NO Release Day Reviewers have addressed 4870X2 Microstutter

1

1erCru

Limp Gawd
Joined
Sep 9, 2006
Messages
217
Seems kind of important. Yes it is a powerful card, but there are a numbers of games that can force the X2 into micro stutter territory, which is from what I understand sub 50fps.
 
This German review addresses it:
http://www.computerbase.de/artikel/hardware/grafikkarten/2008/test_ati_radeon_hd_4870_x2/24

Translated:
http://translate.google.com/transla...adeon_hd_4870_x2/24&hl=nl&ie=UTF8&sl=de&tl=en

The bottom line is that it's still there.

106.png

107.png

108.png
 
It is about 1/2 of what NV does today.... Still pretty damn good.
 
This review of the x2 talks about the stuttering issue.

Here's the specific quote:
Now, a word on something else that is rarely mentioned in reviews: microstuttering. It has long been the bane of many multi-GPU users, and has been one of the many barriers for myself. Does microstuttering exist on the X2? Unfortunately, yes. Again, the lack of a unified frame buffer is no doubt the cause for this. However, there is good news. On the X2, it really isn't that bad. It happens typically within the first minute of loading a map or save game, and then it is gone for good. So, while not perfect, it is an improvement.
Click to expand...
 
Ah, I missed that review. Well an improvement is better than no improvement. Would be nice if more high profile sites would address micro stutter in depth.
 
1erCru said:
Ah, I missed that review. Well an improvement is better than no improvement. Would be nice if more high profile sites would address micro stutter in depth.
Click to expand...

If microstutter was an issue they would have mentioned it.
 
I can tell you that I've noticed microstutter on every multi GPU setup that I've ever had but my dual 4870's are by far the best. The 9800GX2 was bad, 3870x2 was horrendous.
 
It appears the "micro-stuttering" campaign didn't gain the traction NV marketing was hoping for. lol I'm sure they saw the writing on the wall, and realized ATI's approach was the future. Scrambling, they reached into their bag of tricks and found nothing. Enter marketing... Spin ATI's much more efficient and cost effective strategy as inferior due to this phantom stuttering effect, and poof, the sheep follow. Problem is, NV has sheered the sheep one too many times.
 
Nuker said:
Another review shows absolutely no MS on the 4870x2, it would be interesting to know how all of these review sites came up with these numbers...
Click to expand...

depends on how much schwag they wanted for a favorable review
 
Nuker said:
Another review shows absolutely no MS on the 4870x2, it would be interesting to know how all of these review sites came up with these numbers...
Click to expand...

Link please, I've just seen two sites that have even mentioned it, and both said they were both there to an extent.

As for ATI's approach there is nothing revolutionary, sideport is not working, all that has changed vs the 3870X2 is the new bridge chip that has somewhat minimized the stuttering.


 
flippin_waffles said:
It appears the "micro-stuttering" campaign didn't gain the traction NV marketing was hoping for. lol I'm sure they saw the writing on the wall, and realized ATI's approach was the future. Scrambling, they reached into their bag of tricks and found nothing. Enter marketing... Spin ATI's much more efficient and cost effective strategy as inferior due to this phantom stuttering effect, and poof, the sheep follow. Problem is, NV has sheered the sheep one too many times.
Click to expand...
With their SLI boards, I'm sure they want you to think multi-chip/card="micro-stuttering" :rolleyes:
 
AMD stated it might be enabled later if the bandwidth was needed.
 
don't fire up your new 4870x2 and sit there waiting to see micro-stutter. fire it up and assume it's gone. if you concentrate too hard on it you'll trick yourself into seeing it....
 
Nuker said:
http://www.xtremesystems.org/forums/showthread.php?t=194808
Click to expand...


That looks quite promising. Either one of the two sites has a bad way of measuring microstutter (as its pretty well defined, its the variation in rendering time from one frame to the next), OR, microstutter doesn't occur in GRID like it does in those other games, OR, microstutter is system dependant. Take your pick.

don't fire up your new 4870x2 and sit there waiting to see micro-stutter. fire it up and assume it's gone. if you concentrate too hard on it you'll trick yourself into seeing it....
Click to expand...

To me that's just silly :p either microstutter is there or it isn't. If the game is running above 60fps, you wont see it. If you do see it above 60fps, its not microstutter, its some other form of stuttering.

Microstutter is something that means the displayed framerate is actually higher than what you will see with your own eye, as a frameRATE is taken as an average over a given time. So if you see a dual GPU card getting 30fps and a single GPU card getting 25fps, if the dual GPU card is microstuttering by +-5fps, even though your frameRATE is higher for the dual GPU, they're actually performing much the same as each other. Now when I say microstuttering +-5fps, that doesn't mean min = 25fps, av = 30fps and max = 35fps. You could have min = 28fps, av = 30fps, max = 32fps, but still be getting a +-5fps microstutter which means you're effectively seeing min = 23fps, av = 25fps, max = 27fps. Make sense? Probably not :p
 
The human brain perceives the outside world subjectively not objectively. So unless you aren't playing the game and using tools to take measurements instead, you would never know if it's just an area that hit the card hard, or microstutter.

Those of us at XS & B3D have already given up on microstutter being an issue for the 4870X2 weeks ago.
 
