RADEON 4870X2 broken dogtracker?

[y eye]

Originally Posted by beerandcandy

its probably caused by the refresh rate clocking
when the os says 60 HZ it probably isnt really 60 HZ its probably 58.7 Hz or some other close value.
see if the micro stutter follows certain display resolutions and refresh rates
and document what they are

What is the difference between Stutter (micro stutter) and Judder?

[beerandcandy]

judder is a frame drop around the 12th or 13th frame
micro stuttering is the same thing bu attributed to nvidia cards
and stuttering is just frame drops
they can all be linked to the FPS and refresh rate if we make a map of refresh rates and actual clock timings for the refresh rates we could probably figure out what video cards show this behavior at specific display modes

[y eye]

Originally Posted by beerandcandy

judder is a frame drop around the 12th or 13th frame
micro stuttering is the same thing bu attributed to nvidia cards
and stuttering is just frame drops
they can all be linked to the FPS and refresh rate if we make a map of refresh rates and actual clock timings for the refresh rates we could probably figure out what video cards show this behavior at specific display modes

I started wrestling with a Dell 30" that was showing mosquito noise, and edge crawl . After installing Juddertest and Powerstrip I ran ntune with temps realtime and found that juddertest was showing wide bars with irregular gaps. It seems that the fluxuating gaps were indications of stutter/judder . It did not take long to realize that the overclocking of the videocard before the juddertest and PowerStrip install had the GPU and GDDR out of sink but right on the thermal limit, which is what I thought was the way to clock. Resetting the GPU Voltage and viewing Juddertest made noticable changes in the test screen. Resetting the GDDR voltage and viewing Juddertest changes gave a wake-up call and it was not long before my Juddertest was narrow bars evenly spaced. The Mosquito noise and edge-crawl was gone. Then on to Powerstrip and a slow tick up on refresh rates. At 62Hz the colors became vibrant and pastel as in 3D and the text became liquid and 3D and easy on the eyes. The Monitor remains at default 2560 X 1600 and the Video Card is 1920 x 1080 pixels. I ramble here in answer to your explanation it is my total of 3 months trying to tune the Dell. Thanks for your answer it seems right on. I thought that Stutter and Judder might be very similar or the same after beginning to get the Dell under control. Maybe dealing with stability here...

AMD GAME Drivers & Tools

"Launch: AMD OverDrive Video"
by: Sami Maekinen

[m25]

Originally Posted by beerandcandy

its probably caused by the refresh rate clocking
when the os says 60 HZ it probably isnt really 60 HZ its probably 58.7 Hz or some other close value.
see if the micro stutter follows certain display resolutions and refresh rates
and document what they are

I'm not in the mood to dig now but I read this morning a 4870 X2 review that explicitly mentioned that there was no micro shuttering at all during all tests, at least with their sample.

[y eye]

Originally Posted by m25

I'm not in the mood to dig now but I read this morning a 4870 X2 review that explicitly mentioned that there was no micro shuttering at all during all tests, at least with their sample.

May be but I doubt it because there are too many variables...
The Dr. says that Stutter, Studder, Judder, and Micro stutter are the same by definition, and calles them UFO 'Uneven Frame Output'. He offers a home brewed utility to help an individual to test for 'Uneven Frame Output'.

"UFO 'uneven frame output'

Microstutter - fact or fiction? Let's find out once and for all...
There is much debate about 'microstutter' (which should really be called 'uneven frame output' but anyway) when using multi-GPU setups. Many people report that they notice this when using their multi-GPU systems. A few have produced results 'proving' its existence, whereas others swear down that they have never experienced it, and/or question the validity or applicability of the very few numerical results actually available.

Anyway, to help answer the question, I have written a very simple little program which analyses FRAPS benchmark outputs to determine the degree of microstutter present. You can download the program here:

Edit: 12Oct08
The program is no longer available.

Of course there will be some non-uniformity of frame output, even in single GPU setups. Also it is clear that the degree of non-uniformity changes from game to game. So, what would really be interesting is to find out by how much using a multi-GPU setup increases the level of microstutter, if at all.

