Slashdot Mirror


NVIDIA Begins Requiring Signed GPU Firmware Images

An anonymous reader writes: In a blow to those working on open-source drivers, soft-mods for enhancing graphics cards, and the Chinese knock-offs of graphics cards, NVIDIA has begun signing and validating GPU firmware images. With the latest-generation Maxwell GPUs, not all engine functionality is being exposed unless the hardware detects the firmware image was signed by NVIDIA. This is a setback to the open-source Nouveau Linux graphics driver but they're working towards a solution where NVIDIA can provide signed, closed-source firmware images to the driver project for redistribution. Initially the lack of a signed firmware image will prevent some thermal-related bits from being programmed but with future hardware the list of requirements is expected to rise.

192 comments

  1. Alibaba by rsilvergun · · Score: 4, Insightful

    I'm guessing this is a response to Alibaba, where you can buy a $300 graphics card for $100 so long as you're OK with being an $80 card with a flashed bios. Remember folks, if it looks too good to be true it probably is :(.

    --
    Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
    1. Re:Alibaba by ThatsMyNick · · Score: 3, Interesting

      They are selling nvidia cards with a modded firmware? Why? Nvidia is going to change their hardware, and hardware will only accept signed firmware. Fake cards, can choose to simply not do any signature checks on their hardware. Unless the fake cards are real nvidia cards, which for some reason run a modded firmware instead of nvidia singed firmwares, this will have no effect on them.

      This is to simply prevent modding. Modded firmware often pushed the hardware beyond the recommended limits. This is more like some of the android phones only accepting signed firmware.

    2. Re:Alibaba by Anonymous Coward · · Score: 5, Informative

      I don't think you understand what these fake cards are.

      They are legitimate $80 nvidia cards, flashed with modded firmware to enable cores and clock speeds that the silicon is incapable of handling (Hence why the chip became an $80 card in the first place, instead of a $300 card).

      The modder then puts the flashed $80 card on e-bay for $200 and makes a sweet profit.

    3. Re:Alibaba by Anonymous Coward · · Score: 1

      nVidia will often use some of the same hardware in their lower end cards as they do in their higher end cards, with some functionality throttled or disabled. Taking one of these lower end cards, and flashing it with a modified firmware meant for a higher end card will sometimes activate some functionality, but at the least the card will report to the OS that it is a better card than it is. Then you jack up the price and sell it to suckers/cheapskates. Again, if it seems like a deal is too good to be true, it probably is.

    4. Re:Alibaba by Anonymous Coward · · Score: 1

      >nVidia will often use some of the same hardware in their lower end cards as they do in their higher end cards, with some functionality throttled or disabled.

      Not just Nvida. Everyone in the computer industry bins their silicon the same way. Got an i7 chip with a bad core? you turn it into an i5. got an i7 chip that just doesn't clock very high? It becomes a cheap i7 instead of a K series.

    5. Re:Alibaba by Baloroth · · Score: 4, Interesting

      They are selling nvidia cards with a modded firmware? Why? Nvidia is going to change their hardware, and hardware will only accept signed firmware. Fake cards, can choose to simply not do any signature checks on their hardware. Unless the fake cards are real nvidia cards, which for some reason run a modded firmware instead of nvidia singed firmwares, this will have no effect on them.

      That's exactly what they are. It's pretty trivial to take, say, GTX 440, and reflash the firmware to report that it's a GTX660. It's extremely difficult to make a fake nvidia card that isn't actually an nvidia card that actually works as a video card and isn't completely obviously a fake. The story was even on slashdot.

      --
      "None can love freedom heartily, but good men; the rest love not freedom, but license." --John Milton
    6. Re:Alibaba by Anonymous Coward · · Score: 1

      flashed with modded firmware to enable cores and clock speeds that the silicon is incapable of handling

      How do you know the silicon is incapable of handling higher speeds? For all we know, the hardware is capable of the speed, but the official nvidia bios intentionally degrades the speed so nvidia can sell at a lower price.

    7. Re:Alibaba by ThatsMyNick · · Score: 1

      That is interesting. Thanks for the link

    8. Re:Alibaba by K.+S.+Kyosuke · · Score: 2

      NVIDIA did that three times with GT430s. They started selling them as fake GT 520s, then they continued with selling them as fake GT 620s, and today they're selling them as fake GT 730 128b DDR3. They're perfectly happy with fakes on the market, they just don't want anyone else to sell them!

      --
      Ezekiel 23:20
    9. Re:Alibaba by Mashiki · · Score: 5, Informative

      Quite often they're not even unlocking anything. Rather they're doing a dirty hack to change the bios information of the card to display something that it isn't. This isn't all that unfamiliar to those of us who were in the industry back in the mid to late 90's when scammers were resilking(cpu info used to be silk screened on, to counter this it's why all cpu's are now stamped) Cyrix cpu's as AMD and Intel. You only found out what the CPU actually was, when you plugged it into the board and it said "cyrix." And while there are cases of people doing this to binned parts, most of the time the links to enable those pathways are cut before they're made into a gpu to stop people from doing exactly that. And if you're wondering why, it's because Intel ran into a massive problem where fly-by-night companies would unlock the binned CPU, and then actually flashing the microcode to change what the CPU was.

      The cheap and dirty way to unlock CPU's during that time period was to use a graphite pencil across a unfinished path. I think it was pin 14 or 23 on the board. Very nasty problems with Slot 1 cpus.

      --
      Om, nomnomnom...
    10. Re: Alibaba by YaHooL · · Score: 1

      There was this case in which GT440 where modified to report themselfs as GTX660:
      http://m.hexus.net/business/ne...

    11. Re:Alibaba by Anonymous Coward · · Score: 1

      The thing is, nVidia does not sake "fakes". They relabel their own products and guarantee certain specs.

      The 3rd parties sell fakes. They take something like an nVidia card, reflash the bios and resell it as something that is no longer guaranteed by nVidia. They did this before. Take customer card, flashed it as Quadro.

      See the problem?

      Customers can get shafted in the 2nd version and have no recourse. Then because they are ignorant in the first place, blames nVidia for the fuck up. In the 1st version, the customer gets the speced card that works properly.

      So, what's the problem? Nothing really. The nouveau project, as indicated, will probably get signed firmware bits to load onto cards that need it and everything works for them. The counterfeiters no longer can modify firmware to load only wrong hardware.

      It's definitely a win for nVidia customers. They'll be able to verify what they bought is what they have.

    12. Re:Alibaba by Anonymous Coward · · Score: 0

      Evidence? That's a serious accusation.

    13. Re:Alibaba by spire3661 · · Score: 1

      Becasue i trust Nvidia's binning process over some nobody on Alibaba.

      --
      Good-bye
    14. Re:Alibaba by hairyfeet · · Score: 1

      Bimbo Newton Crosby, Nvidia has been getting a LOT of shit about how easy it is to flash the firmware to lie about the card and I am 100% with the users on this as there is ZERO way to tell what chip a graphics card is using without tearing the card apart, voiding any warranty, and possibly damaging the card in the process. After all if I hand you a card with a huge double heatsink and tell you its the latest card, the screencaps show GPU-Z showing its the latest card, how else are you gonna tell what it is without tearing off the sink and seeing what chip is underneath?

      I do think its funny the author tries to frame it as a "freedum!" issue which I would argue it is, only its Nvidia trying to give its customers the freedom from being scammed!

      --
      ACs don't waste your time replying, your posts are never seen by me.
    15. Re:Alibaba by Anonymous Coward · · Score: 0

      well, if it was 0.01 of not voting shares it would be a steal...

    16. Re:Alibaba by Anonymous Coward · · Score: 0

      nvidia bins their chips based on tests and market demand. if you take a chip that performed poorly in the tests, then run a firmware that ignores the fuse settings you can overclock it. And in some cases even enable workstation support like a quadro card (~$500). But the chip is likely to fail to be stable under load when you do such things, and you will be unhappy that you've wasted $100 on an $80 graphics card with no warranty.

    17. Re:Alibaba by Anonymous Coward · · Score: 1

      You can tell by simply running a good benchmark like, say, 3DMark - if the performance doesn't match the known average performance (which is easy to find from reviews or from 3DMark result database) of the model and/or if you see a stability issue, it's a dud.

      All truly "economical" fakes - taking substantially cheaper hardware and modding name strings - would be caught by this. The recent GT440-sold-as-GTX 660 got busted by simple benchmarking proving the card was not performing anywhere near GTX 660 levels.

    18. Re: Alibaba by neoedmund · · Score: 1

      In China, taobao,com, a fake GTX780 can be $20. crazy. I know it's fake one. Bad thing is I cannot tell cards sold in high price is true. Nvidia is to be blamed let such thing happen. Intel CPU cannot be modded like this, right?

  2. One word by Anonymous Coward · · Score: 0

    Robot Chicken!

  3. f**k nvidia... by Anonymous Coward · · Score: 2, Insightful

    "NVIDIA, f**k you!" - Linus Torvalds

    1. Re:f**k nvidia... by smash · · Score: 1

      Pretty much. Also, given the general purpose nature of GPUs now for running "other" code, it's only a matter of time before someone writes malware that lives in your GPU firmware.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    2. Re:f**k nvidia... by Anonymous Coward · · Score: 0, Redundant

      Malware in your GPU? It's a good thing 4chan doesn't have firmware hacking skills; We'd be faced with malware that makes all the textures in GPU memory into goatse.jpg whenever a scene is rendered.

    3. Re:f**k nvidia... by Anonymous Coward · · Score: 4, Insightful

      Nonsense, it's not on Nvidia to stop fake cards, its on law enforcement. All they are doing is fucking everyone into having to use their signed firmware. Its another avenue for NSA style snooping from public and private parties without you ever knowing. No code reviews, no way to know if that signed firmware is actually what you would've compiled from any code snippets they may provide. No way to know if other functions are being executed from withing the code without your knowledge.

    4. Re:f**k nvidia... by Anonymous Coward · · Score: 1
    5. Re:f**k nvidia... by armanox · · Score: 2

      It's on nVidia to keep people from counterfeiting cards, it's on law enforcement to punish those who do. It's one of those problems you attack from both ends.

