Fixing a Bad BIOS Flash on the 4870

[halo112358]

edited: This is resolved, most (all?) of the cards reported here had incorrectly flashed bioses due to the use of DDR5, see W1zzard's excellent fix article over at techpowerup (http://www.techpowerup.com/articles/overclocking/vidcard/152) for a proper bios and full instructions - and GOOD LUCK!

If anyone cares, flashed cards wouldn't initialize ram after a cold boot (and we all know how often we cold boot our machines) so the failures appeared random. Easy mistake to make IMO.

[purecain]

same here m8, i have 2 of em that wont boot.... wtf is going on....

[Gunnz Akimbo]

well, you could try taking all PCI cards out?

[jimmyz]

Be more descriptive , does the fan spin 100%, are any of the red led's on the bottom of the card staying on, do they go on then off while trying to boot, does it seem as though the pc is booting but not the card? any info you can provide would help, it may or may not be dead

[purecain]

3 leds light up instead of 4 , the fans spin as normal... no post so no reflash.... dead as a dodo... but why didnt they do this last night why only after they had been dorment for a few hours.... dont get it m8...why are they preventing post?????????????????????????????????????????????? ???

[halo112358]

I'm pretty sure that's what mine does - I already filed an RMA with the egg, this crap is going back in the box. Maybe I'll buy another when they've sorted this out, maybe not.

[keiths]

What brands?

[halo112358]

Sapphire - but as far as I can tell all of the boards on the market are just rebadged OEM.

[jimmyz]

3 leds light up instead of 4 , the fans spin as normal... no post so no reflash.... dead as a dodo... but why didnt they do this last night why only after they had been dorment for a few hours.... dont get it m8...why are they preventing post?????????????????????????????????????????????? ???



You flashed them already? if your clocks are too high for the voltage it will not post, there is your issue. If you can perform the volt mod they will most likely work, I had a 4850 refuse too boot at 900 , raised the volts and it boots fine, the card with only 1 led is not getting juice, i bet it is your secondary right?

[keiths]

Yeah, that sounds similar to my 940 server board. It has two power leds; one lights up, the other on and off if the 8 pin CPU power connector isn't plugged in.

[purecain]

right np... i have booted xp by disabling a 4870 in safe mode using the 3870 as video adapter... i'll try upping the volts now.. thanks.. let you know what happens in 10 minutes...

[halo112358]

I've got 3 LED's lit up when the board tries to post, then it fails and spins the fan down. If you're looking at the board from the top down the two leds on the right are lit up, then one is dark, then the one on the far left is lit.

[purecain]

yep thats what i had, now after raising the voltage, it still doesnt work.. only now i have 2 red lights flash very quickly... so i'm on my way to fixing it...
best thing to do is flash a stock bios, only problem is that it wont flash the original one as it says its already installed... using latest winflash... atiflash wont work... what else is working in dos for these cards???
right... managed to reflash my 4870 but the bios isnt right, or the flashing process is fubar'd... this is what gpuz reads from the card now...

[Jakalwarrior]

Fastest 4870 ever!
Anywho, why not try flashing in a bios from a different manufactuer so it wont say its already installed? or maybe there is a force command parameter.

[Scubar]

Wow, that makes me abit happier i didnt hit the Buy button yesterday on the HIS crossfire package for £440. I might hold off to see how the 4870X2 does when released next month because the 4870s do look very promising even for the slightly inflated price for launch due to price gouging.

[mr2sw21]

oh man I just bought one yesterday and I hope it is not going to happen like that.

[xMrBunglex]

i just got two 4870's as well. i'll hold off on the overclocking or flashing until the source of this is discovered. i'd hate to kill my shiny new cards. i don't think any of us know much about GDDR5 yet, so i have no idea how hot it runs or how well it overclocks.

[Papu]

unlucky :(

[philbrown23]

I read an article saying that atiflash and winflash have caused crashed on the 4XXX series because of inproper bios coding for gddr5 try the one that is in the bios mod forum on this site, flashrom I heard there have been no problems with it yet and you can force flash the cards back to stock if needed.

[purecain]

can someone upload a stock bios and a link to a bios flash utility that works... i'm getting bored of this...

[jimmyz]

only one I know of so far with a link. good luck. http://www.techpowerup.com/vgabios/index.php?page=1&architecture=ATI&manufacturer=&model=HD+4870&interface=&memSize=0

as for a flash utility. that is another story, if winflash won't do it correctly you may be out of luck, it is the most recent that I know of.

[keiths]

Or ask someone with an unmodded 4870 to dump their stock BIOS. Probably be a good common practice at xtremes to collect stock BIOS dumps. Of course, good common practice to save your own before modding:)

[xMrBunglex]

i have Sapphire 4870's fresh out of the boxes. i'll gladly upload my BIOS if you need. can i use ATiFlash or will that cause problems? i used ATiFlash 3.49 to flash my 3870's, but i only used it once and i need to know the command to dump it. let me know if i can help.

[purecain]

thanks jimmyz i was just flashing that bios, but it had the same results in winflash... gonna try ati flash now
get back to you in a minute mr bunglex...
atiflash doesnt recognise the second adapter(4870) only the first (3870)....

[Dragy2k]

good luck cain....most likely teething probs wit the 4870's :(

[zanzabar]

the thread need the name changed

[GoldenTiger]

Mine is recognized by atiflash with a DOS boot CD, but I flashed back to stock again and it still won't POST with just the card... in Windows it is detected as an HD4800 Series with a yellow exclamation point saying "the device reported problems" in device manager, winflash won't even start now for some reason. I have three lights that go on with 100% fan, then go off and the card spins back down... looking at the card top-down with the IO shield at the top, it's the two right-hand LEDs and then the left one, with the 2nd from left unlit (X-XX) like that. Any thoughts? Is it possible it didn't really flash it like it should have (it went very quickly, a couple of seconds, whereas Winflash before took 15-20 seconds generally)?

[purecain]

no you did everything properly, i just cant believe RBE has 4870 support listed otherwise i would never of tried it... its bugging me that it worked all night.. only showing a problem when i tried to boot my pc today...
thats two cards completely fubar'd.. all be it temporarily.... and the community learns from our mistake and moves on.... lets hope wizard and BAGZZlash hears of our problem and offers some help....

[GoldenTiger]

I don't get why flashing back to my pure-stock BIOS that I had copied from the card itself isn't fixing it... that's what bugs me especially, it's weird :( unless RBE had something very wrong there that caused permanent damage.

[halo112358]

Here's the stock bios from my sapphire - http://halo112358.googlepages.com/bios.rom

I'm pretty sure that using RBE to tweak the fan settings is what caused the card to not post on a cold boot. Funny that it worked fine for a day until I shut the machine off and back on (totally explainable). Should I track down an old PCI video card and use it to boot up and flash the stock bios back onto the board?

If I can get the card working again I'll cancel the RMA and order a crossfire motherboard - I already have another 4870 on the way :D

[GoldenTiger]

Here's the stock bios from my sapphire - http://halo112358.googlepages.com/bios.rom

I'm pretty sure that using RBE to tweak the fan settings is what caused the card to not post on a cold boot. Funny that it worked fine for a day until I shut the machine off and back on (totally explainable). Should I track down an old PCI video card and use it to boot up and flash the stock bios back onto the board?

I already tried using my stock BIOS from my card that I had backed up before editing at all (so 100% unmodified) and my card still isn't booting either :(.

[halo112358]

Well then, I guess we should warn people not to use RBE and winflash to tweak anything on their 4870s :-p

[GoldenTiger]

Yeah, I just don't get why with the *stock* BIOS it isn't booting...

[jimmyz]

no you did everything properly, i just cant believe RBE has 4870 support listed otherwise i would never of tried it... its bugging me that it worked all night.. only showing a problem when i tried to boot my pc today...
thats two cards completely fubar'd.. all be it temporarily.... and the community learns from our mistake and moves on.... lets hope wizard and BAGZZlash hears of our problem and offers some help....

The issue isn't with RBE in case you didn't notice If goldentigers card won't flash back to an unmodded bios then it is the flashing component that isn't working properly with the 4870.

also if you read it clearly says experimental in the revision history for RBE 1.11
there are reports of 4870's dying that were never flashed so it could be the same deal, I lost a 3870X2 ran fine the first day never booted again.

Can somebody with a "dead" card use a multi-meter to test voltage while booting and report back.

[xMrBunglex]

if it was the flashing that caused the problem, the title of this thread is a little unfair.

Purecain, i imagine that person meant a hard mod or pencil mod for the voltage. if the 4K cards are anything like the 3K cards, i don't think changing the voltage through BIOS has any effect on them. at least that's how i remember it - i never messed with the volts on my 3870's.

[keiths]

Isn't there an easy pencil mod that could tried?

[jimmyz]

if it was the flashing that caused the problem, the title of this thread is a little unfair.

Purecain, i imagine that person meant a hard mod or pencil mod for the voltage. if the 4K cards are anything like the 3K cards, i don't think changing the voltage through BIOS has any effect on them. at least that's how i remember it - i never messed with the volts on my 3870's.



Exactly W1zzard said the max bios can give is very little so that is the issue. A hard mod is the only current way to get voltage greater than that.

and the attitude from Purecain can stop now, all I am doing is trying to help, you can chase dead ends all you want but the sooner you realise the flash is the problem not RBE the sooner you will have working cards,(maybe RBE doesn't work for 4870's yet that is irrelevant if the stock unmodded bios won't take, use your head.) I don't care if you listen to me, I will just go play Grid with my cards(they work).:up:

[purecain]

if i cant fix this through rbe and if bagzzlash doesnt update it, which is unlikely . he's probably been working on it like mad as the community needs it.

if they still do not work i'll return them... thanks for the ideas lads but its definatly rbe1.11... i bet the next version fixes it..... remember if bagzzlash has used his own 4870 to test functionality he will be seeing exactly what we are...

[jimmyz]

the issue is BAGZZlash doesn't have a card, he desperatly needs input in the form of bios dumps etc. in order to test, the more the better and also dumps of bios's modded with a complete description and possibly a screenshot of the clocks page open using the bios in the dump. SO that is why it is a work in progress, if you can send me a hi res pic of your card I will show you how to get it booting with a pencil mod, do you have a camera and a MultiMeter?

[purecain]

i cant post on there, i've been a member for 5 years and my account wasnt set up properly... i need to set up a new account, infact will do so immediatly... also from bagzzlash to a forum member over at techpowerup


Even though RBE can open pre-modified BIOSes, this is only recommended for looking up the changes made. RBE uses some bytes inside the BIOS file for adjusting the checksum. The number of those bytes is limited. So, it is recommended to use a "fresh" BIOS for any changes.
The crash however occurred because of a bug in a code line for displaying just an error message for used up bytes. Will be fixed in the next version. Thanks again for telling me this.

Kind regards
BAGZZlash

any ideas why i can pm members but not post...

[GoldenTiger]

My card DID accept the original BIOS in DOS, but it still won't boot.

I don't have a multimeter, and just a point-and-shoot 5mp 3x optical zoom camera... :(. Thank you for the offer though... I wish there were any way to get it going, as it looks like the RMA for a HIS video card (if they'd even accept it) has to go to hong kong, and pay shipping from them back to me :( ! Really silly, that's almost $100 both ways... 1/3 the price of the card, and if it breaks again, another 100...


i cant post on there, i've been a member for 5 years and my account wasnt set up properly... i need to set up a new account, infact will do so immediatly... also from bagzzlash to a forum member over at techpowerup

any ideas why i can pm members but not post...

You probably need to have the activation email re-sent, you likely never clicked on it? *shrug*

[jimmyz]

My card DID accept the original BIOS in DOS, but it still won't boot.

I don't have a multimeter, and just a point-and-shoot 5mp 3x optical zoom camera... :(. Thank you for the offer though... I wish there were any way to get it going, as it looks like the RMA for a HIS video card (if they'd even accept it) has to go to hong kong, and pay shipping from them back to me :( ! Really silly, that's almost $100 both ways... 1/3 the price of the card, and if it breaks again, another 100...



