Cast your vote Do you agree or disagree with GPUBoss?

Thanks for adding your opinion. Follow us on Facebook to stay up to date with the latest news!

Differences What are the advantages of each

Front view of Radeon RX Vega 56

Reasons to consider the
AMD Radeon RX Vega 56

Report a correction
Significantly higher memory bandwidth 409.6 GB/s vs 256.3 GB/s Around 60% higher memory bandwidth
Significantly better floating-point performance 10,544 GFLOPS vs 5,783 GFLOPS More than 80% better floating-point performance
Significantly higher texture rate 329.5 GTexel/s vs 180.7 GTexel/s More than 80% higher texture rate
Significantly more shading units 3,584 vs 1,920 1664 more shading units
Significantly more texture mapping units 224 vs 120 104 more texture mapping units
Much wider memory bus 2,048 bit vs 256 bit 8x wider memory bus
Front view of GeForce GTX 1070

Reasons to consider the
Generic GeForce GTX 1070

Report a correction
Much higher effective memory clock speed 8,008 MHz vs 1,600 MHz More than 5x higher effective memory clock speed
Higher clock speed 1,506 MHz vs 1,156 MHz More than 30% higher clock speed
Much higher memory clock speed 2,002 MHz vs 800 MHz More than 2.5x higher memory clock speed
Higher turbo clock speed 1,683 MHz vs 1,471 MHz Around 15% higher turbo clock speed
Lower TDP 150W vs 210W Around 30% lower TDP

Features Key features of the Radeon RX Vega 56  vs GeForce GTX 1070 

memory bandwidth Rate at which data can be read from or stored in onboard memory

Radeon RX Vega 56
409.6 GB/s
GeForce GTX 1070
256.3 GB/s

pixel rate Number of pixels a graphics card can render to the screen every second

Radeon RX Vega 56
94.14 GPixel/s
GeForce GTX 1070
96.4 GPixel/s

texture rate Speed at which a graphics card can perform texture mapping

Radeon RX Vega 56
329.5 GTexel/s
GeForce GTX 1070
180.7 GTexel/s

floating point performance How fast the gpu can crunch numbers

Radeon RX Vega 56
10,544 GFLOPS
GeForce GTX 1070
5,783 GFLOPS

shading units Subcomponents of the gpu, these run in parallel to enable fast pixel shading

texture mapping units Built into each gpu, these resize and rotate bitmaps for texturing scenes

Specifications Full list of technical specs


Radeon RX Vega 56  vs
GeForce GTX 1070 
GPU brand AMD Nvidia
GPU name Vega 10 GP104
Clock speed 1,156 MHz 1,506 MHz
Turbo clock speed 1,471 MHz 1,683 MHz
Is dual GPU No No
Reference card None Nvidia GeForce GTX 1070 1.5 GHz 8 GB

raw performance

Shading units 3,584 1,920
Texture mapping units 224 120
Render output processors 64 64
Pixel rate 94.14 GPixel/s 96.4 GPixel/s
Texture rate 329.5 GTexel/s 180.7 GTexel/s
Floating-point performance 10,544 GFLOPS 5,783 GFLOPS


Radeon RX Vega 56  vs
GeForce GTX 1070 
Memory clock speed 800 MHz 2,002 MHz
Effective memory clock speed 1,600 MHz 8,008 MHz
Memory bus 2,048 bit 256 bit
Memory 8,192 MB 8,192 MB
Memory bandwidth 409.6 GB/s 256.3 GB/s

noise and power

TDP 210W 150W


Showing 7 comments.
Why tests vega56 vs gtx1080ti? It is very interesting to compare them )
You know that these are benched cards from before proper drivers. Technically prerelease versions. Specs can change before release or within manufacturing quarter. Memory software sucks so compression would slow a card down Schwarzenbach. Nvidia usually uses faster memory than their target card. Cooling efficiency is not about looks, it is materials used in the heat sink like copper transfer plate to high grade aluminum would be best, but the aluminum would be too costly. Most heat-sinks are made of mid to high-mid grade aluminum. The exterior cover is to direct air through the aluminum fins, slight ridges in the aluminum fins would help cool better than most flat fins too. AMD is very slow at maturing drivers, but usually makes the chips age far slower than Nvidia drivers or chips. Will be interesting as the R9 chip drivers are still not fully mature. Somewhat like Microsoft does with Windows without dropping support till fully matured. Nvidia makes too many design changes to do anything close without screwing up some added feature in their newer cards. Over all, who cares as long as it does what you want. Both companies make great cards. Nvidia has the speed and uses cheap and easy programmable stencil filler instead of a more realistic cut pixelation generator that is harder to program fast. These differences show up when blowing up the frame to max then zooming it in. Ether the Nvidia stops short or start looking pretty bad in comparison to AMD Radeon of the closest generation and value point. Probably the reason AMD takes much longer to mature as well. Sometimes the programmers look for stability over performance when there is a deadline. Just something to think about.
Are you stupid or what? Vega has a 2048 bit memory bus, the 1070 256 bit 409.6 GB/s vs 256.3 GB/s so vega has twice the effctive memory speed.
Average over 30 games the Vega 56 is the clear winner, doesn't matter which one is better for DX11 or DX12 outside of "future proofing" in which case vega 56 will only get better.
nvidia has way better memory compression, meaning higher effective memory bandwidth. AMD needs to step up their memory compression and we'd see the pixel rate skyrocket on the Vega 56
The stock cooler for nvidia's cards are way better which allow better thermals and performance. It will be interesting to see how much of a jump vega 56 gets once AIB cards start to show up. May even make the Vega 64 pointless.
Funny, reading this comparison it seems like the Vega 56 should completely destroy the GTX 1070. However when checking actual game benchmarks they are fairly similar. The Vega 56 might gain a lead once the drivers get better optimized. At the moment the GTX 1070 has the edge in DX11 titles while the Vega 56 gets the upper hand in DX12 and Vulkan.
comments powered by Disqus