pouët.net

To Unity or Not to Unity

category: general [glöplog]
Well. I am tired. I see now that he has many contributions, including on demos that I liked a lot.

Anyway, I will accept my defeat. I should not opinate here, anyway. I am so far from you all, I never visited any party, makes no sense to talk anything here.

I will learn to meditate and stop posting shit.
added on the 2020-04-26 07:45:21 by imerso imerso
Quote:
Please, please let's stop this fucking thread. I would simply delete it if I could.
Quote:
I will learn to meditate and stop posting shit.

Please proceed. Like, really.
added on the 2020-04-26 09:24:05 by SiR SiR
Quote:
My impression is that if it's not directly related to graphics, nobody will care. Third-party compressors and music players have been standard in the scene for ages. "Music player" can even be an MP3 decoder outside size restricted compos, because music is not graphics and thus isn't required to be real-time.


I presume the distinctiveness of demoscene products is more relevant aspect here than "real time". It is common to see comments which imply that the speaker values it if music compo entries are made with scene tools. Maybe the style matters here as well: does the final product have some qualities that connect it to the computer culture or not.
Quote:
because music is not graphics

I can see multiple historic reasons for that to be honest:
- Traditionally "computer graphics" was always made by a computer but even "computer music" was usually produced by hardware and tape. While there's been great effort to make computer graphics look more and more advanced, there hasn't been a cultural push to do so for audio - emulating "traditional" sounds via computer has been generally shunned upon.
- Audio hardware is essentially stuck in the 90s with the complexity of a Sound Blaster; even pro sound cards are just essentially more channels and routing and noise shielding, but no hardware-accelerated DSP. Hoping the new console gen changes this but at this point even the newest innovation in audio (RTX Voice) is running on the GPU.
- Let's face it, "CD quality" digital audio (44khz stereo 16bit) and MP3 are standards that have done a pretty good job capturing what is possible in audio; you can gain more fidelity with like 96khz lossless, but I think most people wouldn't notice - comparing that to video there used to be an obvious resolution- and quality-loss before things like H264 came around.
- Audio has a much more prominent uncanny valley, people are more likely to notice when something sounds wrong as opposed to when something sounds great. In that sense, there's more to lose.

So I think just the nature of how audio is in the current day is part of the reason why audio is okay to be as it is, and there's no big push for any sort of innovation there in demos. That being said, I also think that moving away from strictly computer-y sounding music did good for the scene and the versatility of demos with vocals or real instruments made it a better thing.
added on the 2020-04-26 12:08:48 by Gargaj Gargaj
Quote:
- Traditionally "computer graphics" was always made by a computer but even "computer music" was usually produced by hardware and tape. While there's been great effort to make computer graphics look more and more advanced, there hasn't been a cultural push to do so for audio - emulating "traditional" sounds via computer has been generally shunned upon.


I've felt that in 8-bit and 16-bit computer demos and games there was often an assumption that "real" sounding stuff was awesome and many desperate efforts to include lengthy PCM samples of music in products were taken. On C64 efforts for "real" sound were taken despite of the fact that samples were heard correctly only on certain SID revisions. I've assumed that many people who used computers in the 1980s understood the intrinsic value of bleepy sounds only later ...if ever. On the other hand, arcade and PC games seemed to stick to FM chips for a rather long time so the big picture may be mixed.
imerso: aside of my importance in bringing art in the demoscene to higher levels that were previously unseen, i am pretty sure Soundy understood the sarcastic notion of my joke towards him a lot better than you ;)
added on the 2020-04-26 13:54:05 by maali maali
Quote:
I can see multiple historic reasons for that to be honest

I agree that it's mostly a cultural issue, and not one limited to the scene. When the industry introduces new tv/film formats, game console generations, etc., the superficially technical consumer discussions mostly revolve around picture quality, leaving audio for a minority with special interest in surround systems.

In the context of this thread, the point was to illuminate the hypocrisy in demanding that everyone should write their own graphics related code from scratch, while the soundtrack is rendered in professional software. That's like submitting playback of a video rendered in Maya to the demo compo, but of course it's not seen that way due to the abovementioned cultural situation.

