[pcper] frame metering review 690 vs. 7970 CF vs. Titan

[omeds]

Oh, sorry. Didn't know you were going to doublecap your fps.



Diminishing returns and placebo, gaming 101.

Double cap? What?

You clearly have no idea wtf you are talking about, as you already proved earlier in the thread.. you might actually learn soemthing if you took your head out of your ass..

 

[Fx1]

It does add smoothness. Due to the cap, the frames are kind of synced in multiples with the refresh rate. My comment about 120hz was just to show that using a cap without vsync at 120hz makes even 60hz with vsync look juddery, which should look smooth due to vsync.

As for hit detection, as I said earlier you set update rates, packets etc to match your frame rates. This results in the best hit rego when there are updates to match every frame, or every second frame depending on the fps cap you use (125 or 250).

Like I said its not new, its common knowledge, I'm guessing you're a bit of a n00b with online gaming if you havent heard of it before, but thats fine I'm not knocking you, its just your inexperience shows.

This is not true.

For a start without V Sync your monitor and game frames are running totally independent of each other and are never synced regardless of multiples. This is why screen tears no matter what the multiple.

Also i have been online gaming since the days of Subspace and Infantry in the 90's long before FPS was anything more than Duke Nukem 3D.

What i dont do is spend all my time on BF3 forums and CSS sites which is where you have picked up this placebo.

When the human response time is 215ms your not going to get much benefit out of 16ms +/-

 

[omeds]

Ok you and your little side kick are starting to annoy me. There are competitive gamers all over the world who have been using this method for over 15 years... and then you come along, captain n00b who has never even heard of it telling me its a placebo, lol. Its common knowledge ffs, its not like I have told you some unheard of bit of information.

Believe what you like, I'm guessing you dont like the truth because it conflicts with your agenda in this topic of CFX frame time issues, and you know that CFX presents a problem in this scenario so you simply discount it instead of admitting to the issue.

 

[Fx1]

Ok you and your little side kick are starting to annoy me. There are competitive gamers all over the world who have been using this method for over 15 years... and then you come along, captain n00b who has never even heard of it telling me its a placebo, lol. Its common knowledge ffs, its not like I have told you some unheard of bit of information.

Believe what you like, I'm guessing you dont like the truth because it conflicts with your agenda in this topic of CFX frame time issues, and you know that CFX presents a problem in this scenario so you simply discount it instead of admitting to the issue.

There just isnt any science to back up your points.

The only fact is that 120hz monitors can lower your input response times in to the low 20's ms.

It takes your finger 200-250ms to push the mouse button even if you dont have to move the cross hair.

How many times do you think the game engine has updated by the time your finger hits the button?

Even with Vsync the game engine updates every 8ms on 120hz or 16ms on 60hz. Do you honestly think your going to get caught short just because the game engine has to wait possibly 15ms until it can register your click after it took you 250ms to push the button?

 

[omeds]

Its no secret higher frame rates make games feel more responsive. That wasnt even the point of what I was telling you anyway, dont move the goal posts... What I was telling you is about fluidity and minimising tearing when not using vsync.

 

[Fx1]

Its no secret higher frame rates make games feel more responsive. That wasnt even the point of what I was telling you anyway, dont move the goal posts... What I was telling you is about fluidity and minimising tearing when not using vsync.

Yes and you might get benefit up to 120FPS on a 120hz monitor.

250 FPS cap wont do anything.

Screen tearing will also be bad but you might not notice it as much as the lines in the frame(the tear) will be moving twice as fast.

Vsync on 120fps wont make anything feel less responsive and it fixes all stuttering and tearing.

 

[BallaTheFeared]

lol anyways

 

[omeds]

Yes and you might get benefit up to 120FPS on a 120hz monitor.

250 FPS cap wont do anything.

Screen tearing will also be bad but you might not notice it as much as the lines in the frame(the tear) will be moving twice as fast.

Vsync on 120fps wont make anything feel less responsive and it fixes all stuttering and tearing.

