Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Graphics Portables Hardware Games

NVIDIA Launches Mobile Maxwell GeForce GTX 980M and GTX 970M Notebook Graphics 29

MojoKid writes: When Nvidia launched their new GeForce GTX 980 and 970 last month, it was obvious that these cards would be coming to mobile sooner rather than later. The significant increase that Maxwell offers in performance-per-watt means that these GPUs should shine in mobile contexts, maybe even more-so than in desktop. Today, Nvidia is unveiling two new mobile GPUs — the GeForce GTX 980M and 970M. Both notebook graphics engines are based on Maxwell's 28nm architecture, and both are trimmed slightly from the full desktop implementation. The GTX 980M is a 1536-core chip (just like the GTX 680 / 680M) while the GTX 970 will pack 1280 cores. Clock speeds are 1038MHz base for the GTX 980M and 924MHz for the GTX 970M, which is significantly faster than the previous gen GTX 680M's launch speeds. The 980M will carry up to 4GB of RAM, while the 970M will offer 3GB and a smaller memory bus.

From eyeballing relative performance expectations, the GTX 970M should be well-suited to 1080p or below at high detail levels, while the GTX 980M should be capable of ultra detail at 1080p or higher resolutions. Maxwell's better efficiency means that it should offer a significant performance improvement over mobile Kepler, even with the same number of cores. Also with this launch Nvidia is introducing "Battery Boost" as a solution for games with less demanding graphics, where battery life can be extended by governing clock speeds to maintain playable frames, without overpower the GPU at higher than needed frame rates.
This discussion has been archived. No new comments can be posted.

NVIDIA Launches Mobile Maxwell GeForce GTX 980M and GTX 970M Notebook Graphics

Comments Filter:
  • Graphics Card News (Score:5, Interesting)

    by Timothy Hartman ( 2905293 ) on Tuesday October 07, 2014 @09:10PM (#48088445)
    Graphics card news has been virtually the same since 1999 with the exception of not seeing games that meet future gen card specs on launch. The incentive to run your current gen games at faster than eyeballs speed on a laptop has to have lost its charm when people who purchase high end laptops would prefer portability of ultra books and anyone who demands the performance and travels can ship their desktops to the competitions they are in to need them.

    The launches of the laptops that actually support these just felt flat to me from what I saw. Honestly who wants that big a hunk of appliance on their lap at this age? The audience for laptops as big as early 2000 era laptops has to be getting slim. With laptop sales being so flat and performance gamers being what they are it doesn't make any sense for the people who have bought $600 cards every six months to want to be behind again on a $3000 lappy.

    Maybe I'm just getting crotchety and I have a 670M.
    • Many laptops live permanently on a desk, and are lugged at a kitchen table, dining table or a low lying table in front of the sofa at best. Or they're taken to a music performance and sit on a table. It might as well be an Osborne 1. Using a laptop on a lap is silly sometimes (it hurts and the keyboard + trackpad combination is miserable when you need it much)

    • by UnknownSoldier ( 67820 ) on Tuesday October 07, 2014 @09:56PM (#48088609)

      No offense, but you don't know what the fuck you are talking about. There are several orders of magnitude difference between a 1999 GPU and a 2014 GPU, let alone the fixed function pipeline and pixel (fragment) shaders.

      I have a GTX Titan, and I *miss* my old work 17" MacBook Pro laptop -- the bigger screen was _nice_. While I appreciate the new slim MacBook Pro and the high resolution Retina brings I _also_ want the fastest nVidia GPU available so I continue to do CUDA research, aka "Big Data on the GPU." when I'm not at home. Current mobile GPU's suck for performance, but they are _finally_ decent.

      • by tlhIngan ( 30335 )

        I have a GTX Titan, and I *miss* my old work 17" MacBook Pro laptop -- the bigger screen was _nice_. While I appreciate the new slim MacBook Pro and the high resolution Retina brings I _also_ want the fastest nVidia GPU available so I continue to do CUDA research, aka "Big Data on the GPU." when I'm not at home. Current mobile GPU's suck for performance, but they are _finally_ decent.

