Rant on why we need better computers!

[Thomas]

Note This was meant as a reply the "Kometen" thread, but become kinda long so I made a separate topic.

Just have to say a few words about the "we need 1000 times faster computers" that Michael said. I totally agree with this and I think that the current state of 3D graphics is horrible and feels very much like high rez graphics was early 90s. In Penumbra we used very little spice on the graphics and tried to keep it plain and simple, yet an enormous amount of work as gone into making the game look decent and work on as many systems as possible. However, with almost every driver update the games stop working for some and we have to make new patches, etc. Unless you use the most basic 3D stuff (like Torchlight) you will run into trouble.

Now that is just the techie part. Even worse is that the power of people's systems vary A LOT so you have to have all of these different settings to make sure as many people can play it as possible. This makes it hard for devs and I spend countless of hours on these sort of issues. It also makes it harder for users who have to configure their rigs to runs as smoothly as possible. This problem is not that hard to consoles, where hardware is the same, but usually games have to support several platforms and the problem is there again.

Another problem is that even if u have the same sort of hardware, you have to be very careful when making art so that it does not eat up too much resources. I spend countless of hours nagging on artists who fail to understand my lessons in how to make a map efficient. The bottom line is: Artist should not care about this! They should only focus on making the art look nice, end of story. Once you put limits on how to do the art, especially with the vague, very techie instructions that come along with making 3D graphics, the quality of art will suffer. This is huge problem when developing games, and while new techniques do lower the amount of thinking needed, we have a LONG way to go.

Compare to where 2D games are today, where you can pretty much do the art in whatever way you like and it will work. You do not have to put objects so the effectively block visibility, etc and can just focus on one thing: To explore the medium and let your imagination be the limit.

I would like 3D graphics to first of all become so fast and efficient that it handles anything an artist throws at it. To create digital art is to leave boundaries of physical reality behind, and that should be strived for in the creation process as well. Further more I would also like to see it all run on Virtual Machines and not be specific to hardware, exactly like flash is to day. Once you get something running on your computer, everybody should be able to run it and, most importantly, see it they way you intended too (and not fiddle with some settings to make it work). Just see the explosion in 2D games since simple work environments like flash and GameMaker has come out. This vastly different from when I started out 15 years ago, when just having a scrolling screen imposed limit on what art could be done.

Even more important, without some kind of virtual machine system that runs everything, it will be harder and harder to keep the amount of emulators needed to work and I fear that many things created today will not be able to experience 20-30 years from now. This by itself would be a hinder from videogames to be a really powerful medium, it is simply very volatile and gets quickly lost as hardware / software evolve.

[Kjell]

Bit shocked that there is so little love for the craft. When a artist hammers into his sculpture causing it to break, should he complain that marble isn't strong enough .. or get some more practice?

[Thomas]

Bit shocked that there is so little love for the craft. When a artist hammers into his sculpture causing it to break, should he complain that marble isn't strong enough .. or get some more practice?

I think the big difference is how obscure the limitations are and an artist cannot evaluate it as simply as a when sculpting or painting. These constraints are not put on the creation process, but much later and this makes all the difference. When the artist has learned all the tools and starts experimenting, the limit is not what can be creating, but what actually is actually useable. It would be like saying to a painter that certain combinations of paint where not allowed, even though it would be perfectly possible for the painter to use them. Or for certain shapes to not be allowed by a sculptor, even though it was possible to create them.

I agree that it is a very worthwhile craft and interesting problem in keeping these techy constraints, but I fail to see how they can improve artistic qualities of the medium.

[Kjell]

A sculpture can suddenly break after having chipped away the final piece ( it's called poor planning ). Anyway, at least we've got Ctrl+Z

[JordanMagnuson]

Bit shocked that there is so little love for the craft. When a artist hammers into his sculpture causing it to break, should he complain that marble isn't strong enough .. or get some more practice?

I think a lot of us are interested in the "craft" of writing good code, but not as interested as we are in creating the end product that we envision.

At least, that's how it is for me: I'm fascinated by programming, and enjoy an elegant and efficient piece of code as much as the next person, but there's only so much time in a day, and I'd generally rather spend my time working on the big picture--what people are actually going to interact with--rather than trying to improve the efficiency of my code or assets.

[Michaël Samyn]

Elegance in coding is a worthy goal in and of itself. There's no need for computers to be slow to make us write elegant code. Elegant code does not only run better, it also tends to be easier to read and easier to debug. I think if our computers would indeed become 1000 times faster, we would still feel the need to write elegant code. So don't worry, Kjell.

You make a lot of good points, Thomas. I really hope the hardware continues to evolve until it can indeed run anything an artist throws at it. I also believe that this will cause an outburst of creativity.

Sometimes I doubt whether this will actually happen. The move to mobile devices (and web 2.0 too) seems to suggest that the mass market (that drives technical evolution in a capitalist system like ours) has other priorities than speed. And true enough, for most applications, most contemporary computers are fast enough. It's not just for 3D that we need more speed, though. I remember Erik talking about how slow Blueberry Garden ran when he was simply following his own human logic of creating a living world in the computer. That's what I mean by making the computer do the work. We humans should come up with the ideas, and the computers should execute them.

[Thomas]

Worth noting is that it is not only a hardware problem, but also a software one of making intuitive and powerful tools for artists, and also making engines that are intelligent and can figure out how to best render the art. So that the computer figures out the techy stuff and lets humans be creative.

[ghostwheel]

Limitations and boundaries aren't a bad thing. They force you to think creatively and make better art. Unlimited does not mean better. However, sometimes you hit a wall no matter what and need to stretch the limits or push them further. That's the nature of both art and technology.

There was an interesting email exchange last week between a Gizmodo employee and Steve Jobs. There was one quote from Jobs I thought was extremely relevant to this topic:

The time's they are a 'changin, and some traditional PC folk feel like their world is slipping away. It is.

I agree. PCs and PC gaming are going to die. Don't count on better (cheaper) hardware. In 5 years, PC hardware costs will rise and low cost PC hardware will not be available - home PCs will be rare. In 10, there won't be a such thing as a home desktop computer at all. The only place you will find hi-def, photo-realistic 3d graphics will be on consoles and services like OnLive. Of course, slate technology will improve as well so you may see some stuff on those but it will probably lag behind, as portable technology always has.

The technology landscape in changing again. I hate to see desktops go away but it's not really up to me.

[Michaël Samyn]

I think you're being a bit premature. It's not because Apple has good marketing that they will suddenly conquer the world. Apple has had good marketing, and often good products behind it, for decades and they have only been able to get a relatively small market segment. I don't think the iPad will change that.

That being said, I'm not a fan of desktop PCs at all. I much prefer a games console, actually. But games consoles are proprietary devices (like iPads) and therefore the range of entertainment they offer is very limited. If the open environment offered by PCs (open as in allowing self-publishing) ceases to exist, it means we're returning to the age of the television, where the big corporations decide what we should be watching. It's a return to broadcasting: with producers on one end and consumers on the other. I'd rather not.

PC games have been a small niche of the games market compared to consoles for a long time. And they've managed to survive and continue to be the place where most experimentation happens. PC games are good at being small. So I'm not too worried. As long as there is a small place in the world where we can dream, I'm happy.

[ghostwheel]

I think you're being a bit premature. It's not because Apple has good marketing that they will suddenly conquer the world. Apple has had good marketing, and often good products behind it, for decades and they have only been able to get a relatively small market segment. I don't think the iPad will change that.

That being said, I'm not a fan of desktop PCs at all. I much prefer a games console, actually. But games consoles are proprietary devices (like iPads) and therefore the range of entertainment they offer is very limited. If the open environment offered by PCs (open as in allowing self-publishing) ceases to exist, it means we're returning to the age of the television, where the big corporations decide what we should be watching. It's a return to broadcasting: with producers on one end and consumers on the other. I'd rather not.

PC games have been a small niche of the games market compared to consoles for a long time. And they've managed to survive and continue to be the place where most experimentation happens. PC games are good at being small. So I'm not too worried. As long as there is a small place in the world where we can dream, I'm happy.

I might be a little premature. However, I'm not saying the iPad alone is going to conquer all. It's simply providing the model for what an easy-to-use, handheld device should be.

The starting price for a Mac pro is $2499 USD. This is fairly expensive. I think that or an even higher price will be what ALL desktops will go for in the near future. And yes, the big corporations will be the only players. Affordable home computing is going to be netbooks and slates. Low cost desktops are going to disappear. I don't like it either. I hate it. Unfortunately, I have a feeling that is where things are heading. We are in the golden age of indie game development and it will be gone all too soon. Enjoy it while it lasts.

[JordanMagnuson]

Elegance in coding is a worthy goal in and of itself. There's no need for computers to be slow to make us write elegant code. Elegant code does not only run better, it also tends to be easier to read and easier to debug. I think if our computers would indeed become 1000 times faster, we would still feel the need to write elegant code.

I totally agree. I'm just saying that there's a difference between having something working and being able to leave it at that, and needing to go over your code again and again and again to try and optimize it for a slower computer. For people who are more art-oriented than code-oriented, that can be a bit of a damper when developing, in the same way that having to fuss over their assets for slower computers can be.

[Michaël Samyn]

You shouldn't give in to The Man so easily, ghostwheel. Fight for what you believe in!
Even in your Doomsday scenario of corporate-owned slow computing, we'll find a way.
But we're not there yet. We can resist.

[zerojuan]

"if this is easy, then everyone will be doing it.". - some nerd game programmer

Games should be so easy to make everyone can do it. as a programmer concerned with elegant code and tight engineering, isn't that a more worthy goal?

[Kjell]

Anybody can use a Piano, but creating something of quality with it is another story.

There's certainly still room for lowering the entry barrier of software development, but we've already come a long way over the past 20 years. Back then almost every title was written directly "on the metal" using assembly, while these days you can create a fully-fledged 3D game without writing a single line of code.

Doesn't have anything to do with hardware though

[Erik Svedäng]

Doesn't have anything to do with hardware though

Well, that's not true at all. The reason programs like Gamemaker and Unity works is that the slowness of the code they produce (by being easy to use, creating portable code, etc) doesn't matter because our computers are so fast.

A good thing with these new weaker, smaller computers is that they require a lot less energy... just screaming for more power doesn't feel good for me. And even if computers were say 10 times faster I'd still have to optimize eventually; there doesn't seem to be a logical "end point" were enough is enough. My solution is to just be happy with what I have and embrace the limitations. The problem of preservation is way worse in my mind... dunno what to do about that. These two things are kind of related in an interesting way; I know that my games will run faster and faster in the next couple of years but I can't be sure they will run at all in 2020.