Regarding audio hardware being stuck in the 90s, is that really relevant? You can render audio in software, just like non-accelerated PC demos in the 90s rendered graphics in software. Whether there's a point in doing that outside size compos or not is a different discussion, but I don't think it makes sense to claim that Unity gives an unfair advantage while Cubase is accepted.
added on the 2020-04-26 14:55:27 by absence absence
Quote:
Regarding audio hardware being stuck in the 90s, is that really relevant?

Not necessarily directly, just culturally; computer graphics and obviously demos boomed when shaders came around because suddenly everything was plausible. Audio, by contrast, still mostly relies on the CPU or in rare cases the GPU.
added on the 2020-04-26 15:30:13 by Gargaj Gargaj
Interesting topic. Here's some thought..
The way I see it, he scene is first and foremost about being proud of your own achievement, and having to share it with your friends cause it's so cool.
In the golden era, you couldn't even run videos so realtime was a necessity. Then you've got demos that were awesome because they could actually show videos! (9fingers, toaster, etc). I don't know when and how and why they came to put animations into wild category. Anyway.. think of all the YouTube kids, decades ago they'd be sceners (or wouldn't have had computers so would be drawing on papers instead)

I could say that the early scene was united by its tools: making music with computers, making graphics with computers, making animations with computers... And then the computer has been fully democratized. There really is no difference between a scene and non-scene graphic or music anymore, other than who and why it's been done. In a parallel universe, there would be people proud of their AfterFX prods, as well as coders proud of their neural networks or whatever. Like somehow, someone decided the focus should be real-time visuals, so that the coders could stay relevant. And now this is threatened by U/UE.
But those are tools just like AfterFX, or Visual Studio.

If you use Visual Studio to end up doing the equivalent of a shitty AdterFX prod, is it worth it? If you're proud of what you've learned, yes. If you do it just to obey the rules but in your heart you'd go the AFX way, not at all.
added on the 2020-04-26 18:52:08 by BarZoule BarZoule
(rambling continues)
When it comes to videogames, I'm pretty much on the "custom tools" side but for a different reason: generic engines are made to solve everything, instead of your specific needs.
How comes Doom could change the gaming history and be done in roughly 6 months with a super small team? Think of what we're seen in the last 15-20 years. It's common to need at least 100 people over 4 years. The level of detail is going through the roof but it seems the industry is still pushing tech towards graphics instead of world/asset -building. It's kinda silly to not be able to adjust a room layout or race track post-dressing.

Back to demos. For a time, we were ahead of games because we didn't have to care about playability, occlusion, physics, ai, etc, meaning more CPU/GPU and more dev time. Nowadays, we can't rival with armies or outsourced artists and it's hard to do a radically better PBR than UE. But we can still use unorthodox rendering (raymarching, particles) and.. we *can* make ourselves super awesome tools and routines to help make stuff your average 3d modeling tool just can't. And since everything is focussed on realtime it should be easy to keep all the tweaking real-time.

I guess what I'm saying is: don't fear Unity, you can do better!
added on the 2020-04-26 21:20:45 by BarZoule BarZoule
regarding the audio discussion: the stuff that folks like virgill and noby been doing these last few years on 4ks totally deserves extra props.
added on the 2020-04-26 23:31:36 by psenough psenough
Some uninteresting questions I have asked myself, that may not be that uninteresting in the end:

- Ask yourself first, WHY is watching or creating a demo worth your time? I mean, why is really it, that you spend your time creating or watching demos? There is something worth your time there. Do you know what is it?

- Would you value more some demo of you liking if it ran on a less powerful computer? Is there a limit on this when you would stop caring? Would it be more awesome if it ran on a computer so underpowered that it would seem like this was impossible to do, but this person has proven otherwise?

- Would you value less your favourite demo if it needed a computer so powerful and expensive that your home computer wouldn't be able to play it at all? For example if it required one thousand cores with 16 terabytes of RAM? To do exactly the same.

- Generalizing that, would you like even more better visuals in PC demos if they didn't work at all on your sub-6000€ PC? Why? Or why not?

- If you really don't care about requirements, then why do you care if it's real time or not? Just wait 10 or 20 years and voilà, there you have it. Now it's real time.