[y eye]

PCGH discussing with AMD
AMD: No solution for microstuttering in sight
Aug 26, 2008 15:33 - At the Games Convention we had the chance to talk to AMD about the mircostuttering issue. They are aware of the problem but there is no easy solution.
PCGH - News: AMD: No solution for microstuttering in sight
Before/After: This is what a solution for the microstuttering problem could look like. See link above.

The micro stuttering issue reached AMD, but it seems like there are different opinions about the problem. Sasa Marinkovic, Senior Product Marketing Manager of AMD's Graphics Group, doesn't think microstuttering to a problem. In his opinion he "Micro Stall” are not a not a real problem for the customers.

Rene Froeleke, Technical PR of AMD, already made some test in that area. But he is skeptic about to what extent microstuttering can be realized without benchmark recordings (with Fraps for example). Possibly Fraps isn't working accurate and a simple solution would actually not be close at hand anyway.

Only recently an ex AMD employee introduced a solution for the microstuttering. PCGH will closely observe the matter.
(Thilo Bayer)

[beerandcandy]

it shouldnt just be amd having the problem it should happen on all video cards
but its up to the designer to make sure that they tune the clocks to make sure the refresh rates are set so that the issue will only be limited to a couple of modes and not all over the place
i suspect amd didnt do this study since they are the only ones mentioned with this

[y eye]

Originally Posted by beerandcandy

it shouldnt just be amd having the problem it should happen on all video cards
but its up to the designer to make sure that they tune the clocks to make sure the refresh rates are set so that the issue will only be limited to a couple of modes and not all over the place
i suspect amd didnt do this study since they are the only ones mentioned with this

It is not just AMD. The example I have given is from my experience with a single GPU NVIDIA card.

It/they are redefined here as UFO (uneven frame output) for simplicity/focus.

My search and apply... began many months ago after reading a post on prade.de regarding judder.
Their reference was to wave judder on Dual GPU Graphics Cards, both Red and Green.
They offered no definition...

Recently I received an email from prad.de with the following included.
Note:
"Monitor Calibration"
> Digital Quality Tool
Welcome to Cipho

[RKH404]

first off, my hd3870x2 i got the first time was defective, i had to RMA it so there is no way to say that that hd4870x2 wasnt defective as well. Second, dont compare the hd4870x2 to the gtx 280, its almost 150 bucks more, has gddr5, and the fact that its performance level is in a different rank its like compareing a stock clocked 8800gs to a 9800gtx+ on a 1920x1200 monitor. third the gtx280 is a great card, does what a 400 dollar card is expected, the hd4870x2 does too.

[sark.inc]

Originally Posted by RKH404

first off, my hd3870x2 i got the first time was defective, i had to RMA it so there is no way to say that that hd4870x2 wasnt defective as well. Second, dont compare the hd4870x2 to the gtx 280, its almost 150 bucks more, has gddr5, and the fact that its performance level is in a different rank its like compareing a stock clocked 8800gs to a 9800gtx+ on a 1920x1200 monitor. third the gtx280 is a great card, does what a 400 dollar card is expected, the hd4870x2 does too.

i think you miss the point with comparing those kind of cards, it's not about what price they are, it's about how many millions of fps they can do.

I think it's proven true when the 4850 came out and forced nvidia to drop prices by a good 100-200$ NZD

the gtx280 and 4870X2 are on comparable levels, in terms of price and performance.

[y eye]

Originally Posted by Viperabyss

I'm with YMAA on this one. The benchmark results seemed too fishy to be true.

Oblivion at 2560 x 1600


Notice how a much wider 512bit memory bus GTX200s being outperformed by a 256bit memory bus card?

The exact same test done by Anand:


Now, I'm not sure if Anand used Cat 8.6, but I really don't think a simple driver update will suddenly give RV770 cards a significant jump in performance

....................................

Why are you posting here?
Are you the cut/paste Guru?

[Afrodisiac]

Originally Posted by y eye

....................................

Why are you posting here?

Now now y eye, this is a free speech place, and anyone can post anything they want, as long as it does not violate the rules. Let's not create hostility out of nothing


It's why I'm not banned yet, after a whopping 1,400 bull**** posts.