        Why not use the fact that you have two external PCIe connections and well, use it with an external PCIe enclosure?

        (The fact t

        • by jandrese ( 485 )
          An external graphics card is a lot less useful when it's attached to the soda straw of a single PCIe lane. I've seen the idea of external video cards for laptops come up every few years, and the reality of the situation just kills the idea. You have an expensive and awkward dongle that has to be plugged in attached to a bus that's way too skinny to feed it properly and driver issues for days. It's just miserable.
      • He wasn't talking about the relative difference between 1999 and 2014 GPU performance. He was comparing graphics performance on laptops to other hardware available at the same time.

        That doesn't mean I agree with the OP. Mobile graphics has improved to the point that a low end laptop can actually be used to play a fair number of games, without breaking the bank in the process.

    • Does your "portable" LAN rig (including monitor) fit in a rucksack or panniers (bike rack)? Does it have decent h/k speakers and a meaty subwoofer built in for train journeys? Does it boot (and login) in under 30 seconds from power off, less than that from hibernate?

      These trade-offs alone, along with the ability to game or perform (I have an Edirol and portable 12V PA) for two hours without the mains, make the NVidia 9700M GTS seem quite impressive, actually!

      They're bigger because they contain more useful t

    • by Anonymous Coward

      >> Maybe I'm just getting crotchety

      Yup, definitely. Who wants to sit at a desk and play video games when they can enjoy them from the comfort of their couch? I get all the benefits of console gaming (increased comfort level, ability to recline) while maintaining all the benefits of PC gaming. You don't need to spend $3000 on a laptop to play current gen games on high or ultra. I bought a $1.5k Lenovo Y50 (gtx 860m) to upgrade my previous gen Y580 (gtx 660m) because I have no desire to be constrain

    • I am one of (maybe rare) customers for these solutions.

      I'm in my late 30's now and have been and still am a gamer. My work requires me to travel around a lot internationally and I decided several years ago that I won't have a desktop PC anymore. So, whenever I'm purchasing a new laptop, I'm always looking for a best performance-to-weight ratio. 6 years ago that didn't go well at all and I ended up with a 17" monster that worked nicely, but was a bitch to carry around. Three years ago I scaled down to a 15"

  • What about the 7xxM series? 780M being the flagship?

    • how about 870M being BETTER in almost EVERY single parameter than 970M?

      • how about 870M being BETTER in almost EVERY single parameter than 970M?

        The 970M is based on Maxwell instead of the Kepler architecture of 870M. It is going to be a lot faster than the 870M despite having a lower number of shaders and a lower clock speed.

        There are several reasons for this. In Maxwell memory access is more effective: There is a improved framebuffer compression that increases effective bandwidth by around 25%. L2 cache is now 2 MB instead of 768 KB in 870M. The next and likely more important change are more efficient gpu cores. 970M has 10 SMM cores with 128 Shad

    • Rebranding is widespread in the mobile (as in laptop) GPU market. The 7xxM is probably just a rebrand of the 6xxM series.

  • by should_be_linear ( 779431 ) on Wednesday October 08, 2014 @03:05AM (#48089581)
    Good news for developers is: You can connect this to HDMI 2.0 4K TV set in native res (if TV actually support it 60Hz 4:4:4 or RGB mode in 4K, which is not always the case in current range of 4K TVs).
  • It remains to be seen what support these mobile cards have for double precision floating point (DPFP) operations used in scientific programming applications. The Nvidia 950 M card on the 15" Macbook Pro retina has poor support for DPFP operations and it is very difficult to check even a single test case on these machines. The Nvidia 950 M card offers a paltry 23 Gflops while the host i7 on the 15" Macbook pro offers 120 Gflops. See http://tinyurl.com/cuda-on-mac [tinyurl.com] for details. What do Slashdotters recommend

Top Ten Things Overheard At The ANSI C Draft Committee Meetings: (10) Sorry, but that's too useful.

Working...