[Tom's Hardware] CPU bottlenecking/frame latency benchmarks

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

SPBHM

Diamond Member
Sep 12, 2012
5,056
409
126
Exactly that is the problem. Who would play at those settings with 30fps min and 44fps avg? That is way too low for a 1st person shooter, at least for me. And I think I'm not alone with this opinion. The question must be allowed if these settings make sense for the ambitious gamer. In my opinion they do not.



How about at least 50fps sustained? Better 60fps, at least most of the time.

And no, your second statement is completely wrong. It's ultimately about fps, not GPU usage (or CPU usage for that matter).

I have to agree, investing the money needed for the hardware, to play a first person shooter at 30-40 FPS doesn't sound good, further reducing IQ for a higher framerate makes a lot of sense.
 

AtenRa

Lifer
Feb 2, 2009
14,000
3,357
136
I have to agree, investing the money needed for the hardware, to play a first person shooter at 30-40 FPS doesn't sound good, further reducing IQ for a higher framerate makes a lot of sense.

Only for Online Multi-player FPS(First Person Shooters) gaming, in single player 30fps minimum if fine.
 

boxleitnerb

Platinum Member
Nov 1, 2011
2,601
2
81
Only for Online Multi-player FPS(First Person Shooters) gaming, in single player 30fps minimum if fine.

You can only speak for yourself. Your standards seem quite low compared to others. That is okay, but don't force them onto other people.

For example for me, 60fps in SP is desirable, for MP 100+fps. Think input lag. And if people use vsync, anything below 60fps will reduce fps to 30 (if they don't use adaptive vsync, that is).
 
Last edited:

AtenRa

Lifer
Feb 2, 2009
14,000
3,357
136
You can only speak for yourself. Your standards seem quite low compared to others. That is okay, but don't force them onto other people.

For me, 60fps in SP is desirable, for MP 100+fps. Think input lag.

With 60Hz monitors you dont get anything having 100+ fps. If you have 120Hz monitor then 120fps will be your optimal goal.

For online multi-player games like BF3, all you need to have is 60fps constant (60fps cap for 60Hz monitors).

Playing FC3 in single player with 30fps minimum and 45fps avg is perfectly playable.
 

Gikaseixas

Platinum Member
Jul 1, 2004
2,836
218
106
45 fps is somewhat playable but i prefere +60 fps
Anyway it's kind of interesting to see benchmarking methodology changing lately. We are witnessing the evolution of benchmarking.
 

boxleitnerb

Platinum Member
Nov 1, 2011
2,601
2
81
With 60Hz monitors you dont get anything having 100+ fps. If you have 120Hz monitor then 120fps will be your optimal goal.

For online multi-player games like BF3, all you need to have is 60fps constant (60fps cap for 60Hz monitors).

Playing FC3 in single player with 30fps minimum and 45fps avg is perfectly playable.

That is not correct. There is a distinct difference between 60fps and 120fps even on 60Hz monitors. As I said: input lag. With 120fps, controls feel much more responsive and direct, because the time between what you do with your mouse/keyboard and what is shown on the screen is different. The game enige itself updates more quickly. It is dependent on the game engine though. UE3 is a good example where the difference is much better felt than with other engines.

It is playable for you, but again: Other people, other standards. I could not and would not want to play this way. So you should not present your benchmarks and assume they are valid for everyone.
 

Abwx

Lifer
Apr 2, 2011
10,847
3,297
136
With 60Hz monitors you dont get anything having 100+ fps. If you have 120Hz monitor then 120fps will be your optimal goal.

For online multi-player games like BF3, all you need to have is 60fps constant (60fps cap for 60Hz monitors).

Playing FC3 in single player with 30fps minimum and 45fps avg is perfectly playable.


That s not the point..

Whatever shows AMD under a good light is to be dismissed
and downplayed as not significant or representative of real
world conditions by the same people who find settings to be
realistic when top of the notch gfx are used at low resolutions.

I see a poster above complaining that THG created an "artificial"
GPU bounded bench to display their findings yet the same people
never complained that there is artificialy settled CPU bound
tests using ridiculously low resolutions with high end Gpus
as it goes in the direction of their biaising....
 
Last edited:

AtenRa

Lifer
Feb 2, 2009
14,000
3,357
136
That is not correct. There is a distinct difference between 60fps and 120fps even on 60Hz monitors. As I said: input lag. With 120fps, controls feel much more responsive and direct, because the time between what you do with your mouse/keyboard and what is shown on the screen is different. The game enige itself updates more quickly. It is dependent on the game engine though. UE3 is a good example where the difference is much better felt than with other engines.

