GeForce 600M series brings new names, old silicon

Apparently, AMD doesn’t have a monopoly on rebadging old parts. We’ve noticed that Nvidia has quietly added a trio of GeForce 600M-series GPUs to its notebook lineup: the GeForce 610M, GeForce GT 630M, and GeForce GT 635M. These are the first members of the 600M-series product line, and like the AMD Radeon HD 7000M-series products that came out earlier today, they appear to be lower-end offerings based on old silicon. Their specifications are as follows:

  GeForce 610M GeForce GT 630M GeForce GT 635M
Stream processors 48 96 144 or 96
GPU clock speed 900 MHz 672 MHz 675 or 753 MHz
Memory type DDR3 DDR3 DDR3/GDDR5
Memory bus width 64 bit 128 bit 192 bit
Max. mem. bandwidth 14.4 GB/s 28.8 GB/s 43.2 GB/s
Nvidia 3D Vision No Yes Yes

Look familiar? AnandTech points out that these 600M parts are straight-up rebadged versions of the GeForce 520MX, GeForce GT 540M, and GeForce GT 550M. Everything, down to the GPU silicon, shader configurations, and clock speeds has reportedly remained identical. By the looks of it, this launch is little more than marketing sleight of hand by Nvidia.

Just like AMD, though, Nvidia has left room in its model numbering scheme for higher-end offerings based on next-generation silicon. Nvidia CEO Jen-Hsun Huang actually talked up those future products in a recent financial conference call, saying his company had scored "more notebook design wins for the Ivy Bridge cyle than we ever had in notebooks." He attributed that success to Nvidia’s upcoming Kepler GPUs, which were purportedly "designed for intense energy efficiency." Clearly, there’s something to look forward to in the GeForce 600M lineup—it just ain’t here yet.

Comments closed
    • ronch
    • 8 years ago

    Monkey see, monkey do.

    • HisDivineOrder
    • 8 years ago

    And who said that having a virtual monopoly for two decades while building up your own fabrication facilities to be second to none, then paying your two biggest competitors (though both small fish next to you) around a billion each to shut up to the government, was a bad thing?

    Intel is moving to 22nm while their competitors can’t even get 28nm out. Intel has their “3d” fabrication while their competitors WISH they had it. Intel has such an insane head start because of their anti-competitive shenanigans from years ago that not even a complete crapstorm like Pentium IV/5 could keep them down for long. And now that they’re actually moving along with what people actually want in their computers?

    Forget about it. Now look at what Intel did in CPU’s… and apply it to GPU’s. They’re intent on absolutely killing the discrete market for all but the most diehard amongst us. Tick, tock, they don’t care when it comes to GPU’s as they actually recognize they’re behind and they’re working to catch up and in a hurry, too. If they double their GPU performance for a couple of years and begin to say the same as AMD on making CPU’s and GPU’s become more cooperative, sharing memory and resources in ways that are easier and easier for developers to take advantage of…

    There may come a day when developers prefer to work with integrated GPU functions on a CPU over working with a discrete GPU entirely separate from the CPU (aka nVidia). Then again, AMD may be about to implode and abandon the traditional CPU performance market. In which case, Intel’s virtual monopoly becomes a real monopoly… especially if the rumors about MS hobbling ARM’s version of Windows 8 with no desktop option are real.

    Just imagine how much money Intel had to pay MS to make that happen. Easily enough to make nVidia tear its metaphoric hair out in rage as Project Denver becomes DBA (Dead Before Arrival). Add it all together (AMD imploding, Bulldozer’s a dungheap, ARM+AMD+nV having to use inferior fab tech that still isn’t even out when Intel’s already moved on to the next stage, Windows 8 hobbling ARM CPU’s arbitrarily, Intel having settled its problems with the government to remove any threat to them) and, if you’re Intel, you’re all smiles.

    Paying their two competitors around a billion each while giving away trivial concessions (and more importantly NOT enacting less trivial concessions) was the smartest thing Intel ever did. I imagine nVidia didn’t anticipate Intel’s sideswiping them through MS after they agreed to the patent licensing deal. Intel gets what they need (video patents) while “encouraging” (aka paying off) MS to hobble Windows 8 for ARM after going through all the trouble of porting it. Project Denver becomes a huge waste of money and time for nVidia, possibly negating the money Intel paid them altogether.

    Intel is many things, but stupid is not one of them. And their fabrication advantage may well be the single most important advantage that will wind up winning them so many aspects of the future. While AMD and nVidia are busy rebadging their low end GPU’s, Intel will keep showing up with better and better integrated GPU’s, slowly but surely stealing their bottom line while increasing their own. Incredible sleight of hand.

    • ShadowEyez
    • 8 years ago

    Though I’m not an industry insider, they may have capacity and yield issues with 28nm parts, so only the high end and high margin stuff gets the 28nm parts. In other words not the base laptop gpus.

    • Rageypoo
    • 8 years ago

    Do you guys really think this is the top of the line? That they are truly having difficult times making this stuff? there’s way better tech out there that we’re not using, this is all marketing.

      • yogibbear
      • 8 years ago

      They are having difficulty at a MASS scale. Sure you can farm at 10% yields for 1nm parts to shove some super secret chip up your penis, but we’re not all 007 and you can’t sell 5 million cards at a 10% yield.

    • Glix
    • 8 years ago

    Sigh I’ve stopped tracking mobile gpu’s now, there isn’t anything exciting at all happening to them.

    *Feeds trolls* Intel comes out on top?

    • Palek
    • 8 years ago

    Between this and the AMD announcement, it’s pretty clear that TSMC is facing serious capacity constraints at 28nm due to poor yields. The question is: who IS using whatever capacity is available at 28nm, then? nVidia and AMD may be top semiconductor companies, but neither of them are TSMC’s top customer. Whatever business they give to TSMC is [url=http://www.isuppli.com/Semiconductor-Value-Chain/News/Pages/Intel-Reasserts-Semiconductor-Market-Leadership-in-2011.aspx<]dwarfed by Qualcomm[/url<]. Qualcomm [url=http://www.qualcomm.com/news/releases/2011/11/16/qualcomm-unveils-new-snapdragon-mobile-processors-across-all-tiers-smartpho<]announced their next Snapdragon S4 mobile SoC just recently[/url<], and according to the news release S4-based products will appear early next year, meaning that the S4 is well and truly in mass production right now. S4 is manufactured on [url=http://www.qualcomm.com/documents/snapdragon-s4-architecture<]TSMC's 28nm process[/url<]. Qualcomm are almost certainly getting top priority at TSMC, possibly to the point of draining all other clients of any meaningful capacity.

    • ModernPrimitive
    • 8 years ago

    Double whammy today.. oh man ! No surprise from either camp though. I have to say I was expecting a slight possibility of 28nm parts from AMD though, since they have shown a demo a couple of times…. ah well.

    • Hattig
    • 8 years ago

    At least AMD added UVD3 to those low end 7000 series parts that had been missing it previously.

      • BestJinjo
      • 8 years ago

      UVD3 brings almost nothing worthwhile to the table over UVD3. AMD and NV should be disgraced for rebadging. It’s even worse in the case of NV since GTX400 –> GTX500 and now GTX500 –> GTX600 is the same card!

        • willmore
        • 8 years ago

        Uhh, wanna edit that? UVD3 seems to be the same as UVD3. Maybe a ‘2’ in there somewhere?

    • jpostel
    • 8 years ago

    I would venture it has more net gain for laptop makers than nvidia, because I fail to see how this particular kind of marketing nonsense works for nvidia. What is the net gain for nvidia on this? There is no competitive advantage over ATI on this because the numbering schemes are different.

    If I am shopping and see two laptops and compare the specs, I see one has 540M graphics and the other has a 630M. The marketing trick works and I buy the one with the higher number. What does nvidia get out of this? Either way, I am buying a laptop with nvidia graphics, and nvidia’s price to the laptop maker may well have dropped on a per-GPU basis, since it is older, so whats the difference?

    The nvidia chips are already sold to the laptop maker in the first place, so are they actually trying to pull the wool over the manufacturers eyes? I doubt anyone would fall for it, so it seems far fetched. I tend to think this is something the laptop makers ask for so they can faux-differentiate between models where the only other differentiator would be the processor speed. Then the scenario becomes easier to understand, because now I am looking at two laptops where the processors, RAM, etc. are the same, but one has a 630M and the other has the 540M, so I must buy the one with “better” graphics even if it costs more.

      • flip-mode
      • 8 years ago

      [quote<]I fail to see how this particular kind of marketing nonsense works for nvidia. What is the net gain for nvidia on this?[/quote<] I used to gnash my teeth at renaming but there is a plausible benefit to it. Both GPU makers have generational product lines. If they're going to keep a particular chip around for another whole generation then there is a degree of usefulness in repositioning the product in the current generation and, in effect, let consumers make sense of all the products available in relation to each other and even imply a sense of ongoing product support. AMD did exactly this with the 5770 > 6770. In other words, the product is not going away and it has a defined place in the current product generation hierarchy. It's not worth getting worked up about - people who care enough will know what is what and people who don't do the research probably don't really care one way or another.

    • tviceman
    • 8 years ago

    If this is an indication to TSMC’s 28nm process, then Nvidia made a brilliant move by making Tegra3 on 40nm. Even if it isn’t the be-all end of mobile CPU’s, it’s months ahead of any other quad-core mobile CPU which in itself is a huge selling feature to vendors and average consumers.

    • flip-mode
    • 8 years ago

    [url=https://techreport.com/discussions.x/22120?post=600554<]To quote one of our resident Nvidia trolls:[/url<] [quote<]LOL That is all![/quote<]

      • Silus
      • 8 years ago

      Awww looks like the AMD fanboy troll can’t handle that his favorite company did the same!

      • cegras
      • 8 years ago

      To him, this thread doesn’t exist.

        • Silus
        • 8 years ago

        Oh it does! And it suffers from the same problem. Unlike AMD fanboys, I criticize what needs to be criticized and this is no different.

        Not that I expect you do understand any of that!

          • jpostel
          • 8 years ago

          Disclaimer: I am not a graphics fanboy, mostly because I am not a big gamer.

          Does this need to be criticized? I see your retorts to others jabs, but I don’t see your criticism of nvidia.

            • Silus
            • 8 years ago

            You see TR is populated with a majority of AMD fanboys that also don’t seem to like NVIDIA for some odd reason. So if NVIDIA does something wrong, threads about it are flooded with comments about how bad they are and how no other company does what they do. When AMD does the same thing or worse, there’s just a couple of people actually criticizing them for it, while the majority say “meh, this isn’t so bad” or “NVIDIA does it too”.

            So yes, it needs to be criticized when either camp does it, at least to keep the dreadful double standards that plague the AMD fans so much.

            The LOL I was quoted from my number one fan (or so it seems), was directed at the AMD fans that made NVIDIA rebrands look like the end of the world, while AMD rebrands are just “business as usual”. Again this deserves to be criticized, since we, as consumers are not getting anything new. But also again, it’s not the end of the world and it is indeed business as usual. As technology advances there are problems and the solution isn’t always clear, especially when these companies depend on a third party to fab their chips.

            Just to finish this already long reply, I find it entertaining to say the least, that you only ask me that question and not the other two, that aren’t criticizing AMD in the other thread. In case you didn’t notice it or simply didn’t care, I was the one being quoted here from another thread.

            • flip-mode
            • 8 years ago

            That is a lot of talk from someone with his foot caught so deep in his mouth. But, I understand, you’re trying to save face – embarrassed yet still trying to play the victim and the fanboy-blame game.

            • Palek
            • 8 years ago

            There are plenty of snarky comments in the AMD news thread. I think your paranoia is getting a bit out of hand.

      • JustAnEngineer
      • 8 years ago

      Meh. Krogoth is not impressed?

      • willmore
      • 8 years ago

      To followup, How’s the crow? Nice and fresh?

    • derFunkenstein
    • 8 years ago

    If both vendors are doing it, it kinda makes me wonder how TSMC’s 28nm process is doing. If yields are bad, we could be in for a shortage again, and 40nm rebrands might be all we’ll be getting.

      • Bauxite
      • 8 years ago

      28nm is obviously fucked up, long delay, bad economy is a nice disincentive and not a lot of talk from those actually working on it.

      • flip-mode
      • 8 years ago

      Yep. This is not a good sign. Microprocessor manufacturing at nanometer scales certainly cannot be an easy thing, but TSMC just as certainly needs get more realistic with its schedules. I’m sure than Nvidia and AMD aren’t happy about this.

      Intel, meanwhile, looks on, bemused.

        • willmore
        • 8 years ago

        Intel has had its share of issues or are we all agreeing to forget that B2 stepping SB chipset issue?

          • Geistbar
          • 8 years ago

          That was a QA issue, and not one with their process nodes. As far as I know, their 32nm switch happened without any major hitches, and 22nm is set to do the same thing despite including a very new technology in it (tri-gate transistors).

          Intel is “looking on bemused” because their major rivals for revenue at the moment tend to rely on TSMC or GloFo- both of whom have had major issues with their process and yield when shrinking past 45nm/40 nm.

    • indeego
    • 8 years ago

    So this is how they will fight the tablet onslaught: new numbers!

    Good luck!

    • dpaus
    • 8 years ago

    Looks like AMD and Nvidia are having a pre-Christmas re-gifting promotion.

    • yogibbear
    • 8 years ago

    I used to be a Nvidia GeForce 520MX like you, then I took an arrow to the knee

      • glacius555
      • 8 years ago

      I used to post Skyrim memes at every occasion, but then I took an arrow in the knee..

      • Arclight
      • 8 years ago

      If you’re having adventure problems i feel bad for you son, i dodged 99 arrows but my knee took one.

Pin It on Pinterest

Share This