As a full time unity dev for over a decade and a bevy user it’s a fantastic choice. Bevy is coming up FAST and Unity is getting worse every year.

[@lajawi.]

As an upcoming game designer still studying, I despise Unity. I hate it, it's clunky, you can't easily get an overview, and the UI is... outright bad..

[@luk17032000]

Hyped to see Bevy in action!🎉

[@RogueShadowTCN]

Honestly, I read the comments first to see if Bevy was gonna be the thing. I've been trying bevy for just a few days, and it's pretty amazing. I've only done the tinkering for fun, but I've still been able to do things in bevy much faster than unity or godot...

[@boatbomber]

This happens to every gamedev KZbinr. (Dani, Randy, the list goes on.) Rather than just making fun, I wanna take a moment to think about WHY it always happens. Maybe it's because the incentives of the system are rewarding the development of the game, rather than the release. Releasing the game to a flop would mean the destruction of an entire income, since development of a new project may not be a hit like the current one. It'd be risking the videos and the game by releasing. Maybe it's a matter of pressure? All games are released imperfect, but when everyone is looking over your shoulder during the process the flaws are apparent even before release so there's a pressure and need to always keep grinding for the unobtainable perfect game, otherwise people may not like the KZbinr for publishing with the flaws they've been made aware of by the videos outlining them. All that is to say, who knows. KZbin is a complex environment and there's likely no singular factor that causes this phenomenon.

[@MellowKlug]

Fr it's weird that this has happened so many times. Maybe doing 2 hard and time consuming things at once makes them exhausted and they drop one. Dani dropped youtube and in this case it was the game(or the engine idk)

[@TS-by9pg]

When you look at other indie developers that successfully release their games - you just don't see how much effort actually went into it and how many failed projects they had. On KZbin you can actually see all the failed attempts. Maybe because of the pressure from the audience, KZbin developers are more scared to fail and start from scratch. Also, yeah, KZbin definitely pushes you to prioritize wrong things in development. Cool visuals, flashy new mechanics. You can't just slap some programmer art and work on the gameplay

[@endy2629]

Honestly, I think that not releasing games is not just a gamedev youtuber thing, but and indie gamedev thing in general. There are countless unfinished games that aspiring game devs have made which never got into a finished enough state for release. But no one ever sees these unfinished games. When a youtuber is documenting it you do. I think that the number of youtube gamedevs who actually end up releasing a game is probably pretty representative of how many indie games are released, out of the many more that begin development. Now I guess the question is why do so few indie developers end up actually releasing games. I think one of the biggest reasons is because of lack of experience, most of these people haven't done much gamedev before, and, as they improve, they keep wanting to change things and do things better. This often makes it awkward to work on a single project for a long period of time, often being thousands of hours to meet the original vision for the game.

[@boatbomber]

@@endy2629 That's a very good point!

[@sugrado]

The devlogs are the product, not the games. But aarthificial proved that he can code and finish projects (last video), and seems to be focused on the main game.

[@zio]

I've been following Your progress almost since the start and I have to say, that these gamedev videoblogs gave You one of the best possible tools any game developer can get. You've got pretty big audience, that is ready to test, complain and just simply - help You with any single process during the development. I would love to shout out here one channel, which is perfectly corelated with your issues - Wintergatan. Martin creates massive mechanical music instrument and he softlocked on creating components way too good looking instead of being practical in their form. He restarted his whole project and then started creating it again in "form from function" formula, which leads you to making best possible and most optimal solution, instead of focusing on looks only. Your plan with making "base" of a game is the best You can start. Just create something that works, plays nicely, then refine the graphics the way You've learned to during previous processes. And USE Your audience as community, that can help You deliver best gameplay experiences player can get. Good luck, mate. 💪

[@SikerGaming]

I had the feeling in the back of my mind but couldn't tell what it was, thanks for putting it into words. Yeah, this feels exactly like what Wintergatan went through going from MMX to MM3. The only difference is that there won't be a full year of silence like when Martin dropped off the face of the earth after quitting the Marble Machine X. Luckily it doesn't seem like they're going to constantly threaten to quit the project if it's not perfect like Martin has been doing with the MM3. Martin's perfectionism has been an issue. I'm glad Aarthificial has an understanding that the game will "never be perfect", because that means it'll _actually come out_ when it's good enough.

[@StriderGW2]

Wintergatan also came to mind for me, while watching this video

[@Duckamoly]

Learning to actually let yourself stop is such a challenging thing. I wish you luck in anything you do in the future and I'll be here whenever you do decide to share something. Have a great day/night, wherever you are. EDIT: One thing that really applies here is the "Art is never finished, only cancelled" quote. If we keep working on something forever, we'll burn out and never get to put our passion to true use. Also I think I'll take some inspiration from the top down approach and try that more in my current project that I've just started after actually breaking out of my rut and releasing something. Again, I say, best of luck, and take care.

