In 2014/2015, it took NVIDIA 6 months from the launch of the Maxwell 2 architecture to get GTX Titan X out the door. All things considered, that was a fast turnaround for a new architecture. However now that we’re the Pascal generation, it turns out NVIDIA is in the mood to set a speed record, and in more ways than one.

Announced this evening by Jen-Hsun Huang at an engagement at Stanford University is the NVIDIA Titan X, NVIDIA’s new flagship video card. Based on the company’s new GP102 GPU, it’s launching in less than two weeks, on August 2nd.

NVIDIA GPU Specification Comparison
  NVIDIA Titan X GTX 1080 GTX Titan X GTX Titan
CUDA Cores 3584 2560 3072 2688
Texture Units 224? 160 192 224
ROPs 96? 64 96 48
Core Clock 1417MHz 1607MHz 1000MHz 837MHz
Boost Clock 1531MHz 1733MHz 1075MHz 876MHz
TFLOPs (FMA) 11 TFLOPs 9 TFLOPs 6.6 TFLOPs 4.7 TFLOPs
Memory Clock 10Gbps GDDR5X 10Gbps GDDR5X 7Gbps GDDR5 6Gbps GDDR5
Memory Bus Width 384-bit 256-bit 384-bit 384-bit
VRAM 12GB 8GB 12GB 6GB
FP64 1/32 1/32 1/32 1/3
FP16 (Native) 1/64 1/64 N/A N/A
INT8 4:1 ? ? ?
TDP 250W 180W 250W 250W
GPU GP102 GP104 GM200 GK110
Transistor Count 12B 7.2B 8B 7.1B
Die Size 471mm2 314mm2 601mm2 551mm2
Manufacturing Process TSMC 16nm TSMC 16nm TSMC 28nm TSMC 28nm
Launch Date 08/02/2016 05/27/2016 03/17/2015 02/21/2013
Launch Price $1200 MSRP: $599
Founders $699
$999 $999

Let’s dive right into the numbers, shall we? The NVIDIA Titan X will be shipping with 3584 CUDA cores. Assuming that NVIDIA retains their GP104-style consumer architecture here – and there’s every reason to expect they will – then we’re looking at 28 SMs, or 40% more than GP104 and the GTX 1080.

It’s interesting to note here that 3584 CUDA cores happens to be the exact same number of CUDA cores also found in the Tesla P100 accelerator. These products are based on very different GPUs, but I bring this up because Tesla P100 did not use a fully enabled GP100 GPU; its GPU features 3840 CUDA cores in total. NVIDIA is not confirming the total number of CUDA cores in GP102 at this time, but if it’s meant to be a lightweight version of GP100, then this may not be a fully enabled card. This would also maintain the 3:2:1 ratio between GP102/104/106, as we saw with GM200/204/206.

On the clockspeed front, Titan X will be clocked at 1417MHz base and 1531MHz boost. This puts the total FP32 throughput at 11 TFLOPs (well, 10.97…), 24% higher than GTX 1080. In terms of expected performance, NVIDIA isn’t offering any comparisons to GTX 1080 at this time, but relative to the Maxwell 2 based GTX Titan X, they are talking about an up to 60% performance boost.

Feeding the beast that is GP102 is a 384-bit GDDR5X memory bus. NVIDIA will be running Titan X’s GDDR5X at the same 10Gbps as on GTX 1080, so we’re looking at a straight-up 50% increase in memory bus size and resulting memory bandwidth, bringing Titan X to 480GB/sec.

At this point in time there are a few unknowns about other specifications of the card. ROP count and texture unit count have not been disclosed (and this is something NVIDIA rarely posts on their site anyhow), but based on GP104 and GP106, I believe it’s safe to assume that we’re looking at 224 texture units and 96 ROPs respectively. To put this into numbers then, theoretical performance versus a GTX 1080 would be 24% more shading/texturing/geometry/compute performance, 50% more memory bandwidth, and 33% more ROP throughput. Or relative GTX Titan X (Maxwell 2), 56% more shading/texturing/geometry/compute performance, 43% more memory bandwidth, and 42% more ROP throughput. Of course, none of this takes into account any of Pascal’s architectural advantages such as a new delta color compression system.