You probably need to have the activation email re-sent, you likely never clicked on it? *shrug*



do you have a pci video card, don't pack up the card yet you can boot to the pci card and once in windows try the atiWinflash again, that is the preffered method for flashing .

[GoldenTiger]

do you have a pci video card, don't pack up the card yet you can boot to the pci card and once in windows try the atiWinflash again, that is the preffered method for flashing .

I tried atiWinflash but it won't even start like it would before (I am on vista 64, used 2.0.0.7 and it was starting/flashing fine, after this issue it no longer will start for some reason, just opens and closes after a split-second in task manager). Any ideas? I only have another PCI-E video card (nvidia), no PCI one, but my rig has 3 PCI-E slots... that's how I was booting to DOS. Is it possible ATIwinflash stopped loading because of an nVidia card in the system?

[jimmyz]

i cant post on there, i've been a member for 5 years and my account wasnt set up properly... i need to set up a new account, infact will do so immediatly... also from bagzzlash to a forum member over at techpowerup


Even though RBE can open pre-modified BIOSes, this is only recommended for looking up the changes made. RBE uses some bytes inside the BIOS file for adjusting the checksum. The number of those bytes is limited. So, it is recommended to use a "fresh" BIOS for any changes.
The crash however occurred because of a bug in a code line for displaying just an error message for used up bytes. Will be fixed in the next version. Thanks again for telling me this.

Kind regards
BAGZZlash

any ideas why i can pm members but not post...


You can pm him here he is a member and has the same handle. he lives in germany so he is asleep right now. I will be talking to him early in the morning and we will be going over this thread. the issues need to be resolved, nobody should RMA their card till monday anyway and we should know something by noon EST tommorow. I am currently testing states and their effect in full screen and windowed, and what settings affect which etc. so by morning we will know more.

[cstkl1]

can someone upload a stock bios and a link to a bios flash utility that works... i'm getting bored of this...
saphhire hd4870 stock
http://rapidshare.com/files/125529697/RV770.bin.html

[jimmyz]

I tried atiWinflash but it won't even start like it would before (I am on vista 64, used 2.0.0.7 and it was starting/flashing fine, after this issue it no longer will start for some reason, just opens and closes after a split-second in task manager). Any ideas? I only have another PCI-E video card (nvidia), no PCI one, but my rig has 3 PCI-E slots... that's how I was booting to DOS.

right click and run as administrator, if that doesn't work delete the folder and redown load, i had that issue on Vista 64 I am only flashing in xp now. but it works in Vista or XP 64 just quirky.

tommorow if you can get a pci card (local pc shop for 5 bucks used is fine).

[GoldenTiger]

right click and run as administrator, if that doesn't work delete the folder and redown load, i had that issue on Vista 64 I am only flashing in xp now. but it works in Vista or XP 64 just quirky.

tommorow if you can get a pci card (local pc shop for 5 bucks used is fine).

Yeah, I tried running as admin but no go still. There's actually a shop down the road from me, maybe I'll see if I can grab a cheap PCI card. You don't think ATIwinflash wouldn't load from an nvidia card in the rig, do you? In safemode it won't work at all, says must run as admin but rightclick--->run as admin, gives the same message. In normal mode it just does what I described. I tried redownloading and putting in a new folder but it wouldn't load then either.

[jimmyz]

Yeah, I tried running as admin but no go still. There's actually a shop down the road from me, maybe I'll see if I can grab a cheap PCI card. You don't think ATIwinflash wouldn't load from an nvidia card in the rig, do you? In safemode it won't work at all, says must run as admin but rightclick--->run as admin, gives the same message. In normal mode it just does what I described. I tried redownloading and putting in a new folder but it wouldn't load then either.

Try it, got nothing to lose either it works or doesn't.

[GoldenTiger]

Do you think it might run off of a BartPE boot disk (Win32 XP "live" environment)? I'm thinking that's the next best option for the night...

[purecain]

thanks for helping out jimmyz...
changing just the first voltage in rbe changes the lights that appear at boot... all four should light up i think and the two outer lights flash now x..x ...:up:

[jimmyz]

Do you think it might run off of a BartPE boot disk (Win32 XP "live" environment)? I'm thinking that's the next best option for the night...

Could you take the card to the local shop and try on an xp rig, a small shop will most likely lend help, they might be interested in learning a new trick. also think about friends pc's etc. I swear by the winflash method after issues with dos flashes....for mobo's I only flash in dos if possible with cards(ATI anyway) I prefer winflash, If it errors you can load a new one and try again if a dos flash goes bad, you will have to either blindflash or PCI card it.

[Edin]

Hi everyone, I'm new at these forums (by posting) and I just wanna add I apreciate everyone trying to help. I also have bought two of these HD4870 cards (Powercolor brand) and after editing one of the BIOSes with Radeon BIOS Editor and flashing both cards... cards just don't boot any more at all. They were working fine about 1-2h after the BIOS flash and just as I wanted to close the case and wanted to boot again there was just no boot... PC just dead. Fans are working but I get no screen at all.

I have an HD3870X2 running now and 1 of the HD4870 connected, trying to get these cards working again. I don't know where some of you are sugesting to check voltage on the card, all I can say right now is I used the voltmeter and it's showing around 11.99V on the pins on the card where the 6-pin PCI-E power connection is. I don't think it's the PSU, I think it's bad BIOS in the card itself... maybe the flash went corrupted (although unlikely to be the same on both cards the same way) and now it just won't get back to original... I don't know really... I feel regret to doing this right now... but what's done is done...

[erocker]

I'm glad I saw this thread! I'll refrain from flashing with RBE for now, and I suggest everyone experiencing this problem to report it in the RBE thread on www.techpowerup.com. Good luck to you all!

[xMrBunglex]

I used the voltmeter and it's showing around 11.99V on the pins on the card where the 6-pin PCI-E power connection is.

they weren't talking about the power from the PSU. Jimmyz wanted you guys to check your vGPU and vMEM to see how many volts they were getting. i don't know where the contact points are on these cards though. VR-Zone had some pictures a while back but i didn't save them.

[W1zzard]

purecain, whats wrong with your forum account at tpu? i can fix it for you np

if you want to see more 4870 bioses in the tpu bios collection you have to submit them.. takes about 20 seconds using gpu-z. dont submit modded bioses, dont spread the plague ;)

i'll try to get you guys a new winflash/flashrom

[purecain]

just woke up and seen that wizard,:rolleyes: dont worry i wont spread the plague... first thing thats made me smile since my cards died.... even my old lady's getting it in the neck... but i cant tell her why i'm being funny with her because she has no idea i've spent more money on hardware....
i would appreciate my account sorting out, that would be brilliant thakyou...:up:

i'll post as much info as you need, just let me know....

supposed to be taking my old lady away tonight so i'm not gonna be around untill monday....
but i can stall herfor a few hours....

edit- new bios tool version for winflash and atiflash for dll @tpu
unfortunatly the bios succeeds in flashing but the cards remain unusable....
got to go now but will keep checking back on my tocco....

[Edin]

edit- new bios tool version for winflash and atiflash for dll @tpu
unfortunatly the bios succeeds in flashing but the cards remain unusable

I tried it also, flashed again to original BIOS in DOS with Atiflash and still no go...

I wonder, after trying to edit original BIOS and chaning to different clock speeds for GPU and RAM the GPU-Z always shows all wrong information, for example:

Shaders: 0 Pixel / 0 Vertex
Pixel Fillrate (grayed out) Unknown
Texture Fillrate (grayed out) Unknown
Memory Type: GDDR5 ............... Bus Width: 512 Bit
Memory Size: 4095 MB .............. Bandwidth: 5497558.1 GB/s
GPU Clock: 2949673 MHz ........... Memory: 2949673 MHz
Default Clock: 750 MHz ............. Memory: 900 MHz

Sensors page shows:
GPU Core Clock: 200.0 MHz
GPU Memory Clock: 6350.0 MHz
GPU Temperature: 255.0C
Fan Speed: 100%
GPU Load: 200%
Fan Speed: 0 RPM

The card pretty much is very cold, the air coming out is like ambient temperature and heatsink is same. I don't know how GPU Load can show 200% or basically all info is pointless, I think the card very likely had some sort of permanent damage which is why even successful flash with original BIOS doesn't do anything but I so hope I am wrong about this :(

[W1zzard]

try flashing with -newbios (winflash accepts the same command line parameters as atiflash)

the numbers on gpuz are because the card is not initialized, so every register read returns 0xFFFFFFFF which results in those funky numbers. a future version of GPU-Z should detect this properly

[Edin]

Sorry W1zzard, is that to try flashing with Atiflash using this -newbios command (in which format though?) or with AtiWinFlash?

[wanky]

noobs

[W1zzard]

for both? "winflash accepts the same command line parameters as atiflash"

[Edin]

ah right...

[Edin]

Just tried Atiflash from Floppy and I used "atiflash -f -p -newbios 2 biosname.bin" command and it flashed alright but still no change... I wonder if I'm doing something wrong?

[purecain]

noobs

then help the guy, youd rather be a geniuos in idiocy...:down:

@everyone else... right i dont have too much time left but its looking like we really need someone with a working 4870 to flash the card to another working bios. if that card then fails we will know for sure that this problem is with the flashing utilities...

@edin- your doing it right, its very easy and almost impossible to get wrong.... we are going to need some advice from ati.....
i'm still holding a candle for rbe1.12 as hopefully bagzzlash or wizard will gain more insight into the effects of flashing these cards...

ati/amd technical support contact info

United Kingdom
Tel:44 (0) 1276 803299
Monday - Friday: 09:30 - 16:00 GMT

[W1zzard]

Just tried Atiflash from Floppy and I used "atiflash -f -p -newbios 2 biosname.bin" command and it flashed alright but still no change... I wonder if I'm doing something wrong?

that command line doesnt look too valid .. are you sure it flashed?

try: atiflash -newbios -f -p 2 biosname.bin

are you sure your adapter index is 2 ?

the -p and adapter index belong together

[Edin]

Tried also with atiwinflash and same thing... no change :(

Don't know if this is of any significanse but I'll mention it anyway... there was a difference though with running the atiwinflash.exe as regular aplication and flashing the BIOS like that, compared to typing the commands in RUN window... first case the flashing progress bar goes quickly to about 80% then stays there for about a minute before I get successful flash message... while typing in RUN window using "atiwinflash -f -p -newbios 2 biosname.bin" comand the progress bar indicator would continually go to end and repeat about 5-6 times I think... again I got successful BIOS flash message, rebooted and still the HD4870 is dead.

[Edin]

that command line doesnt look too valid .. are you sure it flashed?

try: atiflash -newbios -f -p 2 biosname.bin

are you sure your adapter index is 2 ?

the -p and adapter index belong together

I'll check again with command sequence, and yes it did say BIOS was successfuly flashed (I'll try again anyway) and yes this card has ID 2 (I have HD3870X2 in system as running card otherwise I can't even boot) and I checked this with "atiflash -i" command... be right back with trying to flash with different (your sugested) command sequence...

EDIT: tried "atiflash-newbios -f -p 2 biosname.bin" and again it said successful flash, rebooted and still no change...

[Sleepyzone]

purecain, whats wrong with your forum account at tpu? i can fix it for you np

if you want to see more 4870 bioses in the tpu bios collection you have to submit them.. takes about 20 seconds using gpu-z. dont submit modded bioses, dont spread the plague ;)

i'll try to get you guys a new winflash/flashrom

Is it posible the bios flash burnt something up on the card? :confused: Or maybe AMD put some evil lock on the bios chip.

[Olly_K]

Hi guys, im in the same boat as a lot of you, I've got 2 dead HIS 4870's here.

im about to try a DOS flash......but, do I need a PCI gfx card ? the only other card I have is a 9800GX2, I dont want to mess up the bios on that and have THREE cards borked !

[s1nykuL]

noobs

Very helpful, well done, DUE respect earned :shakes:

Sorry for interjecting here, I am not part of this thread nor I do not have a 4870, but this kind of contribution makes me angry.

