Diminishing returns with advancing 3d hardware?

[Skrying]

Farcry wow'd me because I'd read about it before but it didnt get lots of hype. Boy, when I played it I was amazed though.

I thought HL2's graphics were mind blowing because the textures were just so amazingly good.

Doom 3 sucked, it was and still is ugly.

FEAR seemed like a mixture of Farcry and HL2, good but not a big leap over those two.

 

[Suflex]

My biggest WOW ever was back in 87, the Amiga.

HL2 definitely is a close second just for the shear artistic beauty, F.E.A.R. for developer guts, this game crippled everything and for no reason that is immediately obvious to the average gamer.

 

[Arwin]

My biggest WOW ever was back in 87, the Amiga.

The Amiga for me was more about lots of Wows rather than biggest wow. But I had the Atari ST first, and had a lot of Wows on that as well. I think my biggest wow back then was probably Populous, the first for looks and gameplay, and the second for being so much better than the first. I have played Populous through Steem recently on a PC and it still looks great.

On the Amiga there were some pretty awesome things as well. Strangely enough Pinball Fantasies was one of my biggest Wows there, the sound was just awesome.

On the PC I had some very different wows, like MT32 support in PC games, or the Pro Pinball series (played one of those again recently, still looks amazing), or the many fun ways of dying in Larry games.

On the C64 my biggest wow was probably Elite. That game was so different. I was almost too little to be able to play it, but watched my older cousin play and then played it shortly after on the Atari ST. The dragon in Gauntlett 2 also wowed me, and so did GFA Basic

Xenon 2 also very much wowed me on the ST, as did Gods, and these two games still look great today also.

The point is that you need to find a balance between gameplay, graphics and sound. That's always been the case. People absorb a lot of information visually, so the importance of that shouldn't be underestimated - but it depends a lot on the kind of game you want to offer also. Something like Point Blank on the PS1 was great even if the graphics weren't amazingly good, and that's the kind of feeling I think the Nintendo is trying to latch onto.

But a lot of the complaints here are already being addressed. You saw the EA presentation of how the movement and behavior of the players was greatly improved on the next-gen versions. In the next Indiana Jones the behavior of the AI is absolutely revolutionary, augmented by use of physics on static objects as well.

In fact, I think the next-gen is going to do precisely that - the paralel setup of the new consoles is such that even if you max out your console's power for the best looking graphics, you are still very likely to have spare processing power for AI and physics improvements.

Also, while the AI is currently sufficient for some types of games, it not nearly is for others. No racing games have great AI at the moment for instance. This is something I hope to see addressed in the next generation, but it will still take a good coding effort even with sufficient processing power. There are plenty of other things, but it's important to see that improvements are happening.

And I agree that graphics wise we're far from having reached the maximum. Environments are still often very static and dead. A lot may improve here in the future ... but it remains a matter of balancing different factors - and that balance differs per game also, not in the least because in online games that exclusively work with humans, you don't necessarily need any AI at all.

A valid point by the way is that if you could use the next generation's full power for an SDtv image, you may be able to get much closer to convincing graphics than we get now in HDtv. I would certainly like to see someone try.

 

[_xxx_]

or the Pro Pinball series

Heh, I just reinstalled Timeshock yesterday. Still the best pinball around by far

 

[Blacklash]

Most games are extremely scalable. Many people try to turn everything on max, pile on AA|AF then complain about FPS.

Many folks have native 1280x LCDs these days and you can get a 7900 GT or X1800 XT 256mb which will do very well at that resolution with some level of AA|AF. The X1800 XT is over at the Egg for 219usd right now.

The only time I am dissatisfied with my Crossfire system is when the CPU can't handle the crunch in certain situations and FPS dips. An overclocked E6600 will fix that.

 

[ERK]

Real life does get blurry because of a thing called lens optics.

Get bifocals.


On topic...
As for WOW factor, I think that Ultima IX deserves some mention here, despite all its problems.

ERK

 

[soylent]

The biggest wow factor I've experienced in games was probably ultima underworld(first textured 3d game I played. It even had slopes and pits and was out before wolf3d) or possibly hunter or stuntcar for the amiga 500(3d un-textured games. Hunter was a somewhat open ended FPS with heigth-mapped terrain, usable boats, tanks and choppers).

 

[pascal]

I still remenber the highest impact I have ever had with a new 3D card, the original Radeon 32MB that I bought (IIRC) during 2000.

The Arc demo was impressive. Today you see people using 10 (ten) times more transistors and the wow factor is not much higher than that.

I still dream see a game like the original System Shock using something like the Arc demo graphics level.

What I mean is that the balance is lost somewhere. The use of heavy/unoptimized graphics/game code, lack of gameplay creativity, lack of really good art/aesthetic, uninspiring sound.

What we really need is to use well the hardware available, not develop for future hardware.

edited: also IMHO the low end systems should be better than they are now. The PC is more and more a visualization system and deserve much better integrated graphics and better sound with some eye in the living room too.

 

[KimB]

What we really need is to use well the hardware available, not develop for future hardware.

Well, considering that games take a number of years to release, if you want to use the hardware available at release well, then you need to develop the game for future hardware.

 

[pascal]

edited: My guess today they develop for new top hardware available at or after the release of the game.

 

[soylent]

Sure, it's good to run well on high-end hardware dating a few years back/low end hardware at release. But it's a lot better if also you can continue to scale with high-end hardware for a year or two into the future.

Of course, games that do this get a bit of bad reputation from the 'tards who cannot accept that a new game doesn't run well with max everything on their shiny new hardware.

 

[Parousia]

What I mean is that the balance is lost somewhere. The use of heavy/unoptimized graphics/game code, lack of gameplay creativity, lack of really good art/aesthetic, uninspiring sound.

I think most of the more anticipated games are pretty clean and as efficient as can be in terms of coding. The art and aesthetic parts are probably down to issues with budgets, financially and technically. I think games should most definitely feature better music to heighten suspense (like in movies). I'd agree with you regarding gameplay although I'm not quite sure how much more creative things can be in a first-person-shooter game (certainly most games in this genre should be more challenging mentally IMO but not just in the sense of looking for the next switch to push but maybe with a bit more puzzle-solving elements).

 

[Ailuros]

edited: also IMHO the low end systems should be better than they are now. The PC is more and more a visualization system and deserve much better integrated graphics and better sound with some eye in the living room too.

Hold that thought. That's something I can definitely agree with.

 

[sonyps35]

And here's a Crysis screen..another powerful argument against diminishing returns

I wish that could be done on the next gen consoles. Since I think it was supposedly running on a X1900XTX..I wish they had put that instead of Xenos in X360.

 

[TurnDragoZeroV2G]

Crysis does look quite beautiful.

I'd like to know how many developers agree with your sentiments regarding the GPU, though.

Maybe somebody should make a poll asking developers whether they would prefer, keeping in mind that the 360 will be around for half a decade, an X1900 GPU (offering a higher pixel shader throughput) or Xenos (offering a more robust feature set, i.e. memxport and 16 bilinear filtering units + 16 point samplers).

 

[fearsomepirate]

I was going to start a thread, but the forum software told me to necro this one. So here you go. A friend and were discussing, and we realized...these are all roughly 13-year-gaps in hardware:

SNES -> Xbox 360 -> Today

So if you like, compare Star Fox with Modern Warfare 2 to the latest games. The 5th most popular game on the PC right now is CS: Go, which came out in 2009 and is running on the ancient Source Engine, which was already getting a little long in the tooth in 2009 anyway. If you still don't believe in diminishing returns, that's the equivalent of one of the hottest PC games in 2005, a year after Far Cry, still being Duke Nukem 3D.

And these are 18 year gaps in hardware:

Atari 5200 -> PS2 -> Today

So if you've got a fairly new gaming PC, the time gap between that and the dusty old PS2 in the basement is the same gap as between God of War 2 and Ms. Pac-Man. But IMO it's pretty obvious that GoW2 holds up better in 2018 than any game targeted at early 80s hardware did in 2000. Heck, run the thing in an emulator with AA and high res, and it doesn't look half bad.

 

[KimB]

Wow. This thread is a blast from the past!

There is no question that diminishing returns have been settling in for a long time, and will accelerate going forward. Most of the (graphics) difference between PS2-era games and today is in resolution and detail: most PS2-era games have absolutely laughable detail by today's standards. Environments are flat with few objects. Textures are low-resolution. The idea of making textures look non-flat was in its infancy.

Today you can play games that are so detailed that you can sometimes forget that the game world is made out of polygons in the first place. The polygons are still obvious if you look for them, but it's getting to the point that you sometimes have to look for them.

Those changes are pretty amazing. But they're nothing compared to the changes seen pre-2000. Moving forward, I think the big changes will be a dramatic increase in dynamic environments, especially with regards to fluids and lighting. Those will be amazing (and they've already begun; Horizon: Zero Dawn comes to mind), but the jump from today will not compare to the change from PS2-era games to today.

 

[Frenetic Pony]

Fun to see everyone here compare. But maybe it's something to do with age as well?

I'd have said the point of diminishing returns was mid 2000's, with the likes of Crysis, Uncharted 2, and even Fable 2 which was for me the first stylized fantasy game that could look like it was really getting the sort of visuals it might have wanted to hit.

But with everyone disagreeing, maybe it's just age that really plays into it. At some point you get so used to seeing advancements that it no longer wows you, you just expect it and so don't have as "objective" a viewpoint on how much of a difference those advancements have made.

 

[fearsomepirate]

"Diminishing returns" is a quantitative idea, so it kind of implies the question...how do we quantify this? I came up with a thought experiment that I think could quantify it. Get a bunch of screenshots going all the way back to the Atari 2600, photographs, and artistic works (sketches, paintings, etc). Poll a bunch of people and ask them to rate the quality of the graphics from 1 to 10. Tell them only that "1" means "This couldn't be much worse," and "10" meaning "I'm not sure if this is a video game, a picture, or an artist's work."

Then plot the rating of the picture with the year the game's target hardware was released. Dollars to donuts say you see a curve that visibly increases more slowly some time in the early to mid 2000s. It's not worth the effort for me to do it, but if I was writing for an industry publication, I'd try to put together the poll.

 

[KimB]

"Diminishing returns" is a quantitative idea, so it kind of implies the question...how do we quantify this? I came up with a thought experiment that I think could quantify it. Get a bunch of screenshots going all the way back to the Atari 2600, photographs, and artistic works (sketches, paintings, etc). Poll a bunch of people and ask them to rate the quality of the graphics from 1 to 10. Tell them only that "1" means "This couldn't be much worse," and "10" meaning "I'm not sure if this is a video game, a picture, or an artist's work."

Then plot the rating of the picture with the year the game's target hardware was released. Dollars to donuts say you see a curve that visibly increases more slowly some time in the early to mid 2000s. It's not worth the effort for me to do it, but if I was writing for an industry publication, I'd try to put together the poll.

I would do it in the following way:
New technologies can be categorized into three groups.
1. Technologies which allow new gameplay mechanisms. Examples include first person gaming (e.g. the original Wolfenstein). Fully 3D environments (e.g. Quake), networked multiplayer (e.g. the original Doom).
2. Technologies that allow more in-depth games with more variation (think Donkey Kong vs. Super Mario Bros.).
3. Technologies which improve visual and audio fidelity.

Early on, we had many innovations which were of the first and second type, as many gameplay styles require a certain level of either processing power or storage. As those improved, there was an explosion of gameplay styles. Some of this was due to simple creativity in a nascent creative field. But generally they simply weren't possible until computer tech advanced to a certain point.

Those kinds of innovations have slowed to a crawl. Most such gameplay innovations these days were possible years prior, but overlooked. These days the most significant improvements are of the third type (think 4k, HDR, adaptive sync). These sorts of things can still change the character of games, but mostly they just make the games look prettier.

That's what I mean by diminishing returns, personally.