[@t3ssel8r]

wow, thanks for the brutally honest reflection. I definitely have learned some similar lessons over the past few years as well.

[@RegahP]

Failing is OPTIMAL, my friend. Always know that, failing isn't a bad thing that can happen and that you can come back from, it is OPTIMAL. You HAVE to fail to be better. Your project is visionary, personal, ambitious, and passionate. I know your feelings towards it. Your change of approach is smart, and caring, and you'll fear it less as time goes on. As creative creators, you and me struggle with similar things. All the power to you man!

[@aarthificial]

Thanks man! Really appreciate that!

[@DarkDax]

“No more of this perfectionist mindset” You and me both mate. It’s such a motivation and progression killer. I find myself slipping back into it but I’m going to trying to actively push out of it moving forward. Hope your journey with that has been positive since this video!

[@qwfp]

I was so ready for you to say "that's why I'm switching to Godot". Bevy is a surprise to be sure, but a welcome one! Good luck with Rust. It can be hard at first and overall it's pretty stiff(?), so it's not great for prototyping. Been here from the first video, waiting for the next one. Cheers!

[@aarthificial]

Thanks so much!

[@raremc1620]

I was honestly expecting something like that about godot :D I love rust heavily, though personally for simple and fun games that I want to make, Godot is easily the better choice, since it's so easy to handle. Though it's rust bindings are not finished, and so, I need to use a different language myself. And while godot is easy for something you want to make with simplicity, bevy definitely looked nice for more.. Complex and hand tailored projects

[@catgirlQueer]

as someone who's currently experimenting with some rather custom rendering in Godot (less custom than Astortion, and more suitable for how Godot's rendering pipeline already works, but rather custom nonetheless), if I was working on rendering things in ways as specific as Astortion I probably would just write it myself in something more modular like Bevy

[@raremc1620]

@@catgirlQueer Exactly! To be fair, for a new project of mine, I'm thinking if I want to use a custom render pipeline with godot, or use Bevy to begin with..

[@idedary]

I was in the same spot as you, stuck using my old game engine even thought it kind of sucked and I didn't enjoy it. In the end, I always defaulted to making an abstraction layer, a custom solution to hide the engines' shortcoming. But then I decided to try to learn something new. I got to know Bevy and started porting my project over. Best decision I could have made. Not only did getting familiar with Rust spike my coding expertiese, it also made me feel like my time spent programming libs for Bevy was worthwhile. Rust is an amazing ecosystem and writing an abstractions feels so good. Like building an automated farm in minecraft or playing factorio. This shit just feels right. From the videos, I can tell you are the same kind of person as me. You wont regret learning Bevy. It's the ultimate toolkit and it's source feels like your games code.

[@TheExtremeCube]

I think the comments saying you should just buckle down and finish the game are wrong. If you always fight with your own tool that impedes finishing the game. Finding a workflow that works for you is important. Personally I am excited about this news, both for you as you seem to be eager to use the new tools, and as a viewer, it is exciting to see a gamedev youtuber try a less common "tech stack". Also I think you will probably become a much better programmer using something other than Unity. Good luck!

[@TheExtremeCube]

@@speedboiiii2722 I can see what you mean, but by virtue of the fact that Aarthificial has been working on his game with Unity for a number a years already, I wouldn't say he is scared to start. Yeah, if you are overthinking which tools to use to find the perfect one without actually making anything, just pick something and get started. But it's perfectly fine to realise that the tool doesn't actually serve you as much as you'd like.

[@Danilocked]

You got a problem if you really think that using, or not using, unity is what defines a "better" programmer.

[@TheExtremeCube]

@@Danilocked nah man I left Unity and I no longer have a problem

[@Danilocked]

@@TheExtremeCube Good for u

[@tomoeatsock8794]

@@Danilockedthat’s not at all what he’s saying. He’s saying that if one is constantly struggling against an engine, or that it’s slow, clunky etc, they shouldn’t use it. No ones saying that an engine defines a developer. It just seems like you’re insensitive

[@i_love_python5862]

I hope you don't get exhausted while developing Astortion. I want to see you being passionate about developing Astortion, and if the developing process is painful for you, it would be better to change to a better workflow. I really enjoyed watching your coding adventures, and would wait for more!

