NVIDIA has taken advantage of the ongoing Autodesk University in Las Vegas to launch its fourth Turing-based Quadro: the 8GB RTX 4000. This is a GPU many have been undoubtedly waiting for, as the 4000 series Quadros typically offer some of the best bang for the buck – and in this continued case, a single-slot cooler.
At a price of $900 USD, the premium on this particular Quadro over the GeForce equivalent isn’t as high as normal. Under-the-hood, the RTX 4000 is spec’d similarly to a GeForce RTX 2070. Both have 2304 CUDA cores, 8GB of non-ECC memory, and deliver 42~43T RTX-OPS performance. Of course, the Quadro card will give you a lot more as a workstation user, including a countless number of workload-specific optimizations, and of course, enhanced product support.
With the RTX 4000, NVIDIA says it’s putting real-time ray tracing into the hands of many more developers, artists, and designers the world over. When RTX first launched, most of us wondered how much whittling down of the RT core there would be, but since we’re still at RTX 2070 level between the series, we’re going to have to wait and see if the even smaller Quadros snip something off.
Let’s take a quick look at NVIDIA’s updated lineup:
|
NVIDIA’s Quadro Workstation GPU Lineup |
|
Cores |
Base MHz |
Peak FP32 |
Memory |
Bandwidth |
TDP |
Price |
GV100 |
5120 |
1200 |
14.9 TFLOPS |
32 GB 8 |
870 GB/s |
185W |
$8,999 |
RTX 8000 |
4608 |
1440 |
16.3 TFLOPS |
48 GB 5 |
624 GB/s |
???W |
$10,000 |
RTX 6000 |
4608 |
1440 |
16.3 TFLOPS |
24 GB 5 |
624 GB/s |
295W |
$6,300 |
RTX 5000 |
3072 |
1350 |
11.2 TFLOPS |
16 GB 5 |
448 GB/s |
265W |
$2,300 |
RTX 4000 |
2304 |
??? |
7.1 TFLOPS |
8 GB 1 |
416 GB/s |
160W |
$900 |
TITAN V |
5120 |
1200 |
14.9 TFLOPS |
12 GB 4 |
653 GB/s |
250W |
$2,999 |
P6000 |
3840 |
1417 |
11.8 TFLOPS |
24 GB 6 |
432 GB/s |
250W |
$4,999 |
P5000 |
2560 |
1607 |
8.9 TFLOPS |
16 GB 6 |
288 GB/s |
180W |
$1,999 |
P4000 |
1792 |
1227 |
5.3 TFLOPS |
8 GB 3 |
243 GB/s |
105W |
$799 |
P2000 |
1024 |
1370 |
3.0 TFLOPS |
5 GB 3 |
140 GB/s |
75W |
$399 |
P1000 |
640 |
1354 |
1.9 TFLOPS |
4 GB 3 |
80 GB/s |
47W |
$299 |
P620 |
512 |
1354 |
1.4 TFLOPS |
2 GB 3 |
80 GB/s |
40W |
$199 |
P600 |
384 |
1354 |
1.2 TFLOPS |
2 GB 3 |
64 GB/s |
40W |
$179 |
P400 |
256 |
1070 |
0.6 TFLOPS |
2 GB 3 |
32 GB/s |
30W |
$139 |
Notes |
Doesn’t that RTX 4000 look crazy attractive with that price? Everything above it in the list is going to cost substantially more, so if your needs are really not that demanding, but you appreciate some real horsepower, then the RTX 4000 looks to be a great choice.
This reminds me a bit of when AMD released its Radeon Pro WX 7100. It was a card that cost a few hundred dollars more than the top gaming card at the time, but it allowed people to enter the workstation ecosystem without breaking the bank, or sacrificing real performance. The WX 7100 is still a very good card in many respects, but with RTX on tap, the new 4000-series card is going to be hard to resist. At $900, the RTX 4000 best competes with AMD’s $999 Radeon Pro WX 8200. If the opportunity ever arises, I’ll compare the two, since I already have one half of the equation.
|
NVIDIA’s Quadro RTX Performance |
|
RT Cores |
RTX-OPS |
Rays Cast |
FP16 1 |
INT8 2 |
Deep-learning 1 |
RTX 8000 |
72 |
84 T |
10 Giga Rays/s |
32.6 |
206.1 |
130.5 |
RTX 6000 |
72 |
84 T |
10 Giga Rays/s |
32.6 |
206.1 |
130.5 |
RTX 5000 |
48 |
62 T |
8 Giga Ray/s |
22.3 |
178.4 |
89.2 |
RTX 4000 |
36 |
43 T |
6 Giga Rays/s |
14.2 |
28.5 |
57 |
Notes |
We’ve been comparing GPUs the same way for ages, so NVIDIA decided to spice things up with Turing by giving us some new metrics to worry about. That includes “RTX-OPS”, which effectively represents the aggregate performance of the important performance aspects of the GPU. It’s still important to look at individual metrics, but you’re never going to be sacrificing anything when opting for a higher RTX-OPS option.
If the RTX 4000 sounds like the Quadro for you, you’ll have to wait until next month to snap one up. Interestingly, the RTX 5000 still shows up as a pre-order on NVIDIA’s own website, so this is a bit of a strange rollout if you ask me. The RTX 6000 is currently “sold out”, and the RTX 8000 is also shown as being a preorder. That said, this is only from NVIDIA’s own website; if you source a PNY card from a certified retailer, then you may have more luck.
Addendum: Post originally had some incorrect specs in both tables referencing RTX 5000.