      --
      I'm starting to think GNU is the problem with "GNU/Linux" these days.
    6. Re:f**k nvidia... by west · · Score: 3, Funny

      Nonsense, it's not on Nvidia to stop fake cards, its on law enforcement.

      Actually, if it cuts their into sales because purchasing NVidia is perceived as risky, then it makes complete commercial sense to make changes to protect people who think they're purchasing NVidia. It's straight dollars and cents.

      Now perhaps NVidia is only using this as an excuse to launch their evil conspiracy, but as excuses go, it's completely legit.

      (And while I'd love to make fun of you for the evil conspiracy business, the NSA's actual shenanigans have made that impossible. When the utterly improbable has turned out to be true, the completely ridiculous now becomes only highly unlikely...)

    7. Re:f**k nvidia... by Anonymous Coward · · Score: 0

      What is the problem with using Nvidia's signed firmware? It's not like you need to reflash it to install a driver.

      This is a complete non-issue.

    8. Re:f**k nvidia... by Anonymous Coward · · Score: 0

      Except that they'll just find a way of cracking the protection and doing it anyways. There's just way too much profit in this for people to no crack the protection.

      This is why manufacturers have official distribution channels, if you're buying from a legitimate site,then you've got nothing to worry about, but if you're buying through a 3rd party for a really good price, then you might get screwed. There's not much that nVidia can do about that.

    9. Re:f**k nvidia... by JustNiz · · Score: 1

      >> it's not on Nvidia to stop fake cards, its on law enforcement.

      Its no surprise to me that nVidia feel the need to cover their own asses. Waiting for Law Enforcement to step up is a lame joke in most countries.

      I am disappointed that nVidia (at least so far) apparently haven't given the nouveau project the ability to sign or at least proxy-load their own drivers, but on the other hand even after all this time nouveau is (still) a turd that sucks enough to not even just not crash on some nVidia hardware.

      If this change means as a side-effect that Mint will finally go back to installing nvidia's binary rather than nouveau by default then I'm all for it, as the way it stands now I can't even install mint on my laptop since there is no way to manually override the automatic installation of nouveau, then novueau crashes and locks up the whole CPU before I can even get to a point where I can uninstall it.

    10. Re:f**k nvidia... by Luckyo · · Score: 1

      That is why you make it as hard and expensive as possible to crack the protection. Criminal outfits need to turn a profit. Make counterfeiting process hard enough, and their profits dwindle and they go do something more profitable. You win.

    11. Re:f**k nvidia... by Anonymous Coward · · Score: 0

      If Linus doesn't like it, why doesn't he start up a non-profit and develop his own completely open GPU? That's what I don't get, all of these entitled open source whiners want the companies who spend millions researching and developing these things to do all of the work for them.

      Want a completely open system? Design and produce one yourself.

    12. Re:f**k nvidia... by Anonymous Coward · · Score: 0

      Right on! Why should I have a lock on my door when we could just as well have a police officer guarding my door 24/7?

    13. Re:f**k nvidia... by Blaskowicz · · Score: 1

      In fairness there's something about it in the release notes, and a workaround that seems farily easy (no dicking around in the command line)

      http://www.linuxmint.com/rel_q...

    14. Re:f**k nvidia... by Anonymous Coward · · Score: 0

      F**k them for exactly what ? Requiring the use of binary firmware blobs like both other major GPU vendors do as well anyway (for those who do not know, such manufacturer provided blobs are also used by the open source AMD and Intel drivers) ? Or for releasing the only competently performing and up to date (version 4.5) OpenGL implementation that exists on Linux that is not a joke compared to Direct3D on Windows ?

    15. Re:f**k nvidia... by neoedmund · · Score: 1

      Do you worried about your i7 4790 cpu is actually modded from 80386? maybe not.

    16. Re:f**k nvidia... by JustNiz · · Score: 1

      Thanks for this useful info. I wasn't actually aware that it was a common enough problem that they needed to make a section for it in the release notes, however that just underlines my comments and that it is a singularly inapproriate choice to use nouveau rather than the nVidia blob as the default in a mainstream environment.
      I'm sure that most normal people would easily prioritise robustness, performance and functionality over some purist's anal sense of political correctness.

  4. Fuck That Shit! by Anonymous Coward · · Score: 2, Insightful

    That's the god damn fucking last straw. All these years I thought Nividia was slowly being dragged into the open by Nouveau. Digging their heals in but still an inexorable movement in the direction of the inevitable. But jesus fucking christ this move is such bullshit, 2 steps forward and 5 steps back. No more nvidia for me. They've just made AMD the only choice for graphics cards.

    1. Re:Fuck That Shit! by smash · · Score: 2, Insightful

      You know WHY people implement code signing, right?

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    2. Re:Fuck That Shit! by Anonymous Coward · · Score: 0, Interesting

      Oh please enlighten us of the one true reason why people implement code signing /sarcasm

    3. Re:Fuck That Shit! by goarilla · · Score: 5, Funny

      - To implement a chain of trust.
      - To tighten their control.on the hardware you paid good money for.
      - To fight terrorism, childporn, illegal goods.

    4. Re:Fuck That Shit! by Anonymous Coward · · Score: 4, Insightful

      So that they can pretend to sell hardware while maintaining permanent control of it?

    5. Re:Fuck That Shit! by Luckyo · · Score: 1

      How does ensuring that certain FIRMWARE stays on the card do that, unless they are planning some sort of massive malware conspiracy?

      This argument about drivers I could understand to an extent. But firmware?

    6. Re:Fuck That Shit! by Anonymous Coward · · Score: 0

      https://public.bn1.livefilestore.com/y2podx0T3QYYS7yjO0Z26SEW9ZiqVU_aWlbC-PQUFAS9Hcs_gPmvxIqyzjuAJ8i5RWleCPLDYoS1RcOFtg8kBM0v0aXKn1R1rS_oE-WVeYAdm0/Linus Torvalds F U NVIDIA Wallpaper.jpg?rdrts=86462669

    7. Re:Fuck That Shit! by Anonymous Coward · · Score: 0

      u mad bro?

    8. Re:Fuck That Shit! by Anonymous Coward · · Score: 0

      > This argument about drivers I could understand to an extent. But firmware?

      Firmware is just another kind of driver. Every argument for open drivers applies equally to open firmware. Binary firmware is a midway point, locking it down is an extreme position.

    9. Re:Fuck That Shit! by Anonymous Coward · · Score: 0

      Meanwhile, the adults doing the professional graphics rendering and supercomputing on Linux (The ones who buy those delicious high-margin $5000 Quadros and Keplers) will continue to ignore your screaming fits about the driver that it is always my first job to remove from a new install, and enjoy the fact that nVidia makes best-in-class hardware and provides fully-featured drivers for Linux.

    10. Re:Fuck That Shit! by Anonymous Coward · · Score: 0

      In case you did not know, all open source GPU drivers (including Intel and AMD) require proprietary firmware blobs to work. But I guess facts rarely get in the way of blind zealotry.

    11. Re:Fuck That Shit! by hobarrera · · Score: 1

      Why not Intel? The latest Intel lines are pretty good (I game a lot on one of those), and have their drivers in the upstream linux kernel.
      AMD is simply so awful in terms of drivers, that it doesn't really matter if the hardware is slightly better.

  5. f**k nvidia... by Anonymous Coward · · Score: 4, Insightful

    Yeah. F**k Nvidia for keeping scammers from selling faulty video cards with hacked bios's.

    How dare they protect their brand integrity!

  6. Supply & Demand by Anonymous Coward · · Score: 0

    Look at the bright side. Since this will hurt demand for NVIDIA products, they will be cheaper. Being cheap is a good thing.

    1. Re: Supply & Demand by Anonymous Coward · · Score: 0

      Hurt demand? Because of some obscure shit only losers care about? Keep dreaming. Linux on the desktop yet? :)

    2. Re:Supply & Demand by Anonymous Coward · · Score: 0

      Of the people who use computers, how many use linux, and also need powerful GPUs, and have an ideological objection to closed source drivers?

      Fixed that for you.

    3. Re:Supply & Demand by tepples · · Score: 1

      And Linux gamers will use the proprietary driver anyway because the vast majority of games with professional-class production values are proprietary software, so it's not introducing any extra taint.

    4. Re:Supply & Demand by smash · · Score: 1

      The only people who are going to get butt-hurt over this are a tiny fraction of Linux users who represent a tiny fraction of a tiny fraction of the GPU market.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    5. Re:Supply & Demand by allo · · Score: 1

      Or just want the drivers to work, which might require some patching. okay, the nvidia driver works ootb in most cases, but in the past there were vmware-driver patches for each new kernel.

    6. Re:Supply & Demand by Luckyo · · Score: 1

      Most of people running gaming GPUs on linux are probably gaming.

      That means nvidia's own closed source drivers. Open source drivers are utterly crippled when it comes to gaming, and it would take a huge masochist to use them.

    7. Re: Supply & Demand by Miamicanes · · Score: 1

      Keep dreaming. Linux on the desktop yet? :)

      At the rate Microsoft is going in their mad race to piss off & alienate just about everyone with a high-end workstation (by pushing Windows towards dumbed-down touch-based interfaces), that goal is actually starting to look attainable. Five years from now, one of two things will likely happen:

      * Microsoft will have finally pissed off & alienated enough users for some critical mass of high end desktop/workstation power users to decide Windows is annoying them more than making their lives easier, and vendors like Adobe will notice & release their flagship software for Linux (effectively destroying what little market would remain for high-end Windows applications).

      * Hedging their bets, companies like Adobe will port their flagship apps to Linux... then port them back to Windows with "kde6.dll" as a dependency. IMHO, this is Microsoft's ultimate nightmare scenario. If the apps high-end workstation users care about are all native KDE apps with equally good Linux versions, there's literally nothing left at that point to keep them chained to Windows. They'd basically be running Linux under a Windows kernel through a compatibility thunking layer anyway. ESPECIALLY if the apps are licensed in a way that allows users to buy the app once, then install & run it under BOTH Windows AND Linux.

