The term proxy gets thrown around a lot in this industry. For many, the term is a panacea for everything. “We’ll generate proxies during the shoot…”, “I’ll give you a drive with the proxies…” and “Can they just use the proxies?” are all commonly heard.
It appears people mistakenly think that when they say “I need proxies” they have requested a specific file format, including specifications like size, compression method, frame rate, etc.
It feels like when people ask for just “a QuickTime” deliverable. A QuickTime movie comes in different sizes, different frame rates and even different codecs. For example, I could deliver a QuickTime ProRes, a QuickTime DNxHD, a QuickTime with h.264 compression… the options go on and on. Telling me to create a QuickTime only tells me that the file name should end with “.mov”. (But don’t assume that all .mov files are QuickTimes—they aren’t.)
So my job is to translate “I need a QuickTime” into what the client really wants. It’s usually a pretty quick discussion. Sometimes, if I can’t get a complete answer I’ll deliver multiple options, just in case.
Although I might be reminded of QuickTimes when people talk about proxies, the situation really is different. Proxies are usually representations of camera originals—although there are use cases for proxies of finished shows.
Why is that different? For a couple of reasons. First, proxies are created near the beginning of the workflow, not as part of deliverables at the end. You need to create the right kind of proxies from the start or they might hold up the rest of the workflow.
Second, rather than creating one QuickTime, proxies based on camera originals entail creating multiple files, perhaps numbering in the hundreds. Once again, you need to get them right.
How to do you get them right? Well, you need to understand how they’re going to be used, so you can ask the right questions. That’s what I’ll talk about next time.