New Generation, New Founders Edition

Performance for Turing Unveiled!

At this point, it seems that calling NVIDIA's 20-series GPUs highly anticipated would be a bit of an understatement. Between months and months of speculation about what these new GPUs would be called, what architecture they would be based off, and what features they would bring, the NVIDIA GeForce RTX 2080 and RTX 2080 Ti were officially unveiled in August, alongside the Turing architecture.

We've already posted our deep dive into the Turing architecture and the TU 102 and TU 104 GPUs powering these new graphics cards, but here's a short take away. Turing provides efficiency improvements in both memory and shader performance, as well as adds additional specialized hardware to accelerate both deep learning (Tensor cores), and enable real-time ray tracing (RT cores).

  RTX 2080 Ti Quadro RTX 6000 GTX 1080 Ti RTX 2080  Quadro RTX 5000 GTX 1080 TITAN V RX Vega 64 (Air)
GPU TU102 TU102 GP102 TU104 TU104 GP104 GV100 Vega 64
GPU Cores 4352 4608 3584 2944 3072 2560 5120 4096
Base Clock 1350 MHz 1455 MHz 1408 MHz 1515 MHz 1620 MHz 1607 MHz 1200 MHz 1247 MHz
Boost Clock 1545 MHz/
1635 MHz (FE)		 1770 MHz 1582 MHz 1710 MHz/
1800 MHz (FE)		 1820 MHz 1733 MHz 1455 MHz 1546 MHz
Texture Units 272 288 224 184 192 160 320 256
ROP Units 88 96 88 64 64 64 96 64
Tensor Cores 544 576 — 368 384 — 640 —
Ray Tracing Speed 10 GRays/s 10 GRays/s — 8 GRays/s 8 GRays/s — — —
Memory 11GB 24GB 11GB 8GB 16GB 8GB 12GB  8GB
Memory Clock 14000 MHz  14000 MHz  11000 MHz 14000 MHz  14000 MHz  10000 MHz 1700 MHz 1890 MHz
Memory Interface 352-bit G6 384-bit G6 352-bit G5X 256-bit G6 256-bit G6 256-bit G5X 3072-bit HBM2 2048-bit HBM2
Memory Bandwidth 616GB/s 672GB/s 484 GB/s 448 GB/s 448 GB/s 320 GB/s 653 GB/s 484 GB/s
TDP 250 W/
260 W (FE)		 260 W 250 watts 215W
225W (FE)		 230 W 180 watts 250W 292
Peak Compute (FP32) 13.4 TFLOPS / 14.2 TFLOP (FE) 16.3 TFLOPS 10.6 TFLOPS 10 TFLOPS / 10.6 TFLOPS (FE) 11.2 TFLOPS 8.2 TFLOPS 14.9 TFLOPS 13.7 TFLOPS
Transistor Count 18.6 B 18.6B 12.0 B 13.6 B 13.6 B 7.2 B 21.0 B 12.5 B
Process Tech 12nm 12nm 16nm 12nm 12nm 16nm 12nm 14nm
MSRP (current) $1200 (FE)/
$1000		 $6,300 $699 $800/
$700		 $2,300 $549 $2,999 $499

 

As unusual as it is for them NVIDIA has decided to release both the RTX 2080 and RTX 2080 Ti at the same time, as the first products in the Turing family. 

The TU102-based RTX 2080 Ti features 4352 CUDA cores, while the TU104-based RTX 2080 features 2944, less than the GTX 1080 Ti. Also, these new RTX GPUs have moved to GDDR6 from the GDDR5X we found on the GTX 10-series.

dsc05175.jpg

One of the most significant departures with the RTX 2080 and RTX 2080 Ti can be found in the newly redesigned NVIDIA Founders Edition products.

dsc05176.jpg

Finally moving away from the blower-style cooler, the Founders Edition cards now feature a dual axial fan design, along with a vapor chamber that's the entire length of the card, as well as a substantially redesigned look.

Utilizing this new found cooling capacity, the Founders Editions cards will also be shipped in a factory overclocked state for the first time, 90MHz for both SKUs.

Since NVIDIA is generally the only person that ships their graphics cards at reference clock speeds, it's difficult to predict if we'll see any RTX cards clocked lower than the Founders Editions, but NVIDIA is claiming that these cards are in fact "overclocked." 

dsc05180.jpg

Another addition to the Founders Edition cards and all of the third party RTX cards we've seen so far is the addition of a USB-C connector. This USB-C connector is VirtualLink compliant, a standard developed through coordination with NVIDIA, Oculus, Valve, Microsoft, and AMD to provide a one-cable solution for next-generation VR headsets.

The VirtualLink port is capable of providing 4 Lanes of HBR3 DisplayPort, USB 3.1 Gen 2 connectivity, as well as 27W of power delivery.

dsc05178.jpg

The RTX 2080 features an 8-pin and a 6-pin power connector, while the RTX 2080 Ti moves to two 8-pin connectors.

dsc05187.jpg

The other funky new connector on the RTX cards is NVLink. This communication protocol, previously seen in NVIDIA's Tesla and Quadro products provides a high-bandwidth replacement for SLI and will enable resolutions up to 8K (Single Link on the RTX 2080), and 8K Surround (Dual Link on the RTX 2080 Ti). However, NVLink is only compatible with 2 GPUs, officially ending the days of 3 and 4-way SLI, even if Pascal only featured support in select benchmarks.

Unlike the launch of the Pascal-based GTX 10-series cards, graphics card designs from third-party manufacturers such as MSI, EVGA, and ASUS will be ready and shipping on the same launch date as the Founders Edition (September 20th).

dsc05184.jpg

Here's just a small taste of what's in store for the coming weeks as we take a look at these new third-party designs.

Review Terms and Disclosure
All Information as of the Date of Publication
How product was obtained: The product is on loan from NVIDIA for the purpose of this review.
What happens to the product after review: The product remains the property of NVIDIA but is on extended loan for future testing and product comparisons.
Company involvement: NVIDIA had no control over the content of the review and was not consulted prior to publication.
PC Perspective Compensation: Neither PC Perspective nor any of its staff were paid or compensated in any way by NVIDIA for this review.
Advertising Disclosure: NVIDIA has purchased advertising at PC Perspective during the past twelve months.
Affiliate links: This article contains affiliate links to online retailers. PC Perspective may receive compensation for purchases through those links.
Consulting Disclosure: NVIDIA is not a current client of Shrout Research for products or services related to this review. 
« PreviousNext »