• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

PPU more then 100% faster then CPU in UT3 benchmark here.

“Please Pottsey tell us what the hell the PPU is making the CPU do that takes up 2 more corse of cycles AND still is only 30fps?”
Honestly I don’t know. Other members have posted good theories but I am not sure how we can test them. If you go back to around about page 2 or 3 Marc Fraser posted some good ideas. Well it would be good if we had a quad core and tested to see if its CPU limited like dual core.

It could be anything from a very bad map design and nothing to do with the PPU effects. Perhaps the PPU effects are stressing out the GPU to much so it’s the PPU faults. Perhaps it’s the AI dodging and dealing with all the extra map options.

There are more maps coming it will be interesting to see what if they have all the problems the first two maps have.

Either way UT isn’t a good reason to rush out and buy a PPU.
 
“Pottsey, will you addmit that it MIGHT be the PPU not able to cope? As you seem like that isnt the case but as you say, how do we know?“
It might be but everything points to it being a CPU problem. There isn’t one bit of evidence to suggest it’s the PPU not coping. It could be a problem but what makes you think it is?

We need a CPU that’s not holding back the game and if that fix’s the FPS then its not the PPU. If it doesn’t fix the FPS it might be the PPU.
 
“Wait, if it's a CPU problem - why isn't the PPU doing anything about it?“
It is that’s why the FPS go up by 100%. But its not enough even without the physics strain on the CPU the maps are still CPU limited for some reason.

The question is why is it still CPU limited? What is cuaseing the CPU to run at 100%.
 
Pottsey said:
“"PPU more then 100% faster then CPU in UT3" when clearly it's not.”
26.1 v 12.1 clearly is a 100% faster. It’s more then playable if you have a quad core. If it was unplayable lots of people wouldn’t be playing and enjoying it on the online server. I dont see how 30fps+ is unplayable.
Click to expand...
Don't quote me out of context. As I said, just because you get more FPS in dedicated maps, which run like crap either way, doesn't mean you get double FPS in all of them. The fact that your thread title suggests you do is the reason I said that as it suggest you are indeed a fanboy.

Less than 26 FPS is simply NOT playable in a game like UT and is VERY poor considering the highspec machine them benchmarks were taken on. Especially considering the fact that you don't get any gameplay enhancing physics for the extreme drop in performance. Having quad core does NOT make them any more playable. If you think it is then fine go for it, but please stop posting crap on here about it because no one else thinks so. Also, there are not lots of people playing them, there are only a handful of servers with them on.

Pottsey said:
“On normal maps it makes next to no noticeable difference, so what is the point?”
We have yet to see if that’s true. We need a fair benchmark. Aynway I have been saying for a while the PPU is a luxury item not a must have item. It’s nice to have to boost FPS but you don’t need it for UT. Your better off with it, then without it, but you don’t need it.
Click to expand...
Yet to see if that's true? What are you on about? There are lots of people now who have tested it INCLUDING PEOPLE IN THIS THREAD and you seem to ignore their findings. You DON'T need a benchmark to know that you are not getting much of an improvement. Yes there is one, but it is very small. We don't need an exact number. Benchmarks are NOT IMPORTANT as you don't play a benchmark.

As for your last comment, I'd say you are wrong. You are not better off with it. To buy it is to throw money away. It is next to worthless. 10FPS in a game that does over 60FPS is NOT WORTH IT because it MAKES NO DIFFERENCE to your gameplay. You would be better off spending that £80 on another area of your PC. Hell that is half of a 8800GT!

I fail to see how you can constantly defend it.
 
Better thread title would be "watch how £88 makes maps still unplayable".

Personally i see this -

the PPU DOES help but by such a pathetic amount. 100% is a percent. it means nothing when talking like this. going from 2 to 4 FPS is 100% but its still worthless. If the PPU took a stock physics map from 30/35 FPS up to 50/60 FPS then id agree, but it doesnt. your paying for a product that does help, but doesnt help enough to make it work. In the future the PPU will either take off in a big way, in which case theres a lot to do from all parties. coders, manufacturers et al or it will die and be replaced by GPU and CPU calculations. My bet is on the later.

- Pea0n
 
Very impressive thread, some good info and some good debating, on a lot of forums this kind of thing turns into a proper slagging match, well done for keeping it proper!

I am enjoying playing UT3 very much, I'm not a huge gamer and tend to choose games carefully and play them lots.
However, I do have a PhysX card (lol). Before the pelters come, I probably wouldn't spend £88 on one, if it were under £50 then I may be curious enough to get one as I like "quircky" stuff. I seen the BFG card for sale somewhere on clearance for £53.99, they had two left so I bought both! But cunningly I sold one by auction for £99 + postage. So turns out my card cost me £9, well worth trying for that.
I have been turning the feature in UT3 on, but once I did forget and I wondered why I was gettign a bit of stuttering thinking there was a CPU drain somewhere. But turned the physics on and it played fine. I do think it does take my graphics near it's limits though and will be looking to upgrade that soon. So I think the card does benefit me, but so would a better graphics card!

I'm off to sleep in the bath. lol!
 
Pottsey i have to ask you this... do you see a time when all enthusiasts have dedicated ppu's like all have soundcards....? and do you think that the physX physics effects are demonstrably better than havok or similar physics apis?
 