[@bruderdasisteinschwerermangel]

bevy seems like a solid choice for someone who wants to do the kinds of things you do. Excited for what you'll show us next

[@voxelltech]

Nice to see more Bevy adoptions! Also, using Vello for your UI rendering is certainly a great choice! I made a very minimal renderer for Vello for Bevy and used it to create simple motion graphics. It’s really nice. The only downside is that it’s still very experimental and lots of unsupported features like only round strokes and no HDR 🥲

[@hallwaerd]

I’m sure this was a torturous decision to make, but I’m excited for you and your projects. Starting on the right foundation is key to success, and I’m confident you’ll be able to make Astortion better than it ever could have been in Unity.

[@Waradu]

Always a pleasure to watch your videos. I'm excited for the motion canvas update (:

[@plusplusplus9837]

Good luck with all of this, I've loved your channel since day one. Although I'm just a random on the internet and I don't want to make too many baseless conjectures about what will work for you/what won't, I think you correctly hit on the two most important things being playtesting and actually creating the playable content, so you can always have - even if very very rough - a playable version of the game. The work I've done in manufacturing (and what you'll hear if you listen to any talk from a silicon valley tech bro) has shown me that by far the most important thing is real world validation of your product, and having a really tight iteration loop. If not polishing the systems 100% lets you get to the playtest phase quicker it's worth it. And when you get feedback from patrons, don't just get it in writing, or even worse via a survey (the worst feedback gathering mechanism of all). Have your patrons record their gameplay sessions while they talk out loud, or even better observe them live in a discord stream while they play. Then use your new knowledge to modify the system and retest it the next day, even if requires throwing together a shoddy patchwork change for temporary purposes (luckily Rusts's great type system can often ease refactoring). Just beware that people are bad at verbalizing their problems and even worse at coming up with solutions to their own problems. Furthermore, even if they can do those things it's really really really easy to subconsciously guide how someone will answer by asking a leading question. Questions like "was this level fun" or "what did you dislike about this level" or "how would you improve this level" are often useless dead ends, even though they sound like they will yield helpful feedback. I've often found that it works better to qualitatively dig at the root cause of the problem by observing and asking behavioral questions. For example, "what are you thinking now" when they get stuck, then "why are you thinking that". Even if they're not stuck, just asking "do you have any plans for what you're going to do next at this moment" can show you what they're thinking and if they're having fun, without having to ask "are you having fun" directly and getting an answer that's bad because being able to properly verbalize thoughts and feelings is so difficult. Getting insights laterally will let you dig at the problem without getting misled by what your user (incorrectly) thinks the problem is, then from there you can solve it in a better way. ANyways, I love the internet because it's such a great medium for me to give unsolicitied advice on! You probably nkew all this. But maybe some of it helped. Again, I really like your channel and I'm really looking forwards to seeing what you make, and I really appreciate that you're willing to share it with us even though you don't owe us anything. Also I've been having quite a bit of fun programming in Rust recently (after years of C98) so It's cool to see you make the switch as well.

[@kalvinpearce]

As some who, at work, recently had to completely start from scratch in a different engine after 3 years of development, I just want to say you may be surprised with how rewarding and great this experience will be. When the tooling/workflow you are using feels good and clicks, you can get so productive. The progress my team made in 3 months is already more focused and clear than the past 3 years. I very much hope you will receive the same breath of fresh air from this switch and it will reignite that flame. Also keen to see the rust content ❤️

[@Vorono4ka]

OH YEAH, RIIR! Man, your game visual and mechanics are so cool, I hope you won't give up and will done the game. Good luck!

[@HumanityAsCode]

Bevy is a great choice. I've been having fun using it. You're more than well equipped to write the parts of a game engine that are missing from Bevy

[@planckstudios]

This is inspiring! Schedules change but the clarity of vision you've earned will stay forever. Excited to see the progress

[@imvirpil]

In my experience, starting over isn't usually as intimidating as I thought it was. You get into the groove of it just so much quicker. Yes, even with a new approach like using Bevy. Also, in my opinion, this isn't really 'starting over', it's more of a 'continuing my game development on Astortion', because I see it as a necessary failure. If not for those experiences and mistakes, you would never know how you could actually make this game better (and what you actually want to do), and you've got a clearer vision for the game than ever before, so go for it, and good luck!

[@SnowPeaGames]

Different because I dropped Unity for Godot, but I agree fully. So many things I struggled with in Unity - Input handling, dynamic window resizing - worked almost off the bat. I never realised until I made the switch that I was in an entirely optional fight against the *tools* , not just fighting a necessary part of game development.

