Whoops: Some Radeon HD 4830 cards are gimpy

The Radeon HD 4830 looked pretty good upon its release, but some cards that made it into the hands of reviewers and retailers are, in fact, not what they should be.  AMD has just released a statement about the matter, which says:

AMD has identified that, in addition to reference samples of the ATI Radeon™ HD 4830 boards sent to media with a pre-production BIOS potentially impacting the card’s performance, a very limited number of ATI Radeon™ HD 4830 boards were released to market with the same pre-production BIOS. This is in no way hardware related, and an updated BIOS fully resolves the performance limitation.

Through consultations with AMD board partners, it has been determined with a high degree of certainty that fewer than 400 ATI Radeon™ HD 4830 boards from one AMD board partner, HIS, have reached the market with the pre-production BIOS incorrectly provided by AMD.  As only a small number of HIS-branded ATI Radeon™ HD 4830 cards are impacted, we ask any customers that purchased an HIS-branded ATI Radeon™ HD 4830 to test the board using the GPU-Z utility (available at http://www.techpowerup.com/gpuz). If the GPU-Z utility reports fewer than 640 shaders, please visit the HIS website for information on how to update the card BIOS via a downloadable install utility.

If you’ve purchased one of these cards, you will very much want to check and be sure you aren’t leaving some performance on the table.

As I noted in the comments to our Radeon HD 4830 review, our 4830 card was one of those affected.  I’ve already confirmed a minor performance increase with an updated BIOS. I plan to update the numbers in our review, but doing so may take some time due to other pressing obligations.

Comments closed
    • Damage
    • 11 years ago

    I’ve updated our Radeon HD 4830 with new test results that come from the updated BIOS with the proper number of SIMD arrays and TMUs enabled. We saw minor performance gains across the board for the 4830 with little change in power consumption.

    • moose17145
    • 11 years ago

    Oh wow… read this last night at 1:00 in the morning and didn’t see that the TR 4830 was one of the affected ones. Damn and that card performed so well even while gimped! Now i am just even more impressed by it!

    Yea it sucks a few number of gimped cards made it out into the wild, but flashing their bios is not hard, and even gimped they performed very well. Should they maybe have checked a bit closer to make sure that the correct cards with the correct bios’ made it out the door… maybe. But accidents and shit happen. So im not gonna hold a mistake like this against them. ESPECIALLY considering it’s a mistake that makes their cards look crappier than they really are. NO company is ever going to do that on purpose. So i have to believe it was an honest mistake. Much like shit, they happen.

      • MadManOriginal
      • 11 years ago

      Yea they were gimped but the difference is max 5% according to AT, we’ll see what TR’s updated results show. So it just makes it a bit better but doesn’t greatly change the conclusions, it comes out just enough better to match the price – the 4830 is just about perfectly positioned for price/performance if you count 9800GT MIRs, the only problem I see is the old ‘every $25 problem’ where there’s the next card for just a little more. This close segmentation has it’s ups and downs.

      I do want to see updated power draw numbers, I don’t think AT changed theirs. Techpowerup, who found the problem, shows fairly different power draw between the gimped and full version. When I read TRs review I couldn’t reconcile the power draw rankings with other reviews and this is obviously why.

    • indeego
    • 11 years ago

    /[

    • lycium
    • 11 years ago

    it’ll be interesting to compare the temps before and afterwards πŸ™‚

    • Fighterpilot
    • 11 years ago

    The irony of a well respected industry figure like Dave Baumann trying to patiently explain the facts to TR’s resident NVidia fanboi….
    Please don’t waste any further time on him Dave…he couldn’t care less whether he’s wrong or right…..any anti ATI FUD is what he lives for.

      • ludi
      • 11 years ago

      First, I do appreciate Dave Baumann stopping by and saying something.

      Second, I don’t really appreciate you saying something. Prime may be succumbing to Nvidia fanboy behavior, but at least he still has some sense of propriety.

