Nvidia Geforce GTX 1080 Ti Spotted In A Zauba Shipping Manifest - Features 10 GB Worth of vRAM And A Cheaper Price Tag Than A TITAN-X

Image used for illustrative purposes, not a real Geforce affiche.

The expert folks over at Videocardz have spotted the shipping manifest at Zauba that we have all been waiting for. The anxiously predictable GTX 1080 Ti appears to finally take been spotted in the wild. The manifest itself follows the aforementioned nomenclature that nosotros accept previously seen on Pascal based cards being shipped for testing and features a graphics bill of fare with x GB worth of vRAM.

Nvidia's Pascal 'GTX 1080 Ti' flagship entering - GP102 GPU spotted in the wild with 10 GB vRAM

The story of the GP102 based core started nearly 4 months back and was confirmed by a leaked modify log by the AIDA64 devs. Nvidia has already released 1 variant of the core in the form of the TITAN-Ten for the ultra high finish spectrum but we have all the same to come across i in the usual mainstream "Ti" format. Dubbed the Geforce GTX 1080 Ti (at to the lowest degree till nosotros are sure of the name) the graphics carte du jour will bring the GP102 cadre into the mainstream with a price tag under $k. The aircraft manifest'southward INR value, which is basically the insurance value of the flake, is cheaper than the TITAN-X's. While INR does non directly interpret to MSRP, we tin can safely infer that information technology will be cheaper than the TITAN-X. The exact type of GDDR (GDDR5 or 5X) is not mentioned.

The shipping manifest shows the memory at 10240 MB, which is a solid 10 GBs worth of vRAM. The retentivity omnibus is shown equally a 384 bit part, although it remains to be seen whether the full configuration will be active. As for the core count itself, apart from the fact that this is a GP102 part, we really don't know anything concrete. The TITAN-Ten itself has 3584 cores while the GTX 1080 has 2560. Either the GP102 part will have a configuration that is somewhere in between or it will have the same die equally the TITAN-X (unlikely). There is a very slim possibility that Nvidia decides to go ahead with a full fledged core.

The shipping manifest is publicly available on the website: Zauba.com

On newspaper, the GP100 had a total of 60 SMs or Streaming Multiprocessors. Each SM had a 2:ane ratio of FP32 cores to FP64 cores. This finer meant that you were looking at 3840 CUDA Cores on the FP32 side of things and 1920 on the FP64 side. For a total of 5760 cores, the GP100 clocks in at simply 610mm². Of course, as the more than tech savvy would recall, the P100 Accelerator didn't really use the full GP100 die. It used ane with 56 SMs for a total of 3584 FP32 Cores and 1792 FP64 Cores. The reasons for the lower amount of SM can just be attributed to bad yields (which are expected this early in a node and at this big a size) which is also i more reason why we should non expect the GP100 to ability the GTX 1080 Ti (not a full one at any rate).

Of form, unlike the P100 accelerator, a gaming graphics card similar the GTX 1080 Ti doesn't actually need the FP64 cores. So there is no point in wasting valuable die infinite in hosting DP units. It is with this that nosotros can begin the discussion of what we can await from the GP102. Go along in listen however, that accelerators like the P100 do non have ROPs – so we are not actually looking at a linear pay-off. At that place is also the fact that the GP102 will nearly certainly not be equally big as the GP100. Our usual sources have been very tight lipped about the this detail die simply they did country that it would be "exactly one-half way" between the GP104 and the GP100.

Since the GP100 and GP104 are 610mm² and 314mm² respectively, the GP102 was going to be in the ballpark of 462mm² to 478mm². This was confirmed with the release of the TITAN-X GP102 variant. A fleck of that size was theoretically capable of hosting the full compliment of the GP100 FP32 cluster, or in other words 3840 FP32 CUDA cores. But due to FP64 remaining constant, nosotros are looking at anywhere from 3072 to 3584 CUDA cores. In any example, the amount of CUDA cores nowadays on the die volition exist somewhere around this range (give or take a few SMs). According to our estimates, with that cadre count y'all are looking at power consumption within the range of 270W with GDDR5X. If Nvidia shifts to HBM2 for the GTX 1080 Ti, the consumption should fall within the power budget of 250W. Co-ordinate to our sources, HBM2 will start taping out in the tertiary quarter of this twelvemonth, so it is unclear whether Nvidia volition opt to go with the same.

Nvidia GTX 1080 Ti 'Expected' Specifications Comparision

NVIDIA Graphics Menu	Tesla P100	GTX 1080 Ti*	GTX 1080
Process Node	16nm	16nm	16nm
Transistors	15.3 Billion	ten.viii Billion	7.2 Billion
GPU Dice Size	610 mm2	471mm²	314mm²
SMs	56	60	twoscore
CUDA Cores Per SM	64	64	64
FP32 CUDA Cores (Total)	3584	3072-3584 (TBC)	2560
FP64 CUDA Cores / SM	32	TBD	two
FP64 CUDA Cores / GPU	1792	TBD	80
Memory Interface	4096-bit HBM2	GDDR5X	256-scrap GDDR5X
Retention Size	16 / 32 GB HBM2	10 GB GDDR5X	eight GB GDDR5X
TDP	300W	250W	180W