Games for 2009... And BEYOND!

The place to hang out and talk about totally anything general.
Jordy...
level5
level5
Posts: 2367
Joined: Tue Jun 30, 2009 7:57 pm

Re: Games for 2009... And BEYOND!

Postby Jordy... » Fri Dec 20, 2013 1:13 am

Anyone playing DayZ? Thoughts, impressions, feelings, emotions,... anything? I really enjoyed the premise, but the mod was a bit too bare bones and boring for my taste, how has this changed?
Jackdapantyrip
level4
level4
Posts: 752
Joined: Sat Mar 19, 2011 6:33 am

Re: Games for 2009... And BEYOND!

Postby Jackdapantyrip » Thu Feb 13, 2014 6:30 am

I've currently been playing hearthstone, while some of the strategy aspects are quite deep, not having all the cards available and playing at a high level can be frustrating.

The arena mode is pretty fun.

Also for anybody interested http://www.titanfall.com/beta is taking applications.
User avatar
NeatNit
level5
level5
Posts: 2929
Joined: Mon Jan 28, 2008 2:41 pm
Location: Israel
Contact:

Re: Games for 2009... And BEYOND!

Postby NeatNit » Thu Feb 13, 2014 5:44 pm

At some point in the distant future, games would be rendered and displayed at 5000 fps. It will be glorious.


See you there. *enters freezing chamber*

Edit: I'm bored, and feel like writing something up. Therefore, my view on things:
1. There is a limit to how big a monitor you might want. This limit has been reached.
2. There is a limit to how many pixels a monitor of that size can have before any increase becomes meaningless. Many professionals think 4K (QuadHD) is just that.
3. Computers and processing power is gradually improving, always.
4. So is graphics quality.
5. Eventually, processing power would become so powerful that it would be no problem to create fully detailed images in full 4K resolution without any compromises.
6. Besides framerate, what do we have left to improve in games with all that processing power?

Therefore this will happen.


Why is this awesome? Because of motion blur, or lack thereof. In real life, if you track a moving object with your eyes, it will appear sharp and the entire background will show motion blur to you. Do the same on a computer screen, and the moving object will appear blurry as well, even if no computed motion blur is applied. Now, in real life, if you're looking at something static (let's call it the background) and an object comes flying past, it will be blurry. In a game with no computed motion blur, it would flash in each position on its track with each frame. If you're paying enough attention, you can see details in it even though you shouldn't. If there is motion blur applied to the object, it would look right when you're not following it with your eyes, but when you are it would be even blurrier.

The best way to solve problems is to go crazy with the numbers. Why can't I find any info on a prototype screen/simulation with non-fake ultra-high frame rates?
User avatar
xander
level5
level5
Posts: 16869
Joined: Thu Oct 21, 2004 11:41 pm
Location: Highland, CA, USA
Contact:

Re: Games for 2009... And BEYOND!

Postby xander » Thu Feb 13, 2014 6:10 pm

NeatNit wrote:At some point in the distant future, games would be rendered and displayed at 5000 fps. It will be glorious.


See you there. *enters freezing chamber*

Edit: I'm bored, and feel like writing something up. Therefore, my view on things:
1. There is a limit to how big a monitor you might want. This limit has been reached.
2. There is a limit to how many pixels a monitor of that size can have before any increase becomes meaningless. Many professionals think 4K (QuadHD) is just that.
3. Computers and processing power is gradually improving, always.
4. So is graphics quality.
5. Eventually, processing power would become so powerful that it would be no problem to create fully detailed images in full 4K resolution without any compromises.
6. Besides framerate, what do we have left to improve in games with all that processing power?

Therefore this will happen.


Why is this awesome? Because of motion blur, or lack thereof.