      • PRIME1
      • 11 years ago

      How about the irony of you calling someone a “fanboi”.

        • eitje
        • 11 years ago

        and how ab out the irony of a drawer full of spoons, when all you need is a knife?!

          • ludi
          • 11 years ago

          That IS ironic, I do think.

            • PRIME1
            • 11 years ago

            A little too ironic.

            • moriz
            • 11 years ago

            …don’t you think?

            it’s like RAIN!!!!!!!!!! on your wedding day!
            a freeeeee RIDE!!!!!!! when you’re already there!

            bah, i can’t remember the rest of that chorus πŸ™

    • Fighterpilot
    • 11 years ago

    They should’ve just called the new BIOS a “Big Bang” driver πŸ™‚
    Nice to see the 4830 will perform even better than TR benched it at.
    Seems a little hard to believe that no one at AMD was charged with checking the review samples before they were sent out tho….somebody screwed up there.

    • marvelous
    • 11 years ago

    Β§[<http://www.anandtech.com/video/showdoc.aspx?i=3441<]Β§ We asked, and AMD answered. They did not test the review samples before they sent them out to reviewers. We can usually expect to recieve boards that have not been fully QA'd, as that can take a while, but when ever we get new boards or software companies tend to make sure that what we are getting works right. Apparently AMD was in such a rush to get reviewers parts for launch that they didn't have time to run even a basic check after the cards came back from the factory (which ever company they use to build their reference boards). So the answer is that they were in a hurry and assumed that the correct BIOS would be installed. It wasn't and they didn't catch it. Now, apparently there were some review samples that had the correct BIOS on them. We aren't sure who received those samples, but AMD indicated that it was based on how early the sample was sent out. Those who got later batches were more likely to have boards with the correct BIOS. So why won't this happen in the wild? Because AMD's board partners all QA the boards they sell and because they all had a different version of the BIOS (one that functioned correctly) from the beginning.

    • SpotTheCat
    • 11 years ago

    You sure are hyping something up with all of that ‘I’m busy’ speak.

      • Sargent Duck
      • 11 years ago

      We don’t pay him to sit around and do nothing do we?

        • ssidbroadcast
        • 11 years ago

        We don’t pay him /[

          • Firestarter
          • 11 years ago

          The mindspace occupied by the ads is not cheap :p

          • toxent
          • 11 years ago

          Speak for yourself. Some of us donate happily to this site.

            • DrDillyBar
            • 11 years ago

            indeed

        • jss21382
        • 11 years ago

        depends on whether you donate to the site or not

    • PRIME1
    • 11 years ago

    Kind of an about face from their original statement

    r[

      • SecretMaster
      • 11 years ago

      Don’t try to misinform readers. Yes, that was the initial comment- their preliminary speculation. The key word in what you quoted is that “they believe”, meaning that hadn’t investigated it fully. When they did, they contacted Damage again saying:

      /[<"AMD has indeed confirmed that it shipped reviewers Radeon HD 4830 sample cards with too many units disabled, hurting performance versus the shipping products."<]/

        • PRIME1
        • 11 years ago

        I did not misinform anything, that was a direct quote.

          • SecretMaster
          • 11 years ago

          And you completely ignored the more recent and relevant statement that was issued to Damage. That is how you are misinforming people.

            • TO11MTM
            • 11 years ago

            Maybe he was referring to AMD making an about face?

            • PRIME1
            • 11 years ago

            Exactly…….

      • deruberhanyok
      • 11 years ago

      How is it an about face if they were guessing at the cause, tested it and then realized (and stated) they guessed incorrectly? Sounds to me like the same kind of theorizing and troubleshooting we all do on a regular basis.

        • PRIME1
        • 11 years ago

        First they blame the utility and say their cards are fine, then when it turns out they are selling gimped cards to the public they suddenly change the story.