[dehx]

Any luck guys? I read this thread after modding my bios last night, and I went ahead and shut down .. no sense in preventing the inetivitable...

My 4870 is bricked.. and I don't have a PCI card to boot with... the only one I do have is damaged, so I'm screwed.

Wizz.. what is your opinion? Will any of those new flash tools help?

[Dragy2k]

noobs
atleast hes got the right name ^^

good luck guys hopefully you will have it cracked soon.......maybe some1 at ati will read and point something out

[AriciU]

Did anyone's card die without him messing with the bios and stuff? Is there a problem with them as they come or only if you try to OC thru bios update? This is apparently starting to become a trend unfortunately :(

[wiak]

bah stop bitcing about 4870 going out in less than 24hours when its YOUR own foult that didnt get or read the memo about
"flashing modified bios = asking for truble" ;)

w1zz to teh rescue!

anyway am getting 4870 on monday

[AriciU]

bah stop bitcing about 4870 going out in less than 24hours when its YOUR own foult that didnt get or read the memo about
"flashing modified bios = asking for truble" ;)

w1zz to teh rescue!

anyway am getting 4870 on monday

Unfortunately it seems to be happening out of the blue -> http://www.xtremesystems.org/forums/showpost.php?p=3099037&postcount=17

and

http://www.xtremesystems.org/forums/showpost.php?p=3098971&postcount=3956

I don't think OC'ing thru software should cause this :banana::banana::banana::banana:ing crap. Let's see who else get's "lucky". :(

[wiak]

Unfortunately it seems to be happening out of the blue -> http://www.xtremesystems.org/forums/showpost.php?p=3099037&postcount=17

and

http://www.xtremesystems.org/forums/showpost.php?p=3098971&postcount=3956

I don't think OC'ing thru software should cause this :banana::banana::banana::banana:ing crap. Let's see who else get's "lucky". :(
yes got a point, software oc is alot safer than bios flash, and well i hope you guys get them working again, are you guys sure if they arnt overheating?

[Marvin_The_Martian]

yes got a point, software oc is alot safer than bios flash, and well i hope you guys get them working again, are you guys sure if they arnt overheating?

What's their thermal envelope? I can imagine that great increases in shader counts might mean it will be more sensetive to cooling? Naw come on, if it were so simple Ati wouldn't have released them they would have noticed during QA..

*looks at thread tittle*...

maybe not?

[SNiiPE_DoGG]

well i am positive it was a ram issue in my case (symptoms linked by AriciU) because even when the card that died was working properly ATItool artifact scan gave me errors ever 3 sec's or so. my 2nd 4870 does not have this problem at all so i figure the card that died was bad all along :\

[Marvin_The_Martian]

What I gather is that it might be something about crossfire to since it seems to happen when people change profiles to but I'm ony guessing..

[ceevee]

Unfortunately it seems to be happening out of the blue -> http://www.xtremesystems.org/forums/showpost.php?p=3099037&postcount=17

and

http://www.xtremesystems.org/forums/showpost.php?p=3098971&postcount=3956

I don't think OC'ing thru software should cause this :banana::banana::banana::banana:ing crap. Let's see who else get's "lucky". :(

and this is why i didnt go with xfire. i used to have 2x x1950xtx in crossfire a couple years ago and had nothing but problems.

every time AMD/ATI comes out with a new gen of cards there are people that proclaim that all the problems are fixed and everything is perfect now.

but every time there are more problems.

anyway hope they fix this stuff soon because the cards seemed really great, at least on paper.

[dehx]

Oh, the card was f-in amazing.. I got 9027 points in 3D mark Vantage... it beats the crap out of my OC'd 3870 crossfire... and thats just one 4870.

[dehx]

ok, I put the card in an athlon 64 x2 system w/ a 550W PSU... set the board to do onboard gpu, and it boots with the card in the system... but its not in device manager or anything... atiflash in DOS doesn't even detect it. I beleive the card to be stone cold dead.

[Kensek]

At least ATI Flash sees my Dead Sapphire HD4870s. I backed up the original BIOS before card 1 died.

I reflashed it into the first card also. But, the fan only spins still and there is no Video or POST happening.

That's when I have another model ATI VGA in the Primary PCI-e slot.

[Theta]

Any new information?

Got 4 dead here, as well.

To fuel the fire and speculation, both sets we have died in Crossfire mode. (may be a trend from the looks of it)

[Garrett]

You have 4 dead 4870 cards... omg... :eek:
Just great... I've been looking for 2 of those cards since I want some new toys again... NVIDIA is ridiculously expensive and ATI cards just die for no apparent reason... just perfect :down:

[Edin]

Well aint this just great... I'm loosing any hope these cards will run again since I can see now some of you have dead cards even without flashing the BIOS... very bad neews indeed... looks like it'll be RMA time for me monday morning...

[Theta]

Not trying to stir the pot at all, just curious to see what progress has been made.

Playing with 4850s in the mean time...

Oh, and if this helps... ALL four have three red lights that illuminate upon power on, not all 4 of them.

All exhibit the same no-POST behavior, but can be seen by ATIFlash as a secondary adapter (just checked for the hell of it).

EDIT: HIS brand, BTW.

[AriciU]

Hope all of this gets sorted in one week when i'll buy two of them. It sucks to be a guinea pig.

[Papu]

well none of these problems came up when they were being tested by review sites?

[jimmyz]

This is indeed a bummer of a thread, but keep in mind a problem caused by a bad flash can be corrected by a proper flash. in a little while the software should be out to fix the cards up, more than likely it is going to involve a pci card boot and flash or similar, but it should be correctable.

I feel the culprit is not realising that you can't add voltage to the cards through bios,( only about .01v) so the speeds won't boot. then when the card gets powered down and tries to reboot it lacks the voltage to do so properly. could someone please report back the vcore on boot of a "dead" card. and some hi-res photos of the bottom of the card in the voltage area..

[keiths]

Have there been any cases of this happening to 4850s as well?

[xMrBunglex]

it sounds like these guys need instructions on how to do a pencil mod on these cards so they can see if they boot. then they can just erase the pencil mod after they flash back to a working BIOS.

and for the record, i have two 4870's in CrossFire and everything is working like a champ. i haven't flashed and i haven't tried to OC yet, but i haven't exactly been going easy on them. i've been playing Crysis and CoD4 pretty heavily during our Southern California heatwave and the cards are doing great.

i have two Sapphire 4870's and it looks like they have different BIOSes. the top card will let me underclock its memory to 810 and the bottom card won't go below 900. the top card also idles 5-7° hotter than the bottom card, but that is probably just due to heat rising.

my top card's BIOS is 011.003.000.001 and i don't know how to find my bottom card's BIOS number. i can't check it in CCC.

[Miwo]

I am thinking DDR5 is the culprit

Remember that you are pushing the memory twice as hard with every MHz vs the DDR3 on the 4850s
900Mhz = 3.6GHz
1000MHz = 4.0GHz
1100Mhz = 4.4GHz

Guess killed his 4870 almost instantly by going to 1200MHz
http://www.xtremesystems.org/forums/showthread.php?t=192696

Thats a 33% increase in memory speeds without additional volts. Be reasonable! So perhaps either a quick death @ 1200MHz, or a slower death @ 1100MHz like I see many people doing?

[Kensek]

I don't think so. I only Oc'd 1 card to 790/1050 for 15 minutes.

The second card was never OC'd. On the initial bootup with both cards inserted and powered up and with the cF bridges installed XP would not load.

Hence, I removed 1 card thinking a Corsair 750TX was insufficiently powering the cards. However, after this neither card never worked again.

Though they can be seen through ATI Flacs when another ATI model card is used as the primary.

[keiths]

Been trying to determine GPU vs memory performance in this thread: http://www.xtremesystems.org/forums/showthread.php?t=192690 Could use some more results at higher resolutions and aa to tax the memory, as so far it's GPU dominated, with memory not having any impact at all.

[Kensek]

It's like a circuit breaker has been tripped in these HD4870s. And someone needs to devise a way to reset it.

AS for OCing - I use to run my HD3870s at 890/1360 and never had a problem. But as you eluded to, that wasn't GDDR5 ram.

But reviews say GDDR5 Ram is capable of 5600, which is 1400.

[Edin]

I didn't overclock my two cards either...

[GoldenTiger]

I still don't get why even flashing back properly won't fix this... I just went out and bought a 4850, going to send the 4870 back if no solution is found shortly. I was offered a full refund for it, while I'd prefer to have a 4870 I'm going to just stick with the 4850 if no one figures this out by the time I send it back :(. Monday probably, 4850 isn't as fast as I'd like but *shrug*.

[halo112358]

bah stop bitcing about 4870 going out in less than 24hours when its YOUR own foult that didnt get or read the memo about
"flashing modified bios = asking for truble" ;)

w1zz to teh rescue!

anyway am getting 4870 on monday

It's not just people who've flashed bios, random cards are dying too.

[mr2sw21]

just wonder is it now the card killed the mobo or just the card dead? And how many got a dead card now?

[jimmyz]

It's not just people who've flashed bios, random cards are dying too.

I 've been reading up all day I feel now the GDDR5 is going belly up most of the people have clocked up in ccc or flashed. I think the ram is dying, could be way off but I wouldn't overclock one until we all know for sure.

I also updated the 4850 bios first post to reflect that it most likely will kill a 4870.

Even the people I talked with outside of XS are reporting many dead cards... surely something besides the flash is killing them. most outside of this forum never saw the flash option.

I wish someone would reply who has a MM and a camera. I think we can eliminate some possibilities if we can get some measurements. and short of me bouncing a check to get one I am unable to help, or someone could mail me a dead card to check.

[Kensek]

The Mobo is fine. It can run all of my other VGAs still.

But the VGA cannot dispay or POST. Tried the dead HD4870s in my BlackOps, P5Q3 and IP35 Pro.

[Edin]

Well I've given up on these :banana::banana::banana::banana: cards (can't believe they died just like that) I took'em out of my system gonna pack them up and off back they go where they came from... it :banana::banana::banana::banana:s me now even more that I can't get CCC to start now at all even with just HD3870X2 installed...

[jimmyz]

The Mobo is fine. It can run all of my other VGAs still.

But the VGA cannot dispay or POST. Tried the dead HD4870s in my BlackOps, P5Q3 and IP35 Pro.

We are all still at a loss as to the cause for this.I mean we know ATI must have tested their own CCC yet many cards have died just using it, and some even stock. all we can do right now is try different things and see what may be causing it, so we know where the issue is.

if anyone with a dead card is near Richmond IN I will drive to your house if needed to get a look at the card, please reply.

Unless we eliminate some things from the list we are just spinning our wheels.

has anyone tried flipping the pal/ ntsc switch on the card?
has anyone tried the other port for the monitor?
has anyone tried booting with no power to see if the pwer message shows up?
has anyone tried the command line W1zzard posted a page back?
has anyone tried anything I left out?

any more advice? all input is needed at this point and be honest about clocks and such , we aren't here to judge we just want to figure this out so you guys don't lose your cards. I know as well as anyone what that is like.

[ghost101]

Any new information?

Got 4 dead here, as well.

To fuel the fire and speculation, both sets we have died in Crossfire mode. (may be a trend from the looks of it)

For all 4 to die like that, it must mean that they are failing under certain conditions. For them to die together independently of each other is very unlikely. As you mention, it could be due to crossfire-x, or is there something specific to you?

[Kensek]

I tried - "has anyone tried the other port for the monitor"? Still No Display or Post

[Edin]

My gut feeling now suspects sticker "Made in China" on the cards and such low price could be poor(er) quality components chosen for the card and who knows what else... I am so disapointed in this...

[jimmyz]

I tried - "has anyone tried the other port for the monitor"? Still No Display or Post

could you try all the others, none should cause any problems.

[jimmyz]

4850's have much cheaper components and are also made in china, I haven't heard of a dead one yet.

[SimBy]

My gut feeling now suspects sticker "Made in China" on the cards and such low price could be poor(er) quality components chosen for the card and who knows what else... I am so disapointed in this...

