NVIDIA Unveils Next-Gen Turing Quadro RTX Professional Graphics Cards (hothardware.com) 56
MojoKid shares a report from Hot Hardware: We been hearing a lot about NVIDA's next-generation GPU architecture since late last year, and today NVIDIA is announcing the first products based on Turing. NVIDIA is targeting the professional graphics market first with its new Quadro RTX 8000, RTX 6000 and RTX 5000 GPUs. Turing GPU architecture replaces Pascal, which has served both the consumer and professional markets since 2016. But as its 8th generation GPU architecture, NVIDIA is ushering in a number of advances with Turing and it's billed as the "world's first ray-tracing GPU." When it comes to content creators, NVIDIA claims that with the power of Turing, "applications can simulate the physical world at 6x the speed of the previous Pascal generation."
Getting down to brass tacks, the entry-level Quadro RTX 5000 has 3,072 CUDA cores, 384 Tensor cores, and will come with 16GB of 14Gbps GDDR6 memory. Its ray-tracing performance is dialed in at 6 GigaRays/sec, according to NVIDIA. Both the Quadro RTX 6000 and RTX 8000 have 4,608 CUDA cores and 576 Tensor cores; the only difference between the two is that the former has 24GB of GDDR6, while the latter doubles that to 48GB. Ray-tracing performance for both of these GPUs tops out at 10 GigaRays/sec. NVIDIA is also claiming up to 16TFLOPs compute performance for the Quadro RTX 8000. NVIDIA's new Quadro GPUs will also be among the first to support both USB-C and VirtualLink for next-generation virtual reality and mixed reality headsets. Other VirtualLink backers include AMD, Oculus, Microsoft and Valve. The Quadro RTX 5000, RTX 6000 and RTX 8000 will all be available during the fourth quarter of 2018 priced at $2,300, $6,300 and $10,000 respectively.
Getting down to brass tacks, the entry-level Quadro RTX 5000 has 3,072 CUDA cores, 384 Tensor cores, and will come with 16GB of 14Gbps GDDR6 memory. Its ray-tracing performance is dialed in at 6 GigaRays/sec, according to NVIDIA. Both the Quadro RTX 6000 and RTX 8000 have 4,608 CUDA cores and 576 Tensor cores; the only difference between the two is that the former has 24GB of GDDR6, while the latter doubles that to 48GB. Ray-tracing performance for both of these GPUs tops out at 10 GigaRays/sec. NVIDIA is also claiming up to 16TFLOPs compute performance for the Quadro RTX 8000. NVIDIA's new Quadro GPUs will also be among the first to support both USB-C and VirtualLink for next-generation virtual reality and mixed reality headsets. Other VirtualLink backers include AMD, Oculus, Microsoft and Valve. The Quadro RTX 5000, RTX 6000 and RTX 8000 will all be available during the fourth quarter of 2018 priced at $2,300, $6,300 and $10,000 respectively.
math is hard (Score:5, Funny)
24 * 2 = ?
Re: (Score:2)
Yes, the /. summary is taken straight from the TFA.
So Brandon Hill, MojoKid and BeauHD - please pay a bit more attention.
Re: (Score:1)
Re: (Score:2)
Re: (Score:2)
Hothardware, say not more. The amusing part is they are always behind yet can't even properly copy other peoples published specs. it is 48GB not 96.
Side note: "say not more" ... Grammar and spelling - like gravity, such a bitch.
Re: (Score:2)
Re:math is hard (Score:5, Funny)
Well, that gives you an idea of how this card generation got its performances boost.
Re: (Score:2)
Re: (Score:1)
It's because companies are forced to hire Negroes and other unqualified individuals as part of social engineering experiments dreamed up by SJWs, who they they themselves are unqualified. The blind leading the blind. Negroes are the dead weight that the rest of society has to carry on its back.
Judging by your last two presidents, it's not the blacks that you should be blaming, its the oranges.
Re: (Score:2)
And here I am, out of mod points.
Virtual +1 Funny to you, sir.
Re: (Score:2)
Re: (Score:2)
24 * 2 = ?
I'm guessing it's more like Wayside School math. GDDR6 is the latter for some odd reason, so GDDR6 * 2 = 96GB.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
That's right up there with the names of video standards:
240p - SD
480p - SD
720p - HD
1080p - Full HD ( *should* technically be "2k" referring to the total amount of horizontal pixels - 1920px. )
2160p - 4K Ultra HD
4320p - 8k Ultra HD
Nevermind the fact that the human eye with 20/20 vision couldn't differentiate the pixels of a 65" 1080p TV from 10 feet away, but I digress.
2x24=96 (Score:2)
So does someone need a typing or a maths class?
Seems like a lot of compute for the money (Score:2)
If only I had an application that would be tractable on this sort of hardware. The ray tracing side of things is boring to me, i'd be interested to know how this works out for large matrix operations, or signal processing type applications. Still, more is more when it comes to compute!
Re:Seems like a lot of compute for the money (Score:4, Interesting)
big marix math is something CUDA does well on account of being able to parallelize large amounts of simple operations which is ultimately hosw matrices work , just stepwise working through the steps.
https://www.quantstart.com/art... [quantstart.com]
As for signal processing in general, mostly is a pretty great use, however if its real time audio, there has been traditionally issues with latency shifting blocks in and out of GPU ram. Video for whatever reason is a bit better because the discrete timing between frames are large enough that latency isn't a problem, but audio is continuous so a degree of extra buffering needs to happen. But if I was calculating messy convolutions, or something that might make the CPU sweat, the GPU would still be my first port of call
Re: (Score:1)
Re: (Score:2)
Re: (Score:1)
The pro cards (Quadro, and also some NVS cards) nowadays come with four DisplayPort outputs. The consumer cards tend to have one DisplayPort and a mishmash of HDMI and DVI. That is annoying, since there are converters from DisplayPort to both of these but not the other way round. The GeForce cards also tend to be dual-slot, power-hungry monsters while the lower-end Quadro cards fit in a single slot and don't need auxiliary power.
So I think there are reasons to prefer the Quadro line, apart from the quest
Re: (Score:2)
Ah, I see your mistake here. I bet you've been buying the latest, younger cards instead of the older, more mature ones. My GTX 650 has always been quiet, on time and never gave me any trouble.
Re: (Score:2)
Yes. You run a week long simulation only to find an error caused by unstable hardware.
Re: (Score:2)
I don't believe that Google gets a licensing fee on common math. Yet.
https://en.wikipedia.org/wiki/... [wikipedia.org]
It Figures... (Score:2)
Crap, I just bought a P4000.
Re: (Score:2)
Similar thing here, damnit. I just bought the Bates 4000!
96GB is 48GB (Score:2, Informative)
Note that if you do RTFA, it's still badly explained, wrongly written but you get 48GB on the top end Quadro (not in the article : it's up from the older 24GB on GDDR5, because GDDR6 is available with twice the density)
If you have two Quadro RTX 8000 linked with a bridge, you get a real and genuine 96GB memory space in the same way it happens on modern dual CPU systems. Coherent, seamless but bandwidth is much slower if you go over the border of course. This feature was already available on Quadro GP100 (Pa
We been? (Score:2)