- If you don't care about tools, why do you care if the graphics are generated or not? Can we also buy art? Use a mp3 from freemusic.com or buy 3D models? In fact, if visuals are the rage, why not just go the extra step and use real videocameras to capture real footage? It can't be more real than that. Then use commercial or open source video tools to edit your production. There are many.

- Because, what separates demoscene from, for example, music videos in your opinion?

The problem with PC demoscene is that nobody knows anymore the answer to many of these questions. Nobody knows anymore what a PC demo is, because if there are no limits, there are less and less things to "demo" anymore. It's like having a popular race but some people go on foot, some use bikes, some use cars, some just take a plane. And they are all arguing about who's doing it properly. The answer is nobody, and everyone. There is no debate possible. You are mixing things.

Nowadays when you work on a PC production you don't know what you are competing against, because there are no rules anymore. If you are working on a 68k high end Amiga prod, even if there aren't a 100% clear set of rules you at least know you are targetting about 50mhz, probably 64mb of RAM and you have many examples of what is easy to achieve and what is not. With this sort of oldschool stuff you know the rules, people know what has been achieved until now, and to awe people you are going to have or very original or very skilled.

As I said before, this "debate" about allowing (game?) engines on PC demoscene is not a debate at all. Some people like bikes, some people like skating. If the PC demoscene wants to survive it needs to set limits. It needs rules. Freeze a config for five years. WHATEVER CONFIG. But ONE config. i7, 8 cores, 32gb of RAM, a RTX 2080ti. Whatever! Set rules on tools. Who decides? I don't know. Let those who win the big events vote, they are the "experts", aren't they?. Make a Unity competition, why not? Aren't there so many PC sceners? This is something you can't do on the Amiga or the Atari demoscene, but you surely can on the dominant platform of the last decades. Make an Unreal Engine competition. Don't we have a DOS competition? Why can't there be a "Demos with tools" competition? And a separate "raw" competition where the only tools you have are your own. I don't understand why this seems so outlandish to so many people. Maybe Epic and Unity would even want to toss some money to promote their tools at the biggest events. The games industry has always had a love relationship with the demoscene. I find it hard to believe they wouldn't want to create a source of talented professionals experienced in their tools.

But some poeple will say: What if there are too few entries on each competition?

Well, fuck me. First of all, why do you care? You weren't already competing in the first place, just like nobody is competing against you when they take a car to go to work and you take a bus. Just split the damn money and give everyone their participation trophie. Or toss them to the wild compo. Or make a wild "outliers compo" for these cases and split all the money with them.

Really, why are there people risking their lives by climbing mountains instead of getting there by jumping from a plane or helicopter? Do someone think the climbers and the pilots or parachute jumpers "compete"? No. Is there less merit on flying an helicopter or climbing a mountain? No. Just different. Why can the PC demoscene just understand this? How can PC demoscene even have this sort of discussion? There is really no discussion at all!
added on the 2020-04-27 23:38:41 by Debvgger Debvgger
Quote:
Can we also buy art? Use a mp3 from freemusic.com or buy 3D models?


Back to mediation... =)
added on the 2020-04-28 04:13:48 by imerso imerso
The analogy with climbing is actually quite good.

"I'm doing this because I enjoy it. Not to mention the most important reason for climbing a mountain."
"And that is?"
"Because it's there."
– James T. Kirk and Spock, 2287 (Star Trek V: The Final Frontier)

It looks like the whole discussion is about the fairness of the competition. The whole thing should be about surpassing yourself, not the guy next to you. Even with the same machines, there will be people with more experience, some with a professional background, some with more time, some who can type faster.... there's no end.
added on the 2020-04-28 04:20:52 by BarZoule BarZoule
Quote:
The problem with PC demoscene (...) having a popular race but some people go on foot, some use bikes, some use cars, some just take a plane. And they are all arguing about who's doing it properly.


actually, the only people arguing about that are purist zealots. i would venture (from talks with other active sceners for the past 30 years) that circa 90% of the active demoscene is just happy to see cool stuff being presented on the big screen.

Quote:
If the PC demoscene wants to survive it needs to set limits. It needs rules.


It's been doing just fine without any of that.