Meanwhile like the past Titans, the new Titan X is a 250W card, putting it 70W (39%) above GTX 1080. In pictures released by NVIDIA and confirmed by their spec sheet, this will be powered by the typical 8-pin + 6-pin power connector setup. And speaking of pictures, the handful of pictures released so far confirm that the card will be following NVIDIA’s previous reference design, in the new GTX 1000 series triangular style. This means we’re looking at a blower based card – now clad in black for Titan X – using a vapor chamber setup like the GTX 1080 and past Titan cards.

The TDP difference between Titan X and GTX 1080 may also explain some of rationale behind the performance estimates above. In the Maxwel 2 generation, GTX Titan X (250W) consumed 85W more than GTX 980 (165W); but for the Pascal generation, NVIDIA only gets another 70W. As power is the ultimate factor limiting performance, it stands to reason that NVIDIA can't increase performance over GTX 1080 (in the form of CUDA cores and clockspeeds) by as much as they could over GTX 980. There is always the option to go above 250W - Tesla P100 in mezzanine form goes to 300 W - but for a PCIe form factor, 250W seems to be the sweet spot for NVIDIA.

Moving on, display I/O is listed as DisplayPort 1.4, HDMI 2.0b, and DL-DVI; NVIDIA doesn’t list the number of ports (and they aren’t visible in product photos), but I’d expect that it’s 3x DP, 1x HDMI, and 1x DL-DVI, just as with the past Titan X and GTX 1080.

From a marketing standpoint, it goes without saying that NVIDIA is pitching the Titan X as their new flagship card. What is interesting however is that it’s not being classified as a GeForce card, rather it’s the amorphous “NVIDIA Titan X”, being neither Quadro, Tesla, nor GeForce. Since the first card’s introduction in 2013, the GTX Titan series has always walked a fine line as a prosumer card, balanced between a relatively cheap compute card for workstations, and an uber gaming card for gaming PCs.

That NVIDIA has removed this card from the GeForce family would seem to further cement its place as a prosumer card. On the compute front the company is separately advertising the card's 44 TOPs INT8 compute performance - INT8 being frequently used for neural network inference - which is something they haven't done before for GeForce or Titan cards. Though make no mistake: the company’s GeForce division is marketing the card and it’s listed on GeForce.com, so it is still very much a gaming card as well.

As for pricing and availability, NVIDIA’s flagships have always been expensive, and NVIDIA Titan X even more so. The card will retail for $1200, $200 more than the previous GTX Titan X (Maxwell 2), and $500 more than the NVIDIA-built GTX 1080 Founders Edition. Given the overall higher prices for the GTX 1000 series, this isn’t something that surprises me, but none the less it means buying NVIDIA’s best card just got a bit more expensive. Meanwhile for distribution, making a departure from previous generations, the card is only being sold directly by NVIDIA through their website. The company’s board partners will not be distributing it, though system builders will still be able to include it.

Overall the announcement of this new Titan card, its specifications, and its timing raises a lot of questions. Does GP102 have fast FP64/FP16 hardware, or is it purely a larger GP104, finally formalizing the long-anticipated divide between HPC and consumer GPUs? Just how much smaller is GP102 versus GP100? How has NVIDIA been able to contract their launch window by so much for the Pascal generation, launching 3 GPUs in the span of 3 months? These are all good questions I hope we’ll get an answer to, and with an August 2nd launch it looks like we won’t be waiting too long.

Update 07/25: NVIDIA has given us a few answers to the question above. We have confirmation that the FP64 and FP16 rates are identical to GP104, which is to say very slow, and primarily there for compatibility/debug purposes. With the exception of INT8 support, this is a bigger GP104 throughout.

