| Friday, October 24 2008 |
Earlier this week, when evaluating samples of the new Radeon HD 4830 GPU-based graphics card, W1zzard, who also authors the GPU-Z diagnostic utility, had found an anomaly with the stream processor (SP) counts on samples sent by AMD. The GPU could access only 560 out of its 640 SPs available. This affected the card's performance significantly, and TechPowerUp's findings were validated by several other reviewers with similar samples.
AMD on its part quickly followed up the issue with its engineering department, and released a video BIOS update that fixes the issue and makes available all the 640 SPs. AMD will be circulating this BIOS to all its press contacts and add-in-board (AIB) partners, to make sure the issue is isolated and fixed. The updated BIOS file can be downloaded from here. AMD also asserts that the issue isn't hardware related, and that updating the BIOS resolves the issue completely.
AMD on its part quickly followed up the issue with its engineering department, and released a video BIOS update that fixes the issue and makes available all the 640 SPs. AMD will be circulating this BIOS to all its press contacts and add-in-board (AIB) partners, to make sure the issue is isolated and fixed. The updated BIOS file can be downloaded from here. AMD also asserts that the issue isn't hardware related, and that updating the BIOS resolves the issue completely.
User comments
Great to hear this! I was about to say there are going to be alot of people really P-O'ed about it.
our reviewer W1zzard? more like TPU admin/owner.
glad to see we got this going.
glad to see we got this going.
IMO the only reason this was brought out was because of the TPU review.
It's called accountability!!! AMD tried to pull an nvidia and got caught. If it were never found out do you think that they would have fixed it? "The fix would have been a new 4835 with 700 SPs" with 640 actual!!!
Ponder :toast:
It's called accountability!!! AMD tried to pull an nvidia and got caught. If it were never found out do you think that they would have fixed it? "The fix would have been a new 4835 with 700 SPs" with 640 actual!!!
Ponder :toast:
Way to go W1Z! Catch AMD maybe trying to pull the wool over our eyes.
more like tinternet gpu god :)
worshipping aside does this mean we could mod the bios of these 4830's and turn them into 4870's?
if its only soft modded to disable the cores/pipes w/e its gotta be doable right?
worshipping aside does this mean we could mod the bios of these 4830's and turn them into 4870's?
if its only soft modded to disable the cores/pipes w/e its gotta be doable right?
Doubt AMD was trying to pull a fast one or anything of the sort. The card was meant to beat out the 9800/8800 segment. As it was it would not have been much of a matchup with a defective neutered bios.
W1z great job for helping AMD/ATI solve an issue that could have gotten a lot worse!
W1z great job for helping AMD/ATI solve an issue that could have gotten a lot worse!
by: fitseries3;1032060We know who he is, but when in news, in context of the incident, he's our reviewer, and GPU-Z author. It relates better to the guy at AMD/non-TPU member reading the news.
our reviewer W1zzard? more like TPU admin/owner.
glad to see we got this going.
thats a fast response by amd but i like to see that.
by: btarunr;1032081oh come on... i was just messin with ya.
We know who he is, but when in news, in context of the incident, he's our reviewer, and GPU-Z author. It relates better to the guy at AMD/non-TPU member reading the news.
So this only makes me question if a bios release enabled 640 could another (modded) bios release even more?
by: jbunch07;1032100exactly.......see previous statement :roll:
So this only makes me question if a bios release enabled 640 could another (modded) bios release even more?
or mine :)
I dont think so....the rest of the SP's according to w1zz aparently hard locked kinda like laser locking not the same but same principle...however lowing the SP's is possible because you can tell the bios not to map them. this is post 6666
But did any of these cards make it to retail channels?
If ATi was trying to pull a fast one, I would think they would have at least sent out real cards to the reviewers so they reported good performance.
If ATi was trying to pull a fast one, I would think they would have at least sent out real cards to the reviewers so they reported good performance.
by: Solaris17;1032106you just posted you 6,666th post. :eek:
I dont think so....the rest of the SP's according to w1zz aparently hard locked kinda like laser locking not the same but same principle...however lowing the SP's is possible because you can tell the bios not to map them.
But OT, that's a bummer, Oh well, im sure the 640sp will be plenty to fill the gap!
by: newtekie1;1032107
But did any of these cards make it to retail channels?
