Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

REDCODE COMPRESSION: WEAPON

Kyle Dones

Well-known member
Joined
May 3, 2012
Messages
304
Reaction score
1
Points
18
Age
38
Location
Grayslake, IL
Website
www.kyledones.com
I know I saw the specs floating around and now I cannot relocate the info I am looking for.

At what frame rate and resolution does 2:1 RC open up for availability? Is a larger card necessary?

any other info on REDCODE and compression welcome.

By all means, if you have a thread link that answers this please share.
 
nice table says 5k WS 2:1, I thought i remember reading else where 6k at 2:1 but maybe I was on something.
 
You'd better be able to do 6k 2:1..

With Carbon Fiber or Forged Weapons at speed (24fps) 6K FF, 6K 2:1, 6K HD, 6K WS, and 6K 3:2 can be shot at REDCODE RAW 3:1. 6K 6:5 (commonly used for anamorphic), 6K 4:1, and 6K 8:1 can be shot at REDCODE RAW 2:1.

5.5K WS and 5K can be shot at 2:1 at speed.

These are the numbers if you are using either 1TB or 512GB MINI-MAGs.
 
Based on the metrics of an entropy codec like RedCode, once you get to a 3:1 compression ratio there isn't much left to be gained by going 2:1. Phil might have the data to support, or refute that claim, but the generally accepted point at which J2K derivatives reach parity with uncompressed is 2.5:1. YMMV.

Cheers - #19
 
Based on the metrics of an entropy codec like RedCode, once you get to a 3:1 compression ratio there isn't much left to be gained by going 2:1. Phil might have the data to support, or refute that claim, but the generally accepted point at which J2K derivatives reach parity with uncompressed is 2.5:1. YMMV.

Cheers - #19

I want it all...every crumb of info
 
I want it all...every crumb of info

The only thing I can add to what Blair said is we traditionally understand that 2:1 is "visually lossless" if comparing to a 1:1 uncompressed image. Hard to truly test in the case of how it's implemented in REDCODE RAW, however, we do know that visual quality is effected by the variable Compression Ratios the user can select when recording. Which is why for so long we recommended recording REDCODE RAW between 5:1-8:1 Compression Ratio for typical shooting scenarios. You can go higher and lower for sure without too much penalty. I still see a lot of people shooting 10:1-12:1 out in the wild. I've gone the other way. I'm mostly shooting at 2:1-5:1 on Weapon at speed.

I think one thing that can't be understated is for 3D productions shooting 48-60fps timebase having these lower REDCODE RAW Compression Ratios is a very big deal.
 
Disclaimer: I'm not an expert on compression schemes.

You can get compressed images that are visually and mathematically identical to an uncompressed image. Jpeg2000, which is a similar compression scheme to .r3d, has lossless compression. PNG is actually a better lossless compression choice for real world images, but jpeg 2000 has other advantages, like scaling its resolution. 3:1 might be very near lossless, depending on what was in the frame being compressed.

Compression%20(RGB%2016%20bit%20linear,%20Lossless).png



edit:
this is the 'flower_foveon.ppm' image that compressed losslessly to ~3:1 above. There's not a lot of high frequency data in any given line of pixels, so it can compress to a smaller size.
flower_foveon.jpg
 
Last edited:
Back
Top