      Why KDE, and not Gnome? Licensing & logistics. KDE is Apache-licensed, so there's nothing to stop Adobe from bundling an installer for KDEwin directly into their own installers to auto-install it if the user hasn't done so already. And KDE for Windows already exists in beta form (see: http://windows.kde.org/ ).

      Five years from now, we might not all be running Linux per se... but most of us will probably be running "Winux" (Windows kernel, Linux UI).

    8. Re:Supply & Demand by Anonymous Coward · · Score: 0

      "Cheap supercomputing work"? I rarely see these words used toghether. I suspect a large portion of professional and academic supercomputing is done on NIXes.

    9. Re:Supply & Demand by X0563511 · · Score: 1

      Working with a complex scene in Blender with a Intel "graphics" is about as fun as rubbing sandpaper in your eye.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    10. Re: Supply & Demand by Anonymous Coward · · Score: 0

      What the fuck are you smoking? Five years from now the most likely thing is Windows on 90%+ of desktops, and no one but losers give a shit about Linux (~1% or less of desktops, as usual). No software company has ever and will ever make Windows software with KDE as a dependency. Keep dreaming. Oh yeah, and the so called Linux UI sucks so badly, amateurs are doing such an incredibly shitty job of it, that no one will ever willing use the "Linux UI" unless they hate themselves.

    11. Re: Supply & Demand by Anonymous Coward · · Score: 0

      Linux on the desktop has been and is running just fine been using windowmaker GUI for 14 years...
      People are just parroting how it's not desktop ready, when in fact most parts of UI is really done better
      windows 7 doesn't even have properly working virtual desktops, but hey its the DESKTOP! lol

      games and some proprietary software on otherhand rolls mostly to windows world.... (but bit another topic)

    12. Re:Supply & Demand by Luckyo · · Score: 1

      I imagine working on it with open source nvidia drivers is not going to be much better, just in a different way. One is painfully slow, other is less slow but painfully crashy.

    13. Re:Supply & Demand by X0563511 · · Score: 1

      Yep, unless you've got one of a quite small list of well-supported cards. Those are not so powerful these days, but if you've got one then you're still worlds better off than with an Intel.

      I've tried... two ATI (back then) cards, one I never got to work, and the other found all sorts of interesting ways to crash and malfunction. Have to admit I haven't tried again since - the devil you know and all that.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    14. Re: Supply & Demand by Anonymous Coward · · Score: 0

      Linux on the desktop has been and is running just fine been using windowmaker GUI for 14 years...

      WindowMaker? Seriously? How can you even compare that to anything remotely modern? Gee, what an accomplishment to get an extremely ascetic window manager running.

    15. Re: Supply & Demand by Anonymous Coward · · Score: 0

      Hedging their bets, companies like Adobe will port their flagship apps to Linux... then port them back to Windows with "kde6.dll" as a dependency.

      Now that, right there, is some funny shit.

    16. Re:Supply & Demand by Luckyo · · Score: 1

      Why not just use closed source drivers?

      About the only reason I can see not to is putting ideology ahead of pragmatic solution in a problem where being pragmatic will have little to no impact on ideological struggle for open source. Nvidia won't care either way and won't open source its drivers even if every single linux user were to ask for it. Too small of an audience.

    17. Re:Supply & Demand by X0563511 · · Score: 1

      That's the question I ask myself, and I don't have a satisfactory answer. I use the proprietary drivers.

      --
      For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
    18. Re: Supply & Demand by Truekaiser · · Score: 1

      Not so funny anymore.
      http://www.dpreview.com/articl...

  7. Out of the frying pan... by Anonymous Coward · · Score: 0

    Um. Correct me if I'm wrong, but isn't AMD/ATI even worse with open source drivers?

    1. Re:Out of the frying pan... by guises · · Score: 2

      Yes, you are wrong. Back in 2007 AMD started releasing developer documentation and support for the development of open source drivers. This is the "Radeon" driver that you may see in repositories, and it's pretty good at this point. I don't know if 3D is fully supported, but for desktop stuff it's stable. That's in contrast to the Nouveau open source driver for Nvidia cards, which is reverse engineered.

      What you may be thinking of are the closed source drivers for Linux: Nvidia's closed Linux driver is better than AMD's. AMD's used to be notoriously bad, but it's gotten better over time. To my knowledge it's still not as good as Nvidia's, but they're both usable at this point.

    2. Re: Out of the frying pan... by Anonymous Coward · · Score: 0

      The 3d stuff does work, just not for the latest cards. I had it working for years with my previous card though, well after AMD stopped supporting it.

      In the short term the official drivers seem to do a good enough job, but it's generally better to use the open drivers when possible as they tend to work better.

    3. Re: Out of the frying pan... by Anonymous Coward · · Score: 0

      Is there any point in having the latest card if 3d doesn't work?

      Seems a bit like having a Ferrari that's stuck in first gear, don't you think?

    4. Re: Out of the frying pan... by Xtifr · · Score: 1

      The 3d stuff does work, just not for the latest cards.

      Are you sure? You may need a more up-to-date kernel than your favorite distro provides to support the latest cards, but I certainly had the impression that AMD is actively pushing code into the mainstream kernel. Southern Island and Sea Island chipsets are both supported in 3.16 (and possibly earlier).

      3D performance on many ATI/AMD cards is actually better with the open source kernel drivers than with the proprietary drivers if you have a recent enough kernel, according to some reports I've heard.

    5. Re: Out of the frying pan... by Anonymous Coward · · Score: 0

      Nah. It's just that ATI(AMD) catalyst drivers are pretty much crap under linux and slightly better for windows.

      For older cards, the OSS driver might end up being better than crapalyst.

      Either way, I don't give a flying fsck and moved back to nvidia whose drivers are usually pretty good. OSS is just not a purchasing req for me, especially on things like GPUs and even more so after having lived trapped with a notebook with a shitty ATI GPU and their godawful drivers.

      OTOH up until their last space heater replacement design I had thought that they were pretty good at GPU design, which along with people telling me that oh yeah crapalyst is much better nowadays... well it wasn't, but the hw design of that particular(pre-GCN) was pretty good probably horribly hobbled by their shitastic drivers.

    6. Re: Out of the frying pan... by Xtifr · · Score: 1

      For older cards, the OSS driver might end up being better than crapalyst.

      Actually, it's the newer cards that are reported to have better performance with the latest OSS drivers.

      This is a pretty recent development, though. A little over a year ago, everything I'd heard was in line with what you're saying. But when I heard about all the improvements in the recent OSS drivers, I took a chance and bought a box with ATI, just a couple of months ago, and I must say that it's been absolutely smooth, effortless, and 100% hassle free. With a 3.12 kernel (now upgraded to 3.14). And no catalyst.

      If your experience is older than that, then I think your information may be out of date.

      Not that I'm saying you should switch or anything. No skin off my nose. Just saying your data may be out of date.

  8. Firmware != Drivers by Anonymous Coward · · Score: 0

    Firmware != Drivers

    1. Re:Firmware != Drivers by LordKronos · · Score: 1

      You are right, they are not the same thing. However, some pieces of hardware have no firmware instanlled and require the driver to upload it on boot. I have no idea if that's how nvidia cards work, but I know that years ago when I was using hauppauge analog tv capture cards, this was exactly the case. Instead of flashing the firmware to the card, it just gets loaded every time you reboot. In these cases, people passionate about open source are probably going to want to use open source firmware to go with their open source driver. Or perhaps using the closed source firmware with the open source driver isn't even an option. I really don't know.

    2. Re:Firmware != Drivers by allo · · Score: 1

      the companies have licenses, which prohibit distribution of the fireware, except for some ways they thought of (like download of the official windows driver). So distributing the fireware itself is illegal and you get tools like the bcm-firmware-cutter, which extracts the firmware from the windows driver binaries. This is legal, as long as the user downloads the firmware (so the tool maintainer does not distribute anything from the company).

    3. Re:Firmware != Drivers by Anonymous Coward · · Score: 0

      FWIW, Hauppauge has now moved to digital TV capture cards, and they still work the same way. And as far as I can tell, there never existed open-source versions of the firmware; the open source drivers just uploaded a closed-source binary blob to the card to get it running.

      I don't know if this is the case for Nvidia, but the part about nouveau in the summary suggests it might be.

  9. As Linus said "Fuck you NVIDIA" by Anonymous Coward · · Score: 3, Informative

    Surely it is impossible to have an opensource software if it needs a key to build it into a runnable program?

    I mean you have the binary but you cannot recreate it from the source without that key to sign it with. The key is part of the source and you don't have it.

    1. Re:As Linus said "Fuck you NVIDIA" by mx+b · · Score: 2

      Surely it is impossible to have an opensource software if it needs a key to build it into a runnable program?

      I mean you have the binary but you cannot recreate it from the source without that key to sign it with. The key is part of the source and you don't have it.

      This is pretty much the reason the GPLv3 was written, to take care of this loophole in other licenses. If there are other parts of the GPLv3 that people don't like, perhaps we can update it and make a nice GPLv4, but many people throw the baby out with the bathwater with their hatred of GPLv3. I think having the ability of signing the keys yourself is an important topic.

    2. Re:As Linus said "Fuck you NVIDIA" by Anonymous Coward · · Score: 0

      Given this as Nvidia's reply, I guess "FU" really isn't the last word in any argument.

    3. Re:As Linus said "Fuck you NVIDIA" by Anonymous Coward · · Score: 0

      GPL2 requires "preferred form" for making modifications. This should require the signing key. I prefer not to have source without the singing key.

  10. Supply & Demand by Anonymous Coward · · Score: 0

    I'm pretty sure that the people upset by this will equal less than 1% of their total GPU sales.

    Of the people who use computers, how many use linux, and also need powerful GPUs?

    You're talking about people doing cheap supercompuing work, and linux gamers. Thats the entire market segment affected by this.

    That and the people in china making counterfeit graphics cards.