Having 120fps with 60Hz monitor will not reduce input lag. The monitor will still only show half the frames. You need 120Hz monitor and 120fps to reduce input lag.

In BF3 MT, having more than 60fps in a 60Hz monitor will get you worst gameplay. The secret is to cap at 60fps and not having less than 60fps minimum. Enable V-sync will also increase input lag.

The majority of UE3 games can produce 60fps with lower hardware.

So you should not present your benchmarks and assume they are valid for everyone.


I presented my benchmarks with my settings, just because you dont like them doesnt invalidate them. I guess that you invalidate the majority of the reviews that they have less than 100fps then.

And like it or not FC3 at 45fps avg is perfectly playable for the majority of gamers.
 

AtenRa

Lifer
Feb 2, 2009
14,000
3,357
136
That s not the point..

Whatever shows AMD under a good light is to be dismissed
and downplayed as not significant or representative of real
world conditions by the same people who find settings to be
realistic when top of the notch gfx are used at low resolutions.

I see a poster above complaining that THG created an "artificial"
GPU bounded bench to display their findings yet the same people
never complained that there is artificialy settled CPU bound
tests using ridiculously low resolutions with high end Gpus
as it goes in the direction of their biaising....

yeap, they buy $600.00 GPUs to play at 1024x768 and low video settings.:p
 
Aug 11, 2008
10,451
642
126
That s not the point..

Whatever shows AMD under a good light is to be dismissed
and downplayed as not significant or representative of real
world conditions by the same people who find settings to be
realistic when top of the notch gfx are used at low resolutions.

I see a poster above complaining that THG created an "artificial"
GPU bounded bench to display their findings yet the same people
never complained that there is artificialy settled CPU bound
tests using ridiculously low resolutions with high end Gpus
as it goes in the direction of their biaising....

So many posts get dismissed because some posters repeatedly post gpu bound benchmarks to supposedly evaluate CPU performance.
 

boxleitnerb

Platinum Member
Nov 1, 2011
2,601
2
81
Having 120fps with 60Hz monitor will not reduce input lag. The monitor will still only show half the frames. You need 120Hz monitor and 120fps to reduce input lag.

In BF3 MT, having more than 60fps in a 60Hz monitor will get you worst gameplay. The secret is to cap at 60fps and not having less than 60fps minimum. Enable V-sync will also increase input lag.

The majority of UE3 games can produce 60fps with lower hardware.

No you are wrong, sorry mate. The difference can easily be felt if you are sensitive enough. Just because you aren't doesn't mean others feel the same. I can clearly feel a difference - are you saying I'm making this up?

I've not experienced this "worst gameplay" you are speaking of. Must be a subjective thing.

Of course vsync will increase input lag. I never contested this. I noted that if you fall below 60fps and the game doesn't have triple buffering, you will get nasty fps drops to 30 which is bad. Same goes for 120Hz displays and vsync.

I presented my benchmarks with my settings, just because you dont like them doesnt invalidate them. I guess that you invalidate the majority of the reviews that they have less than 100fps then.

I don't think your benchmarks are invalid. I think you should be more tolerant of other point of views and stop trying to speak for other people. You often say things like "we don't need x fps" or "game xyz is playable at y fps". This applies to you, but not necessarily to everyone else.

As for reviews:
I want to know the full potential of my hardware void of any bottlenecks. The reviewer cannot and should not decide how many fps is "enough".

And like it or not FC3 at 45fps avg is perfectly playable for the majority of gamers.

That's the problem right there, this kind of arrogance. You don't even KNOW the majority of gamers or their opinion on fps. You cannot speak for them.
 

AtenRa

Lifer
Feb 2, 2009
14,000
3,357
136
So many posts get dismissed because some posters repeatedly post gpu bound benchmarks to supposedly evaluate CPU performance.

Wrong, we evaluate Gaming performance.

But what I can tell you is that while I understand the point of low res benchmarks when testing out a CPU, I do wish there were more whole platform tests along with the low res tests when it comes to gaming (testing the games as I would play them may not isolate the CPU, but it would provide numbers to compare based on how I would use two platforms... afterall, none of us game on just a GPU and CPU). Maybe having the PCIE lanes on the northbridge has some effect that helps here?
 

boxleitnerb

Platinum Member
Nov 1, 2011
2,601
2
81
The problem with evaluating "gaming performance" is, that people have different standards. There is not one "realistic" setting that applies to everyone. To assume so, is arrogant and intolerant.


Where is your argument?
 

