Nvidia Geforce GTX 980M Mobility Maxwell Specifications and Benchmarks Revealed

The GTX 980M as you may recall is the 900 series flagship for mobility platform that was released a few weeks agone. Videocardz.com has now found information on the same, via Sweclockers.com, regarding operation numbers and CUDA core count. The Geforce 900 Mobility series has yet to hit the shelves technically so some element of the original apprehension withal remains. NVIDIA Maxwell GM204

The Nvidia Maxwell Geforce GTX 980M allegedly has 2048 CUDA Cores

At present the primary leak by sweclockers.com mentions that the mobility GPU has 2048 CUDA Cores, which would be effectually the same number of SMMs equally the GTX 980 (desktop). The full specifications are equally follows. Allegedly, the gtx 980m volition boast 64 ROPs, 128 TMUs and 2048 CUDA Cores. The Geforce GTX 980M volition feature huge amount of retentiveness as is traditional for laptop gpus, infact information technology volition feature 8GB of GDDR5 memory.

Since we are talking about the mobility platform here, you deceit look the GTX 980M to have loftier clock rates. Fifty-fifty then, it does pretty much fine, with 1038Mhz as the base clock and 1127Mhz as the boost clock. As VCZ mentions, this is about 100Mhz more than the Kepler variants. Interestingly, the retentivity is going to be clocked at 5Ghz instead of the full 7Ghz. This is evidently an try to go on the thermal footprint on the smaller side. GTX 980M Benchmark

Apparently, NotebookCheck lists the GPU equally having 1664 cores. This is a startlingly loftier discrepancy, something we can't really ignore. Differences in clocks we tin can usually ignore or account for, only data found at NotebookCheck completely contradicts the above. In my experience, Sweclockers has proved to be more reliable than the database tables kept past famous sites such as NotebookCheck. Now for the actual benchmarks.Nvidia's mobile partitioning has been slowly but surely catching up with its desktop counterparts every bit the generations and architectures progress. According to a slide by Nvidia, the 980 and 980M are but ~10% autonomously.