  1. You say that there is a resolution at which further improvement is meaningless, and, therefore, no one will develop higher resolutions. Why does the same argument not apply to framerates? Anything beyond 60 FPS is already pretty much beyond what the average person can perceive. Double that to allow for 3D glasses (with frames alternating between the left and right eyes, for instance), and double that again just for good measure. Anything beyond 240 FPS is almost certainly meaningless, as well.
  2. You assume that computing power will continue to increase. It may, or it may not. It is hard to say. We have already pretty much run up against the limits of circuit size, and there are theoretical bottlenecks limiting what parallelization can do. Maybe computers will continue to get more and more powerful, but it is not something that I would count on indefinitely.
  3. Even assuming that computers do continue to increase in power, there are many, many things that can be done with all of that power. A good friend of mine wrote code for his masters thesis which simulates turbulent flow. His model was the Navier-Stokes equations, on a relatively low resolution grid (say 100x100 cells). He was using CUDA, a parallel extension to C which takes advantage multi-processor GPUs. His code took about a week to simulate maybe a minute of time, and his results were a couple of orders of magnitude faster than most previously published results. To simulate in real time would require much, much more powerful computers, and every doubling of the resolution for the simulation requires four times as much power (eight times as much if you want to model three dimensions). I imagine that if computing power ever became sufficient to use such models in games, we would see cycles devoted to such applications. So there are definitely things that could eat up any spare cycles. ;)

xander
User avatar
NeatNit
level5
level5
Posts: 2929
Joined: Mon Jan 28, 2008 2:41 pm
Location: Israel
Contact:

Re: Games for 2009... And BEYOND!

Postby NeatNit » Thu Feb 13, 2014 6:34 pm

xander wrote:
NeatNit wrote:At some point in the distant future, games would be rendered and displayed at 5000 fps. It will be glorious.


See you there. *enters freezing chamber*

Edit: I'm bored, and feel like writing something up. Therefore, my view on things:
1. There is a limit to how big a monitor you might want. This limit has been reached.
2. There is a limit to how many pixels a monitor of that size can have before any increase becomes meaningless. Many professionals think 4K (QuadHD) is just that.
3. Computers and processing power is gradually improving, always.
4. So is graphics quality.
5. Eventually, processing power would become so powerful that it would be no problem to create fully detailed images in full 4K resolution without any compromises.
6. Besides framerate, what do we have left to improve in games with all that processing power?

Therefore this will happen.


Why is this awesome? Because of motion blur, or lack thereof.

  1. You say that there is a resolution at which further improvement is meaningless, and, therefore, no one will develop higher resolutions. Why does the same argument not apply to framerates? Anything beyond 60 FPS is already pretty much beyond what the average person can perceive. Double that to allow for 3D glasses (with frames alternating between the left and right eyes, for instance), and double that again just for good measure. Anything beyond 240 FPS is almost certainly meaningless, as well.
  2. You assume that computing power will continue to increase. It may, or it may not. It is hard to say. We have already pretty much run up against the limits of circuit size, and there are theoretical bottlenecks limiting what parallelization can do. Maybe computers will continue to get more and more powerful, but it is not something that I would count on indefinitely.
  3. Even assuming that computers do continue to increase in power, there are many, many things that can be done with all of that power. A good friend of mine wrote code for his masters thesis which simulates turbulent flow. His model was the Navier-Stokes equations, on a relatively low resolution grid (say 100x100 cells). He was using CUDA, a parallel extension to C which takes advantage multi-processor GPUs. His code took about a week to simulate maybe a minute of time, and his results were a couple of orders of magnitude faster than most previously published results. To simulate in real time would require much, much more powerful computers, and every doubling of the resolution for the simulation requires four times as much power (eight times as much if you want to model three dimensions). I imagine that if computing power ever became sufficient to use such models in games, we would see cycles devoted to such applications. So there are definitely things that could eat up any spare cycles. ;)

xander

First of all, sorry for the stealth edit but I added a few more lined to the first post explaining why more frames would help.

As for #2: I am not too knowledgeable in this area and I might just be naive, but I like to believe that computers will indefinitely continue to be made better.