I'll concede the number of active folks has lowered, the threshold to do something stunningly mind-blowing on the level of what game studios with people working full time on these things is high, but folks still release demos see Revision 2020 PC democompo.

And you can't logically argue that the lack of a common limit is the reason for the decline. The PC is the dominant platform on the demoscene for the exact reason that it's free to explore. People who prefer limits go explore other platforms. Both are fine.

But by all means, feel free to organize a democompo with those suggested rules in mind, who knows, i could be dead wrong and everyone just jump at the opportunity to create a spectacular ammount of new mind blowing entries that hardly anyone has the correct hardware to watch on. The worse that could happen is we have more demos. My opinion is that you might very well have an inflated number of entries on that compo while you're poking people to do stuff and then the interest on the extra niche limitation die out once people realize it's sucking their free time from exploring stuff for what's actually modern.
added on the 2020-04-28 05:04:07 by psenough psenough
Sucking your free time looks like a powerful reason to do something, isn't it?
added on the 2020-04-28 05:16:59 by Debvgger Debvgger
I'm pretty sure that everyone find some powerful reason behind the impulse for doing their creative endeavours.

I see people here arguing about aesthetic criteria. Although perhaps my inclinations lead me near the definition of a purist zealot, I recognize that my love for filesize limitations is based (partially) on aesthetics.

That... and mountains.
added on the 2020-04-28 05:40:44 by ham ham
Quote:
Can we also buy art? Use a mp3 from freemusic.com or buy 3D models?


Quote:
what separates demoscene from, for example, music videos in your opinion?


Demos must be real-time, I guess. That is, seems like the fact that it runs from code is the main difference. Which implies, by inference, that code is actually what defines a demo, in the context of demoscene.

Interesting enough, it seems that using a professional/commercial engine code is absolutely normal because "freedom" and "fun", but using professional music and models is not, because "???".

Most (all?) compos are clear that you must own the demo content, which rules out the possibility of a coder use third-party professional music and models.

Please note that I'm not questioning Unity, Unreal and Notch (very costly, by the way) usage in demoscene competitions anymore. I am just curious about the compos rules. Maybe only my computer have been showing "no remote entries allowed" and "only submit your own work" for decades.

- Right of reply. Stop posting, and I will stop as well.
added on the 2020-04-28 06:27:42 by imerso imerso
Quote:
actually, the only people arguing about that are purist zealots.


This is not only untrue, there are also zealots at both camps.

Quote:
i would venture (from talks with other active sceners for the past 30 years) that circa 90% of the active demoscene is just happy to see cool stuff being presented on the big screen.


Well, and I would venture, from talks with many active party attenders for the past 22 years, that circa 90% of the active party-scene is just happy to see cool stuff being presented on a big screen, but their definition of "party" and "cool" doesn't include anything we would relate to the demoscene at all.

I'm afraid there's always going to be a bigger group of people who doesn't understand why you like your thing or value it at all.

Quote:
Quote:
If the PC demoscene wants to survive it needs to set limits. It needs rules.


It's been doing just fine without any of that.


I sadly have to disagree, but again this depends on how and what are you looking at.

Quote:

I'll concede the number of active folks has lowered, the threshold to do something stunningly mind-blowing on the level of what game studios with people working full time on these things is high, but folks still release demos see Revision 2020 PC democompo.


And why should they care about what are game studios doing? Isn't this an example of "The number of active bike riders has lowered, as planes are faster"? The crazy 256 bytes compo is stunningly mind-blowing. The 8 bit demos are mind-blowing. I know for a fact that some required more work than a few of the games I have got lately.

Quote:

And you can't logically argue that the lack of a common limit is the reason for the decline. The PC is the dominant platform on the demoscene for the exact reason that it's free to explore. People who prefer limits go explore other platforms. Both are fine.


Sorry it doesn't seem logical to you, but I usually try to do my best to only argue logically.

