NVIDIA Offering TITAN RTX With 20% Educational Discount

Posted on May 18, 2019 12:17 PM by Rob Williams

At NVIDIA’s GTC in March, it kind of felt like price-drops were in the air. At the event itself, the TITAN RTX could be found on the floor for 20% off of its original price, representing a $500 drop. Ahead of the event, NVIDIA quietly price dropped some of its top Quadro RTXs, as well. The 48GB RTX 8000 dropped from $10,000 to $5,500, and the 24GB RTX 6000 dropped from $6,300 to $4,000.

This month, the price-drop love continues with the TITAN RTX, but it will only impact those with educational credentials. Offering the same 20% discount we saw at GTC, NVIDIA is hoping that those wanting to accelerate their data science work will be more apt to jump on board, and while $2,000 is still a good chunk of cash for a graphics card, this is the fastest one NVIDIA sells, and it also includes a framebuffer more than twice the size of the top gaming option (the 11GB RTX 2080 Ti).

Here’s the current and last-gen Quadro and TITAN stack to better highlight where the TITAN RTX fits into place:

	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	260W	$10,000
RTX 6000	4608	1770	16.3 TFLOPS	24 GB ⁵	624 GB/s	295W	$6,300
RTX 5000	3072	1770	11.2 TFLOPS	16 GB ⁵	448 GB/s	265W	$2,300
RTX 4000	2304	1005	7.1 TFLOPS	8 GB ¹	416 GB/s	160W	$900
TITAN RTX	4608	1770	16.3 TFLOPS	24GB ¹	672 GB/s	280W	$2,499
TITAN V	5120	1200	14.9 TFLOPS	12 GB ⁴	653 GB/s	250W	$2,999
TITAN Xp	3840	1480	12.1 TFLOPS	12GB ²	548 GB/s	250W	$1,199
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	¹ GDDR6; ² GDDR5X; ³ GDDR5; ⁴ HBM2 ⁵ GDDR6 (ECC); ⁶ GDDR5X (ECC); ⁷ GDDR5 (ECC); ⁸ HBM2 (ECC) Architecture: P = Pascal; V = Volta; RTX = Turing

Compared to the current Quadro line, the TITAN RTX is closest in specs to the Quadro RTX 6000, sharing the same number of cores and framebuffer size. While both cards would largely perform the same in completely neutral tests (as in, no workstation optimizations), the TITAN RTX actually gets a slight edge with a boost from 624 GB/s memory bandwidth up to 672 GB/s.

Here’s another angle:

	NVIDIA’s Quadro & TITAN – RTX Performance
	RT Cores	RTX-OPS	Rays Cast ¹	FP16 ²	INT8 ³	Deep-learning ²
TITAN RTX	72	84 T	11	32.6	206.1	130.5
RTX 8000	72	84 T	10	32.6	206.1	130.5
RTX 6000	72	84 T	10	32.6	206.1	130.5
RTX 5000	48	62 T	8	22.3	178.4	89.2
RTX 4000	36	43 T	6	14.2	28.5	57
Notes	¹ Giga Rays/s; ² TFLOPS; ³ TOPS

Official “RTX-OPS” specs put the RTX 6000 and TITAN RTX into the same performance box, but TITAN RTX seems to gain slightly with Giga Rays/s performance, likely thanks to higher boost clocks. Beyond that, other specs are shared, including deep-learning. Unlike the Volta-based TITAN V, the TITAN RTX does not have unlocked double-precision, but that would be of no consequence for the markets NVIDIA is going after with this product.

We’ve had a TITAN RTX in the lab for a month, and we’re a bit overdue on publishing our review – but it’s coming. We recently benchmarked enough hardware to produce 109 individual charts, so we’re now just waiting for the words to hit the paper so we can publish them. Stay tuned.

If interested in taking advantage of this TITAN RTX educational discount, click the relevant link on this page to register.

Rob Williams

Rob founded Techgage in 2005 to be an 'Advocate of the consumer', focusing on fair reviews and keeping people apprised of news in the tech world. Catering to both enthusiasts and businesses alike; from desktop gaming to professional workstations, and all the supporting software.