  11. Not really new practice for Linux. by Anonymous Coward · · Score: 0

    For my Broadcom wirless card in Linux, the firmware has to be ripped out of the windows driver, and put into a compatible Linux driver layer. Sounds like something similar will happen with Nvidia.

    On the bright side, most Linux distros have a package that does all of this automatically, so it's not so bad from a user standpoint.

    1. Re:Not really new practice for Linux. by tepples · · Score: 1

      Then don't buy Broadcom. Requiring that a signed non-free blob be copied to the device on every boot is fairly strong evidence that NVIDIA and Broadcom have no plans to ever qualify a product for Free Software Foundation's "Respects Your Freedom" certification mark.

    2. Re:Not really new practice for Linux. by Greyfox · · Score: 5, Funny

      On the other hand, nothing tastes quite as good as the tears of an engineering group that put several million dollars into R&D for a DRM scheme, just to have it broken by a Swedish teenager three days after their product goes live.

      --

      I'm trying to teach myself to set people on fire with my mind... Is it hot in here?

    3. Re:Not really new practice for Linux. by Anonymous Coward · · Score: 0

      Well, it has already been determined that NVIDIA uses a symmetric key algo for their firmware signing. Yeah, that's right, symmetric. If there is a single person out there good enough to do some kind of side channel acoustic attack to extract the key from the hardware, poof, all their efforts were for nothing.

    4. Re:Not really new practice for Linux. by Anonymous Coward · · Score: 1

      Yeah, but first sale doctrine says that when you buy something: ITS YOURS! You can do whatever you want with it. Use the brand new car you bought as a boat or airplane? Sure, whatever floats your boat. But a lot of technology companies want to screw you over and say "no, we don't want you doing that with our stuff". A lot of them will make a standard product, then sell it as tiered products by crippling functionality. Enough. AMD has good functionality out of the box with open drivers. The kernel get along with it. No crippled functionality. I had had enough of NVIDIA from my last computer, and when I built this one, I went with an AMD card: and it was borked, and the store had a replacement NVIDIA card. OK, back to NVIDIA (by the narrowest of margins). Next box (and its getting soon now): AMD!

    5. Re:Not really new practice for Linux. by Anonymous Coward · · Score: 0

      Sure it can be worked around (Nvidia is basically asking everyone running en open source operating system breaking there DRM systems), but I will cross every laptop, net-book or phone containing any Nvidia chip from my list of things I would recommend to others. Since drivers keep changing after the hardware is bought, the will stay blacklisted for about 20 years _after_ they have changed there mind an promised to never pull shit like this.

      Note to Nvidia PR-department (and sales, and basically the whole company): take a break. no one is going to believe anything you say or do for the next 20 years. just mot-ball the company, wait till all of the current Nvidia hardware has all ended on a skip, and has been long forgotten, then try again.
      (Note to investors: 19.5 yeas from now would be a good time to to invest in Nvidia.)

  12. No biggie by Anonymous Coward · · Score: 0

    If it's flashable, it is hackable, right? We'll get right on it. They will rue the day when we can print our own cards. Too bad there is no real karma in this world. Then again, if there was, we would all be dead.

    1. Re:No biggie by tepples · · Score: 1

      If it's flashable, it is hackable, right?

      Perhaps the part that verifies the signature isn't flashable. Consider the Wii video game console's boot process. When the Wii is turned on, the first stage bootloader in mask ROM on the I/O processor ("boot0") loads the second stage ("boot1") from NAND flash and verifies its hash against a hardcoded hash stored in one-time-programmable (OTP) memory on the I/O processor. System updates cannot change boot1. Then boot1 loads the third stage ("boot2") from NAND flash and checks its RSA signature by comparing the hash of boot2 to the expected hash decrypted using Nintendo's public key. Try to change boot1, and its hash will no longer match the value in OTP. Try to change boot2, and its hash will no longer match the signature. (That is, unless you have an early Wii whose hash comparison function was b0rked.)

    2. Re:No biggie by Anonymous Coward · · Score: 0

      Sounds too easy. If this were the case, we would have unbreakable encryption, or what's known in the parlance, "perpetual motion". I don't believe it. With adequate sniffing the key will be found, and fairly quickly, if there is real incentive

  13. Not the same, but a subset by tepples · · Score: 1

    If the firmware were on a flash memory soldered to the video card's PCB, there wouldn't be a problem. But a lot of devices that use a proprietary blob omit the flash to save a few cents and expect the driver to copy this blob to the device at each boot. So in this case, firmware is part of drivers.

    1. Re:Not the same, but a subset by Anonymous Coward · · Score: 0

      That is not how Nvidia's or ANY video card firmware works because they need to be active at the moment of power on, before there is even an OS loaded. VBIOS is stored on the card, not copied to VRAM.

    2. Re:Not the same, but a subset by tepples · · Score: 1

      Then why isn't this GPU firmware image stored in the same chip as VBIOS?

    3. Re:Not the same, but a subset by Kjella · · Score: 1

      That is not how Nvidia's or ANY video card firmware works because they need to be active at the moment of power on, before there is even an OS loaded. VBIOS is stored on the card, not copied to VRAM.

      What you say is absolutely true yet grossly misleading and I suspect you know it. Yes, if you boot a machine with no HDD, no OS, no drivers the computer will display something to say "Hey, I have no boot disk" which is obviously built in. To get 2D/3D/video acceleration though you typically need to load a firmware module first, then you can start programming it through the API. As I understand it based on reading about AMD's open drivers which still depend on closed source hardware their opinion it the firmware makes the hardware comply with their "assembler language" GPU API. It won't function without it and explaining the actual bits would mean explaining the hardware implementation which is a tightly guarded company secret. It should also be noted that the firmware doesn't run on the CPU, it runs internally on the GPU so it's a bit like demanding how a RAID card's chip is programmed, not the driver that runs on the CPU but the programming of auxiliary chips. The funny part is that nobody cares if you use an EEPROM to write the firmware blob to that the card will read from. But if you binary dump it directly, then RMS won't be happy. I don't see the big practical difference though.

      --
      Live today, because you never know what tomorrow brings
    4. Re:Not the same, but a subset by Anonymous Coward · · Score: 0

      The VBIOS is the firmware.

    5. Re:Not the same, but a subset by tepples · · Score: 1

      But I was under the impression that using any accelerated features required the uploaded blob.

  14. not a solution at all by Anonymous Coward · · Score: 0

    The whole point of Noveau is to avoid Nvidia's closed proprietary blobs, for those Linux users who value freedom over mere costlessness.

    Presenting a costless closed proprietary blob as the solution is about as incorrect as possible.

    1. Re:not a solution at all by ledow · · Score: 1

      It depends.

      If the blob is only needed to control the thermal logic on the chips, then the vast majority of the open code that's required (interfacing with the buses, talking to the chips, converting and uploading primitives and textures to RAM etc.) is still under the control of the OS.

      I see it as akin to the wireless devices where the region-specific allowed frequencies are used. The operation of the chip - it's loading order, it's interaction and security of RAM, it's performance, it being under the control of the OS, etc. is all there. But the one bit that nVidia can't take a chance on - the dangerous stuff that you shouldn't be playing with anyway if you're concerned about the integrity of the device - is in firmware. Would it be worse if that just existed on a ROM chip and you couldn't change it at all? Probably. At least this can be upgraded to new versions.

      And the requirement for open-source is that the card can play nicely with other code - that it interacts with the OS and other drivers nicely without having to rewrite PCI drivers or hand over full control of DMA of any part of RAM to the device. That's the point of Noveau, that's WHY we want to avoid blobs. And this looks like a blob that doesn't actually matter. It could be on the card, and we wouldn't care. But if the PCI interaction logic or the DMA code that the OS has to execute is just on card ROM, then it's a whole lot harder to make the damn thing work for Linux, etc. instead of just the main market of Windows.

      Similarly, the wifi debacle of a few years ago - no, we don't want to just be passing off random blobs whose "open" parts are to basically hand off complete DMA / USB access to the device and the blackbox of code makes it work. But if we have to have a firmware of just the essential, must not play with, vendor-supplied part that doesn't interfere with operation and we can revert it at any time - that's a tiny, tiny loss for a massive, massive gain: vendor support without fear of litigation if they somehow "approve" our code that might transmit on frequencies it's not allowed to, or fry our cards.

      It's not perfect, but it's a damn sight better than some other companies offer at all. And there seems to be good reasoning behind it. And, in the end, it won't affect your use of the card for any of the primary goals you buy it for, nor will it break when you go to a Linux 3.6 kernel because they never updated the PCI code in the wrapper, or whatever.

  15. Installation Information by tepples · · Score: 1

    Surely it is impossible to have an opensource software if it needs a key to build it into a runnable program?

    Of course you can under TiVo's interpretation of GPLv2, so long as the key is not an executable part of the program. The publisher can apply the signature key as part of linking the executable.

    I mean you have the binary but you cannot recreate it from the source without that key to sign it with.

    You're referring "Installation Information" in GPLv3. GPLv2 refers to something similar in "scripts to control compilation and installation", but it's not nearly as explicit as in GPLv3.

  16. Well that's just it by Anonymous Coward · · Score: 4, Interesting

    I've had it. I don't understand why they don't just release all of the specs of the cards. Why don't they give them away for free? Or provide a 3D-printable download at the very least. Fuck nVidia!

    1. Re:Well that's just it by Wootery · · Score: 1

      I've had it. I don't understand why they don't just release all of the specs of the cards. Why don't they give them away for free?

      This is actually a good question. As I understand it, the answer is that:

      • 1) They don't want to reveal the intimate details of their architectures and/or drivers that they've invested in

      • 2) They don't want to be sued for infringing patents (either by rival GPU companies or, more likely, by patent-trolls)

      Or provide a 3D-printable download at the very least.

      ....what?

    2. Re:Well that's just it by Anonymous Coward · · Score: 0

      ....what?

      You got trolled, that's what. Dipshit.

    3. Re:Well that's just it by Anonymous Coward · · Score: 0

      I've had it. I don't understand why they don't just release all of the specs of the cards. Why don't they give them away for free?

