Optimizing Intermediate Codecs for 4K 60fps Editing in Premiere Pro on Windows
Hey everyone,
For a while, our editing workflow has included editing multiple layers (tracks) of 4K 60fps footage using full-resolution Cineform files. Our editing setup includes a PC with a 14900KS processor, 96GB of high-performance RAM, an RTX 4090, and we're editing directly off of a NAS with 8 high-performance Seagate Exos drives in a RAID 5, along with SSD caching, over a 10Gbit network. Despite having a top-of-the-line setup and optimized Premiere Pro settings, we encounter significant lag as soon as there are a lot of edits.
To clarify, the lag becomes apparent after making a lot of cuts (probably in the hundreds or more), and this is even without any effects applied. When it gets to this point, playback becomes extremely sluggish. Hitting play often results in long delays before playback starts, and sometimes pressing the space bar results in no response at all from the application, leaving us unsure if the key press was registered or not. This makes the process of continuing the edit very frustrating and time-consuming.
Given our performance issues, I've been trying to optimize everything to the best of my ability to narrow down the source of the lag, and one thing I want to do is ensure that we are using the most performant intermediate codec possible to rule that out. Specifically, I am interested in any personal benchmarks or tests comparing timeline performance between Cineform, ProRes, and DNxHR in recent versions of Premiere Pro on Windows systems. Export speed is also appreciated, but the absolute best editing performance within the timeline is what I am most interested in, and storage space isn't a concern in this case. I realize that switching codecs may not fully resolve our issues, but I aim to optimize every aspect of our workflow.
Additionally, I've noticed that a recent beta version of Premiere Pro includes export performance improvements for ProRes and I'm curious if these improvements have any impact on editing performance within the timeline as well.
I've searched extensively for relevant benchmarks but haven't found concrete comparisons, only people saying you should have good timeline performance using any of them, and in every comparison I've found, they typically are more concerned about file size. I feel as though this is a pretty unique case, so even if it's a small performance difference, every little bit helps.
Also, I've considered trying to do my own tests, but I'm not sure how to properly perform the tests to ensure I'm getting accurate results, so any insights, benchmarks, or personal experiences would be greatly appreciated!
Thank you!
