AndreasOberg
Well-known member
- Joined
- Oct 30, 2011
- Messages
- 1,671
- Reaction score
- 31
- Points
- 48
- Location
- Leicestershire, United Kingdom
- Website
- www.obergwildlife.com
Hiya,
I have been reading up more about the GTX cards and thought I would share some thoughts.
So we currently have 2xTitan X cards. They cannot play 8k Red video realtime in Redcine X.
However I have tried with 2xGTX 1080 Ti cards and they can play in Redcine X.
So with the 2080 Ti I think 1 cards could be able to playback realtime with just one card. Probably good with two to help with Neat Video etc.
So about 2080 Ti vs 1080 Ti
First 4352 cuda cores vs 3584 which is 21% decrease
Memory Speed 14Gb/s vs 11GB/s so a 27% increase
Now most know the above but there is one information i recently found on Toms Hardware. And that is that for float 32 operations the previous generation could only do ONE operation at the time while the new can run them in parallell.
"To start, Turing borrows from the Volta playbook in its support for simultaneous execution of FP32 arithmetic instructions, which constitute most shader workloads, and INT32 operations (for addressing/fetching data, floating-point min/max, compare, etc.). When you hear about Turing cores achieving better performance than Pascal at a given clock rate, this capability largely explains why."
https://www.tomshardware.com/reviews/nvidia-turing-gpu-architecture-explored,5801-4.html
I am thinking that most operations regarding red files are float 32 values.
I plan on getting one card, and then maybe get one later if I need it for other operations.
/Andreas
I have been reading up more about the GTX cards and thought I would share some thoughts.
So we currently have 2xTitan X cards. They cannot play 8k Red video realtime in Redcine X.
However I have tried with 2xGTX 1080 Ti cards and they can play in Redcine X.
So with the 2080 Ti I think 1 cards could be able to playback realtime with just one card. Probably good with two to help with Neat Video etc.
So about 2080 Ti vs 1080 Ti
First 4352 cuda cores vs 3584 which is 21% decrease
Memory Speed 14Gb/s vs 11GB/s so a 27% increase
Now most know the above but there is one information i recently found on Toms Hardware. And that is that for float 32 operations the previous generation could only do ONE operation at the time while the new can run them in parallell.
"To start, Turing borrows from the Volta playbook in its support for simultaneous execution of FP32 arithmetic instructions, which constitute most shader workloads, and INT32 operations (for addressing/fetching data, floating-point min/max, compare, etc.). When you hear about Turing cores achieving better performance than Pascal at a given clock rate, this capability largely explains why."
https://www.tomshardware.com/reviews/nvidia-turing-gpu-architecture-explored,5801-4.html
I am thinking that most operations regarding red files are float 32 values.
I plan on getting one card, and then maybe get one later if I need it for other operations.
/Andreas