“which run like crap either way, doesn't mean you get double FPS in all of them. The fact that your thread title suggests you do is the reason I said that as it suggest you are indeed a fanboy..”
But its not crap either way, 30fps+ is not crap. I didn’t think it did suggest all of them. It says in the benchmark here its gets 100% more and it does in the benchmark listed. It turns the map from unplayable to playable.




“Yes there is one, but it is very small. We don't need an exact number. Benchmarks are NOT IMPORTANT as you don't play a benchmark.”
Don’t be so silly. I guess people never look at benchmarks to see how much a FPS difference the next grade up GPU or CPU makes. I guess GPU reviews shouldn’t bother posting exact numbers or do the same test for each card. Lets just go by feel and run the map differently each time instead of the same timedemo.

When you have some people saying the PPU made UT go from shuttering to smooth and others saying not. You need a fair benchmark,

Jamous and many others say its make a difference. Clearly we need a benchmark to see who’s right.





“I'd say you are wrong. You are not better off with it. To buy it is to throw money away. It is next to worthless.”
I didn’t mean from a money point of view I meant from a hardware point of view. As I said before it’s a luxury item and if you can afford it then your better off with it then without.





“10FPS in a game that does over 60FPS is NOT WORTH IT because it MAKES NO DIFFERENCE to your gameplay.”
10fps makes the game smoother, increase’s your reaction time and let’s you aim easier. More FPS are always better and more FPS do change gameplay. Some people say its more then 10fps as well. 70 is better then 60. What about when you dip below 60? You might dip down to 25 without the PPU but stay at 35 to 40 with the PPU.





“If a quad core and PPU and 8800 cant handle the extra effects then please tell me what the hell will?”
They can. It’s pretty clear that setup would be more then playable.




“If the PPU took a stock physics map from 30/35 FPS up to 50/60 FPS then id agree, but it doesnt.”
How do you know? Many people say it does make a diffrance. No one has ran benchmarks to confirm it doesn’t or does,.




“do you see a time when all enthusiasts have dedicated ppu's like all have soundcards....?”
Yes but only if a lot of improvements are made. If everyone had a very good PPU it would be good as physics could be pushed forward.




”and do you think that the physX physics effects are demonstrably better than havok or similar physics apis?!”
Yes it’s better but there is still room fro improvements. There are many phsyics effects we have seen with the PhysX API that we havnt from the Havok API.

Saying that not all physics are better some are the same.
 
70fps is by no means better or makes you better than 60fps. Your eye can't even tell the difference between 30fps and above. We only notice it when the 30fps changes rapidly between 30-60 in different situations, thus noticing the "faster" gameplay.

But the difference between 70 and 60 is impossible to notice. You need super responsive vision to even have a chance at noticing it.
 
That’s a myth that’s been proven wrong time and time again and is very easy to prove wrong. The eyes don’t see in FPS they stream data and the eye can see a difference up to and beyond 200fps.

Unless of course you have an LCD/CRT with a low refrash rate. If your refrash is 60hz then your display cannot show more then 60fps.
 
Pottsey said:
That’s a myth that’s been proven wrong time and time again and is very easy to prove wrong. The eyes don’t see in FPS they stream data and the eye can see a difference up to and beyond 200fps.

Unless of course you have an LCD/CRT with a low refrash rate. If your refrash is 60hz then your display cannot show more then 60fps.
Click to expand...

Do you wanna link where the "myth" is proven wrong?

And an LCD with a refresh rate at 60hz can show more than 60fps. If you put it at vsync, then it cant.
 
“And an LCD with a refresh rate at 60hz can show more than 60fps. If you put it at vsync, then it cant.”
No it cannot. It drops the extra fps without vsync. The counter says 100fps but it’s really showing 60fps. That’s why you get tearing. It renders half a frame or less, drops the rest of the frame and renders another half to fit the extra FPS into 60.

60hrz means the screen is updated 60 times a second. If you get 60 updates a second your limited to displaying 60fps.

I can do better then give you a source I can give you a program so you can run your own blind tests.

http://forums.gameon.co.uk/showthread.php?t=9989
Top post download fpscmp02.zip

Download the above program and set half the screen to 30fps, half to 60fps cover up the FPS number then ask someone outside the room to come in and look at the screen and say which is smoothest. Every time I did it 90%+ of the time they choose 60fps which proves the eye can see more then 30fps. Most other people who run the same test get the same resuilts .

If your refresh is 100hz do 60 and 100. No real point unless you refresh goes that high. But still with a blind test most people pick out 100 from 60.

http://amo.net/NT/02-21-01FPS.html more info

EDIT:
Most people who cannot see over 60fps are the ones who have display set to 60hz. Its pretty neat to see 200fps v 100fps but few people have displays that have 200hz.
 
Last edited:
are there any benchmarks to show everyone here that its 'perfectly playable' with a quad core.

and by that i mean a stock quad core, not a quad @ 3.6ghz or anything. if ive got this right. you're saying you need a quad core cpu AND a ppu to get good framerates.
 
agenia have done NO improvements to there ppu physically at all... i mean how many generation of graphics cards have we had? at least 2
 
You must log in or register to reply here.
Back
Top Bottom