Well you're wrong. There's nothing left to say. 125 and 250 fps caps do have benefits, even on 60hz displays. People have been using these caps for ~15 years for good reason.

Vsync at 120hz does feel less responsive. Sure its better than 60hz but its still not as responsive and "free" as no vsync.

I can see this is getting no where because you are defending CFX, knowing full well this stutter issue presents a problem. If this was a different thread, you might listen.

 

[Fx1]

Well you're wrong. There's nothing left to say. 125 and 250 fps caps do have benefits, even on 60hz displays. People have been using these caps for ~15 years for good reason.

Vsync at 120hz does feel less responsive. Sure its better than 60hz but its still not as responsive and "free" as no vsync.

I can see this is getting no where because you are defending CFX, knowing full well this stutter issue presents a problem. If this was a different thread, you might listen.

No i wont listen because your 125 and 250 frame caps are made up bull shit that have no basis or benefit other than limiting how hard your system is working.

Are you seriously saying that Quake 2 benefitted from a 250 FPS cap? How did 1998 hardware get you to 250fps?

Explain to me how capping your frames at 250 help? Give me the details because i am at a loss how this helps...

 

[omeds]

Lol ffs, you're a total n00b. Yes seriously Quake benefited from a 125fps cap, and then on to 250. Apart from more fluidity, less tearing, and more responsiveness it allowed you to jump higher, move faster, faster rof, better hit rego and other benefits. 125 and 250 still do all those things in the Cod series which is based on the Quake 3 engine. 333 allows you to run silently on some surfaces too.

 

[Fx1]

Lol ffs, you're a total n00b. Yes seriously Quake benefited from a 125fps cap, and then on to 250. Apart from more fluidity, less tearing, and more responsiveness it allowed you to jump higher, move faster, faster rof, better hit rego and other benefits. 125 and 250 still do all those things in the Cod series which is based on the Quake 3 engine. 333 allows you to run silently on some surfaces too.

Except those BUGS were fixed when they capped the fps to 92. These bugs also don't transfer to css or bf3.

 

[omeds]

Yes but the fluidity, reduction in tearing, and responsiveness transfer to any game.

You asked what benefits it provided for quake, and those were the benefits, aswell as any game made on the Quake engine for a decade, including the CoD series.

 

[Fx1]

Yes but the fluidity, reduction in tearing, and responsiveness transfer to any game.

You asked what benefits it provided for quake, and those were the benefits, aswell as any game made on the Quake engine for a decade, including the CoD series.

They dont reduce tearing. Only V Sync can do that.

The rest is placebo. As i have said your reaction times are 250ms give or take so 8ms wont affect anything. In Fact its well within the margin for error of the human response time.

Vsync fixes all stuttering, frame time variances which show speeding up and slowing down of game play and all tearing for a near perfect game play experience. This is FACT not placebo. You can get all these benefits from just 60 FPS. If you need sub 40ms response times then Vsync with 120hz is the answer.

 

[Keysplayr]

They dont reduce tearing. Only V Sync can do that.

The rest is placebo. As i have said your reaction times are 250ms give or take so 8ms wont affect anything. In Fact its well within the margin for error of the human response time.

Vsync fixes all stuttering, frame time variances which show speeding up and slowing down of game play and all tearing for a near perfect game play experience. This is FACT not placebo. You can get all these benefits from just 60 FPS. If you need sub 40ms response times then Vsync with 120hz is the answer.

So in other words, AMD cards in CF need to be run at 60fps locked by Vsync in order to reduce their frametime and latency issues. This also means that if this needs to be done, chances are that Crossfired 7970 GHz Edition cards aren't any faster than SLI'd GTX650Ti Boost (or the like) that doesn't need Vsync enabled for smooth gameplay. Or even a single GTX 660.
Vsync is not an answer when it comes to 7970 being touted as the faster cards. They aren't if they are locked at 60fps in order to have smooth gameplay.
This goes for any level of Crossfired cards.

