• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

*** Official OMG I've finally got my 5850 thread **

Maybe the review sites had a card that never had the voltage thing on, makes the card look good then on the retail model do it so it can't be over volted.
 
What's slightly annoying though is that techpowerup review. They must have had a modded bios to do this but haven't said anything more about it or put it up on their site when I'm sure they know we want it. :confused:
 
Surely they would only allow this for MSI customers though??

Nope. They posted up the modded 5870 bios for all to download and have stated the modded 5850 bios will be up for download soon.

Depends whether you want to flash your 5850 card with another manufacturer's bios tbh.
 
Nope. They posted up the modded 5870 bios for all to download and have stated the modded 5850 bios will be up for download soon.

Depends whether you want to flash your 5850 card with another manufacturer's bios tbh.

That is what I'm not keen on doing because that would screw any chances of warranty, I'm happy with my 800/1200 overclock atm :D
 
My old koolance low profile one been using it for my last few cards still serves me well. I know theres better out there but it still cools a dam site better than any air cooled solution, plus I dont have to keep buying new ones for each card. Just home from work now I swear I'll never get round to benchmarking this thing properly till next week at this rate! Bit worried about my VRM temps though, even though I've put heatsinks on them still seem kinda hot when running furmark (80s) in everest though when gaming only in the low 60s is this OK? Can someone else with the stock cooler still on tell me what everest is saying for you when running furmark/gaming think i'm gunna beef up the cooling on them.

tempted to do that same as my Dtek block is just a standard block and will fit however i am worried that the VRM's will run toasty as well. Will wait and see what full cover blocks come out of the next few weeks and go from there.
 
That is what I'm not keen on doing because that would screw any chances of warranty, I'm happy with my 800/1200 overclock atm :D

Except you can flash your old bios back should you need to.

Plus 1050/1300 looks nice too. It will be the first thing I'll be doing when available.
 
That is what I'm not keen on doing because that would screw any chances of warranty, I'm happy with my 800/1200 overclock atm :D

Not to screw anyone(shop wise) but if the card doesn't work, the card doesn't work and very few if any people will care what bios is on the card.

No where on my system does my computer tell me my card is a sapphire, its a built by ati, ati radeon, or 5850.

If it doesn't even boot, which is generally what will happen with a lot of failing cards, they won't and can't know what bios is on there, if it does boot but is unstable, its still very unlikely they'll actually look for the device ID and double check the bios.

Working in an RMA department is busy, you plug in , you check, you replace or send off, whatever.

Likewise flashing a bios is very unlikely to damage your card in and of itself and even if you flash badly, a blind flash is a fairly easy thing, and reinstalling the original bios isn't particularly hard either, only really not possible when the card won't boot. In which case, they won't see the bios/name of bios anyway.

I'm not one for RMA'ing hardware I've screwed myself, but lets be honest a bios change, and minor overclocking within pretty safe parameters isn't the reason 99.999999999999% of cards dies eventually. If the card dies, it almost 100% would have died had you not changed bios/overclocked, meaning you would have been entitled to a warranty replacement anyway.

A manufacturer, or two, won't void warranty for you improving the cooling either, as long as any damage caused by the install didn't kill the card, which is also very fair, improving the cooling decreases the chance of failure at your own cost, hardly fair to void a warranty due to that.

I had a 9700 die from a waterblock that leaked(cpu block), I didn't replace the card or send for RMA, I did get the cost of the card and replacement waterblock off company I got the faulty waterblock from though(didn't get the card from them).
 
Wanted to make a separate post for this, I, along with quite a few others(check amd's own game forum) have some trouble with the 58XX cards with any driver so far.

Powerplay seems to be the culprit, by default a bunch of people seem to be getting this enlarged cursor(damn massive thing) that you can't change back, and also crashing playing high def video, be it in media player, vlc, some flash players, and in virtua dub and some other things. Well not crashing but gpu usage shoots to 99% and you can't use the computer, ridiculous lag moving mouse and screen responding.

IT would seem to be powerplay because if you soon after reboot use gpu clock tool to set even default clocks(or an overclock, whatever you want) it disables powerplay(CCC and MSI Afterburner do NOT disable powerplay, that later, is very very welcome when drivers are fixed, can overclock and still have powerplay , yay) and sets the clocks you choose all the time, mouse cursor won't bug and high def is supposed to not crash your comp, fixes a few games performance where again powerplay seems to be kicking in killing performance.

Realistically it seems powerplay just isn't kicking in at the right times causing some issue.

About to test the high def playback as for a day now half the time I play back an mkv file it will bug to 99% gpu load and need a reboot.

As quite a lot of people are seeing the issue(not sure but maybe everyone is on windows 7 RC/RTM and mostly 64bit also) and posting on AMD forums they know about it and should be fixed in the 1st proper driver for the cards.
 
but maybe msi bios may not work on other brands?

All the cards are identical, and as someone said, MSI purposefully released their 5870 bios to be used on all cards, not just theres. They'd need their own personal app to refuse flashing to anything without a msi card device ID, they've purposely gone out of their way to make sure it works with all cards so chances of them doing that for the 5850 are almost non existant.

MSi are pretty pro-overclockers, only really late availability of MSI products to the UK makes me end up with other mobo's/gpu's, they've become a pretty decent company over the past several years.

As for why the same chip isn't allowing the same thing to work on 5850's and 5870's, its very possible for the same chip to use different pin outputs on different cards, purely because of the location of the chip and routing of traces throughout the pcb. It may just have been convinient on the different 5850 pcb to use a different pin to connect vcore and uvd compared to the 5870, because the traces are different.

So it could simply the be the case afterburner needs to be writen to change a different output on the 5850 to the 5870.
 
So you can't use powerplay at all then?

If there are some that don't have this issue (which I'm guessing there must be), what's the link between those that do?
 
Could certainly be windows 7 64bit, not like that many people are using it yet. 8gb mem, amd chipset, no idea, really not sure yet.

But I've not been adjusting it, when I was booting, playing games, I had no issue, its only if somewhere along the lines I used high def video along the way, then i'd get issues normally very quickly, sometimes a little after the video.

While a lot of people watch high def tv, again not that many do, I also hadn't been in the mood to play a game the last day or so, while most people get a new card and are testing every game they've ever played to see performance.

The mouse thing was more irritating than the crash though, and so bizarre :p its only maybe 6-8 times as big, hardly massive on the screen but stupidly distracting none the less.

EDIT:_ looking through the thread on AMD forums, though unsurprisingly considering where they posted, most people with the mouse/high def issue all have 8gb mem, win 7 64bit, and surprisingly, 4-5 of us have a 940 BE :p , most likely win 7 64bit being the least important for them to get working first, because its not officially out, get Win XP/Vista working first because, its out.
 
Last edited:
I'll definitely be testing lots of HD video playback tomorrow when I install my 5850 so I'll report back then.

I've added a link to the MSI 5870 bios in the OP just because I had already put the Asus one there. Of course I'll add the 5850 one when it becomes available.
 
So you can't use powerplay at all then?

If there are some that don't have this issue (which I'm guessing there must be), what's the link between those that do?
powerplay works perfectly on my 5850.

State #1 GPU: 725 MHz, Memory: 1000 MHz (Boot)
State #2 GPU: 725 MHz, Memory: 1000 MHz (3D)
State #3 GPU: 400 MHz, Memory: 900 MHz (UVD)
State #4 GPU: 725 MHz, Memory: 1000 MHz
State #5 GPU: 157 MHz, Memory: 300 MHz (IDLE)
 
Powerplay works, but with powerplay on(maybe only in win 7 64bit, and 32bit maybe) it causes the mouse cursor to get huge, and high def playback causes the gpu to shoot to 99% load(even though surprisingly temps don't increase at all) and completely unresponsive desktop requiring a restart to get back to normal.

Since using gpu clock tool, I've now played back 2 vids that previously cause the issue, and no sign of the mysterious and strange large mouse. Same fix has worked for a bunch of people.

Hopefully it won't be there in final release drivers, theres a RC7 for anyone that wants it available direct from MSI's site(link on product page for the 5800's), though apparently has the same problem and no differences anyones noted as of yet.

It could be possible that, well we've see the UVD voltage is changeable as MSI afterburner can change that, I could be wrong but thought the UVD was responsible for video acceleration, so maybe powerplay at the moment is incorrectly dropping voltage to idle, and becomes unstable during high def vid playback because the main core clocks/voltage remain low, and the UVD isn't stable under full load at a reduced voltage? THat would certainly explain it the high def problem, not sure about the mouse cursor but thats also fixed without powerplay.


WHen I'm not using high def vid, powerplay works fine, I get nice low power usage and its stable and I can game fine, just when I playback vid.

Hmm, though saying that, I don't think i've seen speeds as you get in state 3, I haven't seen 400/900mhz at all, so yes, thats probably the issue and some combo of high def is unstable without the clocks/voltage available at the higher clocks.


Gareth, have you tried watching any mkv encoded files, 720/1080 stuff, it would crash for me(or be unresponsive) when using WMP, VLC, MPC. But it might be certain types of video.
 
Powerplay works, but with powerplay on(maybe only in win 7 64bit, and 32bit maybe) it causes the mouse cursor to get huge, and high def playback causes the gpu to shoot to 99% load(even though surprisingly temps don't increase at all) and completely unresponsive desktop requiring a restart to get back to normal.

Since using gpu clock tool, I've now played back 2 vids that previously cause the issue, and no sign of the mysterious and strange large mouse. Same fix has worked for a bunch of people.

Hopefully it won't be there in final release drivers, theres a RC7 for anyone that wants it available direct from MSI's site(link on product page for the 5800's), though apparently has the same problem and no differences anyones noted as of yet.

It could be possible that, well we've see the UVD voltage is changeable as MSI afterburner can change that, I could be wrong but thought the UVD was responsible for video acceleration, so maybe powerplay at the moment is incorrectly dropping voltage to idle, and becomes unstable during high def vid playback because the main core clocks/voltage remain low, and the UVD isn't stable under full load at a reduced voltage? THat would certainly explain it the high def problem, not sure about the mouse cursor but thats also fixed without powerplay.


WHen I'm not using high def vid, powerplay works fine, I get nice low power usage and its stable and I can game fine, just when I playback vid.

Hmm, though saying that, I don't think i've seen speeds as you get in state 3, I haven't seen 400/900mhz at all, so yes, thats probably the issue and some combo of high def is unstable without the clocks/voltage available at the higher clocks.


Gareth, have you tried watching any mkv encoded files, 720/1080 stuff, it would crash for me(or be unresponsive) when using WMP, VLC, MPC. But it might be certain types of video.

I'm not gonna pick the bones out of Your post but You are way off track here
UVD will only work with approved vc1/H264 and codec/players It sounds like Your MKV's (x.264) are trying to call hardware acceleration !
You do not have a powerplay issue of any description ! You have a driver conflict which considering You are running W7 is what You should not be too surprised about
 
Back
Top Bottom