1024*768 is faster than 640*480

[EasyRaider]

Taken from a post on Ion Storm DX2 forum.

you see if you run the game at lower resolution it means the video card will be able to render it too easily causing the CPU to be the culprit slowing everything down.

if the video card can render very easily @ 640*480 therefore the CPU has to try and keep up with the video card in calculating Physics, AI whatever..

but if you set the resolution to 1024*768 it means that the GPU has about the same workload as the CPU so they work together nicely therefore you don't get a bottleneck effect hence you get a higher frame rate.

if you have a fast video card like the gamespy reviewer like a Radeon 9500 series and above i suggest you run the game at 1024*768 you will get better frame rates than 640*480

this is a known fact and if you don't believe me download 3dmark03 and run it, in the CPU test the game test runs at 640*480 and in the normal game test it runs at 1024*768 the frame rate running the same thing @ 1024*768 gets higher frame rates

This is a fact and the gamespy reviewer doesn't know this, i don't know why they made that guy write the review.

this is also the reason why some of you trying to run the demo at 640*480 are saying they get low frame rates even @ 640*480. the reason behind that is as stated.

The logic seems absurd to me, but perhaps I am wrong. Comments?

 

[Tim Murray]

Semi-absurd. He's right that yes, at lower reses it is likely that the CPU rather than the GPU will be the bottleneck. However, there's no performance advantage to having the CPU and GPU "work together better" in 1024x768. You'll get the same framerate and the game will look better, but it won't magically run faster.

 

[EasyRaider]

Thanks, that's what I was thinking, too. But since he supported it with so much "evidence", I felt I had to ask.

 

[Vortigern_red]

this is a known fact and if you don't believe me download 3dmark03 and run it, in the CPU test the game test runs at 640*480 and in the normal game test it runs at 1024*768 the frame rate running the same thing @ 1024*768 gets higher frame rates

I think the cpu test does a bit more than just reduce the res to 640x480 LOL.

Try running the game test one and three at 640x480 and I bet you do not get as low a frame rate as the cpu tests. The logic of this statement is really flawed as in one test it is using all your graphics hardware and in the other it is doing a higher proportion of the rendering on the CPU!

 

[Neeyik]

I think the cpu test does a bit more than just reduce the res to 640x480 LOL.

The first CPU test is simply the first game test run at 640 x 480 and with software vertex processing. The second CPU test does the same but stencil shadows (and therefore, the vertex shader used to determine silhouettes) and PS1.4 shaders (all done using PS1.1) are disabled.

 

[nggalai]

According to this review, DX:IW runs very, very slow at 1024x768, and still doesn't run smooth at 640x480. That's with a ATI 9800TX and 2.8GHz P4 (HT). Might be that 512MB of system RAM isn't enough, but still ...

93,
-Sascha.rb

 

[Tokelil]

Ti22 from that thread made my (boring) night!

 

[nobie]

if the 3dmark CPU test actually does perform that way (faster in 1024x768) it's probably just a quirk in the software (maybe something to do with how they count frames?)

 

[Tim Murray]

no, the 1024x768 tests actually use, like, all of the available video hardware rather than using simply the CPU to render everything.

 

[Pete]

you see if you run the game at lower resolution it means the video card will be able to render it too easily causing the CPU to be the culprit slowing everything down.

if the video card can render very easily @ 640*480 therefore the CPU has to try and keep up with the video card in calculating Physics, AI whatever..

but if you set the resolution to 1024*768 it means that the GPU has about the same workload as the CPU so they work together nicely therefore you don't get a bottleneck effect hence you get a higher frame rate.

Sounds good to me!

 

[Quitch]

One day, someone will manage a good X-Box conversion. Until then, we might be adding another to the turkey list.

 

[Dio]

There can be obscure resonance issues that can make this kind of thing happen - particularly, for example, if vsync or even triple buffer is on.

On any modern high-end card - probably even the 9600 Pro - I'd expect the performance difference between 6x4 and 10x7 to be very small so little resonance issues could have a noticeable effect.

 

[Simon F]

There's also a possibility that the 640x480 refresh rate defaults to some low value because that resolution is hardly ever used these days.

 

[Basic]

Or that he has 640x480 refresh rate higher than 1024x768, and vsync forces framerate to half the refresh rate for 640x480 but not 1024x768.

 

[Dave Baumann]

I'm pretty sure that many high end boards are tweaked to operate optimally under high bandwith loads, which sometimes has a slightly detrimental effect at lower bandwidth loads. Usually the difference is fairly small though.

 

[Rolf N]

Taken from a post on Ion Storm DX2 forum.

<...>

The logic seems absurd to me, but perhaps I am wrong. Comments?

I didn't laugh as hard for months. The flat mate was quite worried 'bout me

Thanks for pointing that out.

 

[Pete]

The only obscure resonance issue I perceive is in that poster's head.

 

[Hyp-X]

Isn't the SiS Xabre faster in 1024 than in 640?
Of course it has a quite different reason...

 

[KimB]

One day, someone will manage a good X-Box conversion. Until then, we might be adding another to the turkey list.

Heh, what, like Knights of the Old Republic?

 

[Kristof]

this is a known fact and if you don't believe me download 3dmark03 and run it, in the CPU test the game test runs at 640*480 and in the normal game test it runs at 1024*768 the frame rate running the same thing @ 1024*768 gets higher frame rates

Well DUH its not running the same thing at all ! RTFM :

The 3DMark03CPU test allows users to evaluate CPU performance for 3D graphics workloads. It is important to note that this test is intended to only measure CPU performance for 3D graphics usage and not for general PC usage. For the latter, a benchmark such as PCMarkâ„¢, available also from Futuremark, is more appropriate.

This test runs game test 1 at a resolution of 640x480 using software vertex shaders. Game test 1 is very amenable to CPU measurement as the materials used are simple and most of the screen consists of singletextured, low polygon background objects. We also run game test 3 at a resolution of 640x480 with the use of dynamic shadows disabled. The optimized code path for 1.4 pixel shaders is disabled, since that feature should not be credited in this test. Note that this test still depends somewhat on the graphics hardware, but these settings reduce the dependency.

The CPU test result can also be used as good measure of the software vertex shader speed. Vertex shading is an important part of 3D game performance. Similar calculations as vertex shading are needed in other parts of 3D game software, such as real-time physics.

People should read documents before making silly claims...

SIGH...

K-