Home / Tech News / Featured Tech Reviews / Nvidia RTX 2080 Ti Founders Edition 11GB Review

Nvidia RTX 2080 Ti Founders Edition 11GB Review

Rating: 8.0.

Following on from the huge success that was the Pascal architecture, today marks the release of Nvidia's Turing architecture into the wild. We have reviews of both the RTX 2080 and RTX 2080 Ti ready for launch – but this particular review focuses on the £1099 behemoth that is the RTX 2080 Ti.

Nvidia has certainly gone with a different pricing strategy with Turing – the GTX 1080 Ti costs £669 , when its supposed successor will set you back an extra £430 at launch. That's an effective 64% generation price hike – so what are you getting for your money?

Foremost among the appeal of the RTX 2080 Ti is the ray tracing abilities. As you may already be aware, we cannot yet test this feature out in actual games – we are still waiting for Microsoft's DirectX Raytracing (DXR) update that is slated for next month. We do have the results from one ray tracing demo, presented later in this review, while of course there are the usual slew of benchmarks as well.

Watch via Vimeo (below) or over on YouTube at 2160p HERE

 GPU RTX 2080 Ti (FE)  GTX 1080 Ti (FE)  RTX 2080 (FE)  GTX 1080 (FE) 
 SMs  68 28  46  20
CUDA Cores   4352  3584 2944 2560
Tensor Cores   544  N/A 368 N/A
 Tensor FLOPS  114  N/A  85 N/A
 RT Cores  68  N/A 46 N/A
 Texture Units 272  224 184 160
 ROPs  88  88 64 64
Rays Cast   10 Giga Rays/sec  1.1 Giga Rays/sec 8 Giga Rays/sec 0.877 Giga Rays/sec
RTX Performance   87 Trillion RTX-OPS  11.3 Trillion RTX-OPS 60 Trillion RTX-OPS  8.9 Trillion RTX-OPS
GPU Boost Clock   1635 MHz  1582 MHz 1800 MHz  1733 MHz
 Memory Clock  7000 MHz  5505 MHz 7000 MHz  5005 MHz
 Total Video Memory  11GB GDDR6  11GB GDDR5X  8GB GDDR6  8GB GDDR5X
Memory Interface   352-bit  352-bit  256-bit  256-bit
 Memory Bandwidth  616 GB/sec  484 GB/sec  448 GB/sec  320 GB/sec
 TDP  260W  250W  225W  180W

Note: GPU-Z has not yet been updated to show all of the RTX 20 series specs.

On paper, the Nvidia RTX 2080 Ti is nothing short of a spec monster. Not only does it sport 4352 CUDA cores – up from 3584 in the GTX 1080 Ti – but architectural improvements (detailed on the next page) mean these cores are even more effective than previous generations. The new GPU is also fabricated on a 12nm process, where Pascal used the 16nm process.

68 RT cores and 544 Tensor Cores also single this card out as a new GPU – Pascal does not have any such cores, and they were only first introduced with Volta's GV100.

In terms of clock speed, we can expect lower speeds than the RTX 2080, which benefits in that regard thanks to its lesser core count, but we can still expect the RTX 2080 Ti to exceed its rated 1635MHz clock thanks to GPU Boost 4.0.

One thing that remains consistent from the GTX 1080 Ti is the amount of VRAM, at 11GB. The RTX 2080 Ti, however, uses GDDR6 memory over a 352-bit bus, with memory bandwidth totalling 616 GB/s.

Become a Patron!

Be sure to check out our sponsors store EKWB here

Check Also

Nvidia GeForce GTX 1650 is now the most popular GPU on Steam

Because of its price-performance ratio at launch, the GTX 1060 quickly became a favourite for …