[@DPedroBoh]

I dont see it as a reset, this is just another part of the process, im glad you learnt so much about the reality of gamedev even though you already knew so much about the technical part. Your new aproach is much better and more realistic

[@pudimlaranja4118]

you're one of my main inspirations for game dev and i have to say that see you talking abt game mechanics play testing and stuff like that really conforts me because I've struggling with this for years i hope u get good progress with you new graphics library, and at least for me I'm not here to play the game of course i will play it when it come out, but i love your videos man keep it up

[@tiileaf]

I wish you nothing but the best of luck. I've enjoyed all of my time with Bevy. You've shown yourself to be an incredibly talented dev, so I can't wait to see what you can do with it. Listen to your guts, not random comments like me. Keep up the incredible work!

[@j.j.maverick9252]

I’ve never restarted an entire game, but many times I’ve decided to rewrite a big chunk of a project, and Every time it goes super quickly! The feeling of relief when you can bin a messy bit a replace it with a solid clean solution is amazing. Wish you all the best with this bold but important change!

[@clayofman-levisiewert8381]

I know its not a candle against the depth of work that you do, but I've found that restarting or heavy refactoring for me has always allowed me to remove the limitations of different systems and relieve a lot of technical and, perhaps more importantly, conceptual debt. I'm really excited to see what you do in bevy and how you explain its workflow!

[@johnlewis794]

Overjoyed to see you switching to my favorite engine! Hope you have fun with it!

[@henryisaway]

I'm looking forward to February! Seeing you develop Astortion gave me inspiration to continue developing my own game, and it still inspires me to this day. I hope Bevy will serve you well - we'll see. Enjoy your break! Happy holidays and looking forward to your progress!

[@Efra_EMX]

Massive good luck for you! You've been such an inspiration with all the cool systems you've made and how well you presented it in a video :D

[@josipkustan]

Acknowledgment of a bad process is the hardest part. Its a hard pill to swallow, but a neccesery one! So I am glad to see you moving forward in a healthy direction! Can't wait to see more!

[@pokefreak2112]

Bevy is definitely not what I would've personally chosen, but I wish you the best and I'm looking forward to seeing how it serves you next year

[@igrb]

SO happy to see this update, really hope everything goes well for you and astortion. Been watching the journey for 2 years now and am so excited to what we'll see with your newfound focus + tools

[@supercyclone8342]

I completely understand your decision. Restarting my metroidvania years into development was a great choice. While I'm still using Unity (I'm considering moving, but restarting again is a tough ask), the restart helped me re-access the fun of my game as well as rewrite a bunch of spaghetti code lol. Another thing that helped a TON was working on a design document whenever I didn't feel like making the game. I didn't make it too formal or anything. Just bulleted notes about player progression, what abilities go where, a rough world map, story, and lore. Now I have a clear vision of what I'm working towards at all times! My main focus was getting a demo of the beginning of the game ready. That way I wouldn't feel tempted to endlessly add new systems. However, now that the demo is public, I'm realizing the importance of making dev tools (in part thanks to you and GMTK). But now I can make tools with that clear vision in mind.

[@Tantandev]

BEVY!? I did not expect that, welcome to the rust cult :)

[@aarthificial]

Happy to see you here! Your recent video about Bevy where you go over Cart's philosophy helped me a lot in making this decision 🦀

[@axxso719]

Can't wait to see what u'll do with bevy and this new tech stack ! Good luck

[@gandev5285]

Bevy is going to have a great future. It's my favorite engine to mess around with. I had no idea coming into this video what engine you would choose and was surprised when you said Bevy. It's nice seeing more adoption for Bevy.

[@Dimon6731]

Good thing I watched this, I know now what I’m signing up for lol HARD FREAKING WORK. But honestly thank you for your transparency. You keep moving forward despite the frustrations and difficulties.

[@SpellMenderDev]

As someone who has become increasingly disenchanted by Unity I'm GLAD you're making this change. I'm also curious to watch your journey with Bevvy! I've been afraid to pick it up.. it's always easier to jump into something once I've seen it work for others.

[@antoniocapo4883]

Bevy feels like a great middle ground between GUI-driven engines, and a "pure-code" approach. It's something I'm really excited about, but finding to be pretty inaccessible. I learned a lot about system design from your Unity devlogs, and I hope you continue that level of detail in the new Bevy videos; it'll be a really valuable resource for people looking to start working with that framework.

[@RenderingUser]

Bevy will get an editor eventually.

[@bilalinci4981]

The thing I'm most upset about is that even though you've learned the unity editor library very well, it won't be very useful anymore (the experience you've gained will undoubtedly work.). I have been developing tools with the unity editor library for a long time. It's hard for me to leave unity. I can't even think about you! I hope all this will get better.