Abwx

Lifer
Apr 2, 2011
10,847
3,297
136
The problem with evaluating "gaming performance" is, that people have different standards. There is not one "realistic" setting that applies to everyone. To assume so, is arrogant and intolerant.



Where is your argument?

Knowing the full potential of a tested system is to spot all
bottlenecks , isnt it ?.....
 

Abwx

Lifer
Apr 2, 2011
10,847
3,297
136
So many posts get dismissed because some posters repeatedly post gpu bound benchmarks to supposedly evaluate CPU performance.

The point of TR/THG tests is to have a measure of an once
obscured parameter and in this respect they did their job
whatever can be said about their methodology.

It would be more interesting to know the causes of the measured
differences than questionning the protocol because the results
are not favourable to one s prefered manufacturer.
 

boxleitnerb

Platinum Member
Nov 1, 2011
2,601
2
81
Knowing the full potential of a tested system is to spot all
bottlenecks , isnt it ?.....

Well, a review is about either CPU or GPU. So the bottleneck should be eliminated. How fast the other component is (GPU in a CPU review) is the topic of a separate review. It is impossible to test both in one review with one setting and obtain complete results for all demanding cases (CPU and GPU bound).
 

SPBHM

Diamond Member
Sep 12, 2012
5,056
409
126
yeap, they buy $600.00 GPUs to play at 1024x768 and low video settings.:p

well, if you like 60fps you can make sound ridiculous using a $200 CPU + $300 GPU, overclocked to play a FPS at 30-45...

the difference is quite obvious on FC3, just go inside your safe house and move the mouse around, after that go outside where you get 30-40, if you can't notice the decrease in smoothness I don't know what to say,

sure 45 AVG is playable, consoles have a lock at 30 and can go under 25 (the slow controls help to hide how bad it is), but it's not ideal for high end hardware in a first person shooter, so I think there is some validity to wanting to lower IQ for a better experience, also since you can't cover the entire game on a basic comparison, knowing how CPUs perform relative to each other on a less GPU bottlenecked condition can be useful I think...

as for over 60FPS, in some games is easy to notice, frames are being rendered faster, some game engines act differently, but you can get an advantage, more updated information, and the display can show parts of different frames at the same time, although of course, is less of a problem compared to under 60FPS performance.
 

Arzachel

Senior member
Apr 7, 2011
903
76
91
That is not correct. There is a distinct difference between 60fps and 120fps even on 60Hz monitors. As I said: input lag. With 120fps, controls feel much more responsive and direct, because the time between what you do with your mouse/keyboard and what is shown on the screen is different. The game enige itself updates more quickly. It is dependent on the game engine though. UE3 is a good example where the difference is much better felt than with other engines.

It is playable for you, but again: Other people, other standards. I could not and would not want to play this way. So you should not present your benchmarks and assume they are valid for everyone.

Unless there is something I'm missing, fps makes zero impact on input latency, because input polling should be decoupled from rendering for even the most basic 3d engines. Furthermore, 60 fps minimum, 1000 fps average should have zero impact on output latency compared to 60 fps minimum, 60 fps average on a 60hz monitor, because it simply can't show any more, everything over that gets discarded. Mind explaining what you mean?
 

Arzachel

Senior member
Apr 7, 2011
903
76
91
as for over 60FPS, in some games is easy to notice, frames are being rendered faster, some game engines act differently, but you can get an advantage, more updated information, and the display can show parts of different frames at the same time, although of course, is less of a problem compared to under 60FPS performance.

Again, how? A properly designed engine has input polling, game logic and rendering separated. What do you mean with "and the display can show parts of different frames at the same time"? This is really starting to sound like placebo to me.

EDIT:Whoops, sorry for the double post
 

AtenRa

Lifer
Feb 2, 2009
14,000
3,357
136
For years, gamers and reviews considered that 30fps should be the minimum for games. Now all of the sudden we need more than 60fps even if we have 60Hz monitors. :rolleyes:

For those that will like to read about input lag

http://www.anandtech.com/show/2803
 

AtenRa

Lifer
Feb 2, 2009
14,000
3,357
136
the difference is quite obvious on FC3, just go inside your safe house and move the mouse around, after that go outside where you get 30-40, if you can't notice the decrease in smoothness I don't know what to say,

I said that 30fps minimums are fine and that the game is perfectly playable with 45fps avg. I have never said that 60fps is not better or that you will not feel the difference between 45 and 60fps.
 

Abwx

Lifer
Apr 2, 2011
10,847
3,297
136
The most striking is that human eye has retina persistence
above 20 FPS , so it s surprising that some people by there
have a vision as fast as CCD devices....