Radeon 5xxx Owners Report Grey Screens/Hangups

[Black$heep86]

I just bought a XFX 5850 working fine on my MSI 790FX-GD70 motherboard with some el-cheapo DDR3 lo. Great card haven't had any issues with it at all so far. It is alarming that AMD has not found a fix for it yet.

 

[cinergy]

I wouldn't be surprised if those couple of dozen gray screen whiners are nvidia PR drones executing phase III of their master plan:

Phase I: Publish Fermi power-point slides with best case theoretical speed figures
Phase II: Pay off major sites to create positive buzz around Fermi
Phase III: Trash ATI cards to stop people buying them
Phase IV: Launch Fermi
Phase V: Damage control and public announcement of Fermi 2 (-> next goto phase I)

 

[swifty_morgan]

When I first got my ASUS 5850 I had no problems with any gaming except for FarCry2. I thought it would need the next driver update to fix..... ( black blotches here and there as the game went on with a little stuttering here and there ).. I used the CCC to over clock the card. Got it to 923/1023 before it wouldn't run stable without using ASUS' tool to bump the voltages... which I didn't do. I manually set the card back to default settings. I never had a problem with it since.... figure that one out for me.

 

[Guest]

This is too damn funny. Good thing I bought a Geforce 285 to wait for Fermi.

 

[mitch074]

I used to have problems with my 4850. It's a reference design form Sapphire (so much 'reference' that the only difference was a Sapphire sticker glued over the reference sticker 😛 ). That card would heat up a lot.

Well, par for the course for a reference 4850 with a reference cooler.

So, the first thing I did was tweak the VBIOS (I run Linux, on which AMD have yet to issue programmable power profiles) in order to get lower idle frequencies (and heat) and higher fan speed on load. That stopped several heat-related crashes and, after some tweaking (finding the best frequencies and voltages) no more corruption.

Said corruption was noise in accelerated 2D in Windows XP. Never grey screens nor bars.

But, starting with Catalyst 9.9, I would get initialization problems under Xorg. Please note that, according to an AMD engineer, most of the driver code is common from one OS to the other, using only an OS-specific intermediary layer. I couldn't solve that problem, no matter what I tried: new PSU, different compile options, disabling ACPI, modifying RAM timings, resetting VBIOS to factory...

What solved the problem was, a different motherboard. I went from an Nvidia 6150 s939 board to an AMD 785G AM3 board (yes, I of course had to change the CPU, but that old X2 3800+ deserved its retirement). And then, most problems stopped - I would still get hangs once in a while on Xorg initialization, but even that stopped altogether with Catalyst 9.12, which came out a few weeks after I had reported the bug on a forum where AMD engineers dawdle. I have yet to try 10.1. Did it come from the chipset, from unstability stemming from the mobo's age, a conflict in the BIOS (which predated even R600 cards)? No idea.

But, I'll say it right here: card drivers are very sensitive pieces of engineering. An update in Win7 could have indeed triggered the bug for some; for others, it might be RAM access; for others yet, a more stable PSU; and for others, a bad mobo/GPU combination. It could even be that your card's manufacturer didn't follow reference designs and get funky with the drivers.

 

[sohei]

hei guys (tom's and others) the cards have warranty, so if anybody has a problem go and use the warranty,

anyway i have a question, does anyone has a problem with win. XP?

 

[sohei]

[citation][nom]cinergy[/nom]I wouldn't be surprised if those couple of dozen gray screen whiners are nvidia PR drones executing phase III of their master plan😛hase I: Publish Fermi power-point slides with best case theoretical speed figuresPhase II: Pay off major sites to create positive buzz around FermiPhase III: Trash ATI cards to stop people buying themPhase IV: Launch FermiPhase V: Damage control and public announcement of Fermi 2 (-> next goto phase I)[/citation]


man this scenario has repeated several times in the last 5 years ...
it's sounds like a cliche but it's so plausible ...after all the anti trust events

 

[Aleksa2009]

Good article, by the way the biggest news by ATI to IT community is
not the new catalyst v10.1 drivers, but a second postponement of 5830 by ATI. The unofficial reason for a 5830 that was initially slated for 25 January is PCB issues :-D.
That is so lame excuse :-D, the 5850 and 5830 are same card, same PCB, cooling, 256bit BUS, DDR5, same GPU just lover clocks of GPU core and memory, the biggest difference is the sticker :-D = 5830 is binned 5850. The 5830 cost AMD same as 5850 so they decided to postpone 5830 as long as possible at this stage for 3-4 weeks to make more money for ATI- AMD and its partners, 5850 earns higher margins for whole ATI distribution channel. This is a result of lack of competition, NVIDIA hurry up with FERMI ;-D.

 

[View1]

I had two XFX5970BE doing all the stuff listed here. Done all the things asked by XFX and AMD/ATI not only once but again & again all the way to doing a complete win7 64bit re-install and one by one driver updates with testing in between only to continue having the same issues. Ended up sending my system to the suppliers who you guessed it did more tests only to come to the same conclusion I did "The cards are faulty”. Also noted that Temperatures in each card got hot hot hot reaching up to 90dec Celsius (single card setting) and we stopped test crossfire at 104dec Celsius for safety reasons.
Throttle back power drops from around 55watts down to 9watts just prior to dropping from 1000MHz down to 300MHz only for a split second (temp would drop 1-4dec C) then back up to 55watt & 1000MHz
I can not understand why AMD/ATI do not want to say anything or do I understand why they will not ???? Money, Money, Money (ABBA)
I have kept all correspondence between both AMD/ATI and XFX
AMD/ATI it’s about time you came clean, there are far to many people around the world with these issues for it not to be both card & driver issues.