[@bilalinci4981]

(unity editor is nightmare)

[@mikkelens]

learning a bad tool can be good experience if you are able to recognize better tools

[@askeladden450]

Honestly, you are just boxing yourself in. What are you gonna do if unity goes out of business? The lifetime of software is limited, and you will be in no mans land when that happens to unity. Gamedevs should focus on general skills and patterns, not hyperfixate on engine specific details. I created a lot of editors in unity, but I focused more on the ux patterns rather than apis. I now use godot and all that knowledge is invaluable and i got more productive within a week in godot than i had ever beeb in unity.

[@toastyseed8311]

I've been following your devlog and also have been looking at Bevy myself! I'm very excited for your future videos!!

[@agmass]

I tried to code a multiplayer system for unity which i had already done with some of my own games although they were made in haxe. It took 4 hours to set up the multiplayer system because things kept breaking and the engine kept crashing making me lose tons of work over and over. For reference, it took me ~1 hour to get it set up in the haxe framework i was using; for the first time ever. That always left a salty taste in my mouth with unity; but i still want to use it since it's really the only viable option i have for 3D games; since my framework only supports 2D renderring

[@emmiasilk9059]

The decision to move away from the big popular engine can be a difficult decision, but it's one I very much respect. In doing so, you face the Sunk Cost Fallacy head on, and say "I will not be a servant to fear". I greatly look forward to seeing you adapt to this new environment. You've got this.

[@damelux3798]

Is that a quote ? It reminds me a bit too much of "I will not fear, fear is the mind killer ..." but it may be my brainworms x.x Also : 🏳‍⚧❤

[@emmiasilk9059]

@@damelux3798 It could be, but I think it's just from my head.

[@notanimposter]

This is great news, because it means we get more devlog videos! A framework definitely seems like the right choice for your development style. I've always had the same problem working within an engine, where I just keep reimplementing everything until I'm not using any of the engine features.

[@thatepicbanana7499]

When I saw that you were using Bevy I audibly yelled NO WAY in the best way possible. I've been following you from the start, you're the only gamedev youtuber that I actually drop everything to watch. I picked up Rust around a year ago and it's quickly became my favorite programming language. I'd love to see how you use it, especially since I haven't seen many big games created with it (probably just due to its youth). Good luck on your journey!

[@noogiesm]

commenting to say i really appreciate the captions! scrapping everything youve made so far is a hard decision to make but sometimes a fresh start is needed. (ive done the same with my own projects. usually its the right choice) good luck!

[@dktheking996]

Letsgooooooo ive been thinking bevy is perfect for your style of development, i hope you enjoy using it!!

[@c0der23]

I just want to wish you luck with this new approach, I love all your videos and how they go into detail without making it too complex!

[@archieb303]

I've seen the pixel art shader video from this channel before, around when it released, and also last night. Loved the visuals so you've earned a sub. It's heartening for me to hear that youve shifted your process over to one I'm using some variant of for my game dev project, and that I'm not the only one who's bounced off of unity. I'm working in gamemaker 2023, something with a lot more structure than what I understand your plan to be, but i've had to build my own isometric renderer already to achieve what i want from my game. It's going to be hard, but seeing someone go through the same makes it less so. Thank you for your update and all your hard work

[@aarthificial]

Thank you too and good luck with your game!

[@IcyFlame100X10]

It sucks that you're having to restart, but sometimes that's how it goes. I'm still eager to see how things go, so I'm here for it either way. Still haven't given gamedev a proper shot myself save for a few small prototypes, but oneday I'll probably give it a proper go and when that happens I hope I can recall all the lessons learned by people like you. Otherwise, I might have to delete large swaths of progress too.

[@john9francis]

I’m super excited about bevy myself, so I got super excited to hear that you’re gonna switch to it! Best of luck to you! Can’t wait to see more videos.

[@gesprengt6561]

When you said you needed an engine which lets you do the rendering yourself, I instantly thought of bevy and I was so excited when I saw I was right. Amazing language, amazing game engine.

[@TheStone190]

To me, and I'd guess a lot of people, it doesn't matter that it takes time and errors are made. We're here to belong on the journey and learn as much form success as from mistakes

[@needleful]

I've had many projects meet a similar fate, where I had more fun designing systems and tools than the actual content, and I wound up with prototypes that had a bunch of interesting ideas but no game. I hope Astortion comes together in the end! I've liked seeing it.

[@b2rad]

Best of luck with the changes. Will definitely be following along the progress. I'm excited to see where you go with the new path!

[@p13_l0v3r]

Hey man, good on ya for putting mental health first. Bevy is nice. Used it for a few games myself. Wish you the best of luck! Been a fan of this channel for a long time. Your videos provided inspiration when I was feeling burnt out in my own game dev journey. Glad you're back and hope you get where you wanna go

[@Skeffles]

Good luck switching to Bevy. I've heard great things about it and I hope it suits your game well!