If ATi was trying to pull a fast one, I would think they would have at least sent out real cards to the reviewers so they reported good performance.
well you can be sure none will be made as AMD has issued the fix to its partners however if some batches were already shipped their is a chance of contamination unless the partners do a mass recall to check the cards.
its just a bios issue whats the point of amd trying to give people less sps? they are on the card which means it costs amd the same. no matter what it hurts amd because the reviews are bad and the people get less performance than they expected and theres no point in that.
by: roofsniper;1032132
its just a bios issue whats the point of amd trying to give people less sps? they are on the card which means it costs amd the same. no matter what it hurts amd because the reviews are bad and the people get less performance than they expected and theres no point in that.
Exactly, what is the point, it is pointless and a silly presumption...
It feels great to be part of a solution to AMD. Thanks for the notice!
Nice to see AMD respond so fast. I highly doubt they were trying to dupe anyone. Everyone has issues and mistakes come up and not all are always caught.
So Wizzard are you going to Re-review the boards since they are now fixed to see if there is any fluctuation in the performance analysis?
Like others have said, it's really nice to see a company respond to an issue like this so quickly and efficiently :)
@eidairaman1: I don't really see any reason to re-review the amd card any further than verifying the number of SP's enabled and maybe running a couple quick benches, since the other card had the full 640 SP's enabled and we have the performance results from that.
@eidairaman1: I don't really see any reason to re-review the amd card any further than verifying the number of SP's enabled and maybe running a couple quick benches, since the other card had the full 640 SP's enabled and we have the performance results from that.
well its on the clause just to make sure, i mean wizzard can run the cards and let us know the results, if they deviate from the originals then something else maybe amiss.
Right, that's why I said just verify the SP count and run a couple quick benches, but I think as long as those benches yield results on par with the other card then it would be relatively safe to assume that it's ok. I'm sure W1z is a pretty busy guy ;)
Question: Is there already a discussion going on about possible energy savings by switching off a shader block? Since in techpowerup and computerbase.de test the unpatched amd radeon 4830 (having 560 shader count) seems to require less power than the powercolor 4830 having 640 shaders.
Could somebody try this out, please? Maybe this could be a new strategy for 4850/70 energy savings/Powerplay? Another interesting way would be to try rivatuner shader block bit masking.What's the real difference in the old and new amd bios?
Could somebody try this out, please? Maybe this could be a new strategy for 4850/70 energy savings/Powerplay? Another interesting way would be to try rivatuner shader block bit masking.What's the real difference in the old and new amd bios?
ulrik just gave me an idea. Would there be any way to "throttle" the amount of SP's enabled for extra power-saving during 2D mode?
how about you suggest that to wizzard for the next build of ATITool, if he will ever get around to deving that program anymore.
meh, I wouldn't suggest since I would never use it (not my house, i don't pay the electricity bill :laugh:)
its teh idle clock of teh amd bios that cuases teh power savings, ask whizzinator
It's quite possible that AMD runs some additional tests to ensure that the review samples are working correctly (as opposed to doing partial batch testing for retail products).
It may be that they were using a variety of BIOS's to ensure that the shaders on the review cards were working correctly and accidentally sent the cards out in the final test configuration (minus one shader).
I really doubt it was any kind of deceipt on AMD's part as it simply does not make any logical sense for them to do that. Their quick response also points to a mistake, not a purposeful act.
It may be that they were using a variety of BIOS's to ensure that the shaders on the review cards were working correctly and accidentally sent the cards out in the final test configuration (minus one shader).
I really doubt it was any kind of deceipt on AMD's part as it simply does not make any logical sense for them to do that. Their quick response also points to a mistake, not a purposeful act.
teh news post is a quote from AMD's press release
Awww man... You guys are defending AMD on this one, it makes me sick. If nVidia did this, you would be bashing the hell out of em. The point is they tried to get away with one and had a BIOS ready in case they got caught, that is why the "fix" came so quickly.:shadedshu
by: iLLz;1032603
Awww man... You guys are defending AMD on this one, it makes me sick. If nVidia did this, you would be bashing the hell out of em. The point is they tried to get away with one and had a BIOS ready in case they got caught, that is why the "fix" came so quickly.:shadedshu
it has been said and said again, the cards have the SP's on them already, it costs AMD/partners the same to produce the card with all of them working or not. so what possible benefit could it bring to BIOS disable some of them?
by: iLLz;1032603Of course they had the correct BIOS availalbe. That is what the cards were supposed to be using. To mistakenly send out samples with the wrong (perhaps test) BIOS is an understandable mistake.
Awww man... You guys are defending AMD on this one, it makes me sick. If nVidia did this, you would be bashing the hell out of em. The point is they tried to get away with one and had a BIOS ready in case they got caught, that is why the "fix" came so quickly.:shadedshu
I've never bashed nVidia and have owned nVidia cards in the past and was pleased with them. It's true that at the moment I prefer ATI cards, but that is due to what I feel are more stable drivers and I also prefer their "once a month" driver releases (it's much less confusing than nVidia's method).