      This is actually a good question. As I understand it, the answer is that:


      •    
      • 1) They don't want to reveal the intimate details of their architectures and/or drivers that they've invested in

      • 2) They don't want to be sued for infringing patents (either by rival GPU companies or, more likely, by patent-trolls)

      3) They have included technologies to which they are legally bound to not make available to others (license restrictions). [This was one of the reasons Sun took years to be able to opensource Solaris, because they had to go back and either get permission, or rewrite, code they did not own. It only took Oracle a few months to close source Solaris.]

      4) Signed firmware (running in trusted environments) can increase the assurance that content protection (HDCP 2.0) is not bypassed, as required by the usual suspects.

    4. Re:Well that's just it by Wootery · · Score: 1

      Good points.

  17. Real Motivation- Video card based malware by Anonymous Coward · · Score: 0

    The real motivation for this is a current generation of nVidia based root kits- they have storage, processing, and access to the system. There is currently, not publicly, a way to dump an image of a card. Shortly a tool will be released that allows DFIR folks to dump an image for analysis.

    1. Re:Real Motivation- Video card based malware by Anonymous Coward · · Score: 0

      This is hardly a new concern. GPU malware has been known about for years. And the same thing is being said about any kind of USB port or device, because they also have so-called firmware. Shouldn't we be thanking nVidia for working to keep our systems secure? Probably, but the open source mob wants everything for free.

  18. Good riddance. by Anonymous Coward · · Score: 0

    I was always torn on buying NVIDIA or not. This, it seems, settles it for me. And no, I won't go to the hassle of finding out whether this particular model is crippled or not.

    1. Re:Good riddance. by T0min · · Score: 1

      I was always torn on buying NVIDIA or not.

      Well, this is hardly a reason to buy AMD hardware instead. They already ship the firmware blobs with radeon driver.

  19. Fuck you, Nvidia by allo · · Score: 0

    Torvalds was right.

  20. Who makes the most FOSS friendls GFX HW? by Qbertino · · Score: 1

    With all this hassle nowadays - I remember the times when nVidia was the only company supporting Linux and was something like the darly child of the FOSS community - which company actually *is* the most FOSS friendly today? Intel? AMD/ATI? Some other company?

    Educated opinions on this needed.

    --
    We suffer more in our imagination than in reality. - Seneca
  21. Nvidia is shooting themselves in the foot by FudRucker · · Score: 0

    the FOSS community wont stand for it, they will just abandon Nvidia and focus on just maintaining drivers for old cards, and put their efforts in to either hacking driver signatures or ignoring the new nvidia cards and focusing on other cards like ATI and Intel

    --
    Politics is Treachery, Religion is Brainwashing
    1. Re:Nvidia is shooting themselves in the foot by mrpoundsign7072 · · Score: 1

      Which is fine, most people will just use the binary driver. At the end of the day I care about stability and performance.

    2. Re:Nvidia is shooting themselves in the foot by _merlin · · Score: 1

      "The FOSS community" is tiny. The vast majority of the time Linux is selected for practical or pragmatice reasons. Most people using Linux aren't using it for ideological reasons. There aren't enough people prepared to boycott NVIDIA over this to make a difference. Also, it's firmware that needs to be signed, not drivers.

  22. Nvidia... by Anonymous Coward · · Score: 0

    Is truly being anti-consumer these days....
    Just like how their Shield Controller, which is pretty much a 360 controller with a touchpad, require a geforce GPU to even work.. can't install the drivers if you don't have the gpu... even if you want to use game streaming, to get your game from your geforced powered PC in another room to a HTPC in the living room and use the controller, you can't. It's BS.

    1. Re:Nvidia... by tepples · · Score: 1

      I thought the SHIELD was also a standalone Android device. You can probably just use any remote desktop app that supports OpenGL.

    2. Re:Nvidia... by Luckyo · · Score: 1

      Shield is an ARM tablet that's shaped like a controller. Being able to stream a game from your GPU is just one of its functionalities.

    3. Re:Nvidia... by Anonymous Coward · · Score: 0

      He's talking about the controller that comes with the SHIELD Tablet, not the SHIELD Portable. Still, I don't see what's wrong with it requiring an Nvidia GPU, since it is an Nvidia product.

  23. that's sorta the problem by rsilvergun · · Score: 4, Insightful

    they are capable for a little while. Usually the 90 days to get out of any warranty work. Maybe a few of 'em even run at the clock freqs without crashing. It's not just clock freq either. Nvidia shuts off broken cores in software. You're games might run but they'll crash a lot. What Nvidia's worried about is that You'll blame them for a buggy card and go buy AMD. It has major brand damage potential especially with Alibaba about to become a household word what with their IPO.

    --
    Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
    1. Re:that's sorta the problem by Luckyo · · Score: 5, Informative

      You don't understand. All companies, AMD and Nvidia manufacture certain chips. Each chip has certain failure rates. When certain amount of cores fails, they are switched off in software and sold for less.

      AMD does this. Nvidia does this. Pretty much everyone making complex chips does this. It's massively uneconomical to throw away an entire chip over partial failures.

    2. Re:that's sorta the problem by Anonymous Coward · · Score: 1

      Yeah, get rid of binning. I'm sure we'll all enjoy paying 300-400% more for our hardware.

    3. Re:that's sorta the problem by Anonymous Coward · · Score: 0

      How is offering discounted cards "ripping customers of oh a scale to large to comprehend"? Sounds like the exact opposite to me.

      Put ur pantz bac ohn.

    4. Re:that's sorta the problem by Anonymous Coward · · Score: 0
    5. Re:that's sorta the problem by HiThere · · Score: 4, Informative

      You aren't understanding. Since it was explained fairly clearly, I'd guess you don't want to understand. But I'll try again anyway.

      These chips are broken. So they are sold cheap. You don't want to pay full price for seconds. Before they sell them, they use software to set the broken parts as not working. Some of them aren't broken enough that you'll immediately notice, but that doesn't mean they aren't broken.

      Usually the breaks are only in one area. Some die didn't burn properly, or traces weren't properly laid down. Whatever. So that area is sealed off. The manufacturer doesn't do a detailed investigation of exactly what's broken, just one that's good enough so they can figure out what needs to be sealed off to have a working chip. Then the sell the working chip (with reduced functionality) for a much cheaper price.

      So if you don't need the full functions of the chip, you can buy the cheaper, reduced functionality, model at a cheaper price.

      IC manufacturers have been doing this since the i8086, or maybe the i80186. (Intel was the first one I ever heard of doing it.)

      This is a deal for those who don't need the functionality of the full model. It also cuts the prices for those that do, as selling the seconds defrays some of the cost of manufacturing.

      Those who are removing the imposed limits and selling the seconds as if they were first quality are the ones who are cheating the customers. They are also impugning the name of the original manufacturer.

      --

      I think we've pushed this "anyone can grow up to be president" thing too far.
    6. Re:that's sorta the problem by davydagger · · Score: 3, Informative

      all of them, this is how binning works.

      https://en.wikipedia.org/wiki/Product_binning

      Chips are designed for the max freq of the specification. If they fail that spec, they are retested at a lower spec, and if they pass that spec, they are sold at that frequency. Why else do you find many diffrent chips in the same family run at diffrent speeds?

      Many times the chip is %100 capable of running at faster speeds, but they had too much of the higher bin, and not enough of the lower bin.

      But yes, taking a chip that didn't pass a higher speed, flashing it to the firmware of its faster/more capable cousin, and then selling it as such is ripping people off.

    7. Re:that's sorta the problem by Anonymous Coward · · Score: 0

      it's called "binning" - they make a batch of whatever type of GPU. They do a series of tests on them. Those that pass, move on, those that don't, are modified to be slower, less cores, whatever. Then those are tested. if those pass, they move on to go in another type of card, etc, etc.

      it's not ripping off...it's how it is.

    8. Re:that's sorta the problem by Uecker · · Score: 2

      On the other hand, it might equally well be price discrimination where they turn off perfectly working features just to be able to sell the same product to different people for a different price. Who knows?

    9. Re:that's sorta the problem by Grishnakh · · Score: 0

      AMD does this. Nvidia does this. Pretty much everyone making complex chips does this. It's massively uneconomical to throw away an entire chip over partial failures.

      Then they need to stop doing this. Chips should be able to run any software you throw at them without having problems.

      If there's bad cores on the chip, there's an easy solution: don't use software to turn off the cores, use hardware. When the chip is still unpackaged, it should be fairly easy to use a laser to disable the core permanently, maybe by setting some jumpers inside the chip or something. Mfgrs should make any mods to the hardware they need to while it's still in the factory; after it's out, it should be able to run any software.

      This is what they get for trying to do everything with software.

    10. Re:that's sorta the problem by TapeCutter · · Score: 2

      Perhaps they should make sure that their products work in the first place.

      That's exactly what they are doing, making sure you get the functionality you pay for. I buy my stuff from reputable dealers, in 25yrs I've had exactly one Nvidia card and one ATI card blow up, every other video problem I've ever had has been software related. Both cards were cheerfully replaced under warranty.

      AFAIK from personal experience the practice of downgrading faulty chips to a lower spec has been around since the days of maths co-processors, probably longer. And no they don't exhaustively test every chip, the grading is done via random sampling at the batch level because, like science, "statistical analysis works".

      --
      And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
    11. Re:that's sorta the problem by Man+On+Pink+Corner · · Score: 1

      What people are missing is that market segmentation is what counts, not how many chips fall into which bins. If the company sells ten times as many inexpensive GPUs as expensive ones, but the yield on the production floor is more like ten good chips for every crippled one, then it's not hard to imagine that most of the cheap cards will end up with perfect chips.

      The market detects this sales strategy as bullshit and routes around it.

    12. Re:that's sorta the problem by Luckyo · · Score: 1

      They work exactly as advertised. They advertise certain amount of cores in certain products. These cores will be functional and enabled. All of them.

      There is no ripping off.

      This is done by all manufacturers. Yes, including AMD. And Intel. And Nvidia. And all of those ARM chip makers. And pretty much everyone else.