          • DaveBaumann
          • 11 years ago

          On Wednesday evening the application was incorrect – we checked and had boards reporting different numbers of shaders but exactly the same performance. By Thursday the application was changed and we validated what he was using to check the number of shaders.

            • PRIME1
            • 11 years ago

            Β§[<http://www.techpowerup.com/articles/other/155<]Β§ A clear performance difference was found prior to the statement.

            • DaveBaumann
            • 11 years ago

            Prime, do you know exactly when I replied to Scott and exactly how long between that and Scott’s post? Do you know what testing had occured prior to that?

            There’s no conspiracy here. We tested on Wednesday evening with one version of the app that showed different numbers on shader on boards that had exactly the same performance. Later on Thursday, in discussion with W1zzard, we validated that the newer version of that app (that is linked in the article) was doing the correct thing and in doing so we looked at exactly what was happening with the boards.

            • PRIME1
            • 11 years ago

            Did Scott email you prior to the post about the Techpowerup article? If so how did he know about the issue? As the post I quoted was in response to the link to that article.

            Β§[<http://www.techpowerup.com/articles/other/155<]Β§ They clearly had the problem nailed and yet a response discrediting their site was the first response. I'm not saying there is any sort of conspiracy, I'm just saying that the first AMD response (blaming the utility and denying a performance issue) was at odds with this most recent post. Simple as that. I'm assuming AMD would not even need to use GPU-Z to know how many shaders their cards are using. Very CYA

            • DaveBaumann
            • 11 years ago

            “/[

            • flip-mode
            • 11 years ago

            Dave, FYI, Prime1 has a looooong history of fanaticism, an not just for Nvidia (hint hint, Dubya directly controls oil prices with a little dial in the Oval Office – yep). He has an equally looooong history of referencing *[

            • PRIME1
            • 11 years ago

            So Techpowerup was able to determine which cards were bad and AMD was not (initially)?

            • flip-mode
            • 11 years ago

            What is so hard to understand about this? TPU used different software than AMD. Not… that… difficult. Is it a tad bit embarrassing? Yeah. Beyond understandable? Not remotely.

            • PRIME1
            • 11 years ago

            That their own software and testing could not figure it out?

            • DaveBaumann
            • 11 years ago

            Again, the boards were were looking at were fine, they were the correct configuration and the correct performance, yet the application was reporting something different.

            • Philldoe
            • 11 years ago

            Don’t bother trying Dave, he’s already fallen to the dumb side.

            • PRIME1
            • 11 years ago

            Fair enough. I can agree to disagree. I was not looking to drag this into a huge fuss. I just thought the initial quote kind of made Techpowerup look bad without doing some more research first.

            It’s good that AMD did come out and make the new statement and that the issue seems pretty limited.

            • flip-mode
            • 11 years ago

            Yeah, it happens. What is so hard to believe about that? And AMD comes out and tells us about it, unlike Nvidia who hides every possible thing.

            This is a couple of individual cards rushed out to benchmarkers not being QA’d quite right. It isn’t cards going to consumers. Your devotion and worship of Nvidia is messing with your senses.

            • PRIME1
            • 11 years ago

            l[

            • cegras
            • 11 years ago

            You mean the 400 that were released to HIS and then had very clear instructions posted about how to fix this problem?

            • PRIME1
            • 11 years ago

            You mean how flippy said none made it to market?

            :rollseyes:

            • cegras
            • 11 years ago

            Okay, sure, if you like to take more stock in people’s statements then the company’s official one, you go ahead and keep doing that.

            Your whole point was incredibly inane and off topic anyways. None the less, you’ve successfully diverted everyone off on a tangent.

            That doesn’t change the fact that the 4830 beats the 9800 GT. Enjoy. PhysX is currently stupid and useless, and folding doesn’t have the same subjective value for everyone, so by just judging via a performance metric ..

            • PRIME1
            • 11 years ago

            l[< PhysX is currently stupid and useless<]l Odd that's how most people feel about your posts...... Coincidence?