Tudz said:
Microstutter is something that means the displayed framerate is actually higher than what you will see with your own eye, as a frameRATE is taken as an average over a given time. So if you see a dual GPU card getting 30fps and a single GPU card getting 25fps, if the dual GPU card is microstuttering by +-5fps, even though your frameRATE is higher for the dual GPU, they're actually performing much the same as each other. Now when I say microstuttering +-5fps, that doesn't mean min = 25fps, av = 30fps and max = 35fps. You could have min = 28fps, av = 30fps, max = 32fps, but still be getting a +-5fps microstutter which means you're effectively seeing min = 23fps, av = 25fps, max = 27fps. Make sense? Probably not :p
Click to expand...

Either that was a really confusing way of explaining it, or your understanding is not the same as mine. Microstutter, as I understand it, is when the time between consecutive frames varies significantly. So the time between frames looks like this - 5ms - 7ms - 20ms - 6ms - that's exaggerated, but you get the idea. That gives the impression of stuttering even if the average frame rate is high.
 
Forceman said:
Either that was a really confusing way of explaining it, or your understanding is not the same as mine. Microstutter, as I understand it, is when the time between consecutive frames varies significantly. So the time between frames looks like this - 5ms - 7ms - 20ms - 6ms - that's exaggerated, but you get the idea. That gives the impression of stuttering even if the average frame rate is high.
Click to expand...



Yeah I did explain it badly :p sorry. You're kinda right and kinda wrong though. Microstutter is more like 20ms - 30ms - 20ms - 30ms -20ms - 30ms. So you get a displayed framerate of 25fps, but its effectively more like 20fps.

Thats what I meant when I said something like 25fps +-5fps.

You wont see "stuttering" unless the framerate is so low that stuttering becomes apparent, in which case the game would be unplayable either way It wouldn't be called "micro"stuttering if you could see it, by definition "micro" means you can't see the actual stuttering :p
 
Denamian said:
This review of the x2 talks about the stuttering issue.

Here's the specific quote:

Quote:Now, a word on something else that is rarely mentioned in reviews: microstuttering. It has long been the bane of many multi-GPU users, and has been one of the many barriers for myself. Does microstuttering exist on the X2? Unfortunately, yes. Again, the lack of a unified frame buffer is no doubt the cause for this. However, there is good news. On the X2, it really isn't that bad. It happens typically within the first minute of loading a map or save game, and then it is gone for good. So, while not perfect, it is an improvement.
Click to expand...

So just add a "unified frame buffer" and some dilithium crystals.
 
It is not a matter of being able to clearly see it or not. If it exists, then it impacts video quality negatively, regardless of if you can "see it" or not.

That said, this card is definitely a step forward, but microstuttering still exists to the extent that it should remain part of the discussion.
 
cadaveca
Considering I took this issue on full force, and now it's gone out of hand, and I see now people complaining of other problems, and calling it microstutter, I wish I could kill the beast I created!
Click to expand...


Nope. not if FPS is below monitor refresh....then the problem is something else.

Microstutter is exclusively:

high-framerates drawn, less displayed. You cannot measure Microstutter below monitor refresh..if your eyes are sensitive enough, you'll see issues at anything below monitor refresh, and it won't be due to the hardware design, it will most likely be software limitations.

people don't understand that the Crysis video I posted was with 60+++FPS, and they also seem to have neglected to read that the problem is @ 1280x1024, while 1680x1050 runs butter-smooth. The problem is evident only when framrate is above monitor refresh, and the system is obviously not gpu bottle-necked.

Like DevilMayCry4.... approx 300FPS, still stutter, @ 2560x1600. This is what happens when the crossfire connection is not capable of feeding rendered frames to the primary gpu...and does not show itself until the bandwidth of those rendered frames exceeds the bandwidth of the interconnect.

Anyone talknig about Microstutter @ 60FPS is just on the hypewagon, because it looked fun, but they do not know where teh wagon is headed...


Provides a good chance to see which reviewers are worth thier salt, and which are not, IMHO. I've yet had anyone contact me from online media asking about the issue...so I'm not too sure that they are even aware of what the problem is, nevermind that most don't have the hardware to properly create the issue.
Click to expand...
 
Sigh, can any of you guys actually notice difference between two frames that are spaced at 30 ms apart?
On the other hand what you guys should really take a look at is that 8xaa comes free with 4870x2 in crysis
 
Well Tudz example is a little flawed since 25ms spacing would be 40fps not 25fps.

And yea when you are following a moving object on the screen the difference between 30ms and 20ms spacing is noticeable. And what would be the point of buying one of these cards if you can't tell the difference between 50fps and 33fps?
 
flegg said:
Well Tudz example is a little flawed since 25ms spacing would be 40fps not 25fps.

And yea when you are following a moving object on the screen the difference between 30ms and 20ms spacing is noticeable. And what would be the point of buying one of these cards if you can't tell the difference between 50fps and 33fps?
Click to expand...

So this is not even about micro stuttering, the issue is more about the average fps you're getting. Maybe turn down the res or the eye candy and this ""microruckler"" will disappear. The way I see it people are just putting too much load on their system or gpu and expect to see perfect movie quality picture.
 
You must log in or register to reply here.
Back
Top