2

With its new cameras, Canon is introducing a C-RAW compressed RAW format. The compressed RAW files can be processed on the computer in a manner as flexible as with losslesly compressed RAW files.

Apparently, the compression ratio is approximately 1.75:1. So, the disk space savings are huge. Also, the burst buffer of the camera is increased by at least 1.75x, perhaps even more (effectively) because the camera will start to flush the buffer to the card immediately and not when the buffer is full. The continuous shooting speed with full buffer will be increased too by 1.75x if card-limited.

The C-RAW will be stored in the Canon's new .CR3 file format. This might suggest that C-RAW files cannot necessarily be opened by open-source software. However, the recent Canon cameras anyway always use the .CR3 file format, even for losslessly compressed RAW files, so this counterargument might not apply.

C-RAW seems to use a lossy compression algorithm. However, Bryan Carnathan investigated its quality, and the lossy compression does not seem to be an issue. The differences between RAW and C-RAW shots seem to be mostly in the random noise. The dynamic range is not hurt like it is with JPEG.

Because there does not seem to be a RAW → C-RAW converter one could run on a computer, selecting C-RAW (plus JPEG too if not doing buffer-limited high speed shooting) on the camera seems to be the most sensible choice.

So, my question is: is there any valid reason to not choose the C-RAW option on recent Canon cameras?

  • The answer is in the question. If you use FOSS, you cannot use CR3 currently. – xenoid Apr 25 at 10:53
  • @xenoid I think you'll find I mentioned those new cameras use CR3 for both compressed and uncompressed RAW. – juhist Apr 25 at 12:50
  • 2
    My valid reason to not use it is, I want the complete untouched raw data, end of story. – Alaska Man Apr 25 at 17:27
  • 1
    @juhist .cr2 and .cr3 (non C-RAW) raw files are all losslessly compressed, just not with a lossy compression algorithm like C-RAW. – Michael C Apr 25 at 17:40
4

Is there any valid reason to not choose the C-RAW option on recent Canon cameras?

That's kind of like asking, "Is there any valid reason to not choose the JPEG option on recent digital cameras?"

Of course the answer to that is, "Not unless you need the complete raw data."

Likewise, the answer to your question is, "Not unless you find yourself in a situation where the complete raw data will allow you a better result in post processing than a lossy compressed C-RAW file will."

Granted, the edge cases where this will be the case will be far more extreme than the number of cases where a losslessly compressed raw file will be superior to a jpeg file for further editing.

Of course, the biggest thing preventing an easy answer to this question is that you usually do not know until after the fact exactly when you need every scrap of information contained in a lossless raw file. And therein lies the rub: by the time you know whether you need a lossless raw file or whether a minimally lossy C-RAW file will do, you're already past the point at which your decision has already been made.

So how do I decide before I shoot?

Experiment with some non-critical shooting situations that represent the kinds of things you shoot that require the most extreme adjustments in your raw editing workflow. Situations where you know that when you capture the shot you're going to need to push the shadows significantly in order to expose low enough to not blow highlights as you capture the scene. Shoot scenes with C-RAW, then shoot the exact same shots with the exact same exposure settings using the lossless compression option.

Then see how much difference it makes when you get on the computer to do your raw development. Do you:

  • Get near identical results that are visually indistinguishable from each other?
  • Have to do more work to get to that point with the C-RAW file? Or do the same adjustment steps get near identical results? If more work is required, is the extra time worth the storage cost savings and decreased buffer depth when shooting?
  • If there is an observable difference, is that difference worth the burst rate penalty for what you're shooting? Or is the increased buffer capacity when shooting worth more to you?
  • If there is an observable difference, is that difference worth the storage cost penalty for what you're shooting? Or is the reduced per-shot storage capacity needed worth more to you?

These are individual judgements that only you can make. The answers may well be different for you than for me or anyone else.

Other things to consider:

  • Do you often find yourself needing to radically correct for missed exposure or rapidly changing lighting conditions in fast moving shooting environments?
  • Do you find yourself shooting in marginal light that will require a lot of color correction in post? Or do you usually shoot in fuller spectrum light, either natural or that you provide?
  • Do you usually shoot static scenes that allow more preparation, careful exposure metering and adjustment, etc?
  • Do you shoot hundreds or even thousands of frames per shoot? Or only a handful?
  • Are you able to pass the increased storage costs of larger files to your clients? Or are your shots only for your own personal usage?
  • Is the shooting situation a "mission critical" one? Wedding? Graduation ceremony? Other event that must be captured effectively in one take?

You might decide that some situations would benefit more from the increased security of having the complete raw data at your disposal, or you might find that it never makes a difference for what you shoot.

| improve this answer | |
2

My answer will be very much opinion based. But first let me clarify one point: Canon RAW is also compressed, lossless.

And for me size is not so important, at the end those days storage is cheap, 12TB external disk cost around 260 euro and you can store there 300k 40MB images. For cloud storage you can use (for example) Amazon Prime Photo, which is around 70 euro/a. 64GB SD card can store 1600 40MB images and it cost 20 euro.

P.S. I have no proof of this but IMHO lossy compression need more computer power compared to lossless. So (if this is true) you can make more photos with one battery.

| improve this answer | |
  • 1
    True but smaller files means faster transfers (write time to the SD cards, uploads to the cloud). – xenoid Apr 25 at 10:51
  • 1
    I usually find that when shooting long bursts, I'm card-limited and not battery-limited. But yes, I agree that storage is cheap nowadays and will become cheaper in the future. – juhist Apr 25 at 12:51
  • @xenoid, true. But valid if you shoot in burst and do this often. If I do something like I will use appropriate camera, 7D Mark II, 1D Mark II, III... And will use CF cards. About upload I upload once and I'm fine to wait a bit more :) – Romeo Ninov Apr 25 at 13:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.