1

I have a RAW file from a 24 megapixel camera. The RAW file is about 25MB in size, looking at it in OS X. When I load the RAW into Photoshop and then look at Image > Image Size in CC 2014, Photoshop reports a size of 58MB. I find this dialog useful as it allows me to assess how big I can print at various DPI but I wonder about why the file size has now doubled?

I am guessing that this is because Photoshop has to hold the data in an RGB color space vs the native bayer sensor format of the camera.

Interestingly, when I save the file as a PSD, file size grows to 110MB. I am considering a higher megapixel camera but I shudder to think what the end file sizes will be.

Extra info

Just as an exercise, I pulled down a 7360 x 4912 RAW from a Sony A7R. RAW file size is 37.7MB. In Photoshop, Image > Image Size reports a file size of 103MB. Saving as a PSD, 254MB.

1

First of all, size in MB doesn't directly mean anything for print production. Resolution means everything.

When you open raw image in Photoshop, it saves in memory not only raw image itself, but also a copy of the image that you made while import with Camera Raw, so, basically you have original and Camera Raw version of the image in memory. That's why the size is approximately double.

When you save psd file, if you save it with support for previous Photoshop versions, file size can grow up dramatically. So if you don't need support for previous versions of Photoshop, always remember to uncheck that option upon saving.

If you want to keep your files in raw, I suggest you to use Digital Negative (DNG) file format. It's native for Adobe and in the same time uses losless compression.

  • "Maximize compatibility" is also required for Lightroom; basically it creates a hidden stamp visible layer that can be used in programs that don't support layers (and I really, truly hope that people are using working layers for cleanup/d&b and adjustment layers rather than direct manipulation of the background layer until they're sure they're done and everyone with a stake in the image is happy with the result). – Stan Rogers Jan 22 '15 at 0:16
  • That's true, but I don't think it's the case this time, because we are talking only about the image size. – mrserge Jan 23 '15 at 9:38
  • The point is that while the option will essentially double the file size of a flat image, it's not something you can dispense with simply because you don't need to interop with older versions of Ps - there are other, more important reaasons to keep the option checked. – Stan Rogers Jan 23 '15 at 14:56
  • I heard only one important reason - compability with Lightroom. And it's not essential for a lot of users. And in case with raw photos from camera I still prefer to use dng, not psd file format. Just because dng is made axactly for this reason and psd is not. – mrserge Jan 24 '15 at 15:29
  • Lightroom is only one of many photo-management (or image-management) software packages; there is also the little matter of output to and interop with prepress software, etc. And DNG is not at all appropriate for works in progress (pending approval, even if you are the only approver), unless you're one of those self-important "straight photographers" who never actually experienced what went into darkroom photography. – Stan Rogers Jan 29 '15 at 18:23
1

RAW formats differ quite a bit even within manufacturer.

Canon CR2 for example appears to use anywhere from 2 to 4 greyscale "lossless jpeg" (non-linear quantization?) channels to encode the data. It is not stored as RGB in the manner of a typical TIFF or JPEG, nor are the channels themselves necessarily the same size.

Even though they may be storing in floating point or 12-16 bit integer formats (i.e. more storage), the packing and encoding methods used can still account for the smaller file sizes you are describing.

reference: Understanding What is stored in a Canon RAW .CR2 file, How and Why (IFD#3 is the RAW portion)

0

It depends on whether you need to save memory or not. If you need to save memory you may need to compress the file at some point. That's like if you make a music file on some kind of music software, the file's going to be huge, but with much detail. Once you compress it you will most likely lose some of the vibrant detail. Changing it is to another format is like putting algorithms together, since it's not compressing the file, you're making more code. I always use MPEG streamclip, it's just an application with many formats meant for converting things directly so you don't have to go through all of that trouble.

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.