It's been proven a million times that limits foster creativity, and the lack of them end up badly. Also seems to be fairly proved that our sense of self-worth is closely related with how successful we are when overcoming problems. So taking this into account it seems like a sensible theory to me that all human activities follow some recognizable trends and patterns. One could predict many people migrating from the demo compo to the intro compo. It could even happen that the demo competition ends up becoming the entry point for beginners as they would not need to take any limits into account at all, have 3D engines ready to be used for free, etc, before reaching some level of skill that allows them to start working on smaller prods. The same can be theorized for the set of computing platforms. Starting on the PC then migrating to the Amiga or the C64 compos. Because overcoming limits is fun. Will this happen? We'll see. Please note I'm not in any way saying there are not nice PC demos, that they don't require a ton of skilled work or that there isn't interesting hardware present. I love coding shaders and the vast amount of processing power is super useful.

Quote:

But by all means, feel free to organize a democompo with those suggested rules in mind,


Yes, of course :-)

Quote:

My opinion is that you might very well have an inflated number of entries on that compo while you're poking people to do stuff and then the interest on the extra niche limitation die out once people realize it's sucking their free time from exploring stuff for what's actually modern.


Well, being "modern" is not the thing, or Smart Watch / Phone demos would be skyrocketing. It's not like the PC architecture has changed that much compared with other systems in the past years.

Also, as I said in the post before this, I don't see "sucking the free time" as anything wrong, but quite the opposite. If it's sucking your free time then maybe it's just you liking what you are doing, aren't you? Why would it be better to code shaders instead of an Atari 2600 prod? Because "jobs" / "money"? But if there are no limits, why would you in fact want to burn your time coding shaders instead of browsing some shader gallery and picking the perfect one for the job? Just as you can get a ready to go engine you can get your ready to go shaders. Where is the limit when there are no limits?

Interestingly only a few people arguing against using, or not using, Unity/UE/Godot/whatever have in fact experience in doing both things. Why not just try all the available options? Coding your own 3D engine is really useful when later using Unity. Creating some non trivial stuff with Unity makes you aware of a ton of things. Try to do something equivalent with Unreal Engine. Then use Godot. And now code another 3D engine with everything you have learnt. And now make another for the Amiga :-)

Wouldn't it be great to have this many compos? Maybe the number of prods would jump. Why not?
added on the 2020-04-28 06:37:07 by Debvgger Debvgger
Quote:
Demos must be real-time, I guess.


I agree with you. And that's why there are separate and very healthy music / graphics executable compos :-)

Quote:
Interesting enough, it seems that using a professional/commercial engine code is absolutely normal because "freedom" and "fun", but using professional music and models is not, because "???".


One could also argue about downloading shaders from ShaderToy / GitHub / etc.

Quote:
Please note that I'm not questioning Unity, Unreal and Notch (very costly, by the way) usage in demoscene competitions anymore.


I see questioning that perfectly reasonable, to be honest.

Quote:

- Right of reply. Stop posting, and I will stop as well.


I will stop now as I just needed a break from a very long and tiresome task :-)
added on the 2020-04-28 06:52:48 by Debvgger Debvgger
Many if not most democompos require only that you have the license and permission to use assets like music and graphics. My Revision entry had only third-party graphics in it, all either public domain or CC licensed, and in the beginning it also had CC licensed music.
added on the 2020-04-28 07:41:57 by Preacher Preacher
Quote:
- Generalizing that, would you like even more better visuals in PC demos if they didn't work at all on your sub-6000€ PC? Why? Or why not?

BB Image
added on the 2020-04-28 09:15:21 by Gargaj Gargaj
@gargaj joke is on them though, because i totally watched that on a 386DX40 back then and was mindblown regardless ^^
added on the 2020-04-28 09:18:36 by HellMood HellMood
@Debvgger: Thank you, some good input in your comments.

Quote:
Quote:

If the PC demoscene wants to survive it needs to set limits. It needs rules.


It's been doing just fine without any of that.


@psenough: Nope it hasn't. If it weren't for the old farts like fairlight, cncd, cocoon and asd (and that's about it already) pc demo compos in recent years look pretty bleank, if you would be honest with yourself. There are some exceptions here and there (yeah VX2), but in general the quality and technical level in most releases is just mediocre at best (see latest evoke compo for a good example).
added on the 2020-04-28 09:46:08 by arm1n arm1n
And what are you doing to change it?
added on the 2020-04-28 09:53:38 by Gargaj Gargaj

login