Now, better use for the extra power... Sadly, you have a point there. I just wish SOME developers will try it.
User avatar
xander
level5
level5
Posts: 16869
Joined: Thu Oct 21, 2004 11:41 pm
Location: Highland, CA, USA
Contact:

Re: Games for 2009... And BEYOND!

Postby xander » Thu Feb 13, 2014 6:46 pm

NeatNit wrote:First of all, sorry for the stealth edit but I added a few more lined to the first post explaining why more frames would help.

There were actually several stealth edits. I started replying before you wrote your list. :P

NeatNit wrote:As for #2: I am not too knowledgeable in this area and I might just be naive, but I like to believe that computers will indefinitely continue to be made better.

The speed of light should ultimately become a problem, no?

NeatNit wrote:Now, better use for the extra power... Sadly, you have a point there. I just wish SOME developers will try it.

Some do. I mean, haven't you seen IV's raytracing bootloader for Darwinia. ;)

xander
User avatar
NeatNit
level5
level5
Posts: 2929
Joined: Mon Jan 28, 2008 2:41 pm
Location: Israel
Contact:

Re: Games for 2009... And BEYOND!

Postby NeatNit » Thu Feb 13, 2014 7:02 pm

Seriously though, if you think "anything beyond 60 FPS is already pretty much beyond what the average person can perceive", then you missed my stealth-edited point. You need to take a look over here: http://www.testufo.com/

This is your screen showing what is input as a crisp pixel-perfect UFO flying past your screen at a constant speed, at your setup's highest possible framerate. As you instinctively follow it with your eyes it just turns into a smudge. I am assuming that your monitor doesn't have any of that fancy black-frame-insertion or anything of that sort. All I'm saying is that it would look sooo much better with actual data in between. Think of it: if even at 3000 pixels/sec each tick would only be a single pixel difference or even less. That is what I want.
User avatar
xander
level5
level5
Posts: 16869
Joined: Thu Oct 21, 2004 11:41 pm
Location: Highland, CA, USA
Contact:

Re: Games for 2009... And BEYOND!

Postby xander » Thu Feb 13, 2014 7:31 pm

NeatNit wrote:Seriously though, if you think "anything beyond 60 FPS is already pretty much beyond what the average person can perceive", then you missed my stealth-edited point. You need to take a look over here: http://www.testufo.com/

This is your screen showing what is input as a crisp pixel-perfect UFO flying past your screen at a constant speed, at your setup's highest possible framerate. As you instinctively follow it with your eyes it just turns into a smudge. I am assuming that your monitor doesn't have any of that fancy black-frame-insertion or anything of that sort. All I'm saying is that it would look sooo much better with actual data in between. Think of it: if even at 3000 pixels/sec each tick would only be a single pixel difference or even less. That is what I want.

It may be what you want, I simply question your ability to actually perceive the difference. As I said, 60 FPS seems to be pretty close to the limit of human perception. Double it for good measure. Hell, increase it by an order of magnitude, and call it 600 FPS just to be sure. I doubt that there are many people (if any at all) that could tell the difference. Frankly, I can't tell the difference between 60 FPS and 120 FPS in your example.

xander
User avatar
shinygerbil
level5
level5
Posts: 4667
Joined: Wed Dec 22, 2004 10:14 pm
Location: Out, finding my own food. Also, doing the shinyBonsai Manoeuvre(tm)
Contact:

Re: Games for 2009... And BEYOND!

Postby shinygerbil » Fri Feb 14, 2014 7:20 pm

NeatNit wrote:At some point in the distant future, games would be rendered and displayed at 5000 fps. It will be glorious.


