The AMD Radeon RX 5700 XT and RX 5700 Review: Navi Renews Competition
anandtech.com> AMD is also launching their own “Anniversary Edition” version of the 5700 XT, which features a factory overclock
What does factory overclock even mean? Doesn't overclocked mean clocked over what you get from the factory? Wouldn't "factory overclock" just be "higher clock"? Are you paying $50 extra for someone else to push a button for you?
If it’s like CPUs, they’re using units from better performance bins.
If you buy a regular version, it might be stable at this same overclock on stock cooling, but it also might not. And even if it does work, it’s possible the better binned units have headroom to go even higher.
Generally "factory overclock" refers to board partners shipping GPUs at higher clocks than specified by the GPU vendor. i.e. I have a Vega 56 card which AMD specs at 1474Mhz but the default boost clock on mine was 1596Mhz.
The term is rather silly when the card is coming directly from AMD though.
It is important to Note, both the 2060 Super and RX5700 has similar transistor count, ( 10.8B vs 10.3B ). So RX5700 isn't winning because it has new node and stuffing in more transistor.
I am wondering on the state of GPU Compute on Mac, CUDA, OpenCL, ROCm. Apple has now abandoned OpenCL and working on their Metal Compute, which is ( I think ) only available on iOS. CUDA is not available on Mac, and ROCm is only available on Linux.
> RX5700 isn't winning because it has new node and stuffing in more transistor
Sure, I guess, but it is probably still winning because it is manufactured on a new node - transistor count isn't the only thing that matters.
Performance is frequently power-limited, so because 7nm is more efficient it can achieve higher performance.
If you took the same design from 16nm (or 28, ...) and simply shrunk it to 7nm, you could achieve higher performance with the 7nm version - even though the number of transistors hasn't changed.
Metal compute is available on macOS as well. As far as I can tell, it’s been available since at least 10.11.
Thanks I was not aware of that. So Intel and AMD has to write drivers specifically for it?
You can use Intel PlaidML right now on OSX to utilize your metal supported GPU for machine learning. I have tried the python library with Keras and my GPU (Geforce 760) with no issues.
That is good news. On the other hand compared to Zen2 the power usage is kind of disappointing. Navi is taking similar or even significantly higher power while using 7nm vs. nvidia 12nm.
Why there is a dimple on a fan shroud of XT model, and no dimple on a baseline model?
I imagine the outer packaging is reflecting what you paid for, much like how a performance trim on the engine of a car would be accompanied by outer letters and extra styling.