 

[Silverforce11]

So in other words, AMD cards in CF need to be run at 60fps locked by Vsync in order to fix their frametime and latency issues. This also means that if this needs to be done, chances are that Crossfired 7970 GHz Edition cards aren't any faster than SLI'd GTX650Ti Boost (or the like) that doesn't need Vsync enabled for smooth gameplay. Or even a single GTX 660.
Vsync is not an answer when it comes to 7970 being touted as the faster cards. They aren't if they are locked at 60fps in order to have smooth gameplay.
This goes for any level of Crossfired cards.

If your monitor is higher than 60 hz, you can use radeon pro to set the frame rate to what you want, it will completely smooth out frame intervals.

If your monitor is 60 hz, then anything higher is irrelevant and a waste of resources to have it run at 90+ fps without vsync and have massive screen tearing. IF you are those who are sensitive to input lag with vsync on, then radeon pro can set the desired frame rate with smooth intervals to match your monitor without vsync.

It's an awesome tool and these features need to be officially supported in CCC imo.

 

[BallaTheFeared]

Based on the Tom's article that took a 7990 down to 40 fps right?

 

[Keysplayr]

If your monitor is higher than 60 hz, you can use radeon pro to set the frame rate to what you want, it will completely smooth out frame intervals.

If your monitor is 60 hz, then anything higher is irrelevant and a waste of resources to have it run at 90+ fps without vsync and have massive screen tearing. IF you are those who are sensitive to input lag with vsync on, then radeon pro can set the desired frame rate with smooth intervals to match your monitor without vsync.

It's an awesome tool and these features need to be officially supported in CCC imo.

Anything higher than 60fps is irrelevant? So all the benchmarks from all the years past never meant anything in terms of which cards were truly faster?
I NEVER use Vsync and rarely do I see any tearing. Maybe it's because I don't use and AMD card. In fact, I can't remember any instances where I see any tearing.

Point. If you need to use Vsync to get a smooth gameplay experience, then you might as well go with much cheaper, less powerful GPUs. Right? I mean, why should we spend 800+ dollars to play at a capped 60fps. Might as well stick to single AMD GPU or low end Xfire.

I think it is very interesting and resourceful, this avenue you have taken to downplay this particular problem AMD is currently having. But all you're offering is a band-aid that completely neuters the GPU's power reducing ones cards to much lower levels.
I can't see settling for that. Radeon Pro is a third party utility that , as you put it, is to be used by people who are not lazy.

Ridiculous.

 

[omeds]

They dont reduce tearing. Only V Sync can do that.

The rest is placebo. As i have said your reaction times are 250ms give or take so 8ms wont affect anything. In Fact its well within the margin for error of the human response time.

Vsync fixes all stuttering, frame time variances which show speeding up and slowing down of game play and all tearing for a near perfect game play experience. This is FACT not placebo. You can get all these benefits from just 60 FPS. If you need sub 40ms response times then Vsync with 120hz is the answer.

No it does reduce the appearance of tearing, as the frames are kind of synced with the harmonic frequency of the refresh rate at those caps.

Tearing Interaction between fps and Hz (Harmonic Frequencies)
another consideration, if you got a powerful GPU that runs capped out at a game's framerate limit (e.g. source engine games with a configurable fps_max) .... then if you play VSYNC OFF, you _really_ do not want a frame cap (fps_max) that has harmonic frequencies between fps and Hz. For example, an fps_max of 60, 120 or 180 for a 60Hz display, especially if you have a graphics card (e.g. Titan). The reason is you will have nearly-stationary or slowly-moving tearlines, as the splice of the previous frame cuts into the next frame at fairly synchronized intervals. For example, I can see two very clear (nearly-stationary) tearlines during fast turns in an older Source Engine game (without AA) when I configure it to a fps_max 240 on my 120 Hz display, because my GTX 680 can easily run capped-out at 240 fps. And likewise, you will have more visible tearing on a 60 Hz display if you cap at 59/60/61 (one persistent tearline) or at 119/120/121 (two persistent tearlines) or at 179/180/181 (three persistent tearlines) assuming your GPU is powerful enough to always run fully capped out at the fps_max. Instead you prefer tearing to faintly & randomly go all over the place rather than being obnoxiously stationary or in the middle of your screen. Uncapping is better (e.g. fps_max 999), or setting an odd number as high microstutter-free value as you are able to (e.g. fps_max 317), can significantly reduce the appearance of tearing.

