We never touch anything that came from a Red, for instance, or from dSLRs - all our work is DPX or ProRes, from film scans. Having a few different source formats to use would be great. When the test structure is defined it won't be a problem getting the source clips we need. I think with the combined experience here, a small series of test should be possible to define… as long as egos don't get in the way of reaching a consensus.

I'd welcome a 'representative' test just to find out where the bar is at. While such a suite should be mainly test that stress to the max, I know that I will never use 66 blur nodes anytime soon. The purpose with this test would be to check if a certain computer or configuration can be expected to work in light to medium scenarios.

I'm thinking something like Juan's 'Blockbuster tutorial' setup. It could also include a 'threshold' test with a 4k source clip that had a pretty standard, but not over the top, node structure. The current SC is almost completely gfx limited, no? We could have tests that layer source footage to also put pressure on hard drives and raw footage that need real time decoding that tax the CPU(s). I don't think we need too many source clips, as they can be reused in various tests.

We could make it a small suite of 4-5 tests that tests different components of a well configured workstation. How do you guys feel about starting a discussion about the creation of a new 'Standard Candle'.

Davinci resolve 12.5.1 download free