ATI Deliberately Retards Catalyst for FurMark

[W1zzard]

3D-Analyze using nvidia ID

doesnt 3d analyze reduce the general performance of the "card" ? if yes then that's why you see the lower temps. more overh ead in 3d analyze, less time to render, less fps, less load, less heat

you lose ballpark 10% FPS and 10% power consumption accordingly. isn't that to be expected?

 

[DarkMatter]

3D-Analyze using nvidia ID
Ambient temp 25C
GPU 42C
GPU VRM 70C
GPU Fan speed 4364
GPU VRM 65.03A

4870
Ambient temp 25C
GPU 42C
GPU VRM 75C
GPU fan speed 4365
GPU VRM 77.23A


From the looks of things it appears that when I trick Furmark into believe it's using a nvidia card using a nvidia card's vendor and device ID:
-the vrm temps are lower
-GPU VRM is 12.2A lower
Now can we honestly believe that this is only related to how the 4870 is manufactured?

What's your point? Under Nvidia ID performance is significantly lower too, thus everything else is lower too.

TBH are you trying to make a point based on what? You tricked the benchmark to think it was a Nvidia card and you didn't expect something strange to happen? For instance, because the benchmark thinks is a NVidia card, probably isn't loading all the shaders because Nvidia cards don't have 5 ALUs per SP... :shadedshu

EDIT: Wizzard beat me to it.

 

[EastCoasthandle]

An easier way to answer it is to post results of a nvidia card (9800 GTX/GTX+, 260/280) and compare the results of the gpu temps, vrm temps and vrm amps.

I didn't do this to compare scores but to see if a difference was found in temps/amps.

 

[btarunr]

Don't scores relate to amperage anyway? Aren't they directly proportional?

 

[EastCoasthandle]

Don't scores relate to amperage anyway? Aren't they directly proportional?

Unless I missed something aren't there differences in results between one make/model video card and another? If so, wouldn't the results be different anyway regardless of amperage?

 

[btarunr]

Aren't you using the same card to emulate / "trick" FM to seeing a NVIDIA card? As W1z said there's a loss of ~10% score, as also amperage/VRM Temps.

 

[EastCoasthandle]

Aren't you using the same card to emulate / "trick" FM to seeing a NVIDIA card? As W1z said there's a loss of ~10% score, as also amperage/VRM Temps.

I didn't know that a 4800 series was suppose to give the same score when the ID is changed to another. I see your point of view on this. However, I do not believe that the scores are suppose to be the same. However, I am curious as to what a 9800 or 260 GTX results temps/amp/etc would be when using furmark. Do you (or anyone else) mind posting them?

 

[DarkMatter]

I didn't know that a 4800 series was suppose to give the same score when the ID is changed to another. I see your point of view on this. However, I do not believe that the scores are suppose to be the same.

No, scores are not supposed to be the same, as I said probably there are many ALUs sitting idle. But, it doesn't matter. You don't seem to understand that when the card is not stressed under 100% load, the card requires less amperage, will consume less, heat less and even maybe require less voltage.

When disabled or when at idle the SPs act like resistances. High resistances when at idle and almost infinite resistance when totally disabled (I think only GT200 GPU can do this right now though, not sure). Now when voltage is constant (as is the case) the higher the resistance the smaller the amperage it will be.

 

[DrunkenMafia]

It baffles me how these companies (ATI) think they can do something like this and no one will catch on.

They should be fixing the overheating problem rather than trying to slow the card down under our noses.

thats just bad business practises

 

[HAL7000]

Well my 2 4850's from diamondmm have been running great. I bought the pair for 138 each after the rebate. They run hot but I don't OC either.

So is this limited to powercolor and their card bios? The life of a graphics card isn't that long anyways. With improvements, upgrades and so on, this bench means nothing IMHO.
But if it isn't powercolor and it is ati and their drivers ,well lets hope they fix the dam problem and quick.

 

[PVTCaboose1337]

Just an FYI, once the fan fix is initiated, which I bet not many users do, the temperatures are pretty good! 38c on stock cooler, idle, and 80f room temp. Pretty nice if you ask me! 40% fan btw.

 

[newtekie1]

Just an FYI, once the fan fix is initiated, which I bet not many users do, the temperatures are pretty good! 38c on stock cooler, idle, and 80f room temp. Pretty nice if you ask me! 40% fan btw.

Idle temps mean nothing.

 

[Kursah]

Idle temps mean nothing.

Very true...many like to brag of it, hell even I do from time to time...my GTX260 idleing at 44-45C means nothing in reality...it's the load temps and stability that should truly matter. With the cooler temps up this way...my GTX260 hits 57C folding and 60-61C gaming/benching/stress testing @ 80% fan speed.

It's a bummer ATI decided to do this driver "un-tweak"...but it's just a benchmark in the end...gaming performance is what SHOULD matter to most of you out there...if you're a bencher, then obviously these drivers will be skipped or maybe a "patch" will be created to override what has been done.

 

[candle_86]

This only confirms something that I already suspected and that was almost confirmed already by the temperatures of all R7xx cards and the low OC capabilities. Ati clocked this cards very high in order to own this round and also used very cheap cooling to make the cards cheaper, for the same reason.



That's something hard to tell, but as of now it seems that the failure rate is not very high (=excesive to the point to annoy consumers), even though it's definately high above the average.

its nearing the 7900GT failure rate

all this tells me is ATI is scamming there drivers, and they can't cope with 100% load on the GPU. I think ill get either a 9800GTX or GTX260 and not have to worry that in a year or so other games will be just as intense and smoke my card to ash.

 

[eidairaman1]

So are they gonna try this on other boards aswell, because remember powercolor is having a rough time with their New line of boards.

It is a known flaw that some models of the Radeon HD 4800 accelerators fail oZone3D FurMark, an OpenGL based graphics benchmark application that has found to stress Radeon HD 4800 series far enough to result in over-heating, artifacts or even driver crashes. The Catalyst 8.8 drivers have found to treat the FurMark executable differently based on its file-name. Expreview tested this hypothesis by benchmarking a reference design HD 4850 board using Catalyst 8.8 driver, with two runs of FurMark. In the first run, the test was cleared at a low score, much lower compared to those of whatever successful runs on older drivers could churn out. Suspecting that the driver could be using some sort of internal profile specific to the FurMark executable, Expreview renamed the furmark.exe file, thereby not letting the driver know it's FurMark that's being run. Voila! the margin of lead the renamed FurMark executable gave over "furmark.exe" shows the driver to behave differently. A shady thing since Radeon HD 4800 almost became infamous for failing at FurMark, and at least passing it with a low score seemed better than failing at it altogether.

Expreview caught this flaw when testing the PowerColor Radeon HD 4870 Professional Cooling System (PCS+) when odd behaviour with the newer driver was noted. Successive BIOS releases didn't fix the issue, in fact, it only got worse with erratic fan behaviour caused due to a "quick-fix" BIOS PowerColor issued (covered here).




Source: Expreview

 

[Kursah]

My Powercolor x1950Pro Extreme 256mb is still running strong in a friends' rig after almost 2 years...wonder what happened...

 

[candle_86]

they wanted to catch up to NVidia and in return didnt plan the cooling well enough.

 

[bangmal]

LOL.
I see that all the long fraustrated nv boys who bought the overpriced GTX200 series are crawling out of the woodwork to celebrate.

It is amusing to see all the nv girls are having orgasm at the same time because a shady software and a shady website.

Anyway, I ran a test because of this "news":
My setup:
An overclocked 4870 running at 820/1000 using stock cooler
Furmark 1.4.0
Windows Vista Sp1

after running two tests for 1000 seconds, i took two image.

The first image was before the name change.

The second image was after "FurMark.exe" changed to "fuckmark.exe"

As you can see, the temperaure, the load, the VDDC current remain almost the same.
Also notice that the fan is even not spinning at half of the max speed.
so much for the "overheat"
So much for the "ATI Deliberately Retards Catalyst for FurMark"
cheers

 

[Kursah]

Yes but search the forums...there have been at least 2 threads I can recall about this issue with HD48xx cards in the last month.

I'm not a fanboy, though you show your true colors...and maybe need to take a brake from the PC...but I get what makes sense for me and my budget...I spent less than an HD4870 on my GTX260 and have had nothing but an excellent experience with it. The HD's just seemed too finicky to me, fan speed issues, driver issues, temp issues, mediocre cooling, it's gotten better on most fronts and can only improve and even with the issues I found to be "issues" for my "preference" those cards still kick ass and are only improving, for the most part.

Kind of a crap time for news like this to come out tho...almost in-opportune for ATI, hopefully this isn't some sort of scam as I like ATI, I've had many ATI and NV cards...but this kind of crap is unnacceptable from either side...the GPU needs to perform, not be retarded with drivers so less users complain...both sides have done it in one way or another...it will continue, just depends on what seems like the biggest deal or is most noticable at the time I guess.

 

[candle_86]

i doubt its a scam honestly, not from faud anyway, he is AMD bias'd already.

 

[btarunr]

@bangmal: More number of people have verified this than you disproved it. You've not even shown which catalyst driver you use.

 

[bangmal]

Driver Packaging Version 8.522-080731a-067980C-ATI

Catalyst® Version 08.8

Provider ATI Technologies Inc.

2D Driver Version 7.01.01.809

2D Driver File Path /REGISTRY/MACHINE/SYSTEM/ControlSet001/Control/Class/{4D36E968-E325-11CE-BFC1-08002BE10318}/0001

Direct3D Version 7.14.10.0603

OpenGL Version 6.14.10.7873

Catalyst® Control Center Version 2008.0731.2322.39992
-----------------

@btarunr, people do not report when their cards are working perfect. I admit, I have too much time tonight

 

[btarunr]

They don't, but even when a handful of them do, it becomes a cause for concern. Especially when websites report this flaw and publish it. It's a known and verified fact that PowerColor HD 4870 PCS fails FurMark more times than it passes it, and that successive BIOS didn't fix it, at reference fan speeds the card can't take those temperatures, and so Catalyst 8.8 turns down performance.

 

[bangmal]

PowerColor HD 4870 PCS is using a very different cooler, i guess that is the problem.
I was using a GTX 260, overclocked at 720/1440/1200, it gets a significant lower score than a stock 4850 at furmark, does it mean nvidia deliberately turns down performance on furmark too?

It looks to me it is nothing more than a bug in the furmark, ATI fixes it in the driver update, thats all. Some people, especially the nv fanboys are making a mountain out of a mole hill.

 

[btarunr]

We're not comparing GTX 260 to HD 4850 here, we're comparing the same card performing differently with different drivers on a benchmark, more so, performing differently with the same driver with the application renamed. Don't make pointless accusations without looking at sources, it's not just Expreview, several sources I talked to point to the same thing. The mountain is because that's a significant performance hit just to let the card sail through.