If nVidia release review samples with a similar issue, I would give them the benefit of the doubt until it was proved otherwise. I, personally, believe in "innocent until proven guilty".
Just my 2 cents.
kreij you are with ATI because you feel more stable drivers and i am with Nvidia because IMO i feel the drivers are more stable. I have seen so many threads about driver issues with ATI (as with Nvidia) that is what turned me away from ATI for now
by: p_o_s_pc;1032641LOL ... that just goes to show you that we each perceive things differently.
kreij you are with ATI because you feel more stable drivers and i am with Nvidia because IMO i feel the drivers are more stable. I have seen so many threads about driver issues with ATI (as with Nvidia) that is what turned me away from ATI for now
I am thankful that there is an nVidia (and ATI) so that we as consumers have a choice, and the competition is good for us all.
We have seen what happens when there is no competition and a company justs sits back and does nothing new (no, I'm not naming any names ;) )
by: iLLz;1032603what? that doesn't even make any sense? why would amd release a card and pay money to have the sps on there but not use them? thats completely idiotic. thats like amd releasing a cpu that is supposed to clock at 3.0ghz but they clock it at 2.0ghz. because then people will get it and see that it sucks and go to intel. same with ati people might get it then return it and buy the nvidia offering. it would confuse me greatly as to why amd did this on purpose. especialy when they send the same exact versions to reviewers who should the performance of the card. and ontop of that what are the chances of every single person that reviews and buys the card to think that it has the correct number of sps? gpuz shows the sps right there and everyone knows what they are getting. i don't understand why anyone could possibly think that amd did this on purpose. to me it just seems like people wanting to start something to bash amd. :mad:
Awww man... You guys are defending AMD on this one, it makes me sick. If nVidia did this, you would be bashing the hell out of em. The point is they tried to get away with one and had a BIOS ready in case they got caught, that is why the "fix" came so quickly.:shadedshu
Nvidia has been called out on things, often. Such as the Crysis driver tweaks, and the 16bit vs 24bit. None of their cards untill the late 8800 series had the true precision ATI did.
ATI made a mistake, not overpromising hardware and underdelivering, but with a faulty BIOS, perhaps the man in charge of it placed a wrong character in the BIOS. What this does tell us is that a BIOS hack could be possible.
ATI made a mistake, not overpromising hardware and underdelivering, but with a faulty BIOS, perhaps the man in charge of it placed a wrong character in the BIOS. What this does tell us is that a BIOS hack could be possible.
by: Steevo;1032665yea but nvidia was rigging things and this is a mistake that hurts the company. which just makes me so clueless as to how people can actualy think that amd is trying to pull a fast one when in no way at all it helps them.
Nvidia has been called out on things, often. Such as the Crysis driver tweaks, and the 16bit vs 24bit. None of their cards untill the late 8800 series had the true precision ATI did.
ATI made a mistake, not overpromising hardware and underdelivering, but with a faulty BIOS, perhaps the man in charge of it placed a wrong character in the BIOS. What this does tell us is that a BIOS hack could be possible.
read
comprehend
post
comprehend
post
Just to add, AMD loses nothing by selling HD 4830 with its proper 640 SPs, just as they're not going to be upping profits/cutting mfg costs by cutting the SP count. This was an accident, it happened to be spotted on early, and AMD released a fix within 36 hours. Please don't make this a "if it were NVIDIA" discussion, don't play fortune-teller. Back to topic.
competition is good for market but cheating not included,thats illegal,i n this care indicated to "that",so let see who is right,who is wrong,lets wait n see
Some guy on newegg posted a review saying he unlocked the 800 shaders. I noticed it when I was purchasing the card last night. He doesn't say how he did it though of course, just says google to find out. I did that, but I didn't find anything. Anyone else better at googling things want to give it a shot?
He's lying. You can't bring 800 SPs for a HD 4830 by means of any softmod/bios. A HD 4830 is a RV770 with two ALU clusters (160 SPs) disabled by a physical method, not masked. After physically disabling the ALU clusters, they accordingly give out the BIOS. Some of the cards AMD sent out had 2 clusters disabled physically, but tables for three clusters were missing from the BIOS...hence the 560 SP HD 4830 anomaly. A simple BIOS fix solved, and enabled the extra ALU cluster that got disabled accidentally.
Nice job W1ZZ and nice job AMD! I'd say that upped the "stock" of TPU with AMD. :toast: