Use of computer resources by NLEs

[Michael Gissing]

Personally I use PC and WIN OS and yes I have a reasonably powerful graphics card. Resolve is said to work better with NVIDIA cards but in terms of comparison on Macs, I have been told by other editors on Cow forums that FCPX is best on Mac followed closely by Resolve and Premier is coming last at this stage.

Bear in mind also that Resolve is doing incredible processing during Color grading. But in terms of just plain editing with the same codecs, it is not far behind FCPX which makes sense as FCPX is written for Mac hardware and has the home advantage. Resolve has improved so much in the past two version (15 & 16) that I have been able to postpone the upgrade of my GPU.

iMacs and their GPUs are under powered for editing 4k with modern codecs.

[Marc Wielage]

Jason Bowdach has a good webinar on how to optimize Resolve performance on different systems over on MixingLight.com:

https://mixinglight.com/color-tutorial/performance-optimization-in-davinci-resolve-webinar/

The reality is that 4K Raw is tough anywhere. Regular 4K is demanding. You can get it to work to a point by using Optimized Media and tweaking the Performance settings in Resolve. Another method would be to use proxy transcodes for the edit, and then finish on a more powerful system and relink back to the camera originals.

[Robert Withers]

Thanks everybody for the helpful posts. I’m not editing 4K.
You would think there would be hard data somewhere showing use of computer resources by NLEs, compared.
But maybe not.
Best,
Robert

Robert Withers

Independent/personal/avant-garde cinema, New York City

Currently experimenting with iMac 2019, Catalina
3.6 MHz 8-Core Intel Core i9
16 GB Memory
GPU Radeon Pro 575X (4 GB)

[Michael Gissing]

The problem with hard data is very few editors do proper benchmarking so there is a lot of anecdotal evidence and opinion. Typical benchmarks are not much use for actual edit scenarios plus all it takes is a single bottleneck to change performance. All the chain of data needs to be beyond capable of moving modern codecs from drive to CPU, GPU and RAM. A single stumble like insufficient GPU RAM can bring a powerful system crashing down as can hard drive speed.