Everything is made in China today my friend :rolleyes: HD4850 also and it has no issues.

I don't wanna rub in the salt but it is your fault. You flashed the cards with edited bios and the result is a dead card.

We can debate if the card would die anyway even without the flash. I don't think it would, but I sure hope you guys can figure it out :shrug:

[Edin]

Ah what ever... if it was the flash I think card would stop working straight after the flash, not few reboots and testing after... and I said it's just my gut feeling, no scientific reasoning on it ;)

Anywho, if I canget it working gain I'm not gonna return it (logically) and if I can help I would like to so I took a high res screenshot of the back of the whole card... image size is 3.15MB 3888x2592 pixels... I have a multimeter so if someone need me to send him the image I'll email and if they can point me out what points on the card to check I'll measure it...

[jimmyz]

Ah what ever... if it was the flash I think card would stop working straight after the flash, not few reboots and testing after... and I said it's just my gut feeling, no scientific reasoning on it ;)

Anywho, if I canget it working gain I'm not gonna return it (logically) and if I can help I would like to so I took a high res screenshot of the back of the whole card... image size is 3.15MB 3888x2592 pixels... I have a multimeter so if someone need me to send him the image I'll email and if they can point me out what points on the card to check I'll measure it...

email to jimmyz@insightbb.com


same for any of you. please contact if you need help, I will do what I can to help out.

[Kensek]

has anyone tried flipping the pal/ ntsc switch on the card?
has anyone tried the other port for the monitor?
has anyone tried booting with no power to see if the pwer message shows up?
has anyone tried the command line W1zzard posted a page back?

1) I see No switch on the card
3) Tried No 6-pin hookup and all i GET is a BIOS beeps. It is either 1 long and then 3 shrt beeps. Or 4 short. But the first one is only a little longer than the last 3.

What command line wizzard wrote?

[cantankerous]

This is why I always cringe when I see cards using this Quimodo crap. I remember hearing of all the issues this brand of ram caused on the 8800GT cards on release. They are just no where near as robust as Samsung branded ram. Once the 8800GT starting using Samsung ram the problems were gone. ATI should have stuck with Samsung ram I feel. Just saying.

[Edin]

Just now I plugged one of the HD4870's in the system (HD3870X2 is still set in primary PCI-E slot and monitor is connected to it and CAT's 8.6 are installed (but for some reason CCC won't start now at all???)) and I have no 6-Pin PSU cables connected to it... system booted into windows... I got no warning about cable not being plugged in... the HD4870 hardware got recognised and windows started installing the driver for it... and then whole system just stopped and monitor screen got black... same as always... I have to now boot in Safe Mode, disable the HD4870, to be able to boot into windows again... or uninstall Catalyst drivers... what a pain...

EDIT: like all of you I also get only 3 red lights lit up on system restart (on tha image light that doesn't lit up is labeled D601 - second from top - located right side, right next to the 6-Pin PSU connectors) and this seems to regardless if I have the PCI-E power cable connected or not...

[Dragy2k]

:) cross fingers

Long live Nvidia
RUN!
not needed and not usefull..........time for a troll cull !!
any news on possible fix im getting my single 4870 soon!

[jimmyz]

Just now I plugged one of the HD4870's in the system (HD3870X2 is still set in primary PCI-E slot and monitor is connected to it and CAT's 8.6 are installed (but for some reason CCC won't start now at all???)) and I have no 6-Pin PSU cables connected to it... system booted into windows... I got no warning about cable not being plugged in... the HD4870 hardware got recognised and windows started installing the driver for it... and then whole system just stopped and monitor screen got black... same as always... I have to now boot in Safe Mode, disable the HD4870, to be able to boot into windows again... or uninstall Catalyst drivers... what a pain...

EDIT: like all of you I also get only 3 red lights lit up on system restart (on tha image light that doesn't lit up is labeled D601 - second from top - located right side, right next to the 6-Pin PSU connectors) and this seems to regardless if I have the PCI-E power cable connected or not...



can you upload a pic of the other side also naked please I'm not seeing the chip I need.

[SNiiPE_DoGG]

This is why I always cringe when I see cards using this Quimodo crap. I remember hearing of all the issues this brand of ram caused on the 8800GT cards on release. They are just no where near as robust as Samsung branded ram. Once the 8800GT starting using Samsung ram the problems were gone. ATI should have stuck with Samsung ram I feel. Just saying.

very true


so guys i earlier reported that one of my HIS 4870s has major artifacting issues on desktop but is still "functional"

now i have even better news! my other "working card" has now started to exhibit the same symptoms you all are having... no post and my DFI x38 poster LCD gives me post code 26 in the post where jimmyz asked all those questions i tried all those things and no luck.

when i have my 3870 in my systems posts and boots for about 30 seconds then goes black screen

here is the COMPLETE list of things i did to the card that will not post:
-used ATIWinflash and RBE to edit the fan profile on my card, never changed and clock settings or voltages
-Used ATItool to check for artifacts and temp issues
-clocked the gpu to 790, no isuues with that (many reboots at that frequency)
-clocked the ram to 1060 ----> after i confirmed no artifact issues at this speed i rebooted my computer, this is when post code 26 started happening and all went to hell.

Any info you need jimmy, maybe a high res pic of my card (even tho you have one already, i have an EOS 1ds Mk III so i can pretty much get the most detailed pics around of whatever you want.


update on the fragmenting card: i believe this card to be truely defective from the start when i put it in for some fixing attempts to day it reached incredibly high temps without ever reaching the desktop (nearly burned my hands, must have been over 100c) and i truly cannot see how it ever passed QA :shakes::mad:

[Edin]

can you upload a pic of the other side also naked please I'm not seeing the chip I need.

comming right up...

[Calmatory]

You have 4 dead 4870 cards... omg... :eek:
Just great... I've been looking for 2 of those cards since I want some new toys again... NVIDIA is ridiculously expensive and ATI cards just die for no apparent reason... just perfect :down:
Guess BIOS flashing is "no apparent reason". :rolleyes:

There is a reason why it voids warranty, just like overclocking and voltmodding does.

[SNiiPE_DoGG]

Guess BIOS flashing is "no apparent reason". :rolleyes:

There is a reason why it voids warranty, just like overclocking and voltmodding does.

read what jimmyz said, many people who dont even know how to flash a bios are getting dead cards, flashing is not the cause and it certainly doesn't fix it

[keiths]

Except, as has been repeatedly pointed out, there have been cards that haven't been modified or overclocked that have met the same fate.

[Big.Wayne]

the thread need the name changed

if it was the flashing that caused the problem, the title of this thread is a little unfairYup that title could cause unnecessary worry amoungst the faint hearted who just spent they last buck on a shiney new HD4870! :shocked:

"4870 randomly bit the dust in less than 24 hours :explode2:"

I don't call a BIOS flash random, although there may be other reasons, the flash has gotta be the prime suspect. As they say "Mess with the Bull, You’ll Get the Horns"


It's not just people who've flashed bios, random cards are dying too.
Well as we all know some computer hardware has a habit of just dying with no apparent reason and I expect a very small percentage of HD4870's will give up the ghost pretty quickly.

I do sympathize with you to some extent but from what I have read it seems you may have had a hand in the demise of your new card. I feel really sorry for the guys/girls who receive HD3870 cards that are DOA!

Personally I would never dream of flashing a brand new card? and I suggest to anyone reading who likes an easy life to let the card run at stock for a few days at least before abusing it :slapass:

Anyway good luck in sorting things out, worst case scenario is you will have to RMA the card and miss out on a little gaming or benchmarking, no one dies!

[jimmyz]

read what jimmyz said, many people who dont even know how to flash a bios are getting dead cards, flashing is not the cause and it certainly doesn't fix it

At this point even speculation is fine, any input might be the one that triggers a solution, also any bump that keeps this thread being seen is good, until we have solid answers any guess is welcome. and the more people who see this before flashing the better. it may be completely unrelated but it is definitely a prime suspect right now, not so much the flashing but the raising of clockspeed. like a cop solving a murder we need facts from witnesses, please explain all symptoms and be honest.

[dnottis]

Full size pics of HIS 4870 can be found here -swapped the cooler for a TT Duo Orb running great.


http://3dxtreme.net/other/HIS%20HD4870/full%20size/IMG_5286.JPG

http://3dxtreme.net/other/HIS%20HD4870/full%20size/IMG_5293.JPG

[Edin]

Here imags of HD4870 from bottom and top side (click on image for full size)

http://www.3sqn.com/members/images/HD4870_Top_Small.jpg (http://www.3sqn.com/members/images/HD4870_Top_Full.jpg)

http://www.3sqn.com/members/images/HD4870_Bottom_Small.jpg (http://www.3sqn.com/members/images/HD4870_Bottom_Full.jpg)

[Solus Corvus]

Is everyone with dead cards having the LEDs light up differently then under "normal" opperation? Do we know what the different LEDs represent?

[101]

Oh, and if this helps... ALL four have three red lights that illuminate upon power on, not all 4 of them.


Sorry, but that comment cracked me up...
http://img212.imageshack.us/img212/1315/redringofdeathto2.jpg

[jimmyz]

thank you for the picture un fortunately the fb pin appears to go through the board. I will need to look at it with a little more sleep under my brain...lol. I will check it all out in the morning, good news is the chip is a legendary Volterra, the same one as the hd2900 and the 3870X2 so we know the pin-out at least. there is a small through hole near the second pin from the bottom left that should be the FB that is where to check resistance. I will be on in the morning armed with coffee, check back then (about 8hrs from now) I know you are in australia.

I feel I have a good grip on the cicuit layout but I'm no engineer, seems there is some components missing on one of the 2 v-dimm circuits which is odd. I will think about it overnight and be on in a few hours. good news is I don't see any physical damage.

BTW you could host on imageshack and just post the thumbnail, then the page will load faster, and we will be able to get the full size by clicking on the pic. but I do love the clarity , nice pics.

[101]

In all seriousness I REALLY hope this is not some sort of temp cycling issue associated with the RoHS compliant BGAs. Anyone have a card fail that was only ever water or high end air cooled? Perhaps the stock cooling is grossly inadequate and holding these parts back. If the core temps are easily hitting 80ºC+ there is probably a fair amount of mechanical stress on the package interconnect.

[mr2sw21]

Full size pics of HIS 4870 can be found here -swapped the cooler for a TT Duo Orb running great.


does that mean once we get aftermarket heat sink the problem solve? Any one try it on their "dead card"?

[twohype]

I'm pretty sure that's what mine does - I already filed an RMA with the egg, this crap is going back in the box. Maybe I'll buy another when they've sorted this out, maybe not.

If someone flashes a working card and volt mod's it with a modified bios from a tool which ATI "NEVER" indicated was safe for a 4800 series card - Why is that ATI's or NewEgg's issue? :rofl: If you can sucker NewEgg :yepp: more power to you :shrug: Of course I'm not saying that you modified your card since obviously yours was DOA:up:

[Edin]

does that mean once we get aftermarket heat sink the problem solve? Any one try it on their "dead card"?

The card is simply not running at all.. the only thing working on it seems to be the fan... the heatsink (stock) is completely cold (as metal is sold on room temperature) so it doesn't matter what heatsink you put on it...

[Edin]

If someone flashes a working card and volt mod's it with a modified bios from a tool which ATI "NEVER" indicated was safe for a 4800 series card - Why is that ATI's or NewEgg's issue? :rofl: If you can sucker NewEgg :yepp: more power to you :shrug: Of course I'm not saying that you modified your card since obviously yours was DOA:up:

The card was idling around 80C for me with stock fan/heatsink/BIOS settings so dont you think that's a bit high? Also I've done BIOS flashing quite a few times before for motherboard and VGA and never had something like this happen... also the flashing tool I used had a statement it suported HD4850 and HD4870 otherwise I wouldn't have used it... also the card was working fine even AFTER the BIOS flash and stoped working only 1-2h later... and you know, it would be more helpfull if you kept your useless comments for yourself... you got a working HD4870? If yes and you are happy with it then good for you, if not... don't comment on things you haven't had experience with... maybe if you did get one and it would die on you "justlike that" you would be here talking other way around...