The only things he fails to mention is that the best numbers to use are 125, 250 and so on, instead only citing 317.


Secondly, vsync does not automatically fix stuttering in MGPU, I know this from first hand experience and I actually asked on another forum why that is. This is the response I was given from someone who deals with AMD on a adaily basis, assists or attends their press events, and also happens to review for the ATi fan site, Rage3 D

timing.


vysnc doesn't time the start of the frame render, it adjusts the timing of the display of finished render.

so when have an offset of say 5ms between GPU1 starting a frame and GPU2 starting a frame, and displaying every 16.6ms, you've got GPU 2 rendering a frame based on actions that happened before you saw frame 1. By the time you see and react to frame 2, it's 33ms since frame 1 began rendering.

This is why its advised to run a frame cap in conjunction with vsync in troublesome games.


Thirdly, vsync at 120hz still produces noticable input lag. Yes its much better than 60hz but its still there and noticable.


So.. now that you have learnt some facts, I think its time to move on.

 

[omeds]

Thought I would add this:

Do insane framerates really look better?
Yes, if frame render times are consistent. That's if you prefer VSYNC OFF -- then believe it or not, 500fps (consistent frame rendertimes) at 60 Hz can looks much better than ~60fps at 60 Hz. You get many MUCH-smaller splices of many different (fresher) frames into one refresh. The tearline offsets become very tiny as a result. (top part of refresh being almost 16.7ms ago, bottom part of refresh being nearly 0ms ago). Here, this is a situation where a 1000 Hz mouse makes a quite noticeable fluidity difference; since the more accurate mouse position updates result in more consistently and smaller offset splices during insane-high framerate VSYNC-OFF gaming.

I am someone who is sensitive to tearing even if fps > Hz (even at 300fps @ 120 Hz). This is what I've discovered that greatly reduces tearing, at least in games with consistent frame render times (e.g. Source engine games).


If you dont believe him, take it up with him, but he is simply explaining what has been known in online gaming communities for ~15 years.

http://forums.anandtech.com/showpost.php?p=34693183&postcount=3

 

[Fx1]

Thought I would add this:

Do insane framerates really look better?
Yes, if frame render times are consistent. That's if you prefer VSYNC OFF -- then believe it or not, 500fps (consistent frame rendertimes) at 60 Hz can looks much better than ~60fps at 60 Hz. You get many MUCH-smaller splices of many different (fresher) frames into one refresh. The tearline offsets become very tiny as a result. (top part of refresh being almost 16.7ms ago, bottom part of refresh being nearly 0ms ago). Here, this is a situation where a 1000 Hz mouse makes a quite noticeable fluidity difference; since the more accurate mouse position updates result in more consistently and smaller offset splices during insane-high framerate VSYNC-OFF gaming.

I am someone who is sensitive to tearing even if fps > Hz (even at 300fps @ 120 Hz). This is what I've discovered that greatly reduces tearing, at least in games with consistent frame render times (e.g. Source engine games).


If you dont believe him, take it up with him, but he is simply explaining what has been known in online gaming communities for ~15 years.

http://forums.anandtech.com/showpost.php?p=34693183&postcount=3

Name one game you can run at 500 fps that was released in the last 2 years.

Fact is that most modern games cant be ran at more than 60-80 FPS on Ultra with 1 GPU. So this point is moot.

It doesnt matter what size the tears are you can still see the horizontal line moving down the screen. BF3 explosions are the worst.

i just set v-sync on Ultra and capped my FPS to 59 and i have no tearing and no input lag. I have been testing this all morning and i can tell you now that the input lag you experience is NOT related to Vsync. Its related to the game locking the FPS at 60 frames which is NOT the correct value. The refresh rate of a monitor is 59.94hz which seems to cause lag when locking to 60fps.