Meanwhile we have a die size for GP102: 471mm2, which is 139mm2 smaller than GP100. Given that both (presumably) have the same number of FP32 cores, the die space savings and implications are significant. This is as best of an example as we're ever going to get on the die space cost of the HPC features limited to GP100: NVLInk, fast FP64/FP16 support, larger register files, etc. By splitting HPC and graphics/inference into two GPUs, NVIDIA can produce GP102 at what should be a significantly lower price (and higher yield), something they couldn't do until the market for compute products based on GP100 was self-sustaining.

Finally, NVIDIA has clarified the branding a bit. Despite GeForce.com labeling it "the world’s ultimate graphics card," NVIDIA this morning has stated that the primary market is FP32 and INT8 compute, not gaming. Though gaming is certainly possible - and I fully expect they'll be happy to sell you $1200 gaming cards - the tables have essentially been flipped from the past Titan cards, where they were treated as gaming first and compute second. This of course opens the door to a proper GeForce branded GP102 card later on, possibly with neutered INT8 support to enforce the market segmentation.

POST A COMMENT

228 Comments

View All Comments

  • ImSpartacus - Friday, July 22, 2016 - link

    That's not a bad estimate, unfortunately...

    I'd aim for September though. No need to dive deep into the architecture. Just a bigger gp104.
    Reply
  • B3an - Friday, July 22, 2016 - link

    Titan X is nothing more than a great example of how much Nvidia are cunts.

    Instead of simply releasing the 1080 Ti, they first milk as much money out of idiots who will buy the Titan X, because these people are too impatient to wait for the PURPOSELY delayed 1080 Ti version - Which will no doubt offer the same performance with a slight overclock.

    None of these cards should exist in the first place though. They should just release the top-end cards first. Like what used to happen years ago (excluding dual-GPU cards). A 1080 Ti straight out of the gate at the price point of a 1080.

    But Nvidia just want to keep ripping off the consumer, and it will continue to happen until AMD can get their act together and make a good GPU for once that actually matches or beats Nvidia's best offering.

    It's exactly the same thing happening with Intel, because AMD are even worse at competing on CPU performance there, so you get these ridiculously priced Intel CPU's.

    So fuck AMD for incompetence, and fuck Nvidia for just being all-round anti-consumer pro-proprietary cunts.
    Reply
  • Beerfloat - Friday, July 22, 2016 - link

    Don't buy it, then. What's it to you that they run their business cleverly? Reply
  • damianrobertjones - Friday, July 22, 2016 - link

    Things like this NEED to be said. Reply
  • smorebuds - Friday, July 22, 2016 - link

    The market is what determines the price of this product, not Nvidia. If AMD can come out with something that even remotely competes with this, that's all it would take for Nvidia to price things more competitively.

    It's like complaining about BMW's being too expensive becase you can't afford one. There's a market for them, you're just not a part of it.
    Reply
  • fanofanand - Friday, July 22, 2016 - link

    Apparently early-adopters is a foreign phrase to B3an......you have ALWAYS paid a price premium to be among the first to grab any new tech. Reply
  • Murloc - Saturday, July 23, 2016 - link

    the same goes for books.

    Hard covers cost a fraction more to manufacture, but the books are much pricier.

    This is because the fans will jump on it ASAP even if they have to pay 30$.
    Then the other people who don't buy it at 30$ will buy the pocket edition for 15$ a few months later.
    Reply
  • althaz - Friday, July 22, 2016 - link

    If you choose not to buy nVidia, you are also choosing not to buy a high-end GPU.

    AMD haven't been competitive in this space for years.

    So I continue to think nVidia are a pack of dickheads - but I'll still buy their product, because it's the best. If AMD has caught up next time I need a graphics card (I just got my GTX1080), I'll look very hard at them.
    Reply
  • qwerty dvorak - Friday, July 22, 2016 - link

    What's with all this self-entitled whining? It's a freakin' super high-end gaming device - if you don't like the price, don't buy it. They are a business, they don't owe you anything. They do owe their shareholders something though, and that is to extract as much money as possible out of the market. Reply
  • Ahnilated - Friday, July 22, 2016 - link

    Actually they DO owe the customer a lot. Without the customer they would have no business or share holders to worry about. Reply

Log in

Don't have an account? Sign up now