After a long trail of teasers, NVIDIA today debuted a trio of new graphics cards: the RTX 2060 Super, RTX 2070 Super and the RTX 2080 Super.
The super branding is more like the old Ti branding which NVIDIA used in the past when it comes to these new cards. Let’s take a look at the new features of the cards and how they have changed from their older brothers and sisters.
RTX 2060 Super
The RTX 2060 Super leads off the new updated cards at $399 This is $50 more than the regular RTX 2060, which will still be sold, and packs specs more like the old RTX 2070.
The most important change over the RTX 2060 is on the memory front. The 2060 Super packs s8GB, vs the RTX 2050 6GB. Plus it sports an improved 256-bit memory interface, this is an increase of 33-percent to 448 GB/sec over the RTX 2060’s 192-bit memory interface.
There is also a raw TFLOPs improvement of 7.2 at FP32 vs 6.5. All updates should give owners a 15 percent of performance over the base model. NVIDIA says this puts the RTX 2060 Super almost on par with the original RTX 2070 and beats GTX 1080 performance of the last generation.
On the ray-tracing front, NVIDIA claims an increase from 5 Giga Rays to 6 Giga Rays. But they did not share how they got to those numbers.
For all the updates, the new card does increase the power draw: NVIDIA says the Super card will pull a peak of 175 watts vs. 160W power envelope of the RTX 2060.
RTX 2070 Super
The RTX 2070 Super is priced the same $499 as the original RTX 2070, which is being phased out of the product line.
This card looks like a 2080 with some part of the chip deactivated. You again get an extra four SMs (streaming multiprocessor), with 256 CUDA cores (up to 2,560 total) and 16 more texture units (up to 160) over the first RTX 2070
NVIDIA claims the RTX 2070 Super is 16% faster than the standard 2070 on average, which has it outperforming the GTX 1080 Ti, but still behind the original RTX 2080. Ray tracking of the RTX 2070 Super edition is 7 Giga Ray” vs. the original’s six.
Like the RTX 2060 super, the RTX 2070 Super peak power draw has increased over the first variant. it is now 215 Watts, up 40W from the RTX 2070 175W.
RTX 2080 Super
The RTX 2080 Super will replace the first RTX 2080 and costs the same at $699 .
The new version adds two extra SMs over the original – so 128more CUDA cores, 16 Tensor cores, eight texture units and two RT cores
The 2080 Super’s memory data rate has risen 11% from 14Gbps to 15.5Gbps, increasing total memory bandwidth to 496.1 GB/s. from the original card’s 448 GB/s.
The new card is rated at 11.1 TFLOPs, up from 10 TFLOP of the RTX 2080 base version. Many third party versions of the RTX 2080 beat the base cards number with their own tuning of the chipset.
NVIDIA says the new card is “faster than the Titan Xp but testing need to be done with games to verify that. Per the specs 12.1TFLOPS for the Xp vs. 11.1TFLOPS for the 2080 Super, but The XP is based on the the last gen Pascal architecture, so could be an app by app fight between the two cards.
NVIDIA is quoting the same “8 Giga Rays per second” figure for both the original and new 2080. We here at ToT are not fully sold on the Giga Rays numbers in the first place.
Again the power draw also goes up from 215 Watts on the RTX 2080 to 250W on the new Super version.
Specs Recap
With the all the cards above there are two DisplayPort 1.4 interfaces, HDMI 2.0b, and the rare DVI-DL port. Plus there is a VirtualLink over a USB Type-C connector – even on The RTX 2060 Super. Plus, you can use all four display outputs simultaneously for multi-monitor setups.
Until we do testing on these new cards, they are look like midcycle performance bumps coming at a time to take some thunder AMD’s away from AMD Navi GPU launch on the 7th.
| GeForce RTX 2060 Super | GeForce RTX 2070 Super |
GeForce RTX 2080 Super | |
| Architecture (GPU) |
Turing (TU106) | Turing (TU104) | Turing (TU104) |
| CUDA Cores |
2176 | 2560 | 3072 |
| Peak FP32 Compute |
7.2 TLFOPS | 9.1 TFLOPS | 11.2 TFLOPS |
| Tensor Cores |
272 | 320 | 384 |
| RT Cores |
34 | 40 | 48 |
| Texture Units |
136 | 160 | 192 |
| Base Clock Rate |
1470 MHz | 1605 MHz | 1650 MHz |
| GPU Boost Rate |
1650 MHz | 1770 MHz | 1815 MHz |
| Memory Capacity |
8GB GDDR6 | 8GB GDDR6 | 8GB GDDR6 |
| Memory Bus |
256-bit | 256-bit | 256-bit |
| Memory Bandwidth |
448 GB/s | 448 GB/s | 496 GB/s |
| ROPs |
64 | 64 | 64 |
| L2 Cache |
4MB | 4MB | 4MB |
| TDP |
175W | 215W | 250W |
| Transistor Count |
10.8 billion | 13.6 billion | 13.6 billion |
| Die Size |
445 mm² | 545 mm² | 545 mm² |
| SLI Support |
No | Yes | Yes |
The RTX Card Lineup As Of Now
| Old GeForce RTX Line-Up |
Availability |
New GeForce RTX Line-Up |
Availability |
| GeForce RTX 2080 Ti FE – $1200 GeForce RTX 2080 Ti Partners – $1000 |
Now | GeForce RTX 2080 Ti FE – $1200 GeForce RTX 2080 Ti Partners – $1000 |
Now |
| GeForce RTX 2080 FE – $800 GeForce RTX 2080 Partners – $700 |
Out of Production | GeForce RTX 2080 Super – $700 | 7/23 |
| GeForce RTX 2070 FE – $600 GeForce RTX 2070 Partners – $500 |
Out of Production | GeForce RTX 2070 Super – $500 | 7/9 |
| GeForce RTX 2060 Super – $400 | 7/9 | ||
| GeForce RTX 2060 FE – $350 GeForce RTX 2060 Partners – $350 |
Now | GeForce RTX 2060 – $350 | Now |