[@alevanderBatman]

I will personally be just as excited as I've always been for your great videos!

[@eglaiosdeminecraft9259]

2:18 "If something's not fun or challenging to play, it's not gonna be in the game" You most likely already know, but things that are not fun nor challenging are sometimes the cherry on top of some games for their potential to change pace. For example, the stairs scene in FF7 where the team keeps climbing the stairs of Shinra tower and Barett complaining was surprisingly fun to me. I can't tell exactly, but it felt like it acted as a way to build tension up as well as increase the anticipation of the player. Since it's very "braindead" and repetitive, I also think this helped cool my mind down to better enjoy what's coming up. Plus Barett complaining himself confirms to the player that this part was meant to be annoyingly long, and it might also have increased my attachment to Cloud, like devs are not only giving me the epic story and combats, but also the boring and insignificant tasks like those, because that's also part of what the party has to go through. Another situation, in DMC3 where it's all about crazy sick fighting and combos, there're the soul eaters that only reveal themselves when you don't look at them, and the fallen that keep flying around, are a pain to reach and take ages to kill. Despite provoking a "bro, not these again" every time they show up, they still force the player to switch to different strategies so they don't just keep rushing and juggling with all enemies, which offers a noticeable change of pace. On the opposite, I started Far Cry 2, and despite being filled with "fun and challenging" stuff, I'm becoming bored rather quickly. Sure, the map feels big and every enemy camp is carefully crafted with machineguns, garages, explosive barrels, snipers, etc. but... that's the thing. Every time I meet enemy structures, I know the devs built them to be as "fun" as possible. I know there'll be plenty of convenient stuff to exploit, and I know there's ammo and heal on every dangerous spot so I always get rewarded for beating it. As a result, I expected the game to be fun, it is indeed, but then I'm never feared, anxious or wondering because the devs worked very hard to make sure to eradicate deception. Everywhere you spend time to go, there's a reward. If you see something a bit unusual, 95% there's a diamond briefcase in there. Even the 2 times I died, my mate straight up reanimated me to make sure I never face the slightest deception. At the end, my biggest deception of Far Cry 2 is that there's never deception. Other than gunfights, there's no risk to get punished for anything, so every rewards feels like "I knew I woulda got something regardless" rather than "Yeah! I risked to waste time / ressources for that!". Well uh... had nothing else to do tonight...

[@aarthificial]

Absolutely agree but also notice that it all depends on the definition. You yourself labeled the FF7 example as "fun" And DMC3 forcing the player to change strategies is exactly what I'd label as "challenging" Though at the surface they may sound repetitive or annoying, when put in context they end up being exactly what we as players love. And these are the sorts of things I'll be absolutely looking for.

[@Agent40infinity]

You were a massive inspiration for me when designing and developing complex systems, it's a shame to see you move away from Unity. I hope you don't end up spiraling like every other game development content creator.

[@wChris_]

Nintendo also develops games without any visuals at first. Aka the gameplay comes first and if something isnt fun, it gets scrapped.

[@AberrantDroid-je6om]

I've followed Astortion since the beginning, and I will still continue to follow and support it. I myself don't understand much of gamedev as I do not have a tech background. But I wish you the best of luck in your endeavours and hope that you take the best decisions

[@auseawesome9112]

I was thinking 'surely something like godot' but wow, you went with something fully custom like Bevy. Hopefully the workflow suits you and works for the game you want to make. Thanks for making such amazing content, tools, and explaning concepts in such an interesting and intuitive way. I look forward to see what you do in the future

[@Nekkka]

In any case, I will always expect cool projects and videos from you!

[@Birb64]

Thank you so much, this has helped me understand how I should treat my game in development.

[@mikkelens]

When you said you are planning to use bevy I literally gasped! I have been using unity as an amateur for quite a few years now and have felt a lot of the struggles you have, although in a lot less demanding sense (I’ve made a lot of game jam/school projects, but not a suuper long running one), and I have also been looking into rust/bevy for a more elegant and conceptually minimimal development experience. You have taught me how to clean up a lot of my unity systems and I am extremely excited of the possibility of this happening again with bevy in rust!

[@petarpetrov6385]

Tbh I've been watching your videos specifically, because of the unique systems you are able to create and teach about! So the idea of you having more freedom to make said systems without the limitation of a pre-made engine is exciting!

[@ballchen4768]

This is soooo relatable to me right now :D I made the same decisions you are describing about 3 weeks ago and I feel like now I actually make progress. Perfectionism is hard to deal with sometimes.