            • mattthemuppet
            • 11 years ago

            ooh, I spy a kettle!

            • cegras
            • 11 years ago

            I’m as bad as PRIME1?

            What a horrible insult .. QQ!

            • cegras
            • 11 years ago

            l[

            • flip-mode
            • 11 years ago

            Dude that was awsome Prime1. You could totally dunce a -[

            • flip-mode
            • 11 years ago

            There is a difference between intentional obfuscation and honest mistake. I made the latter, and forgot about the mere less-than-400 cards that went to HIS. This means some cards went to consumers with the wrong BIOS. Did all less-than-400 go or was it less than less-than-400? Dunno. This makes it even more embarrassing for AMD and a little more of a headache. But I still see nothing more than an honest mistake resulting from being in too much of a rush.

            Nvidia tries to do a media blackout regarding an enormous problem with their chips failing and you stomp around chanting “there’s nothing to see here, the problem is smaller than it’s being made to seem”. AMD makes an embarrassing but honest mistake by putting the wrong BIOS on a small number of cards and you start with the trolling. It’s pathetic, it’s lame, it’s fanatical, it’s fanboyish, it’s totally obvious.

            • Flying Fox
            • 11 years ago

            At least it is better than Asus shipping Netbook/Notebooks with viruses or cracked software (well, the latter may not be a bad thing lol). πŸ™‚

          • l33t-g4m3r
          • 11 years ago

          It is a small amount with the pre-production bios, easily fixed with an update.
          The cards are not actually gimped.
          I bet most of them were recalled before being sold too.

          IMO, you are trying to spin it worse than it actually is.
          It’s better to play it safe and have everyone check, than to cover it up and pretend it doesn’t exist.
          (like failing cards.)

          • Flying Fox
          • 11 years ago

          The keywords here are “they believe”, meaning it was a guess. Of course when the guess is wrong they need to correct it. Calling them “blaming the utility” is a bit strong, because it was a guess as others have pointed out. If “they said the utility was reading it wrong…”, then it will be a flat out blame.

      • Meadows
      • 11 years ago

      Stop caring about what AMD thinks, right now you’re supposed to throw a fit that their cards are even better than originally.

    • ReAp3r-G
    • 11 years ago

    well at least it’s not so bad, since it’s just a faulty BIOS which the board partners will rectify on their own with little or no problems at all πŸ™‚

    • pogsnet
    • 11 years ago
      • FireGryphon
      • 11 years ago

      They get points for this one, but *always* honest? I dunno.

        • A_Pickle
        • 11 years ago

        Yeah. πŸ˜€

        It changes when they’re on the short stick, performance-wise. I remember reading on the Inquirer about how Intel had published this white paper espousing the wonders of the Pentium D 955 Extreme Edition, and how it was the fastest thing since sliced bread against AMD’s FX-62. Obviously, the Inquirer went to town on the obviously false article…

        …but I want to say I recently saw something published by AMD comparing their Phenoms to Intel’s Core 2 Quads, and theirs were coming out on top an unrealistic number of times. ‘Course, as I have no source or any idea where I might find that, I fail. πŸ™

          • MadManOriginal
          • 11 years ago

          That’s true, it’s not as if the marketers are always truthful or at least non-deceptive. There is a difference between fudged benchmarks, or ones that are cherry-picked or distorted, and this case where AMD said they i[

            • Draxo
            • 11 years ago

            “said they believe it’s an error with GPU-Z”

            I would only say a statement like this if.

            1. The specs for the product were fixed.
            2. These are samples/products from one board partner.
            3.You may try all you want but you have no control over someones quality.

      • derFunkenstein
      • 11 years ago

      …after they’re dishonest. πŸ˜‰

    • bjm
    • 11 years ago

    Other pressing obligations… like a Nehalem review?!

      • charged3800z24
      • 11 years ago

      Most likely…I’d like to see Deneb/shanghi through in there too.. but probably won’t..

Pin It on Pinterest

Share This