[ghost101]

The card is idling around 80C for me with stock fan/heatsink/BIOS settings so dont you think that's a bit high?

ATI picked the fan profile and if they dont think its high, nor do I. the majority of people who buy these as well as OEMs will sell this and no one will ever know what the temperature is and the card will live out its life. It is high if you plan on overclocking, but then...



Also I've done BIOS flashing quite a few times before for motherboard and VGA andnever had something like this happen... also the flashing tool I used had a statement it suported HD4850 and HD4870 otherwise I wouldn't have used it... also the card was working fine even AFTER the BIOS flash and stoped working only 1-2h later

Doesnt matter. Your warranty is null and void and so isnt ATIs problem. Also the tool clearly stated that hd4xxx support is experimental.



... and you know, it would be more helpfull if you kept your useless comments for yourself... you got a working HD470? If yes and you are happy with it then good for you, if not... don't comment on things you haven't had experience with... maybe if you did get one and it would die on you "justlike that" you would be here talking other wy around...

Erm, he can say what he wants.

[Theta]

If someone flashes a working card and volt mod's it with a modified bios from a tool which ATI "NEVER" indicated was safe for a 4800 series card - Why is that ATI's or NewEgg's issue? :rofl: If you can sucker NewEgg :yepp: more power to you :shrug: Of course I'm not saying that you modified your card since obviously yours was DOA:up:

Okay, this BS is not helpful at all.

Some of us are trying to get to the bottom of why this is happening. In some cases, this has NOTHING to do with flashing. We had 4 fail on us, and they were simply put into Crossfire mode.

I appreciate that some may have modded, etc... But stop threadcrapping for the sake of us that actually would like to get this solved.

[ghost101]

Okay, this BS is not helpful at all.

Some of us are trying to get to the bottom of why this is happening. In some cases, this has NOTHING to do with flashing. We had 4 fail on us, and they were simply put into Crossfire mode.

I appreciate that some may have modded, etc... But stop threadcrapping for the sake of us that actually would like to get this solved.

I dont think he meant to do that. But when some people do flash the bios and then file an rma it results in higher prices for everyone. So I agree with him.

But if you genuinely didnt do anything, an rma will fix things surely?

[Theta]

I dont think he meant to do that. But when some people do flash the bios and then file an rma it results in higher prices for everyone. So I agree with him.

But if you genuinely didnt do anything, an rma will fix things surely?

I do understand that, but please realize that an RMA won't help one bit.

Our distributors are out of stock, and NewEgg isn't exactly stocked fully on these.

I am interested in seeing what can be gleaned from all of this...

I do want to point out that several of us were not modding the BIOS, however. ;)

[Edin]

Here one though on my end on maybe what made my cards die... when I flashed the BIOS I did not change any GPU/MEM clock or voltage values, just the MIN and MAX temperature value for the fan profile...the fans were working much faster after the BISO flashing and the GPU idle temperature reported by CCC was much lower (abour 47C insteadof inital 79C). I flashed the second card afterwards also and its fan was spinning faster also and I could feel the aircomming at the back of the card was getting colder and cards heatsink (the big heavy heatsink acting metal plate) was no longer burning my finger (initially I could not hold my hand on it for more than 1-2sec, it was that hot)

Then I modified the BIOS again and set the MIN temp just bit higher as the fas were too noisy now even at idle... after second flash I was happy with result, shut down windows, switched off the PSU in power sitch and connected big 200mm fan mount on side of the case and closed it... wanted to boot again and then it was dead.

Right now, the fan on the card is spinning quite fast. I tried multiple times to flas with original BIOS and every time it said it was a successful flash. I think my only hope of card recovery is new flash utility... just don't know if I can wait very long, and who knows maybe the cards are not recoverable at all any more... I think W1zzard is our only hope :)

[ripken204]

how much heat can gddr5 take?

we should assume that ATI is telling the truth that the 4870 gpu can take 80C idle, but woudlnt that mean that the gddr5 would be under ALOT of heat as well?
if some people have reported that just touching the heatsink burns their finger.. then imagine what the ram feels like..

[jimmyz]

Here one though on my end on maybe what made my cards die... when I flashed the BIOS I did not change any GPU/MEM clock or voltage values, just the MIN and MAX temperature value for the fan profile...the fans were working much faster after the BISO flashing and the GPU idle temperature reported by CCC was much lower (abour 47C insteadof inital 79C). I flashed the second card afterwards also and its fan was spinning faster also and I could feel the aircomming at the back of the card was getting colder and cards heatsink (the big heavy heatsink acting metal plate) was no longer burning my finger (initially I could not hold my hand on it for more than 1-2sec, it was that hot)

Then I modified the BIOS again and set the MIN temp just bit higher as the fas were too noisy now even at idle... after second flash I was happy with result, shut down windows, switched off the PSU in power sitch and connected big 200mm fan mount on side of the case and closed it... wanted to boot again and then it was dead.
Right now, the fan on the card is spinning quite fast. I tried multiple times to flas with original BIOS and every time it said it was a successful flash. I think my only hope of card recovery is new flash utility... just don't know if I can wait very long, and who knows maybe the cards are not recoverable at all any more... I think W1zzard is our only hope :)

RBE has a warning about using a fresh bios to mod each time, don't re-use the same bios, mod a fresh copy each time. in order to correct the checksum the program needs to change other entries in the bios in order for the checksum to match. so always use a fresh bios each time you create a modded bios. just make a copy of the original or dl from the TechPowerUp VGA bios database.

[Edin]

Yeah I made that bit of an error as I didn't use original BIOS to made the second flash. I don't know if that's what caused my cards to die, I wouldn't be surprised if it did make some problem, but all other guys didn't do this and they still have same problem. I did download original BIOS file from TechPowerUp VGA bios database and I flashed both card with it on multiple occasions (DOS and windows) but yeah, as you know already it didn't fix it :(

[Calmatory]

Jimmyz, go get some sleep man. :D

All the defect cards are from HIS?? Or are there other brands aswell? Asus? Club3D?

Has anyone tried to contact AMD/ATI about this issue yet? What about brands?

[Edin]

Powercolor brand here...

[jimmyz]

Yeah I made that bit of an error as I didn't use original BIOS to made the second flash. I don't know if that's what caused my cards to die, I wouldn't be surprised if it did make some problem, but all other guys didn't do this and they still have same problem. I did download original BIOS file from TechPowerUp VGA bios database and I flashed both card with it on multiple occasions (DOS and windows) but yeah, as you know already it didn't fix it :(

I doubt the flash was involved period. I think the ram or some other cause is to blame, lots of stock cards are dead too.

[ripken204]

the flash may have just caused the problem to come out quicker.
this really needs to be resolved, you guys with problems should call amd and complain.

[Calmatory]

the flash may have just caused the problem to come out quicker.
this really needs to be resolved, you guys with problems should call amd and complain.

Only those who didn't overclock or flash bios.

[halo112358]

I wish someone would reply who has a MM and a camera. I think we can eliminate some possibilities if we can get some measurements. and short of me bouncing a check to get one I am unable to help, or someone could mail me a dead card to check.

What do you want? I've got both handy. Sorry I didn't reply sooner, I've been kinda busy the last couple of days.

My card is a Sapphire by the way, I've got a powercolor en route too. I tried to reach ATI support on Friday, but they close at 12:30pm pacific time on fridays :p.

[Theta]

My company is running a number of stock 4870s, and so far, 4 have died. Furthermore, several other users here did NOT flash yet, and still had this happen. My posts have reflected these facts... Please contact Sylvan if you read the "entire thread" - you need a comprehension tweak.

Thanks for your addition... Here's the door. :down:

[Wesker]

I've got two HD4870's on order, and I'm entitled to my own opinion and speculation.

[Theta]

Bump for back on topic!

Morning W1z, any further insight? :)

[Wesker]

Bump for back on topic!



OK, then can I ask you how your 4 cards died?

Did you have them setup in CrossFireX? Also, did the cards die after a cold boot?

[NKD]

I Have been reading the post, and I swear instead of finding a solution, there are some users that are not working on solving the problem.

1. who cares if he flashed the bios, we need a solution, so if you can't help please don't turn this thread into a war.
2. Newegg or Powercolor or whatever company is not going to be in any damage if it is indeed a bad bios flash(just like you said), the manufacturer will probably just flash it back and rebrand it.
3. when someone rma's a card it is not a scam, alright the man paid 300 plus so what if he flashed the bios, no one a noob here. I have flashed numerous nvidia cards and even I have even flashed an 8800 ultra with the 8800gtx bios, why? because I had an ultra and I got a gtx a long time ago for dirt cheap. and the cards are no longer in sli, but they work like a charm.
4. if it is just a bad bios it the card should have no damn problem when you flash it back to original bios.
5. there is a lot of hard work that is involved in this thread, so please don't try to hijack the thread with useless comments. there is no solution in those comments.
6. It does not take a rocket scientist to flash a vga bios, so stop blaming the guy.
7. may be it is the bios tool that we need that would be able to do a good flash, so please chill, and let the people in this thread work towards a solution which in turn is only going to help us.

yes my powercolor radeon hd 4870 just died as well, and yes I flashed the bios, but it worked for two days, so yes it could be the bios tool, gddr5 god knows what, no problems gaming, and I only adjusted for fan speed that is it, and it worked fine, was idling at 55c and load 63c, so for it to work for 2 days and just take a crap does not make any sense, plus it was running hell of a lot cooler, and It was overclocked to 790/1050 with autotune, because I just wanted the damn thing to run cool, I know what I am doing so I don't need anyone to tell me that I am stupid, and yes I did do a stupid thing, do I think it could be because I flashed it with edited bios, probably, may be not. but running at better temps than stock setting for two days and for it to not just boot, it is pretty clueless.

I might have had a bad flash and or the card might be just dead, I tried to flash back to original through dos, it said the flash went fine, but still the same thing. It is useless telling me that I killed my card, because I am man enough to admit that I did, if it is proven that I did it. But that does not mean that we don't need to know the culprit here, all I want to know why it worked for two days and than decided not to boot, all this with 25c less temps. Please help and don't flame because it does not help, and yes I know how to flash a card.

My sig is old so don't even look at it, lol.

[Wesker]

My thoughts were that early BIOS files for the HD4870 were the culprit behind the dead cards, but Theta says he has 4 dead cards without any BIOS flashes applied.

Also:
http://www.xtremesystems.org/forums/showpost.php?p=3099037&postcount=17

EDIT: Theta had 2 sets of 2 HD4870's in CrossFire.

Could CrossFire also have something to do with these dead cards?

[Theta]

OK, then can I ask you how your 4 cards died?

Did you have them setup in CrossFireX? Also, did the cards die after a cold boot?

Please do note that this was all posted previously in this thread, but I will post it again for you...

Each set of 4870 HIS cards died shortly after being placed into CrossfireX mode, and exhibit the now famous no-post, 3-LED scenario.

So far, these were the only two sets that were put into CrossfireX.

Not flashed, not overclocked, not overvolted... So...

Back to the topic at hand. Hopefully there is an answer for this, because I just don't want to believe it's horrible QA related to 4870 Xfire. ;)

After we hear back from W1z, I may just flash one of our tested working 4870s to see what happens.



Could CrossFire also have something to do with these dead cards?

I think that's the root cause for non-flashed deaths, looking at several posts from non-modded failures.

[NKD]

I doubt crossfire had anything to do with it, because I really don't know how that would be possible, but than again anything is possible, but yeah my card had the same bios, I just edited the fan speed, worked for two days and than the 3 led power. the second led from the left does not light up.

[Theta]

I doubt crossfire had anything to do with it, because I really don't know how that would be possible, but than again anything is possible, but yeah my card had the same bios, I just edited the fan speed, worked for two days and than the 3 led power. the second led from the left does not light up.

Yeah, I find it hard to believe as well...

The only thing that is the same across the playing field of errors here is the power LED similarity...

We're all in the same boat, somehow... Let's hope the 3-LED startup does not mean the gDDR5 is dead, etc. ;)

[bill_d]

I doubt crossfire had anything to do with it, because I really don't know how that would be possible, but than again anything is possible, but yeah my card had the same bios, I just edited the fan speed, worked for two days and than the 3 led power. the second led from the left does not light up.

do the 3 led just flash on start up or stay on, my card is stock and working fine but same leds just flash on start up about 1/2 a second flash on left to right 1, 3 and 4
2 stays off

[eleeter]

My company is running a number of stock 4870s, and so far, 4 have died.

What kind of company, and what are you using them for?

[Crankybugga]

1. who cares if he flashed the bios, we need a solution, so if you can't help please don't turn this thread into a war.
Well i for one care!
Like many others we need to know why this is going on, and if people are fudging their 4870s with a pretty stupid bios flash, then that is very rellevent.




2. Newegg or Powercolor or whatever company is not going to be in any damage if it is indeed a bad bios flash(just like you said), the manufacturer will probably just flash it back and rebrand it.
You are joking right?




3. when someone rma's a card it is not a scam, alright the man paid 300 plus so what if he flashed the bios, no one a noob here. I have flashed numerous nvidia cards and even I have even flashed an 8800 ultra with the 8800gtx bios, why? because I had an ultra and I got a gtx a long time ago for dirt cheap. and the cards are no longer in sli, but they work like a charm.
Not sure where this is going...



4. if it is just a bad bios it the card should have no damn problem when you flash it back to original bios.
So why havent you done that?


yes my powercolor radeon hd 4870 just died as well, and yes I flashed the bios,
Now i see...
Why was i not surprised to finally get to the bottom of your post to read you killed your vga..and oh....you flashed the bios..
Be interesting to see what the ratio is of dead 4870s flashed and not flashed, because at this stage id say the flashers are leading :clap:

[NKD]

Well i for one care!
Like many others we need to know why this is going on, and if people are fudging their 4870s with a pretty stupid bios flash, then that is very rellevent.



You are joking right?



Not sure where this is going...



So why havent you done that?


Now i see...
Why was i not surprised to finally get to the bottom of your post to read you killed your vga..and oh....you flashed the bios..
Be interesting to see what the ratio is of dead 4870s flashed and not flashed, because at this stage id say the flashers are leading :clap:

now was all that really necessary, I really wanted to point out that yes I killed the card if that makes you happy, it worked for two days and than it died, now if you can't explain why it died after two days, and tell me what to do, I would be more than happy, it is no help pointing to people for what they already know of. I really don't know what you are getting out of this, I never blamed anyone but myself so for you to come out and point that out like I did not admit it, does not really make any sense. I will not respond to your post again, and you should not really respond to mine if you dont wanna help.

[NKD]

this is the first time after two years with in my last flash that a card worked like a champ for two days and than just died. that is all that makes me suspicious. I flashed the card does not mean that it was a bad flash, I never had artifacts and I was gaming for a few hours, no freezing or anything. my last flash was 8800 utlra to a gtx and that card still works.

I have flashed cards for my freinds with OC'ed bioses and they never had any problems. But I am still open to any conclusions, I guess that is just the way it is.

and the leds on the hd 4870 do not stay on, they flash when I turn on my computer.

[Martin.v.r]

I RMA a 1000W Antec Quattro it kill,s a 3870x2 / 4850 wrong W/AMP on the PSUen???
death after restart,again graphics error before this.

[SimBy]

1st of all you should all spread the warning around for ppl NOT to either flash, overclock or do anything non stock. Thats the least you can do.

Then if the cards continue to die even stock, things could get more interesting.

Would be nice if some1 compiles all the deaths. What brand were they, were they stock or any changes were made to bios or overclocked.

Somehow I have trouble believing cards 'just die' for no fing reason.

[NKD]

Now Please instead of telling me its my fault I flashed, if it happened right after the flash I would understand but 2 days is a big time between flash and gaming for 8-10 hours without a problem, and than going to a dead card on boot up.

and Yes to be on the safe side I would really recommend to everyone that do not mess with this card so we can all isolate this problem. just remember that these cards did not finalize until computex and they were bearly available, so this might be an early culprit.

[eleeter]

Now Please instead of telling me its my fault I flashed, if it happened right after the flash I would understand but 2 days is a big time between flash and gaming for 8-10 hours without a problem, and than going to a dead card on boot up.

What do you think is more likely, a defective card that just happened to die after you flashed it, or that the flashed BIOS caused some glitch or damage and rendered the card dead?

[STEvil]

Neither.

Stock cards are dieing as well as flashed cards.

The problem is the GDDR5 as far as I can tell. No 4850's have died (without heavy mods) and they are identical other than the memory.

Clap your hands for Quimonda... more crappy ram from them!

[NKD]

What do you think is more likely, a defective card that just happened to die after you flashed it, or that the flashed BIOS caused some glitch or damage and rendered the card dead?

LOL, alright may be a glitch happened. I just don't want anyone else going through this I guess, Hey I can't even RMA my card since I got this card of someone, so I would have to buy another one, but anyways it could be that the card just died, or yea there was a glitch in the bios. Anything is possible, but it is hard for me to believe that it would work so good after a flash, and than die. The symptoms seem to be same as the other cards, that weren't even flashed.

[SimBy]

Well of course messing with stock settings/flashing with experimental beta flashers is primary suspect for most deaths for now. With exception of those cards that 'just died' stock.

The other could be GDDR5. GDDR5 is the biggest difference compared to HD4850 which I don't think I've seen much if any reports of dying.

Were there any reports of HD4870 cards dying in reviewers hands?! I mean those guys surely put them through its paces! But what they did not do is flash bios. :shrug:

[eleeter]

Stock cards are dieing as well as flashed cards.

I did a Google and the only thread that came up about 4870's going dead was this one. If this really is a problem, you would think other people would be talking about it.

[NKD]

Neither.

Stock cards are dieing as well as flashed cards.

The problem is the GDDR5 as far as I can tell. No 4850's have died (without heavy mods) and they are identical other than the memory.

Clap your hands for Quimonda... more crappy ram from them!

you are right that is the only difference between the two cards, so lets see what is going on, may be there was a reason the official launch was pushed to july 8th. It might be the ram, it is all speculation unless someone sheds more light on it. so there might not have been a great quality control, because we all know that these are the first wave of the cards, the samples weren't even available until few weeks ago, heck some of the review sites even admitted recieving samples a week before june 25th launch date.

whatever it is, if it is a bad bios flash there is no reason for us to not be able to flash it back. bios might be corrupt but it is highly unlikly it damaged the card from ever being flashed again.

[Solus Corvus]

There does seem to be some correlation between flashed cards and dead cards. But it isn't anywhere near perfect and we are just guessing that is the culprit. An even stronger correlation is between an altered LED pattern and the failed card state. It would be wise to consult ATI and find out what the LEDs represent. It may very well be that in some cases the user is at fault - but we can't be sure without more information.

And even if the users are at fault blaming them doesn't really help them at all. What they need is information because that's what they need to make the decision of how to proceed. We are all aware of the risks in modifying hardware, there is no need to berate each other for trying something a little extreme or risky. ;)

[eleeter]

What they need is information because that's what they need to make the decision of how to proceed. We are all aware of the risks in modifying hardware, there is no need to berate each other for trying something a little extreme or risky. ;)
I applaud people that are flashing and willing to take the risk. And I sympathize with them, I've killed expensive hardware before it's not a good feeling.

But we have people tossing into the mix that stock cards, never overclocked, are going dead for no apparent reason. That is confusing things.

[SimBy]

I guess finding out what those different LED states at POST mean would be nice.

[eleeter]

It might be the ram, it is all speculation unless someone sheds more light on it. so there might not have been a great quality control, because we all know that these are the first wave of the cards, the samples weren't even available until few weeks ago, heck some of the review sites even admitted recieving samples a week before june 25th launch date.
I don't know where you are going with this, but you're not making sense. Of course review sites had cards prior to launch. How do you think they get the reviews ready for launch day? It takes time to do a review, why wouldn't they have cards before the offical launch? What do you think the NDA is for?

whatever it is, if it is a bad bios flash there is no reason for us to not be able to flash it back. bios might be corrupt but it is highly unlikly it damaged the card from ever being flashed again.
You say there is "no reason" but you can't make that claim, bad BIOS code can most certainly kill hardware, especially when it powers up/initializes.

[NKD]

There does seem to be some correlation between flashed cards and dead cards. But it isn't anywhere near perfect and we are just guessing that is the culprit. An even stronger correlation is between an altered LED pattern and the failed card state. It would be wise to consult ATI and find out what the LEDs represent. It may very well be that in some cases the user is at fault - but we can't be sure without more information.

And even if the users are at fault blaming them doesn't really help them at all. What they need is information because that's what they need to make the decision of how to proceed. We are all aware of the risks in modifying hardware, there is no need to berate each other for trying something a little extreme or risky. ;)

I agree, even if I am at fault I would love to know that why the bios flash acted good for two days and than my card died. and what the led pattern means.

[NKD]

I don't know where you are going with this, but you're not making sense. Of course review sites had cards prior to launch. How do you think they get the reviews ready for launch day? It takes time to do a review, why wouldn't they have cards before the offical launch? What do you think the NDA is for?

You say there is "no reason" but you can't make that claim, bad BIOS code can most certainly kill hardware, especially when it powers up/initializes.

we could argue all day, but that won't find a solution. I am getting myself another card soon if there is not a new flash tool. so lets just wait and see, all I was trying to say that the cards were finalized not long ago, so there might be some issue. thats all. there is no need to go any further in this, as it won't do any of us any good.

[NKD]

LOL, I was atleast happy to see those good temps for two days, the damn cooler is pretty good, only if ati had set the fan speed to 31% all the time, I would not have flashed, and may be it would be fine, or may be it would be dead, that is a mystery to me, other than the part that my card is dead. and Yes I am laughing at myself.

[SimBy]

LOL, I was atleast happy to see those good temps for two days, the damn cooler is pretty good, only if ati had set the fan speed to 31% all the time, I would not have flashed, and may be it would be fine, or may be it would be dead, that is a mystery to me, other than the part that my card is dead. and Yes I am laughing at myself.

For all those who want to increase the fan speed without the bios hack you could try this:

http://forum.beyond3d.com/showpost.php?p=1183676&postcount=207

At least until we find out wtf is going on with these card deaths.

[SimBy]

http://www.rage3d.com/board/showpost.php?p=1335517981&postcount=6

[W1zzard]

could someone describe the LED configuration before and after the card dies? what are the differences?
to find the bios flash i described earlier search the thread for -newbios
can someone with a bad card measure vgpu and vmem please?
can someone with a working card save their bios using winflash latest version (http://www.techpowerup.com/downloads/1122/Winflash_2.0.1.2.html)
Look on the PCB for a chip PM25LV512A or PM25LV010A (8 legged IC). 512 = 512 k ROM, 010 = 1 MBit ROM



if there is not a new flash tool.
there won't be a new flash tool soon. i already posted the lastest yesterday @ tpu downloads but its changelog says no fixes for rv770

[Edin]

hey W1zzard...

1) when I first installed the cards there are 4 LED ligths that would lit up (LED's are located right next to 2x 6-Pin power connectors) and now only 3 of them will flash once just when I press the power button (starting up) or when system restarts. I can't remember exactly but before I think the lights were lit up more then once during bootup (someone with working card can double check)

2) I tried the original BIOS and tried the -newbios command and it made no difference on my end

3) On the picture you provide to measure the voltage on HD4850 there are no such solder points (card layout on HD4850 is different from HD4870) so I don't know where to measure the voltage. You can see image of HD4870 (top and bottom side) I took this morning, here: http://www.xtremesystems.org/forums/showpost.php?p=3100026&postcount=126

[W1zzard]

1) do you have a photo of that area by any chance? or a link to a highres pic of the card? none of the lights stay on ?
3) sorry i confused 4850 with 4870. does someone have the link to a 4870 voltmod article?

[Edin]

1) do you have a photo of that area by any chance? or a link to a highres pic of the card? none of the lights stay on?

yup, look at this pic (top/left side right next to 6-Pin connectors)
http://www.3sqn.com/members/images/HD4870_Bottom_Full.jpg

LED's are labeled D1602, D601, D1601 and D1603. The one that doesn't lit up now is labeled D601.

[DemonEyez]

This is why I always cringe when I see cards using this Quimodo crap. I remember hearing of all the issues this brand of ram caused on the 8800GT cards on release. They are just no where near as robust as Samsung branded ram. Once the 8800GT starting using Samsung ram the problems were gone. ATI should have stuck with Samsung ram I feel. Just saying.
Why do people immediately blame Qimonda for all the problem. With the 8800 GT's going bad people said it were the Qimonda chips crapping out, why was it then, that with the 8800 GTS which used the same Qimonda chips, but were connected to a beter power/vrm circuit (3-phase power on the GTS opposed to 2-phase on the GT) didn't have those problems.
Until there's some clear evidence it's all pure speculation

[r4st4m4n]

yup, look at this pic (top/left side right next to 6-Pin connectors)
http://www.3sqn.com/members/images/HD4870_Bottom_Full.jpg

LED's are labeled D1602, D601, D1601 and D1603. The one that doesn't lit up now is labeled D601.

Not sure if that helps but I noticed that in the 4850/4870 voltmod thread,one of the Vgpu mod possible point is labeled C601 ... It may indicate that the LED D601 and Vgpu are somehow related :shrug:

Here are the posts about C601: http://www.xtremesystems.org/forums/showpost.php?p=3064667&postcount=35
http://www.xtremesystems.org/forums/showpost.php?p=3093392&postcount=108

[W1zzard]

Not sure if that helps but I noticed that in the 4850/4870 voltmod thread,one of the Vgpu mod possible point is labeled C601 ... It may indicate that the LED D601 and Vgpu are somehow related :shrug:

Here are the posts about C601: http://www.xtremesystems.org/forums/showpost.php?p=3064667&postcount=35
http://www.xtremesystems.org/forums/showpost.php?p=3093392&postcount=108

smart idea. one of the 4 leds is the VDDC power good LED.
the top one and the bottom one are for the EXT power connectors afaik.
i would assume the 4th one is some temperature too hot indicator.

is D601 always lit on a working card or always dim?

[Jakalwarrior]

I have nothing to add here but im going to throw out a guess so if I'm right in a few months I can gloat or something.

I'm guessing it is possibly the error correction stuff on the RAM allowing cards to seem functional at high memory clocks meanwhile the ram is way past "stable". Perhaps they can even get past internal QA like that.
I remember reading some review that said the ram would correct errors when unstable and overclockers would have to look out for lowered scores. That leaves me thinking though that a small to moderate amount of instability would be nearly indetectable without a direct measure of the memory bandwidth.
Remember how they also said the ram auto negotiates some of the stuff with the controller on startup? maybe the ram keeps running at the higher clocks while correcting the errors, but on reboot it cant negotiate the settings with the controller?

[wiak]

I have nothing to add here but im going to throw out a guess so if I'm right in a few months I can gloat or something.

I'm guessing it is possibly the error correction stuff on the RAM allowing cards to seem functional at high memory clocks meanwhile the ram is way past "stable". Perhaps they can even get past internal QA like that.
I remember reading some review that said the ram would correct errors when unstable and overclockers would have to look out for lowered scores. That leaves me thinking though that a small to moderate amount of instability would be nearly indetectable without a direct measure of the memory bandwidth.
Remember how they also said the ram auto negotiates some of the stuff with the controller on startup? maybe the ram keeps running at the higher clocks while correcting the errors, but on reboot it cant negotiate the settings with the controller?
and the flash messed that up
might be

[jimmyz]

lights should all be off on working cards so the lights that stay on indicate a problem, unless ATI has switched it up this time, anyone with a working card please confirm.

[Mechromancer]

What brand cards are having these failures? Is it one specific brand, like HIS, or are these failures over all brands? If those of you with failures could post your card's brand that would be helpful. I'd like to determine the circumstances behind these failures and the only way to do that is to catalog all the details.

Please post this info:

GPU Make & Model:
GPU BIOS version: (if you have it)
Motherboard Make/Model/BIOS:
Display Driver version used:
Failure under Crossfire?: (yes/no)
Other Details: (any notable details before/after the failures)

Those are just a few variables I could think of that would cause this problem. I'd like us to be able to distinguish between a GPU issue like a board defect, or an outside cause in these particular cases.

[Logos]

What brand cards are having these failures? Is it one specific brand, like HIS, or are these failures over all brands? If those of you with failures could post your card's brand that would be helpful. I'd like to determine the circumstances behind these failures and the only way to do that is to catalog all the details.

Please post this info:

GPU Make & Model:
GPU BIOS version: (if you have it)
Motherboard Make/Model/BIOS:
Display Driver version used:
Failure under Crossfire?: (yes/no)
Other Details: (any notable details before/after the failures)

Those are just a few variables I could think of that would cause this problem. I'd like us to be able to distinguish between a GPU issue like a board defect, or an outside cause in these particular cases.

correct me if I'm wrong but I thought all brands provide reference boards at the moment, so isolating one brand in case of failure doesn't really make sense...other parameters, the one you mentioned yourself, are all valid and interesting.

[Marvin_The_Martian]

Both the people on the Folding Forums and offcourse [H] think each and every one off is lying and your cards all deid because you overclocked them to much or flashed them badly, and you know what proofs it? Because this is XS :shrug:

But one off them had a point, they tested folding on those cards before they were released and I need to say 24/7 folding is heavier on a card I think then what 80% off this forum ever puts their cards through ( no wait it is xs so make it 50% ). They had no issues with them dying, what changed? I posted on the folding forum that instead of blaming it on overclocking without a second thought they should try to help Wizzard vind out what happend. If it's not a hw failure but user caused then that will come out to, but what I see is people claiming cards died without flashing or overclocking and that doesn't sound like a user mistake.

On the other hand, not to many threads about them yet. If people who don't have issues far outweigh those who do, it should be clear to because I'm not sure if it's because their still hard to get in some places or because it's just so few people with problems. So maybe we also should do a count off good vs dead cards here?

Edit: nm on that last thing, http://www.xtremesystems.org/forums/showthread.php?t=192906

Edit2: Perkam wrote everyone who has a dead card flashed it. So, who's lying here? If you flashed and leid you should be banned imho, abusing a forum to try and get an rma :shakes:

[mascaras]

HD4000 séries FAN FIX ( without flashing Bios ) >> http://www.xtremesystems.org/forums/showthread.php?p=3100889#post3100889


:up:

[Mechromancer]

HD4000 séries FAN FIX ( without flashing Bios ) >> http://www.xtremesystems.org/forums/showpost.php?p=3100837&postcount=4045


:up:

Very nice!

I'm starting to have doubts as to the quality of this Qimonda GDDR5. Apparently the 8800GT had problems due to their memory as well. I've also read a bit about how GDDR5 works and its NOTHING like the previous generations of memory. It varies clockspeed, voltage, and other parameters automatically to resolve signaling issues. Screwing with the BIOS is probably messing with this function in some way.

GDDR5 Articles:
http://www.extremetech.com/article2/0,2845,2310000,00.asp
http://www.eetasia.com/ART_8800503578_499486_NP_74d933d1.HTM

Read up on how "signal training" works for yourself and lets discuss it. I'm no memory expert so I won't pretend to know how it all works. It just seems like these BIOS mods are the culprit.

[C'DaleRider]

HD4000 séries FAN FIX ( without flashing Bios ) >> http://www.xtremesystems.org/forums/showthread.php?p=3100889#post3100889


:up:

Good link.....that's almost exactly like the profile edit that was used on the 3870's to modify the fan speeds, among other things. It's what I used on my 3870 to spin up the fan....and worked like a charm.

Like was said, easiest and safest way to modify your fan speeds without touching your BIOS, which seems to be a very dicey and potentially expensive proposition right now.

[WiCKeD]

Qimonda = Infineon. It's been used on ATi cards forever.

So far I have only seen 1 user beside Theta report dead cards without a bios flash and that also happened when he entered Crossfire.

[Solus Corvus]

I get a HIS 4870 monday. Thank you SimBy and Mascaras for the profile hack. Seeing these problems I won't be overclocking or flashing the card right away ( not until there are proper tools and full cover water blocks, down the line ), but I will be using the fan speed hack to keep it cool at least.

With my luck it would be the high power draw fan that is killing the cards. :smoke: :rofl:

[Theta]

Little more detail about our 4 dead 4870s:

1) Lived for just over 48 hours from arrival.
2) Were retail boxed / non-review / HIS brand.
3) Lived for 2 days of testing, heat levels ~ 90C @ load.
4) Died approx. 2 hours after enabling CrossfireX.
5) Died after 3 reboots each between XP 32, XP 64, and Vista x64.
6) No BIOS mods, no fan mods, SmartDoctor not used.

Background information:

1) Still have ~10 working 4870s, none put into CrossfireX yet, due to symptomatic failure of others. Waiting to see...
2) No problems with any 4850s in any configuration - Tri and Quad CrossfireX tested.


Looked back at the packing slip, and the HIS lot was received from the same origin... Distributor is thinking that a bad batch may have leaked, due to a large number of HIS already coming back. I do know that these are reference cards, but it is conceivable that a bad batch slipped into production.

Lastly, I am intrigued by one thing here - flashers and non-flashers seem to be reporting the same events, down to a gnat's ass. Same no-post, same LED, same symptoms. Perhaps the flash brought about something that can die, or has the potential to kick in. I'll leave it in the hands of you all to figure out, just wanted to add a few last details.

These are getting shipped out on Monday, and we'll be hoping for the best on the others that we have so far. Good luck to everyone. ;)

[OndraSter]

Can you download BIOS from dead HD4870?

If yes, try to flash the original BIOS and then download it again to check, if flash went OK or bad and get known if it is because of BIOS or damaged card.

If it's bad idea, sorry for this. ;)

// edit: at least, that this isn't doing HD4850, because I want to buy one of them.

[Solus Corvus]

If I had a dead card that died without a flashed bios, I wouldn't flash it and risk losing the ability RMA it.

[Mechromancer]

Little more detail about our 4 dead 4870s:

1) Lived for just over 48 hours from arrival.
2) Were retail boxed / non-review / HIS brand.
3) Lived for 2 days of testing, heat levels ~ 90C @ load.
4) Died approx. 2 hours after enabling CrossfireX.
5) Died after 3 reboots each between XP 32, XP 64, and Vista x64.
6) No BIOS mods, no fan mods, SmartDoctor not used.

Background information:

1) Still have ~10 working 4870s, none put into CrossfireX yet, due to symptomatic failure of others. Waiting to see...
2) No problems with any 4850s in any configuration - Tri and Quad CrossfireX tested.


Looked back at the packing slip, and the HIS lot was received from the same origin... Distributor is thinking that a bad batch may have leaked, due to a large number of HIS already coming back. I do know that these are reference cards, but it is conceivable that a bad batch slipped into production.

Lastly, I am intrigued by one thing here - flashers and non-flashers seem to be reporting the same events, down to a gnat's ass. Same no-post, same LED, same symptoms. Perhaps the flash brought about something that can die, or has the potential to kick in. I'll leave it in the hands of you all to figure out, just wanted to add a few last details.

These are getting shipped out on Monday, and we'll be hoping for the best on the others that we have so far. Good luck to everyone. ;)

I've read about HIS having issues on OCF too. Also I think I saw a post about a guy having problems with a Visiontek too.

I would like to know what motherboard you tested these cards in. The make & model of the board may have an affect too (just a guess).

[Marvin_The_Martian]

Can you provide OP with the batch number so if other come in with the same issue he will be able to check if it's indeed a batch issue and get the proper responce going as quick as possible? Just a thought.

[NKD]

same with my card, when it worked all foru leds flashed, now the D601 is the only one that does not flash, and yea they never stayed on when it was working either, they would only flash once when you turn on the system. As I said I can't rma the card even if I had a chance to do it I would not, but it is none of my business on what others do, because it is their card. I would most likely buy one next month again if I can't get it to work.

[W1zzard]

can someone please save a 4870 bios from a working card using winflash and post it?

[Kensek]

I have my original sapphire HD4870 BIOS but it was saved with AtiFlash 3.59.

http://www.keepandshare.com/doc/view.php?id=660440&da=y

I removed the file since it has been determined that AtiFlash and WinFlash are truncating the BIOS file.

[W1zzard]

thats fine, link please

[NKD]

I have my original Powercolor bios saved with GPUZ if that works let me know I can send that to you.

[W1zzard]

for this test save with winflash / atiflash only, so we can see if there is some kind of issue between gpu-z bios reading and ati's software

[Kensek]

I edited my earlier post with the link in it W1zzard.

[spajdr]

another dead card by not flashing bios
http://www.nvnews.net/vbulletin/showthread.php?t=115579

[NKD]

oh ok, I really didn't save with winflash, so I guess that is out of question, thanks to all of you for your help.

[W1zzard]

ok looks like the bios images are NOT 1 mbit as a poster on rage3d claims, so thats not the reason for the problems

[W1zzard]

can anyone with a borked card measure the voltages on the board? i marked the test spots for you

http://img.techpowerup.org/080629/Capture389785.jpg

on your dmm put the black probe to ground (like case metal or the metal piece on the graphics card) and the red one to what you want to measure

yellow area and green area should be either mvddc and mvddq memory voltage (dont know which is one because i have no board). any of the pads in one color will work fine, it is safe to short between same colored pads

the blue area is vddc (gpu core voltage). as you can see two solder pads go together (white boxes on the pcb), one is ground, the other one will be vddc. test both, the one with the 0 volt reading is the gnd one. dont create shorts between those pads.

if you have questions ask before causing more damage

[Logos]

another dead card by not flashing bios
http://www.nvnews.net/vbulletin/showthread.php?t=115579

card is not dead, read the thread again, user has solved his problem :D

[W1zzard]

card is not dead, read the thread again, user has solved his problem :D

gogo XSers check your cables. lol. afk football, hope the great german nation pwns the spanish

[NKD]

card is not dead, read the thread again, user has solved his problem :D

that is right, his issue was totally different, it was wasn't a good idea to call it a dead card, now back to topic, cheers.

[NKD]

gogo XSers check your cables. lol. afk football, hope the great german nation pwns the spanish
LOL, I tried everyway I can imagine, the D601 led never turns on.

[Logos]

gogo XSers check your cables. lol
his card is not dead, that's a fact, period. Who said a particular found cable issue would give a solution to others, who? :rolleyes: It just did matter to mention that it was not a new case of HD48xx failure, that's all.

[XCheater]

fortunately I go with 4850 first
anyway after I have collected my money, I'll go with X38 and a 4870 of necessary :D if there is no more dead 4870 anymore, or I'll grab another safer 4850

[Dave Baumann]

ok looks like the bios images are NOT 1 mbit as a poster on rage3d claims, so thats not the reason for the problems
The ROMs on these boards are 1Mb ROM's, used for the first time used on AMD/ATI PC desktop boards.

[GoldenTiger]

W1zzard, I have my HIS 4870 original BIOS saved (unmodified, from before my card died). I'll upload it somewhere in a minute and edit with link.

EDIT: http://files.filefront.com/his4870stockrom/;10854276;/fileinfo.html

EDIT2: File removed as it is truncated per this thread.

[Wesker]

Very nice!

I'm starting to have doubts as to the quality of this Qimonda GDDR5. Apparently the 8800GT had problems due to their memory as well. I've also read a bit about how GDDR5 works and its NOTHING like the previous generations of memory. It varies clockspeed, voltage, and other parameters automatically to resolve signaling issues. Screwing with the BIOS is probably messing with this function in some way.

GDDR5 Articles:
http://www.extremetech.com/article2/0,2845,2310000,00.asp
http://www.eetasia.com/ART_8800503578_499486_NP_74d933d1.HTM

Read up on how "signal training" works for yourself and lets discuss it. I'm no memory expert so I won't pretend to know how it all works. It just seems like these BIOS mods are the culprit.

The 8800 GTS also had Qimoda chips, and I don't see anyone complaining about them. The 8800 GT was plagued by an inadequate VRM for its given specifications.

Further; Qimoda supplies GDDR chips for many other cards, all without any issue. That's beside the point that; to even suggest that engineers at either Qimoda or ATI/AMD didn't put the GDDR5 samples through rigorous testing is way off base.

[Dave Baumann]

W1zzard, I have my HIS 4870 original BIOS saved (unmodified, from before my card died). I'll upload it somewhere in a minute and edit with link.

EDIT: http://files.filefront.com/his4870stockrom/;10854276;/fileinfo.html

Whatever tool you used to save that with it is not saving to the correct size; the ROM image should be 128K.

I've just looked at all the BIOS's collected on W1zzard's download site, and again they are all incorrect sizes and the BIOS is truncated. W1zzard I would urge you to remove those BIOS's

[NKD]

hmmm, if it is indeed the wrong bios size, I am with you o that one DAVE, we should stop it from spreading in any and every way possible.

[Marvin_The_Martian]

Yeah I posted in the other thread he should try to get in touch with Wizzard at techpower if he can't reach em here. On the other hand, I would be surprised if Wizzard wasn't certain when he said the bioses were not the problem. Only other thing could be that the download server is truncating the downloads of Dave Baumann?

[W1zzard]

The ROMs on these boards are 1Mb ROM's, used for the first time used on AMD/ATI PC desktop boards.

Dave, the BIOS dump using winflash/atiflash is 62.976 bytes. i thought 1 Mb ROM too. that's why I specifically asked for Winflash/ATIFLASH dumps.

so far nobody reported if the actual bios chip is 512k or 1M.

[Dave Baumann]

Only other thing could be that the download server is truncating the downloads of Dave Baumann?
No, because I checked those at both TechPowerup and the filefront one GoldenTiger uploaded, referenced in the earlier post (http://www.xtremesystems.org/forums/showpost.php?p=3101448&postcount=225); neither of them are the same size at the production BIOS.

[Dave Baumann]

so far nobody reported if the actual bios chip is 512k or 1M.
Its 1Mb on all HD 4870's (HD 4850 is still 512Kb).

I have no idea how old the versions of WinFlash you are using are. They almost certianly haven't been updated specifically for RV770.

[W1zzard]

the bios files are 1 Mb indeed. i'm taking the 4870 bioses offline right now

http://www.techpowerup.com/downloads/1122/Winflash_2.0.1.2.html is the winflash those people used .. no newer version, feel free to check yourself dave

[Mechromancer]

the bios files are 1 Mb indeed. i'm taking the 4870 bioses offline right now

http://www.techpowerup.com/downloads/1122/Winflash_2.0.1.2.html is the winflash those people used .. no newer version, feel free to check yourself dave

This is good to know. I'm now more concerned with the people that say they simply crossfired their 4870s and experienced a failure. Or were these reports misleading?

[Marvin_The_Martian]

This is good to know. I'm now more concerned with the people that say they simply crossfired their 4870s and experienced a failure. Or were these reports misleading?

I feel that if they were misleading those people's accounts should be banned here, they give XS a bad name trying to aquire an easy rma by using a forum to make it look like they didn't do anything wrong themself.

I still have some hope in the good of people, so I'm not definetly saying they were lying but I would like to have this found out without a doubt.

[SimBy]

I guess the question now is if truncated bios indeed caused many of this deaths, can the cards be revived with correct bios or are they fubared.

[ripken204]

I guess the question now is if truncated bios indeed caused many of this deaths, can the cards be revived with correct bios or are they fubared.

if a proper bios flashing tool comes out then you would assume so.

[W1zzard]

ok football is over, back to work. i need volunteers with a working 4870 and one person with a non-working 4870 .. can be the same person ;)

contant me on instant messenger asap

[GoldenTiger]

Whatever tool you used to save that with it is not saving to the correct size; the ROM image should be 128K.

I've just looked at all the BIOS's collected on W1zzard's download site, and again they are all incorrect sizes and the BIOS is truncated. W1zzard I would urge you to remove those BIOS's

I used ATIwinflash 2.0.0.7 for it under Vista 64.


I feel that if they were misleading those people's accounts should be banned here, they give XS a bad name trying to aquire an easy rma by using a forum to make it look like they didn't do anything wrong themself.

I still have some hope in the good of people, so I'm not definetly saying they were lying but I would like to have this found out without a doubt.

I fail to see why someone would lie on a forum where they are anonymous about what caused their card's death... if they were going to lie why not lie to the manufacturer or vendor for RMA? I told the situation to my vendor and was offered an RMA for a full refund which was lucky as heck on my part.


EDIT: I am taking my truncated BIOS down from filefront now, since I assume those who needed to look at it did and it is clearly not proper.

[STEvil]

ATI Tray Tools can dump bioses as well, but dont know if it works on 4850/4870.. doesnt want to work on my X1950XTX.

[GoldenTiger]

W1zzard, did anyone with a non-working card contact you yet on messenger? I am installing MSN messenger now.

[W1zzard]

nope ..

[GoldenTiger]

OK, I'll be done with installing in a few minutes... should I pop my 4870 in alongside my current card (working 4850) after it's done?

[xMrBunglex]

ok football is over, back to work. i need volunteers with a working 4870 and one person with a non-working 4870 .. can be the same person ;)

contant me on instant messenger asap

MSN is the only messenger i use. i just sent you an invite (i think. i don't use it very often). i have two working 4870's and i'd like to help as long as they don't end up dead.

[GoldenTiger]

MSN is the only messenger i use. i just sent you an invite (i think. i don't use it very often). i have two working 4870's and i'd like to help as long as they don't end up dead.

Reading a BIOS should never harm anything for you, so don't worry about that... it's only if you write to it that you can have something happen.

[Marvin_The_Martian]

I fail to see why someone would lie on a forum where they are anonymous about what caused their card's death... if they were going to lie why not lie to the manufacturer or vendor for RMA? I told the situation to my vendor and was offered an RMA for a full refund which was lucky as heck on my part.

EDIT: I am taking my truncated BIOS down from filefront now, since I assume those who needed to look at it did and it is clearly not proper.

Well some people seem to thing that if their is some forum saying it's not an user problem, it might be easier talking with a retailer/manufacturer about an rma?

I applaud your honesty and you been rewarded for it I see, but not everyone is as honest as you. Again, it's almost like I'm saying I'm certain they lied, I'm not. But to litle cases are known to warrant someone saying that these cards are dying from nothing, that's all I do know for certain. It's been a day since this has been going on here on XS, other forums picked up but I can only see one new case being reported, for me that's pretty conclusive but you seem to think otherwise and again even for your trust I applaud you. I just hope you're right.

[wiak]

I don't think so. I only Oc'd 1 card to 790/1050 for 15 minutes.

The second card was never OC'd. On the initial bootup with both cards inserted and powered up and with the cF bridges installed XP would not load.

Hence, I removed 1 card thinking a Corsair 750TX was insufficiently powering the cards. However, after this neither card never worked again.

Though they can be seen through ATI Flacs when another ATI model card is used as the primary.
if am not mistaking crossfire will make both cards run at same speed unless both are clocked, so here might be the problem?

[W1zzard]

got the 128k image now .. testing if it revives the card

[Kensek]

What Flash Utility do you use and where is the 128k image file.

I'll dig out my 4870s again

[wiak]

guys, i have a idea
whats the serial numbers, manufacturer and manufacturer dates?
and where did you guys buy them from to?
then we might find out something

[W1zzard]

the fix works. i'll need another volunteer with a broken card. instant message me :)

[Kensek]

Don't have IM - but set to try

[SimBy]

Very glad to hear it m8 :clap: :up: