Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
Graphics Software Hardware

NVIDIA Launches Five New Mobile GPUs 67

Engadget is reporting that NVIDIA has released five new mobile GPUs to fill some imagined gap in the 200M series lineup. These new chips supposedly double the performance and halve the power consumption of the older chips, but still no word on why they think we need eight different GPU options. "The cards are SLI, HybridPower, CUDA, Windows 7 and DirectX 10.1 compatible, and all support PhysX other than the low-end G210M. Of course, with integrated graphics like the 9400M starting to obviate discrete graphics in the mid range -- even including Apple's latest low-end 15-inch MacBook Pro -- we're not sure what we'll do with eight different GPU options, but we suppose NVIDIA's yet-to-be-announced price sheet for these cards will make it all clear in time."
This discussion has been archived. No new comments can be posted.

NVIDIA Launches Five New Mobile GPUs

Comments Filter:
  • Finally (Score:5, Interesting)

    by Yvan256 ( 722131 ) on Monday June 15, 2009 @12:46PM (#28337635) Homepage Journal

    Finally, news about low-power GPUs with decent capabilities.

    I'm sure hardcore gamers prefer bleeding edge hardware news, but for the rest of us, heat dissipation and power requirements are beginning to be a nuisance more than anything else. I'm sure 99% of computer users would be fine with a dual-core Atom CPU and one of those new GPUs.

    • Re: (Score:3, Informative)

      by Anonymous Coward

      Finally, news about low-power GPUs with decent capabilities.

      I'm sure hardcore gamers prefer bleeding edge hardware news, but for the rest of us, heat dissipation and power requirements are beginning to be a nuisance more than anything else. I'm sure 99% of computer users would be fine with a dual-core Atom CPU and one of those new GPUs.

      I have a duel core atom, and it sucks for flash. Its really sad that you can have the best video solution in the world paired with these and video ends up being the thing that suffers the most.
      Once we get HTML 5, and video on the web migrates to a non-CPU based video system that will be true though.

      • Re: (Score:2, Interesting)

        I have a duel core atom, and it sucks for flash

        Probably cuz it's tired from fighting in one-on-one combat with the GPU all the time. I recommend getting an Atom that works with its GPU [tomshardware.com].

        • I have a duel core atom, and it sucks for flash

          Probably cuz it's tired from fighting in one-on-one combat with the GPU all the time. I recommend getting an Atom that works with its GPU [tomshardware.com].

          Your link says nothing about the GPU in the Ion chipset (GeForce 9300) helping Flash video in any way (it doesn't). Yes, we all know Ion's GPU accelerates the codecs used in Blu-ray (H.264, VC-1, MPEG-2), but the Atom has to do all the work when it comes to Flash (and it sucks).

          Here's a much better link that explains how the Atom (single and dual core) does with Flash on the Ion platform at different resolutions: Zotac's Ion: The Follow Up - Watching Flash Video on the Ion [anandtech.com]

          Summary: single-core Atom on Io

      • I can stream HD video (movie trailers from apple's site) on my netbook. There is a dual core atom? Where? It looks like dual core cause of hyper threading, but it is still a single core CPU. Well, the 1.6Ghz one in the netbooks that I have seen are anyway. The trailers streamed and played fine under linux (ubuntu 8.10 and ubuntu remix), OSX 10.5, XP Pro, and win 7. I needed to install the proper player to view them, but once installed no issues.

        The flash websites (which drive me crazy in a bad way) have no

    • Re: (Score:3, Interesting)

      by MBGMorden ( 803437 )

      I think there's some misunderstanding between "hardcore gamers" and people who the Atom CPU is viable for. The Atom is a wonderfully efficient chip, and I'll concede that it's probably good enough for most "mundane" computing tasks. However, it's not good for ANY level of traditional (and by traditional I mean something that uses some level of 3d acceleration) PC gaming. I'd also question it's usefulness for things like video encoding. That's not a high end or odd application anymore. My mother (who is

      • Re:Finally (Score:5, Interesting)

        by Hurricane78 ( 562437 ) <deleted&slashdot,org> on Monday June 15, 2009 @02:03PM (#28338715)

        The Atom is a wonderfully efficient chip

        No, it's not. It's a wonderfully feature-less chip, with everything possible off-loaded into the northbridge. Which is why the NB looks like the real CPU, when you look at the board.

        If you want wonderful efficiency, look at those new smartbooks that were show in a recent /. article. They take 1-2 watts, and play full-hd and hardware accelerated flash.
        I rather stack 10 of those, than buying one Atom chip (with the same power usage).

        I just wish someone would offer bare-bones ARM modules that you could take as much as you wanted of, and stick them together to form a desktop computer. maybe even have a special module that you could take out as a smartbook. Throw in some GPUs, and maybe an SPU (sound), or whatever you like.
        Of course Windows would -- as usual -- just choke and die, but Windows and Smartbooks do not fit anyway (yet). It's all Linux in its many forms (including Android).

        I for one, would love to have a desktop system, that is essentially a more tightly integrated blade rack with a fast backbone bus.

        • I was going to post exactly what you said, but you seem to have covered all the bases. It bears repeating though; Atom is not a wonderfully efficient chip at all, it just consumes less power compared to other x86 chips but I would guess that its performance per watt is actually lower than other recent x86 CPUs, and definitely much lower than an ARM CPU. I think it's much more feasible to go from a sensible, low power architecture like ARM and try to increase performance, than to take a complex, power-hungry
        • No, it's not. It's a wonderfully feature-less chip, with everything possible off-loaded into the northbridge. Which is why the NB looks like the real CPU, when you look at the board.

          Are you sure it isn't only about the NB not being as power-efficient yet? I wonder if there is anything more "off-loaded" than with any other CPU.

        • by matmota ( 238500 )

          You can get a MIPS-based desktop system with 72 processors that consumes 300 Watts, from SyCortex. They call it their Deskside Development System [sicortex.com] for their bigger parallel computers, and they say it does have a fast backbone bus.

          It does run Linux, but at $23,695.00 (48 GB RAM) it's not, I suspect, what you were asking for. I would also like some cheap barebones I could just go on populating with CPUs as I wanted.

          The GP might like SGI's Molecule [gizmodo.com] better though, it being Atom-based: 5000 chips, that's 10000 co

        • by renoX ( 11677 )

          >>The Atom is a wonderfully efficient chip
          > No, it's not. It's a wonderfully feature-less chip, with everything possible off-loaded into the northbridge.

          Well this depends what you compare an Atom to, compared to many other x86 it doesn't off-load anything more in the northbridge..
          Compared to an SOC or an ARM, sure.

        • Throw in a Burroughs Large Systems B5000 type CPU, HyperTransport backplane, Radix tree/string optimized GPGPU, possibly with basic dataflow capabilities for dynamic transcompilation (i386, anyone?), storage and networking, hide it all under the LLVA interface, embed Linux in the motherboard firmware, and Intel and IBM are gonna take lessons from you.
      • by Kjella ( 173770 )

        Atom + nVidia ION does full 1080p decoding and is capable to running a 3D desktop with any wiggly effects you might want. That covers a lot of ground in my book. Gaming and video editing is at the opposite end of the scale for me, having a HD cam it's one of the things that really give the machine a workout. But it's rather specific in either you got it or you don't. If you don't, and most people I know only have digicams, then Atom will do you just fine. For gaming, go dualcore + fast GPU, for video encodi

    • As an owner of an Asus EEE BOX 206 with an ATI HD video card, I could only agree that it would suite the needs of most users if Adobe would get up off their but and create decent GPU offloading capabilities into Flash.

      The EEE has an Atom and draws 19W max. It plays DVD's just fine. Not being able to stream YouTube or HULU really sucks, though.

      I question if producing 8 different chip sets is as cost effective is perhaps producing three? The more quantity you can produce of a single chip, the cheaper man

      • by Pulzar ( 81031 )

        I question if producing 8 different chip sets is as cost effective is perhaps producing three? The more quantity you can produce of a single chip, the cheaper manufacturing becomes, right?

        We're still talking about two (maybe three) different ASICs, all packaged/fused into different products. Having multiple packages still costs some money, but being able to hit the sweet spot of every market segment is worth it.

    • Re: (Score:3, Interesting)

      by AmiMoJo ( 196126 )

      Let's just hope they fixed the manufacturing problems that are still dogging them.

      I work fixing PCs for business and the public, and we have seen over 120 HP laptops with nVidia chipsets that have failed in the past six months. Usual symptoms are no video output (but otherwise boots), wifi card dropping out or just completely dead and not POSTing.

      HP will do anything to get out of fixing the problem, which they won't even admit exists on most affected models. There is a website (http://www.hplies.com/ [hplies.com]) org

      • Getting off topic, but I just got an HP replaced for that reason (dead nVidia chip). (I'm an nVidia snob, which is why that lappy had one of their chips to begin with.) If you have a bad HP, take the advice at that site [hplies.com], and get a case manager. Using regular support, we had to send it in 3 times to get a working (though down-specced) machine. But once we got a case manager, they sent a new machine.

        • by AmiMoJo ( 196126 )

          We managed to get three machines fixed by HP.

          1. Took 8 months, went back and forth several times because they initially installed the wrong motherboard (similar spec but lacking a HDMI port). It was a US model and that seemed to confuse them. So much for a world wide warranty.

          2. Was fixed the second time it went to them, but for some inexplicable reason came back with a cracked copy of Vista installed on the HDD. Luckily we imaged the drive before sending it off so we were able to restore it. No idea how or

  • Yes these are nothing special in the big picture. But the pricepoint could be extremely low for all we know. I'll bet this is an effort to put Nvidia chipsets in an entire generation of netbooks -- from which Nvidia has been excluded in favor of integrated graphics.

  • by Vigile ( 99919 ) * on Monday June 15, 2009 @01:04PM (#28337891)

    This piece has more commentary on the release as opposed to regurgitating specs: http://www.pcper.com/article.php?aid=732 [pcper.com]

    It looks like this new architecture is going to be quite different than the desktop counterpart.

  • Suicidal NVIDIA GPUs (Score:4, Interesting)

    by madnis ( 1300099 ) on Monday June 15, 2009 @01:10PM (#28337971)
    So has NVIDIA fixed their bump-material problem, or can I expect one of these GPUs to croak after 6 months like the my laptop's 8400M did?
    • Re: (Score:3, Interesting)

      by ledow ( 319597 )

      Oh, wow. Thanks. I've never heard of this and just had my new laptop repaired with what appears to be an identical problem.

      It was a Clevo with a 9300M on it and the symptoms sound exactly the same - 6 months in, the graphics starting playing up to the point that the computer just hung if you touched the keyboard or moved it in any way, always with graphical corruption, and sometimes Linux/Windows would just carry on regardless, but with corrupt graphics. Sometimes there'd be a kernel panic or freeze, but

    • Re: (Score:3, Interesting)

      by Hurricane78 ( 562437 )

      Well... they already killed themselves with their naming scheme changes. Re-labeling things so that you are pretty much guaranteed to feel ripped off when buying one of their cards, because it is just the same old shit with a new name, does not essentially make them trustworthy, or me wanting to buy anything from them.

      Unfortunately, ATi's current generation is completely incompatible with Linux, (Not compatible to current kernel interfaces [>=2.6.29], massive tons of things that make it crash, composite

    • Re: (Score:3, Informative)

      This is a very important issue for anyone looking into nVidia chips, and I for one won't buy one until it is ascertained whether this issue has been fixed. Apparently the problem was even in a chip as low performance as the Geforce Go 6150 IGP, which is pathetic. An IGP should never have overheating problems, what the hell nVidia.
  • the one (Score:2, Funny)

    by gellern ( 1045842 )
    8 GPUs to rule them all and in the darkness bind them! i guess their strategy in current market is: can't convince them? confuse them!
    • and in the darkness bind them

      Are you saying they will come bundled with Doom 4?
      Better hope they integrate all eight at the same time. Or you might end up with the set of your keyboard LEDs having a higher resolution (and being brighter anyway).

  • by BobMcD ( 601576 ) on Monday June 15, 2009 @01:26PM (#28338147)

    NVIDIA is filling in what it presumes to be holes in its next-generation GPU lineup, adding the 40nm G210M, GT 230M, GT 240M and GTS 250M, with GDDR3 memory ranging from 512MB to 1GB, to its existing GTX 280M, GTX 260M and GTS 160M laptop graphics cards. Apparently the new cards sport "double the performance" and "half the power consumption" over the last generation of discrete GPUs they're replacing. The cards are SLI, HybridPower, CUDA, Windows 7 and DirectX 10.1 compatible, and all support PhysX other than the low-end G210M. Of course, with integrated graphics like the 9400M starting to obviate discrete graphics in the mid range -- even including Apple's latest low-end 15-inch MacBook Pro -- we're not sure what we'll do with eight different GPU options, but we suppose NVIDIA's yet-to-be-announced price sheet for these cards will make it all clear in time.

    Look at the words changed:

    [what it presumes to be holes] becomes [some imagined gap]

    [Apparently the new cards sport "double the performance" and "half the power consumption"] becomes [These new chips supposedly double the performance and halve the power consumption]

    [we're not sure what we'll do with eight different GPU options] becomes [still no word on why they think we need eight different GPU options]

    and [but we suppose NVIDIA's yet-to-be-announced price sheet for these cards will make it all clear in time] gets completely omitted...

    WTF?

    • Also, what's up with calling 9400M midrange in the same article that the faster 210M is called low end? And why is an article that mentions 4 new GPUs labeled as introducing 5 new GPUs in the title?

  • by slyn ( 1111419 ) <ozzietheowl@gmail.com> on Monday June 15, 2009 @01:32PM (#28338241)

    So I was looking around after seeing this earlier to try and make sense of what older generation codenames match to the newer generation codenames, and found this: http://www.nvidia.com/object/geforce_m_series.html [nvidia.com] (scroll down).

    Basically it goes GTX > GTS > GT > GS > G

    The old 9400/8400 line has become the 210/110
    The old 9600/8600 line has become the 230/130
    The old 9800/8800 GT/GS has become the 250/150
    And The old 9800/8800 GTX/GTS has become the 280

    There are a few other cards that fall in the middle of categories, but that seems to be the basic gist of it as far as I can tell.

    Heres another useful resource for comparing mobile gpu's: http://www.notebookcheck.net/Comparison-of-Graphic-Cards.130.0.html [notebookcheck.net]

    • I wonder if these codes promote fanboyism. You've learnt the code, you know the lingo, you buy the card that you know. Accepting that the other side might just have something better this iteration would require turning in your secret decoder ring.

    • The old 9400/8400 line has become the 210/110
      The old 9600/8600 line has become the 230/130
      The old 9800/8800 GT/GS has become the 250/150
      And The old 9800/8800 GTX/GTS has become the 280

      You mean the GTX 280M is not based on the desktop GTX 280, but the previous-generation 9800/8800? Death to NVIDIA!

      I'm kidding, of course, but this is a long-time pet peave of mine. The GeForce4 MX was based on GeForce2 technology. The Radeon 8000 was not a DirectX 8/OpenGL 1.4 GPU like the rest of the 8000-series. This shit continues today with these NVIDIA mobile GPUs.

  • by rickb928 ( 945187 ) on Monday June 15, 2009 @01:39PM (#28338309) Homepage Journal

    - Intel threatening an all-in-one smartphone chipset

    - ARM showing up everywhere, netbooks coming soon, hopeful big battery life gains and HD playback

    - Microsoft feeling left out of the smart- market. (I know, insert favorite pun here)

    - Android liking its chances in the netbook market

    - AMD looking at netbooks for growth

    It's wonderful. I may yet get a netbook with 8+ hrs battery life, touchscreen, and I can settle for a Bluetooth headset profile connection to my smartphone in my pocket.

    Now, gimme the 8' screen that folds out to 8"x14", and a swiveling keyboard. Woot. And that 700MHz thingie that is supposed to make broadband ubiquitous... For under $300, and less than $40/mo for the Interwebs.

    I'll buy it.

  • by sootman ( 158191 ) on Monday June 15, 2009 @01:48PM (#28338463) Homepage Journal

    Fuck Everything, We're Doing 5 GPUs [theonion.com]

    Would someone tell me how this happened? We were the fucking vanguard of graphics cards in this country. The GeForce was the card to own. Then the other guy came out with a three-GPU card... Well, fuck it. We're going to five GPUs.

    Here's the report from Engineering. Someone put it in the bathroom: I want to wipe my ass with it. They don't tell me what to invent--I tell them. And I'm telling them to stick two more GPUs in there. I don't care how. Make the GPUs so thin they're invisible. Put some on the bracket. I don't care if they have to cram the fifth one in perpendicular to the other four, just do it!

    I know what you're thinking now: What'll people say? Mew mew mew. Oh, no, what will people say?! Grow the fuck up. When you're on top, people talk. That's the price you pay for being on top. Which NVidia is, always has been, and forever shall be, Amen, five GPUs, sweet Jesus in heaven.

    (Hey, Slashcode, why won't you format <i> or <em> inside <blockquote>?)

    • by gringer ( 252588 )

      (Hey, Slashcode, why won't you format <i> or <em> inside <blockquote>?)

      Because you should be using <quote> instead, which does support that formatting.

  • we're not sure what we'll do with eight different GPU options

    yeah because theres hardly any options in the desktop market...

  • I apologise for being a little behind the curve here, but can someone tell me if laptop graphics cards have standard sizes and interfaces nowadays, preferably with useful links? I always thought laptops were more of a custom build than your everyday PC.
    • They call them cards but that's just to differentiate them from the integrated GPUs. I don't think you can actually replace them.
    • Re: (Score:3, Informative)

      by Atriqus ( 826899 )

      Sure, here's a link that'll send you in the right direction.

      MXM [wikipedia.org]

  • Yawn, mobile GPU's. Where are the times of the huge graphics revolutions like when the GeForce just came out?
  • then I'll pass. The upcoming ATi updates are rolling in OpenCL which allows for us to have cross-compilation even if Nvidia thinks everyone's gung-ho about CUDA.
    • by Anonymous Coward

      If they aren't OpenCL compliant, ... then I'll pass. The upcoming ATi updates are rolling in OpenCL which allows for us to have cross-compilation even if Nvidia thinks everyone's gung-ho about CUDA.

      AFAIK, Nvidia released OpenCL drivers that run on-top of the nvidia cuda runtime
      http://www.nvidia.com/object/cuda_opencl.html [nvidia.com]

      Since all recent Nvidia chips are CUDA enabled, they are by default also OpenCL enabled.

Every nonzero finite dimensional inner product space has an orthonormal basis. It makes sense, when you don't think about it.

Working...