Is this right- ProRes file size consistency?

So I have two files. One a ProRes 422 quicktime from a ProRes (proxy), the other a ProRes 422 quicktime exported from the baselight.

Same duration (slightly over 10 mins) , same dimensions (1080p), same frame rate (23.98) however the baselight version is roughly 2.5gb bigger, and the "data rate" is about 20Mbit/s greater than the ProRes422 QT file I exported from FCP.

Is this normal? I thought the codecs data rates were set, so two files same duration/size/frame rate would have the same data size and rate if they were in the same codec?

Quote

I thought the codecs data rates were set, so two files same duration/size/frame rate would have the same data size and rate if they were in the same codec?

Nope. ProRes is a variable-data-rate codec. Here's a fun little read, if you'd like more details.

Amazing that there can be a20%-25% difference in size. I always thought it was variable... but 20-25% feels like an entirely different codec. Quite interesting.

Did I read you right, that you took a ProRes 422 (Proxy) file then converted it to ProRes 422? Not a big surprise that it should be much smaller, then. You already threw away tons of detail when creating the ProRes 422 (Proxy) version, and that's detail that doesn't have to be encoded when you reencode as ProRes 422.

Sidebar: Dear Apple, I like ProRes very much. I think it's neat. But the flipping naming is ridiculous. ProRes 422 is fine; ProRes 422 (HQ) sucks a little bit. But it was ProRes 422 (LT) and ProRes 422 (Proxy) that made me realize that you're just messing with us now.

Yea, created from Proxies so yes, we threw a ton of information away to begin with. I'm just surprised it amounts to that much really.

We needed to do a side by side comparo on the Baselight to standard 422 and from our Proxy to standard 422.

422 is what the promo department is requesting. re.

Oh, I wasn't criticizing. What you did wasn't unreasonable. I was just saying that since ProRes Proxy tops out at 36 megabits, while ProRes 422 can go all the way up to 145, it shouldn't be terribly shocking that ProRes Proxy converted to ProRes 422 should be a lot smaller than the same shot encoded straight to ProRes 422.

Do the math for me? What's the percentage difference in either overall size or average data rate? (The differences should be very close to equal.)

>But it was ProRes 422 (LT) and ProRes 422 (Proxy) that made me realize that you're just
>messing with us now

You missed the last one... ProRes-4224444! It's a telephone number!



www.strypesinpost.com

I'm gonna start dropping references to Prores 423 into conversation just to screw with people.

I'm going with ProRes 422.1



www.strypesinpost.com