 

[Guest]

I had a similar problem with my ati 4850 last year. Its seems that the problem was the motherboard (M3N-HT) with the 4850 running together. The video card worked fine with other motherboards and the m3n-ht worked fine with a gf9600gt, so i changed the motherboard for a p5q and the problem was solved.

 

[Lewis57]

Hm damn, I hope they get this sorted because i was seriously about to jump onto the ATI bandwagon. This sounds odd, there so many factors involved. I hope this doesn't proove to be too costly for ATI, but if they don't get it sorted soon, i'm afraid to say i'm going with nVidia for another 2 years.

 

[tarc2000]

I used to hv colored flicker and hangs, but once I update my Motherboard BIOS, the problem had been solved.

My System Spec:-

Sapphire 5770 1GDDR5
AMD Phenom II X4 955
Asus M4A77TD Pro
Mushkin DDR3 1333 PC10666
Cooler Master Silent Pro M 600W
WD 1.5TB (32MB)
OS: Ms Windows 7 32bit

Hopes this may help ppl who try to find a solution.

 

[Guest]

i think there's a driver problem, my 4870 downclock when i play borderlands since i updated drivers, before the update, it was ok

try using catalyst 9.9 with 4xxx series to verify that
5xxx i think that is something like that too, but older drivers don't support them

 

[invlem]

I'm leaning towards a driver update issue, ATI's been huge on the power conservation (a good thing), maybe they were a bit too agressive? My card still works, but I haven't updated my drivers in over a month.

 

[Glacier]

Damn, I hope my 5850 doesn't get these problems.

 

[fozzie76]

Crap.. I just got my XFX 5870 on Tuesday. I had to mod the crap out of my set up so I just started gaming last night (Mass Effect 2), but so far everything is okay. I did not OC the board as I don't own any games that it won't run at 1920x1080 resolution maxed out at factory clocks.

 

[Nossy]

Well, I just pop in a 5770 into my HTPC and it does take noticeably longer to load into Windows Vista 64. Before I was using an IGP GF9300n

 

[mustwarnothers]

I'm running a Sapphire Radeon HD 5770, on Win 6 64-bit with the Core and Memory significantly overclocked, and fan speed set at 50% for gaming.

I haven't had a single crash or color issue and it's been pushed with Modern Warfare 2, Dragon Age Origins, Mass Effect 2, all with nearly top settings.

Probably just an issue in the manufacturing process somewhere? Bad heatsink-chip placement?

 

[kinney]

[citation][nom]Atranox[/nom]So my newly built rig - HAF 922, PII 955 C3, Sapphire 5850, 4Gb G.SKILL Ripjaws, 750W Corsair, etc - is at fault, running at stock speeds? I find it highly suspect that my PC is having the same problem as quite a few other people with setups that aren't even remotely similar in many cases. It's happened to people on AM3 builds, 1156 builds, 1366 builds, with every RAM brand, with different PSUs, etc. I've tried 3 different Catalyst drivers, have flashed the BIOS, undervolted, overvolted, etc - nothing fixed the crashing.I'm Prime95 stable for 24 hours with absolutely no problems, but any time I play a game I crash to a gray screen? That doesn't seem right to me.[/citation]

Please try the new bios, 12.014.000.005.035264. Check your current bios using GPU-Z (and back it up!) Do not forget to uninstall the ATI drivers before you do the bios update, or you'll bork your OS install. Worth a try, but note that if it's in the release notes, then it's acknowledged and all you can really do is wait (officially).

I don't have the issue with my card, but interested to see if this would affect someone having the issue. Yeah, I do think it could be your rig in some way though regardless.. but I'd like to try help you fix this. Post back here after the bios update. PM me in the forum if I can be any further assistance.

 

[WINTERLORD]

i have a xfx 5850 black edition and i heard a terrible schreaching while playing a game last night not sure if its the card though or that i had the bus set to x4 for some reason.

 

[decembermouse]

Interesting. So this problem is supposedly about the "5xxx" series- but half of the model numbers reportedly experiencing problems are from the 4xxx series?! These have been out for what, a year and a half? Why would this suddenly be happening to them now, at the same time that it's happening to the 5xxx series? I'm thinking a Windows Update is the only thing that would tie these two series together. Otherwise we'd have seen this way earlier, a year and a half earlier in fact.

 

[decembermouse]

Interesting. So this problem is supposedly about the "5xxx" series- but half of the model numbers reportedly experiencing problems are from the 4xxx series?! These have been out for what, a year and a half? Why would this suddenly be happening to them now, at the same time that it's happening to the 5xxx series? I'm thinking a Windows Update is the only thing that would tie these two series together. Otherwise we'd have seen this way earlier, a year and a half earlier in fact.

 

[j_chatelain]

To anyone having this problem - Try overclocking the gpu and underclocking the memory.

I'm running a 5870 XXX just fine at 900/1245, but crash at stock speed of 875/1300.

 

[konjiki7]

[citation][nom]miladk[/nom]Hi again, this is funny:i have increased the PCIE Voltage from 1.5 to 1.55 (with Asus AiBooster) and have no problems anymore, the game work fine and vlc player too.Really funny issue !Please test it, maybe it would help you other guys out too ?[/citation]

On second thought my buddy had same problem.... His oc wasn't stable with the 5870 but it was with the 4870...

Long story short it was a matter of the gpu pushing the limits of his cpu and a cheap power supply.lol So increasing the pcie voltage makes sense i can see that working.

 

[gerrardk]

Jane, you better check your sources before you publish an article.

NGOHQ said something different and even included workarounds to solve this issue: http://www.ngohq.com/news/17176-workarounds-for-radeon-hd-5000-series-vertical-stripes-issue.html