      Example: you have a plant that makes hypothetical chip with four cores. You know that 25% of all chips that come off the assembly line have 4 working cores, 25% have 3 working and one faulty, 25% have 2:2 and 25% have 1:3.

      You market four separate products and call them i4 which is advertised as a four-core, i3 which is advertised to have three cores, i2 with two cores and i1 with a single core. You price them accordingly. Chips with 4 functional cores are named i4. Chips with 3 functional cores get software that disables the core that was deemed faulty and sold as i3. Chips with 2 functional cores have two faulty cores disabled in software and are sold as i2. And the remaining chips get 3 faulty cores disabled in software and are called i1.

      Customers get exactly what they want - a chip with just enough cores to suit their needs, at appropriate price. Manufacturer doesn't have to throw away 75% of the production because 75% of it doesn't have 4 functional cores. Everyone wins.

      The problem with this scenario is that quite a few of those faulty cores can still retain partial functionality - they just don't pass stress tests done by manufacturer. Then scammers buy i1, i2 and i3 chips and try to re-enable those faulty cores. If they pass even basic checks (and sometimes even if they don't), they relabel the chips as more powerful i4, i3 and i2 and resell them. This is indeed a scam. These chips will typically crash a lot due to faults in re-enabled cores.

      And that is exactly what nvidia is trying to prevent here. Third party scammers scamming innocent customers by flashing the card with firmware that will attempt to re-enable broken cores, so it appears to be a more expensive card than what it actually is. This has occurred recently in Germany.

    13. Re:that's sorta the problem by Luckyo · · Score: 1

      That is what is typically done. This didn't protect against such scams either however, as people did things like manually redrawing bridges on chips to disabled cores and so on.

      And in many cases, this is impractical, and it's much more practical to simply have software disable the access to those extra cores. With firmware signing, it's likely safer than burning hardware bridges.

    14. Re:that's sorta the problem by Grishnakh · · Score: 2

      > This didn't protect against such scams either however, as people did things like manually redrawing bridges on chips to disabled cores and so on.

      That's because they were doing things outside the chip packaging, such as putting SMD components (jumper resistors for instance) on the top of the package. It's not hard so solder simple SMD components with a soldering iron, though it is a little harder than reflashing some firmware.

      If they make the modifications on the chip die, before packaging, that's going to prevent almost anyone from re-enabling features that were disabled at the factory. Cutting a chip open, making modifications at the microscopic level, then putting it back together so it isn't obvious that it's been tampered with, is not an easy task, or something that a guy in his garage can do.

    15. Re:that's sorta the problem by Luckyo · · Score: 2

      Why would they need to go to such uncomfortable lengths, which would likely make end product cost more for the consumer when same can be reliably done in software on a massive scale with no need to modify physical production?

      The trend for last two decades has been to do things more efficiently in software instead of modifying hardware whenever possible due to inherent added flexibility and cost savings of the latter approach.

    16. Re:that's sorta the problem by Kjella · · Score: 1

      What people are missing is that market segmentation is what counts, not how many chips fall into which bins. If the company sells ten times as many inexpensive GPUs as expensive ones, but the yield on the production floor is more like ten good chips for every crippled one, then it's not hard to imagine that most of the cheap cards will end up with perfect chips. The market detects this sales strategy as bullshit and routes around it.

      The truth is somewhere in between. For example the nVidia GTX 980 now sells with 16/16(?) SMMs enabled for 16*128 = 2048 cores. The GTX 970 sells with 13/16 SMMs for 13*128 = 1664 cores. It is extremely unlikely that no actual cards have 14 or 15 working SMMs. Card makers probably do some more binning to see which chips they can up their OC editions and which they put in their reference editions too. The question is how good is your validation versus their validation, if it runs through 3DMark okay does it mean it's good? Or is it going to start misrendering or locking up the card or bluescreen the machine? There's a real cost to answering "Is this caused by my overclock?" even when the overclock turns out to not be the problem. If I had more time to swap for my money perhaps the answer would be different, but I run at stock speeds and I expect the manufacturer to make sure it runs flawlessly at that speed. I agree that sometimes it might be the manufacturer shaping the bins to fit the market, but what's it worth to you to take that chance? I mean, I seriously doubt a manufacturer bin down all their chips. I expect some of the GTX 970 chips to actually have just 13/16 working SMMs.

      --
      Live today, because you never know what tomorrow brings
    17. Re:that's sorta the problem by Uecker · · Score: 2

      Mod parent up.

      There also better solutions than code signing if it were only about preventing people from getting silicon with broken cores sold as fully functional: Just publish the test result with the serial number.

    18. Re:that's sorta the problem by Asgard · · Score: 1

      Then you'd have people ransacking stores looking for serial #'s that test above their price level, buy them all up and resell them after unlocking them. Instead, perhaps publish a serial #/model catalog. That works so long as the serial # on the card is relatively tamper-evident, and the manufacture has to be ok with essentially exposing their exact manufacturing numbers. Probably not especially palatable.

    19. Re:that's sorta the problem by Uecker · · Score: 1

      They could be open and sell each chip at its fair market price for its level of functionality. Chips sold at the wrong price level can only happen if they *do* price discrimination. On the other hand, if product binning were the only reason as claimed by some posters this could not happen - there would be no chips sold at the wrong price.

    20. Re:that's sorta the problem by gman003 · · Score: 1

      Others have tried arguing with you, but I'll make this very plain and answer your questions bluntly, since you seem to lack some of the knowledge they assume you have.

      It's not a ripoff - they are selling consumers exactly the performance they are promising.

      Manufacturing isn't perfect. They test at the factory to see if each chip can run at the promised core count and clockspeeds. Fairly often, particularly with top-of-the-line chips, a few cores will be broken, or unstable at the specified clock speeds. They are unfit for sale as the originally-designed product. Rather than throw it away, they disable whatever is broken (either in firmware, or by blowing fuses built for this purpose on the processor) and sell it as a lower-cost, lower-capability product. This is standard procedure for everyone. Just off currently-sold chips:

      For Nvidia:
      The 970 is a 980 with 3 SMMs disabled (out of 16)
      The Titan is a Titan Black/780 Ti with one SMX disabled (out of 15).
      The 780 is a Titan Black/780 Ti with 3 SMX disabled (out of 15)
      The 760 Ti is a 770 with one SMX disabled (out of 8)
      The 760 is a 770 with two SMX disabled (out of 8)
      The 745 and 750 are 750 Tis with one SMM disabled (out of 5)

      For AMD (GPUs):
      The 290 is a 290X with four CUs disabled (out of 44)
      The 280 is a 280X with four CUs disabled (out of 32)
      The 265 is a 270/270X with four CUs disabled (out of 20)
      The 260 is a 260X with two CUs disabled (out of 14)
      The 240 is a 250 with one CU disabled (out of 6)

      Note: I wanted to include AMD CPUs as well, but I can't find perfect info on their CPUs. They are clearly using binning like everyone else (probably more, if their Phenom II days are anything to go by), but I can't tell you exactly which ones are stripped-down versions of which.

      For Intel (CPUs):
      The 5920K is a 5830K with 12 PCIe lanes disabled (out of 40). Both of those *might* be 5960Xs with two cores (of eight) disabled.
      Every current desktop i5 is an i7 with hyper-threading disabled. Likewise, any current desktop Celeron or Pentium is an i3 with hyper-threading disabled. For the most part though, Intel only bins based on clock, not cores - if it's a low-clock version of a given chip, it likely tested unable to run at higher speeds with stock voltages and cooling.

      Oh, and every single PS3 processor had one SPU disabled out of 8. Processors with all 8 functional were used in certain IBM servers, amongst other things.

      If you want a car analogy, imagine you were sold a car with a 4-cylinder engine. You check later, and find a 6-cylinder engine block, but two don't have piston heads in them and don't run. When you get some spares and try to run them, you find the two cylinders have completely busted sealing, and they contribute no power, only noise and pollution and a nasty rumble.

      When run exactly as you were promised it would run, it works perfectly. The extra cylinders affect nothing, because this is a metaphor and the actual physics of a car don't apply. They used a part to a higher-end car that would not work in said higher-end car, but they neither told you that it would, nor charged you as if it did. They actually probably charged you slightly less than if they had built it as a four-cylinder engine to begin with.

      For this story, imagine some unscrupulous car dealer (also known as just "a car dealer") took that car, put the pistons back in, and sold it to you as the higher-end car without letting you test-drive it to find out that it doesn't actually work, only letting you pop the hood to see that it has all six cylinders. The car manufacturer then changes their procedures so that instead of simply removing the pistons, they actually fill the broken cylinders with steel to prevent it from even pretending to work.

    21. Re:that's sorta the problem by tlhIngan · · Score: 2

      Chips are designed for the max freq of the specification. If they fail that spec, they are retested at a lower spec, and if they pass that spec, they are sold at that frequency. Why else do you find many diffrent chips in the same family run at diffrent speeds?

      Many times the chip is %100 capable of running at faster speeds, but they had too much of the higher bin, and not enough of the lower bin.

      But yes, taking a chip that didn't pass a higher speed, flashing it to the firmware of its faster/more capable cousin, and then selling it as such is ripping people off.

      Except in the world of GPUs, there are enough "crazy people" out there who want the best of the best. So much so that the top bin is almost always empty - so you'll never have top-end chips binned as lower spec ones.

      At best, you'll find possibly the low end chips that could be mid-range chips, but given the low end generally isn't too popular when mid-range chips are the most common and most desired.

      Shortages of the top-end cards isn't unheard of - either people who are still trying to make a go at it for bitcoins, or gamers. (And given the price of the high end, they could come down a bit before binning takes place - they're still big profit centers).

    22. Re:that's sorta the problem by sexconker · · Score: 0

      In what fucking way are they not functioning properly as-released?
      Are you claiming that nVidia is sending out cards that just don't fucking work?
      They pulled some dirty shit with bumpgate, but what you're alleging is even worse.

    23. Re:that's sorta the problem by Anonymous Coward · · Score: 0

      I don't know - it was kind of called for. You were being a bit of an idiot.

    24. Re:that's sorta the problem by Anonymous Coward · · Score: 0

      We know. There's no guessing here. This is stuff you can test right now with software. You don't even have to know what your doing, the software to view and fiddle with that kind of stuff has been made so easy that even a caveman can shoot himself in the foot with it.

    25. Re:that's sorta the problem by Cruciform · · Score: 1

      Maybe if it was explained with a less technical product they'd get it.

      It's like when you go into Bulk Barn and there are broken chocolate bar bits for less than the cost of a wrapped, branded chocolate bar.
      They are still yummy, but not able to be sold for full price. So they sell them at a reduced price to get some return on them.

      Now I'm craving chocolate.

    26. Re:that's sorta the problem by davydagger · · Score: 1

      What people are missing is that unscrupolous dealers are selling cards that aren't not tested at full specs as cards that *are*.

      You have a pretty big "if" statement. We have no clue on nVidia's yield rate is, and especially not compared to their sales model. What we do know is that at least *some* of the chips will not funciton as advertised.

      which gets us back to why nVidia is signing firmware. They don't want to be associated with products of unknown quality, sold by people who bypass their Quality Control proccess, designed to make sure *every* product that leaves nVidia's line *performs exactly as specified, first time everytime*. nVidia is more concerned with overall quality than performance. Something that will build their brand reputation.

      I don't like what nVidia is doing, but I do understand why, and I'm not going to resort to conspiracy. nVidia recently has been good working with the nouvea open source team, and I hope the experiance continues and we can work something out where we can help fight against counterfeit cards, save nVidias reputation

      AND AT THE SAME TIME

      provide a commitment to high quality Free software drivers to provide a quality error free experiance for people running Free software, and make gnu/linux installs *just work* out of the box.

    27. Re:that's sorta the problem by Shirley+Marquez · · Score: 1

      Another cause for selling chips as less costly models is thermal. A given chip might have all its cores working, but if you run them all at full speed the chip consumes more than its rated power. So you turn off some of the cores and/or lower the clock speed and sell it as a lesser model.

      Nothing new under the sun here; as you point out, CPUs and RAM are also graded that way. The top performers get sold as expensive models, the ones that can't reach the top speed or consume too much power at that speed are downgraded. Sometimes one or more CPU cores are disabled, like the three core processors that AMD sold at one time; they actually had four cores on the die. Some 486SX chips from Intel were actually 486DX chips with floating point units that didn't work. (They also made 486SX chips with no FPU on the die; if you bought one you had no way of knowing which type you would get.) The difference with GPUs is that their large core count makes a much wider range of degrading possible; you can turn off 5%, 10%, 25%, or even 50% of the cores.

      Late in the life of a product line, companies may sell fully functional parts as lesser models. If production has improved to the point where nearly all your chips work completely, you may have to cut the price on some of them because there are customers who aren't willing to pay the higher price.

      In some cases, the modded boards will work properly. The customer might be lucky enough to get one of the boards with a fully functional chip, or may plan to use the board with an aftermarket cooling system. But there are no guarantees. Any reseller who puts up modded boards for sale is committing fraud. But individual customers should have the right to experiment with equipment that they own, and NVidia's action is also causing collateral damage to the open source movement. I don't think NVidia is actively hostile to Nouveau, they are just indifferent about it.

    28. Re:that's sorta the problem by lsatenstein · · Score: 1

      they are capable for a little while. Usually the 90 days to get out of any warranty work. Maybe a few of 'em even run at the clock freqs without crashing. It's not just clock freq either. Nvidia shuts off broken cores in software. You're games might run but they'll crash a lot. What Nvidia's worried about is that You'll blame them for a buggy card and go buy AMD. It has major brand damage potential especially with Alibaba about to become a household word what with their IPO.

      Why would I not buy AMD anyway?

      --
      Leslie Satenstein Montreal Quebec Canada
    29. Re:that's sorta the problem by exomondo · · Score: 1

      On the other hand, it might equally well be price discrimination where they turn off perfectly working features just to be able to sell the same product to different people for a different price. Who knows?

      Does it matter? The product is different, whether the inaccessible bits are working or not is irrelevant. It isn't price discrimination because the product is different.

    30. Re:that's sorta the problem by Anonymous Coward · · Score: 0

      Maybe if it was explained with a less technical product they'd get it.

      It's like when you go into Bulk Barn and there are broken chocolate bar bits for less than the cost of a wrapped, branded chocolate bar. They are still yummy, but not able to be sold for full price. So they sell them at a reduced price to get some return on them.

      Now I'm craving chocolate.

      I still don't get it. Please come up with a car analogy.

    31. Re:that's sorta the problem by Uecker · · Score: 1

      On the other hand, it might equally well be price discrimination

      So what you're saying is, if you sold a computer on ebay and said "This computers CPU fan is bad - you MUST replace the CPU fan before use!!!" you are admitting to criminal fraud simply because I should somehow expect a working CPU fan?!

      How on earth does that even make sense?

      How does you reply even make sense? This has nothing to do with what I wrote.

    32. Re:that's sorta the problem by Uecker · · Score: 1

      On the other hand, it might equally well be price discrimination where they turn off perfectly working features just to be able to sell the same product to different people for a different price. Who knows?

      Does it matter? The product is different, whether the inaccessible bits are working or not is irrelevant. It isn't price discrimination because the product is different.

      The real questions are: Should the consumer (or some reselller?) be allowed to turn the working features back on? Is it OK if NVIDIA tries to prevent this with code signing? If the features are actually broken, than these questions are of less relevance.

    33. Re:that's sorta the problem by exomondo · · Score: 1

      The real questions are: Should the consumer (or some reselller?) be allowed to turn the working features back on? Is it OK if NVIDIA tries to prevent this with code signing?

      Well I would say yes to both questions. nVidia can "turn them off" however they like and the user can try to turn them on however they like.

  24. Doesn't look unreasoanble (so far) by marcansoft · · Score: 3

    So, they're locking out things that can brick the card (flash ROM/fuses, screw up thermal sensors) and apparently a hint of OS security (the Falcons that respond to userspace commands can no longer access physical memory, only virtual memory). The latter sounds somewhat bizarre, considering the firmware should be fully under the control of the driver, not userspace (I guess/hope?), but not unreasonable. Maybe there are software security reasons for this.

    Nouveau is free to continue using its own free blobs or to switch to nvidia's. If they start adding restrictions that actively cripple useful features or are DRM nonsense, then I would start complaining, but so far it sounds like an attempt at protecting the hardware while maintaining manufacturing flexibility for nvidia. This isn't much different from devices which are fused at the factory with thermal parameters and with some units disabled; the only difference is that here firmware is involved.

    NV seem to be turning friendlier towards nouveau, so I'd give them the benefit of the doubt. If they wanted to be evil, they would've just required signed firmware for the card to function at all. The fact that they're bothering to have non-secure modes and are only locking out very specific features suggests they're actively trying to play nicely with open source software.

  25. Gluglug by tepples · · Score: 1

    Probably whatever GPU is in a Respects Your Freedom certified laptop such as the Gluglug X60.

  26. Not a big deal... by JumboMessiah · · Score: 4, Informative

    Andy Ritger at Nvidia is already in talks with Ben Skeggs and Martin Peres with Nouveau. They're are going to hash out the details at XDC2014. The impact for Nouveau is in the packaging and distribution parts of the cycle, not development. Also, it was Nvidia who reached out to Nouveau, not the other way around. Nvidia has their reasons for doing this, but it's not an anti FOSS thing. It's more likely one of the more sane reasons posted above.

    So everyone just relax their sphincters a bit....

    1. Re:Not a big deal... by Anonymous Coward · · Score: 1

      The thing is _THEY HATE_ the open source drivers, since the existence of the open source drivers mostly kills the bussines around the drivers. You know, there's a reason why exist programs like "The way is meant to be played", the "GeForce Experience" or recently the "Gameworks" library - program. It's all about bussines you know, and bussines means money. Nvidis's employees are quite good monetizing the graphics drivers.

      Enter the open source graphics drivers, and POOF!!!, all the revenue they get from graphics drivers is GONE. Yes, Nouveau development nowadays isn't a serious threat nowadays, but in the future could be, so the best way to kill the threat is to interfere at the lowest level possible: Firmware.

      Considering this, is easy to see why they are doing this: They need to protect the investment they are doing in their Graphics drivers. They got developers that have to pay bills, mortages, colleges, food, bonuses, and whatever they need to.

      Is this good or bad??, for Joe User doesn't mean shit, since they only care about something can be visible onscreen. For Joe "Games" Developer, this means the "Graphics Mafia" got more power to screw around their games. For Joe "FREEDOOMMMM!!!" Wallace, this is indeed bad news, but the Wallace Family is getting smaller every day, since people like to live with their chains and jailed ("Malodorus" Stallman viewpoint, not mine) as long as things just works.

      Still, we have Intel and AMD, but for now, none of them can even compare the quality of Nvidia's offering. AMD is getting there though, at slow, but quite safe peace; and Intel just need to make more speedy hardware, and capable drivers.

    2. Re:Not a big deal... by Anonymous Coward · · Score: 1

      I don't have to consider anything, Nvidia are a hardware company. That's where they make their revenue. If you believe for one second that they make a net profit on their driver development, then you either need to get out of the basement or bid on buying this bridge to Russia that I'm selling. The drivers are there to enable the hardware to be sold. It's the selling of the hardware that pays the bills and all. If Nouveau equals parity with the proprietary drivers without Nvidia having to disclose it's secret sauce, then so be it. They now get a boost in the FOSS market without expending labor on it. Win win.

      And, to be specific, the signing doesn't have jack shit to do with drivers. It's the firmware images blotted to the card at initialization. Nouveau or any FOSS camp don't write the firmware, nor would the particularly want to for a GPU. The signing will allow Nouveau to distribute the firmware directly without having to first rip it from the Windows driver. So we will actually benefit from the expanded firmware distribution.

    3. Re:Not a big deal... by Anonymous Coward · · Score: 1

      Actually, the nouveau devs DO write their own firmware for the cards. The amount of firmware code that runs on modern NVIDIA cards is quite significant, and a large portion of it was reversed and rewritten. That is part of why people are so uppity, free firmware was an option before Maxwell, but now it might not be.

    4. Re:Not a big deal... by Anonymous Coward · · Score: 0

      Is this a troll or are you just stupid?

      They do not sell their drivers. They sell hardware. They also spend considerable amount of money on their GameWorks etc programs *to sell more hardware*. Main differentiating reason right now to buy NVIDIA is "has working drivers that work with games on day 1". If a bunch of OSS hippies want to put effort into a parallel driver development effort to have a "free" driver, all the more power to them - if anything, such effort may uncover things that help the development of the closed source driver as well and it may help sell hardware to OSS hippies that can't stand having non-free drivers on their system.

  27. It totally depends what you want by dutchwhizzman · · Score: 2

    Intel has I believe all their Linux drivers fully open sourced. However, they're not really fast compared to AMD or NVidia. AMD has two driver versions, their closed source catalyst driver and the open source one. The catalyst driver is much faster, energy efficient and can do more tricks than the open source one. NVidia is sort-of supporting Nouveau and has their own binary driver as well. The "sort of supporting" is much limited compared to the amount of AMD is pouring in the open source version of their drivers, but it has improved greatly recently.

    Depending on what you are looking for in terms of bang for buck, speed or features each of these might be "the best solution" for your needs. If you want CUDA or openCL, you'll be looking at closed source though, there's no serious support for open source drivers for relevant hardware (yet).

    --
    I was promised a flying car. Where is my flying car?
  28. Read Only Memory by Mr.+Slippery · · Score: 2

    Once upon a time, there was this stuff called "Read Only Memory". Not EPROM or EEPROM, but ROM. Once it was created you couldn't change the contents of it.

    If I was worried that scammers were going to take a board that I was selling as a Whizzo rather than a Whizzo Plus because it didn't meet Whizzo Plus specs, and flash it as a Whizzo Plus anyway to rip off customers, I'd put "Hi there I'm Whizzo serial number 987654321 born 2014-09-24-18:58:56 GMT at the Utopia Planitia assembly line, signed <digital signature>" somewhere in a bit of that old-fashioned Read Only Memory soldered to the board in a tamper-resistant manner, and also have that serial number etched into the board.

    --
    Tom Swiss | the infamous tms | my blog
    You cannot wash away blood with blood
    1. Re:Read Only Memory by Anonymous Coward · · Score: 0

      And what happens when a firmware bug crops up? They recall all of the cards or mail out new ROM chips for customers to swap out instead of issuing a simple firmware update?

    2. Re:Read Only Memory by jonwil · · Score: 1

      They only need a very small amount of actual unchangeable memory. Do it like Microsoft did on the XBOX 360 and have fusible links on-die on the GPU, when the card is manufactured, the fusible links are blown to store the ID of which GPU it is in a way that cant later be altered.

  29. Fuck That Shit! by Anonymous Coward · · Score: 0

    > They've just made AMD the only choice for graphics cards.

    This behavior isn't exactly new. Nvidia has been getting more closed and as a result more consumer unfriendly for some time now.

  30. Time for a Covenant with their Consumers... by Anonymous Coward · · Score: 1

    If Nvidia is really only doing this to stop 'counterfeit' chinese cards, they should provide a public covenant that they will release signing keys necessary for consumers to reflash their own hardware once the market for counterfeit cards has sufficiently dwindled (say two card generations, or two years, whichever is longer.) This has all the benefits of eliminating counterfeits where it counts, while also ensuring to the consumer that they will have full control over their hardware once enough time has passed to eliminate the profitability for counterfeiters and the related support issues for Nvidia.

    Thoughts and/or email campaigns to get this into place?

  31. Pro level cards? by Anonymous Coward · · Score: 0

    In the past it has been possible to flash your card between variations of the same silicon with some luck. Obviously, you can't invent shaders that aren't there, but some of the "Ti" or "Z" or whatever models were often simply validated at the BIOS.

    Most importantly, I suspect that nVidia wants to protect the obscene margins of the pro-level cards. Supposedly, these are locked at the driver level with much lower double floating point data rates for the consumer models (and other quirks, like high-bit color, specific optimizations etc). I don't know if it's possible to flash a consumer card into the equivalent Quadro, but that would make a huge difference in $$$$.

    Anyway, whatever the motivation, any sort of locking that restricts user freedom bothers me.

  32. Actually... by Anonymous Coward · · Score: 0

    While there are legal requirements to DOCUMENT that the odometer has been changed, you are fully within your rights to replace the odometer at any time during which you have the car. The only restrictions are: If the odometer is removed or the value cannot be verified to match the original mileage of the car, the car will recieve a mark on it's pink slip stating the odometer has been tampered with.

    Other than that, you can. It might bite you in the ass in other legal regards (taxes and some things that base their value on odometer readings), but at least in parts of the US it's not illegal or outside of your rights/abilities to do so.

  33. easier way by Anonymous Coward · · Score: 0

    If Nvidia were only concerned with fakes they would only check the signature at boot time and throw a "genuine" or "unauthorized" message on the screen. That would solve 100% of the stated problem with zero problems for Linux or anyone else. The Linux devels do not need to reflash the firmware in the card, only load code into the ram.

    Since I'm not willing to believe Nvidia is so stupid that they don't also realize this, I'm left with evil as their intent.

  34. Intel also requires signatures in firmware by Anonymous Coward · · Score: 0

    From what I've heard, Intel GPUs are not available outside intel CPUs. And current Intel CPUs can't even access RAM until some propietary and signed code is run. That happens in coreboot, UEFI, BIOS or the like, and it can potentially be used to set the stage for the next computing to be under control of the key holder (that'd be Intel or their government or their attackers or some such). I think the signed code is called MEI or something else.
    Intel publishes linux GPU drivers and that makes it easy for linux users who don't mess with BIOS code. But that's just because SMM or whatever they need may already have been take care by the signed firmware.

    AMD had propietary bios for the GPU run by an interpreter, but not such propietary signed code in the CPU boot. But recently they start including an "PSP" smaller CPU which will check signatures on boot firmware before
    starting the main CPU and leting it run this boot code. They used to provide good documentation (I think they still do?).

    ARM varies, but they have this TrustedZone posiibility to allow DRM, remote attestation and so on. Some vendors seem to violate GPL, others (Freescale) give good documetnation, most everyone buys components
    somewhere and ships functionality that only works with propietary drivers, and free software reverse engineers achieve what they can.

    I thought Nvidia didn't have this, just propietary drivers and unhelpful to free software (less lately, even people claiming full 100% free software running on their tegra K1 board?).
    But now I heard this.

    Even if you find a CPU that does not require propietary, signed software, good luck finding a complete system that doesn't ahve a GPU, wifi chip, embedded controller or something
    else run propietary blobs and having full access to RAM and stuff.

    Of course you have to trust the builder of the hardware that runs your code, but I cherish those days when you didn't have to trust it to always keep their keys secure, their governments
    benign, their future boards honest... Those days when hardware was hardware and didn't refuse to run your software.

    It used to be difficult to try to save money when you saw the adverts, but lately they're find the cure to consumism.

    I should include some links to explain better all that, but it's late, and I'm lazy, feel free to look up the keywords. It is so difficult to find hardware designed to be trustable, and
    the crowdfunding for the dozens of units people buy is so expensive, and features so lacking compared with the adverts for untrustable hardware that not buying anything
    is just cooler nowadays.

    Not that vendors notice me. The crowds keep buying their shackles like they were candy (phones, computers, software, media, whatever it is, even cars are now going to be remote
    controlled by the government and hopefully only the government).

  35. Hopefully this doesn't affect the VBIOS by Anonymous Coward · · Score: 0

    That's the only way to get resolutions not from the mid-'90s in VESA modes.

    And VESA's the only way to get rock solid (albeit very slow) graphics on a server without the nVidia binary blob periodically causing triple page faults in your kernel.

  36. Such things should be forbidden by Casandro · · Score: 1

    After all its artificially limiting what you can do with the hardware. Plus it'll mean you'll have to run closed source firmware from the manufacturer on the device, which means that it'll probably contain malware. Why else would you distribute software in object code only? (No, competitors probably have reverse engineered it years ago.)

  37. Intel also requires signatures in firmware by Anonymous Coward · · Score: 0

    Of course you have to trust the builder of the hardware that runs your code, but I cherish those days when you didn't have to trust it to always keep their keys secure, their governments
    benign, their future boards honest... Those days when hardware was hardware and didn't refuse to run your software.

    It used to be difficult to try to save money when you saw the adverts, but lately they're find the cure to consumism.

    I meant the boards of directors, the executive staff, people who decide what to put in the firmware updates and hold control over the keys to pick what to sign with them,
    not the circuit boards they sell.

  38. My sphincter has this for NVIDIA: "yes,A big deal" by Anonymous Coward · · Score: 0

    It's still a big deal. Having to pass trhough a company means not everyone can analyze and experiment, contribution is harder, less freedom so less usefulness, and of course MS can buy NVidia and decide no more
    signatures for nouveau (or anything else can happen). There's a big difference between not disclosing information on what you sell and activing selling things built to stop working when someone fidns out how it works and tries something different.

    Giving correct information about the board is easy enough by putting some values on some ROM easy to access from the drivers. If sellers or users choose to disregard this information is easy enough to point them to
    the on-board info and tell them it's not designed to be sold or used that way. If they fear that someone might alter the board to include different information in those ROMs then they can use whatevver system they
    use to protected the public root key the board will use to verify the signature.

    Including signature verification in the board is too complex and unnecessary for the stated purpose. It is useful to implement DRM, remote controlled devices, content that fails to run if your system is not
    running what a third party wants, etc. Or just convenient for selling that control over users to investors, governments, or in the black market. But it is not necessary for warning users of unadvised firmware, drivers
    or operation modes.

    There's a reason they choose to stop working instead of turning off a "system ok" LED or whatever whenever a signature verification fails. They want control.

  39. Blacklist Nvidia by Anonymous Coward · · Score: 0

    I blacklisted nvidia when buying the last computer and will continue that.