See you there. *enters freezing chamber*

Edit: I'm bored, and feel like writing something up. Therefore, my view on things:
1. There is a limit to how big a monitor you might want. This limit has been reached.
2. There is a limit to how many pixels a monitor of that size can have before any increase becomes meaningless. Many professionals think 4K (QuadHD) is just that.
3. Computers and processing power is gradually improving, always.
4. So is graphics quality.
5. Eventually, processing power would become so powerful that it would be no problem to create fully detailed images in full 4K resolution without any compromises.
6. Besides framerate, what do we have left to improve in games with all that processing power?

Therefore this will happen.


Why is this awesome? Because of motion blur, or lack thereof. In real life, if you track a moving object with your eyes, it will appear sharp and the entire background will show motion blur to you. Do the same on a computer screen, and the moving object will appear blurry as well, even if no computed motion blur is applied. Now, in real life, if you're looking at something static (let's call it the background) and an object comes flying past, it will be blurry. In a game with no computed motion blur, it would flash in each position on its track with each frame. If you're paying enough attention, you can see details in it even though you shouldn't. If there is motion blur applied to the object, it would look right when you're not following it with your eyes, but when you are it would be even blurrier.

The best way to solve problems is to go crazy with the numbers. Why can't I find any info on a prototype screen/simulation with non-fake ultra-high frame rates?


Short answer, as xander has already stated: accurate, detailed simulation. Physics, instead of "Ragdoll Physics(tm)!"

I believe that there is still a lot of room for more processing power in the form of multi-core/parallel processing/clusters/whatever you want to call it. (Just not the "cloud". Urgh.)
Jackdapantyrip
level4
level4
Posts: 752
Joined: Sat Mar 19, 2011 6:33 am

Re: Games for 2009... And BEYOND!

Postby Jackdapantyrip » Sun Mar 23, 2014 10:28 pm

A cool business model that Iv could follow..

https://www.unrealengine.com/

Also that chaos remake will be sweet!

Echoes+ was released on the binaryzoo forums for anybody interested.
User avatar
shinygerbil
level5
level5
Posts: 4667
Joined: Wed Dec 22, 2004 10:14 pm
Location: Out, finding my own food. Also, doing the shinyBonsai Manoeuvre(tm)
Contact:

Re: Games for 2009... And BEYOND!

Postby shinygerbil » Mon Mar 24, 2014 6:47 pm

By the way,

Dark Souls II
User avatar
bert_the_turtle
level5
level5
Posts: 4795
Joined: Fri Oct 13, 2006 6:11 pm
Location: Cologne
Contact:

Re: Games for 2009... And BEYOND!

Postby bert_the_turtle » Thu Mar 27, 2014 3:33 pm

xander wrote:Frankly, I can't tell the difference between 60 FPS and 120 FPS in your example.

In such artificial testing scenarios, I can tell the difference between 60 FPS and 120 FPS very clearly, and the difference between 120 FPS with regular backlight and 120 FPS with the backlight pulsed to the refresh rate (LightBoost). But that's where it ends. There is no visible motion blur left, so no way I'd be able to tell the difference between 120 FPS+LightBoost and 240FPS+Lightboost. Well, telling the difference may be possible, but caring about it enough to get the twice as powerful PC required to achieve it? Not for me, not until game developers have really run out of things to waste CPU and GPU power on, anyway. And we'll hit the hard limit of how many numbers a reasonably sized box can crunch before that happens.

Essentially, the benefits of high refresh go down with the inverse of the refresh rate. Cost goes up linearly. It's hard to imagine that leading to a sweet spot by any measure upwards of 200 FPS (per eye).
User avatar
xander
level5
level5
Posts: 16869
Joined: Thu Oct 21, 2004 11:41 pm
Location: Highland, CA, USA
Contact:

Re: Games for 2009... And BEYOND!

Postby xander » Thu Mar 27, 2014 4:23 pm

bert_the_turtle wrote:
xander wrote:Frankly, I can't tell the difference between 60 FPS and 120 FPS in your example.

In such artificial testing scenarios, I can tell the difference between 60 FPS and 120 FPS very clearly, and the difference between 120 FPS with regular backlight and 120 FPS with the backlight pulsed to the refresh rate (LightBoost). But that's where it ends. There is no visible motion blur left, so no way I'd be able to tell the difference between 120 FPS+LightBoost and 240FPS+Lightboost. Well, telling the difference may be possible, but caring about it enough to get the twice as powerful PC required to achieve it? Not for me, not until game developers have really run out of things to waste CPU and GPU power on, anyway. And we'll hit the hard limit of how many numbers a reasonably sized box can crunch before that happens.

Essentially, the benefits of high refresh go down with the inverse of the refresh rate. Cost goes up linearly. It's hard to imagine that leading to a sweet spot by any measure upwards of 200 FPS (per eye).

Of course, if leave in the rest of the text which you chose not to quote, you'll notice that I allowed for the fact that I may be less able to see the difference than others, and offered an order of magnitude increase to 600 FPS as being (almost certainly) beyond the perception of any person. The point being that there is limit to human perception vis-a-vis framerates, and processing power is better used on other things (physics modeling, for instance).

xander
User avatar
NeatNit
level5
level5
Posts: 2929
Joined: Mon Jan 28, 2008 2:41 pm
Location: Israel
Contact:

Re: Games for 2009... And BEYOND!

Postby NeatNit » Thu Mar 27, 2014 6:33 pm

The problem I was trying to "fix" can best be seen here (fine under default settings): http://www.testufo.com/#test=blurtrail
Ignore the intended purpose of this test.

If you have a CRT or a LightBoost (or equivalent) screen, and you follow this line with your eyes, then it would appear exactly as it's supposed to appear: a single, thin line moving across the screen. But what if you keep your eyes static while the line moves past?

In real life, an object moving past your vision would be blurred. For example, if you're in a car driving past a fence, and you're looking at something behind the fence, then the fence blurs past so as to become partially transparent to you. But, if you were to follow the fence with your eyes, you could examine it in great detail (providing you're not going so fast as to be unable to focus on it for enough time).

On a computer screen, even at 120fps with LightBoost, you can only have one or the other. Either the fence is drawn blurred out + transparent to allow you to look beyond it, or it's drawn frame-by-frame as a static fence to allow you to track it with your eyes. In that case, when you look behind the fence, it just flashes annoyingly. You can't have both.

My intention was to have a framerate so high, so far beyond human perception of motion, that by rendering every frame as a static image, it would still allow you to look beyond the fence and it would blur past your vision.
User avatar
xander
level5
level5
Posts: 16869
Joined: Thu Oct 21, 2004 11:41 pm
Location: Highland, CA, USA
Contact:

Re: Games for 2009... And BEYOND!

Postby xander » Thu Mar 27, 2014 8:18 pm

NeatNit wrote:My intention was to have a framerate so high, so far beyond human perception of motion, that by rendering every frame as a static image, it would still allow you to look beyond the fence and it would blur past your vision.

I don't think that anyone misunderstands your intention. I just think that your expectations regarding human perception are a bit high. While I personally can't tell the difference between 60 FPS and 120 FPS in the example you give, I am more than happy to believe that others probably can (I am terribly remiss both in keeping my lens prescription up to date and in keeping the lenses of my glasses clean, so everything probably looks a bit blurrier to me than to you). That being said, my intuition is that 60 FPS is getting near the ballpark of human perception---particularly outside of laboratory conditions, when a user/player is probably not specifically looking for the kinds of artifacts that you are describing. Given an order-of-magnitude fudge factor, I am confident that you would hit the limit of human perception with near universality---call it 600 FPS. After that point, there is no reason to keep throwing resources at framerate, and it seems more logical to create more detailed graphics (though, again, we are getting close to the limits of human perception), better physics simulations (PDEs are hard, so I doubt that we will ever cease needing more powerful computers to give us better simulations), and so on.

xander

Return to “Introversion Lounge”

Who is online

Users browsing this forum: No registered users and 21 guests