[@Szystedt]

Personally I’m not one to care about the time to release, and in general I feel like this approach will make for a better game! Excited to see where the project goes, I wish you the best of luck! :)

[@FugaceFugite]

Personally I think for 2D games going for the custom engine approach is a fairly solid choice. They often require a lot of special, unique handling of things, and game engines are typically not made for that. Had you not been versed in engine dev at all, it'd be a harder choice to take, but it's still doable to learn a framework like Bevy or SFML in a couple months or even less. I think in your case it's the right call. If you still can't get used to Unity after years of using it, then it simply isn't the tool for you, at least when it comes to Astortion. Even then, going from a lot of prototyping and exploration of your game's concept in a game engine to your own custom solution is not that rare, even for 3D games. By this point you probably have a good understanding of what the different systems in your game will be, what structure you want them to have to continue working on the game, what pipeline the graphics and assets should use etc. Having treaded this path before, my advice is to not skip steps while implementing your custom engine. A well thought out and designed engine while unlock more doors that you'll ever expect, while a rushed or unsuitable design will only impede on your ability to do what you want, much like Unity is doing to you from what I understood. (Side note that I personally really like Unity, but felt the same constant fighting the engine when working in Unreal Engine. A given tool won't fit everyone, and it shouldn't have to)

[@bo00t]

I am very excited to see what you make with Bevy! I made the same switch about a year ago due to my love of Rust, but I still haven't really seen any big gamedev youtubers using Bevy.

[@dgoberna]

Exciting!! Best of luck, and enjoy the trip!

[@alex-howard-media]

Hey! I recently used Motion Canvas to develop music notation and animation software, if that’s something you’d be interested in adding to the software as a whole. It’s super niche, but it let me explain musical concepts and animate notes onto a staff, keyboard, and other functions like that. It was a really neat experience. I’m excited to see the updates you’ll make!

[@uniworkhorse]

Game development is so curious to me. I've never worked on a project that lasted so long it seems so daunting!

[@FelipeAndrade10]

Nice, I am looking into Bevy for a long time and I am exited to follow your project

[@kkibi]

yay you're using bevy!!!

[@donatoclemente4421]

Cant wait to see bevy stuff! Never used it so interested in hiw you use the workflow.

[@bubu88134]

Remember to not feel like you wasted the time you spent, since not only you learned from it (which is arguably the more important part) you created youtube videos and grew your channel with the examples and gamedev videos you made. Keep it up

[@mfalk6263]

Not to sound like an ass, but I kinda expected something like this to happen eventually. Making games is really fucking hard. On the other hand you seem to have the tenacity to eventually make it happen. Also having a big YT channel helps I imagine.

[@gabrielmfern]

everybody going with bevy

[@tiroforeal]

Mad respect to you for taking a step back and readjusting yourself on your dev journey. It would be so easy to continue being stubborn, forcing yourself to drag the game dev process through a game engine that wasn't beneficial for you just because you had already put in so much time into it. In the end of the day you shouldn't think about it as progress lost since you learned so much up until this point and have a clearer vision of your goals now. I wish you the best going forward and hope the new shift in mindset allows you to create something that you will be the most proud of. Love your work, keep going man

[@fruitspunch]

Actually, I have to disagree here. To you it might seem, that you have nothing to show, but during your time with Astortion you not only learned valuable new skills, but you also created videos for other people to learn what you discovered. This made me think about aspects and problems in gamedev that I never considered. Not only that, but you also created tools in the process that other can use for their ambitious projects, like the animator and motion canvas. To say that you have nothing to show is simply false. You have created, you have educated and you have sparked curiosity and interest. That's a lot to for anyone to accomplish, let alone somebody that has "nothing to for it" Your work is great, the presentation is always on point and your voice and narrative style makes it a great and soothing experience. Well done and best of luck to your future goals and ambitions!

[@garyantonyo]

I am here for the videos cuz you do cool stuff and make interesting tech. If a game comes out the other end then all the better but I personally am not disappointed when you don't have some tangible product since I am here for the journey. Good luck, have fun, and make sure you don't accidentally burn out :)

[@thomaspeterson5547]

I've also been trying to make the jump from Unity to Bevy, so this is really exciting to hear. ECS is fascinating, but there isn't too much out there on practical Bevy architecture / render graph code. Looking forward to seeing how you approach it!

[@guyofek3583]

Love your videos and am very excited about the switch to Bevy ❤. Big fan of the engine. Good Luck!