 

[Fx1]

So in other words, AMD cards in CF need to be run at 60fps locked by Vsync in order to reduce their frametime and latency issues. This also means that if this needs to be done, chances are that Crossfired 7970 GHz Edition cards aren't any faster than SLI'd GTX650Ti Boost (or the like) that doesn't need Vsync enabled for smooth gameplay. Or even a single GTX 660.
Vsync is not an answer when it comes to 7970 being touted as the faster cards. They aren't if they are locked at 60fps in order to have smooth gameplay.
This goes for any level of Crossfired cards.

Dumb post.

Of course they are the faster card. They can play on higher settings and still hit 60 FPS. There IS NO BENEFIT to running a GPU at 80 FPS when your monitor can only display 60 FPS. Screen tearing is obvious to all but the blind.

 

[omeds]

Name one game you can run at 500 fps that was released in the last 2 years.

Fact is that most modern games cant be ran at more than 60-80 FPS on Ultra with 1 GPU. So this point is moot.

It doesnt matter what size the tears are you can still see the horizontal line moving down the screen. BF3 explosions are the worst.

i just set v-sync on Ultra and capped my FPS to 59 and i have no tearing and no input lag. I have been testing this all morning and i can tell you now that the input lag you experience is NOT related to Vsync. Its related to the game locking the FPS at 60 frames which is NOT the correct value. The refresh rate of a monitor is 59.94hz which seems to cause lag when locking to 60fps.

You dont need to run 500 fps, it was just his example. 125 and 250 do the trick too.
Who said anything about a single GPU? We are talking about MGPU and microstutter issues here, so the point is not moot.
59 fps still has input lag, but yes its greatly reduced, thats why I use it in a lot of single player games as I mentioned several times already, plus it has the benefit of helping with microstutter that can appear even when vsync is used. Its still not an option for me and many others in online fps.
Yes it does matter what size the tears are, as high frame rates reduce it to the point the appearence of tearing is greatly reduced.

Face it, you're wrong, and are just now arguing for the sake of it.

 

[Keysplayr]

Dumb post.

Of course they are the faster card. They can play on higher settings and still hit 60 FPS. There IS NO BENEFIT to running a GPU at 80 FPS when your monitor can only display 60 FPS. Screen tearing is obvious to all but the blind.

So I am right in saying you and Silverforce share the same opinion.

 

[Fx1]

You dont need to run 500 fps, it was just his example. 125 and 250 do the trick too.
Who said anything about a single GPU? We are talking about MGPU and microstutter issues here, so the point is not moot.
59 fps still has input lag, but yes its greatly reduced, thats why I use it in a lot of single player games as I mentioned several times already, plus it has the benefit of helping with microstutter that can appear even when vsync is used. Its still not an option for me and many others in online fps.
Yes it does matter what size the tears are, as high frame rates reduce it to the point the appearence of tearing is greatly reduced.

Face it, you're wrong, and are just now arguing for the sake of it.

What game can you run 250 FPS? COD4? LOL who plays that any more. CS:GO? most boring game ever.

BF3 will never hit 250fps because its CPU bound. Far Cry 3 wont hit 120 FPS without turning off MSAA. Crysis 3.... Lol good luck hitting 60fps consistant.

Are you playing at 1280x768 just to get frame rates into the hundreds? On what games

1200p screens will never hit 120fps. 1440p wont either and 1600p LOL good luck with that.

So the magic number is 60 FPS. This is the target FPS for every game that comes out and every screen from 1080p to 1600p to eyefinity. This is the FPS that the dev's design their games around. Even 120hz is designed for 3D and not 120 FPS gaming.

So the aim of the game is Stutterless and tear free gaming thats smooth and enjoyable. 59 FPS V-Sync offers all of this.

You CANNOT see input lag on the above setup. I have tested for the last 4 hours.

 

[Fx1]

So I am right in saying you and Silverforce share the same opinion.

What opinion?