Kepler-based Mobile GPUs
We get hands one with the new NVIDIA GeForce GTX 780M mobility enthusiast GPU.
Late last month, just before the tech world blew up from the mess that is Computex, NVIDIA announced a new line of mobility discrete graphics parts under the GTX 700M series label. At the time we simply posted some news and specifications about the new products but left performance evaluation for a later time. Today we have that for the highest end offering, the GeForce GTX 780M.
As with most mobility GPU releases it seems, the GTX 700M series is not really a new GPU and only offers cursory feature improvements. Based completely on the Kepler line of parts, the GTX 700M will range from 1536 CUDA cores on the GTX 780M to 768 cores on the GTX 760M.
The flagship GTX 780M is essentially a desktop GTX 680 card in a mobile form factor with lower clock speeds. With 1536 CUDA cores running at 823 MHz and boosting to higher speeds depending on the notebook configuration, a 256-bit memory controller running at 5 GHz, the GTX 780M will likely be the fastest mobile GPU you can buy. (And we’ll be testing that in the coming pages.)
The GTX 760M, 765M and 770M offering ranges of performance that scale down to 768 cores at 657 MHz. NVIDIA claims we’ll see the GTX 760M in systems as small as 14-in and below with weights at 2kg or so from vendors like MSI and Acer. For Ultrabooks and thinner machines you’ll have to step down to smaller, less power hungry GPUs like the GT 750 and 740 but even then we expect NVIDIA to have much faster gaming performance than the Haswell-based processor graphics.
Even though AMD continues to offer compelling mobile GPUs, NVIDIA has been dominant in the market of discrete graphics for notebooks. Since the launch of Kepler last year NVIDIA’s market share has ballooned to about 70% according to NPD results. NVIDIA even claims that they have more than 99% of the design wins for Haswell notebooks integrating addition GPU resources. That may explain why the flagship mobile system for AMD’s Radeon HD 8970M is a Trinity-powered MSI laptop rather than anything using Intel’s platforms.
NVIDIA does have other advantages other than performance and design wins in the mobile space that AMD is trying to catch up to. Optimus Technology, used to lower power consumption while the GPU is not needed, is still the best software option for users that want a combination of battery life and gaming horsepower in mobile form factor. The AMD Enduro technology has improved quite a bit but I am waiting for my 8970M test bed to arrive before making any kind of claim to matching the levels NVIDIA has hit with Optimus.
Another advantage NVIDIA has held for a long time is on the driver front – most gaming laptops were able to use NVIDIA standard drivers to improve performance and compatibility on a much faster cadence than depending on laptop OEMs. Companies like Dell and Sony are notorious for simply leaving drivers stagnant for the lifetime of a system which is fine for general computing but not so with gaming. Drivers are required for new game releases and performance improvements can be drastic from one release to the next and NOT having access to those drivers is a huge penalty for mobile gamers.
Add in GeForce Experience, software that attempts to make game setting configuration as simple as turning on a game console, while maintaining flexibility for users that want it. Game support is growing for GFE but in my use I found it to be very helpful for beginner or busy PC gamers that want to get the best experience out of PC games without as hassle. For notebook gamers that is likely even more important.
Alright, gentleman, it’s been done. Desktop GPU in a notebook. Good game everyone, good game.
Nothing more can be done; laptop gaming has reached the point where it has the same performance as a desktop. (No, not the same performance, but daaamn it’s getting there.)
Are you out of your mind
Are you out of your mind lol?
you cant begin to compare the two.
Yeh maybe a recent mobility graphcics card is at the same performance as one say, 2 – 3 years ago of a pc card
You do realise that the 780m
You do realise that the 780m is a desktop 680 to be able to fit that in a laptop nowadays is incredible
E3 AND a gpu review, clearly
E3 AND a gpu review, clearly there is nothing Ryan cannot do.
I would like to see the other mGPU’s also put to the test, but I guess that’ll probably happen by proxy with upcoming notebook reviews.
Also, the link under Metro is for Dirt3.
Thanks for the note on the
Thanks for the note on the Metro page; updated!
This has become my favorite
This has become my favorite GPU review site due to the great accuracy and detail of latencies and actual frames displayed.
PC PERSPECTIVE has really raised the bar on GPU gaming testing. Lets hope the tech world takes notice.
Mi desarrollador está
Mi desarrollador está tratando de persuadir
convencer que me mude a la red a partir de PHP.
. Siempre me ha disgustado la idea debido a los gastos
My page :: memorias usb de figuras en queretaro
I’m concerned for AMD. I hope
I’m concerned for AMD. I hope they get well soon.
Oh, and nice review.
Radeon 8xxx series soon no?
Radeon 8xxx series soon no? Only time will tell. 😉
hmm interesting for a almost
hmm interesting for a almost 2K dollars sounds good for me but how about the temp. of this ?
I’ve got to add, love your
I’ve got to add, love your reviews, the frame latency adds so much depth to your reviews. Average FPS only reviews hold no muster to your own reviews.
Keep the great reviews rolling.
And yet when i play skyrim
And yet when i play skyrim and look down from dragonreach in wightrun i dip down to 15 fps on my reference gtx 780…i have no idea
Can you guys include or add
Can you guys include or add in a release date for it, because I’ve been trying to find it all over the “interwebs” and I just can’t seem to find it anywhere at all.
I am running a 780m in a msi
I am running a 780m in a msi Gt70. Metro last light drops to around 28fps with MSAA 4x on. Need to drop to 2x for smoother ride. Metro 2033 runs worse than last light so clearly improved coding.