[@nikmirza6811]

04:33 I would advise against Blender for level design, assuming you're making 3D/2.5D games. I do recommend using something like TrenchBroom for 3D level, and LDtk for 2D. There is a lot of QoL feature that Blender does not provide for level design. Pain point is texturing/uv map, TrenchBroom workflow would definitely help but since you're using Bevy/Rust you might need to build an importer.

[@aarthificial]

The game is 2D but it's not based on a grid. Does LDtk support things like procedurally generated meshes that follow a spline, for example?

[@nikmirza6811]

LDtk is cool for how it organized the level. It also had entity that you can set custom properties to and place your custom object implementation on the map. But if the case is for real-time level editing, it should be worth the effort to build a custom level editor. Haven't seen many small indie gone that route so far.

[@Nyubug]

​@@aarthificialYou could use Blender geometry nodes for that, but I assume you are aware of this.

[@bilel114]

Good luck with the changes man. I'm excited to see the new approach.

[@BluePhoenixGames1]

I would consider that using a framework (that in your research isn't stable yet) will probably also hit its own kinds of bumps and limitations when needing to implement the specific details of your project. No engine or framework is perfect but there will be different things to work around

[@laundmo]

there is a noticeable difference though: bevy, and other open source game frameworks, are far more "hackable" than commercial engines. easier to rip out parts or implement your own. he isnt even using the entire bevy engine, just the ECS, window management and asset system. renderer? audio? all custom (at least the plan) - this is the power of modularity

[@runnkorad]

I was already very impressed with astortion. To hear that you're optimizing it with a new engine and better decisions makes me excited for how much cooler it'll be!

[@warrenhenning8064]

Bevy bevy bevy bevy! Tantan has definitely done a great job hyping up Bevy. Good luck.

[@xavierlanglands9486]

I've been making games for a long time now, and I've run into this issue many times, where I focus too much on creating new, cool systems while neglecting the other aspects of the game. One thing I've started doing to combat this is having a theme of the month I work towards. It helps you hone in on parts of your game that you feel are neglected/need improving, while letting you build some cool systems corresponding to that theme. For example, my theme for this month is meta progression, so I got to add some cool achievements and base building for my game. The only cons to this approach is that developing a vertical slice takes a lot longer, and that it can be hard sometimes to stay focused on one theme when many parts of the game need tending too. But, it's helped my projects become a lot more focused and keeps me interested in them for months now. Whatever you choose to do, I hope you find something that works and we get to play Astortion!

[@WuchtaArt]

Glad to hear about the changes! Hopefully this will help you finish the game. Good luck!

[@0-M72-0]

It's amazing for me to hear you say you've made really bad decisions in the past, when I have seen you reshape unity to your desire in a way I had never seen anyone do before, and it's honestly sad to see you leave this version of Astortion and Unity after two whole years of perfecting it, but I have learnt to trust in you as you understand how you want tomake your game like no one else. I just hope some of the modifications you have made for unity along these years will still be useful for you in this new one.

[@Hersatz]

Regardless of how many time we start a project over, if we keep looking back at our old decisions when it comes to code architecture, we'll quickly deep dive into an endless loop of "This code is bad, I should start over". Wanting to implement custom tool for most of everything can also be a really deep well to never get out of, regardless of the engine. I think the best approach is to do something that's fun to play and build up from there. Refactor the ugly code that made the fun be fun so that it's on par with the best practices and iterate the design until a nice, tight, and solid game comes out.

[@dungensAreDragons]

This was the single most important decision you could make. god speed! my excitement has increases significantly!

[@NICK....]

Good luck with the new engine!

[@aarthificial]

Thanks!

[@oprelwork]

Sounds like a really solid plan! I've been working on a commercial indie project in Unity for the past three years and I'm also heavily considering moving to Bevy for the next one. I do have some concerns about doing a rewrite in another engine before finishing your prototyping/design stage. You're in a position where you have a bunch of building blocks already built along with the sometimes mediocre built-in Unity systems. That sounds like a much healthier place to quickly iterate on design compared to a completely blank slate. Quickly being able to throw stuff in and not worrying about having actually decent architecture is super valuable for prototyping. Whichever way you end up going, best of luck!

[@jojojojoost3507]

I'm all here for your very neat implementations of different systems in your game, so I'd say the more freedom the tool gives you the better :) On a side note I'll use this as a reccomendation for bevvy and might try it out myself. Already had tried to get into both rust and gamedev seperately a while back

[@embersworkshop]

I love this approach, I've spent years working on projects and thinking the tech had to be there in full before I could move on. My recent game I'm developing like you described: always playable, and any improvements are merely improving the game, but the game is technically always publishable since it's always playable. Makes it much easier to reason against scope creep because you can playtest to see how good the game is without that new idea or system.