1

Is there a performance difference in Lightroom when editing (lossless or with losses) compressed RAW files instead of uncompressed RAW files? Even when it might not be noticeable?

I could imagine that compressed files (lossless and with losses) need to be uncompressed first which might take up some performance. On the other hand uncompressed files are a lot bigger so maybe more data needs to be put through.

4
  • What cameras produce uncompressed raw files? Pretty much all of them compress the data, it's just a question of whether they use lossless compression schemes or lossy (to one degree or another) compression schemes.
    – Michael C
    Aug 22, 2020 at 16:39
  • "Performance?" Defined in terms of exactly what? Various optical image quality characteristics (contrast, resolution, edge smoothness, tonal smoothness)? How long it takes various computers with different hardware configurations to open the file and display an interpretation of it on your screen?
    – Michael C
    Aug 22, 2020 at 16:44
  • @Michael C almost every higher end nikon camera lets you choose between compressed/ lossless compressed and uncompressed RAW files. Also: Performance in terms of workflow. Loading times, export times, editing speed and so on ...
    – Arjihad
    Aug 22, 2020 at 16:46

1 Answer 1

2

If you're talking about how long it will take your machine to open and display an interpretation of the information in the raw file it's most likely fairly negligible. Of course it would depend on many variables:

  • Image size in terms of numbers of photosites (you don't really have "pixels" until after demosaicing). A 9000 x 6000 pixel image will take more processing than a 1500 x 1000 pixel image file would.
  • The number of distinct values within the file. If the entire image is a uniform color and brightness, a highly compressed file could be processed faster than an uncompressed one because there is much less data to read and process before converting the information to a raster format.
  • The hardware configuration of the computer doing the processing. Both in terms of the speeds at which various parts run but also in terms of how those parts- storage, memory, processor, GPU, etc. - are connected. Is the processor multi-core? Multi-threaded? How many threads? How many reads does the memory do per cycle? Do you have enough RAM to load the entire image file into memory without using a swap file? How much memory cache does the processor have?
  • The software being used and the routines and algorithms it uses to convert the raw data to a raster image sent to your monitor. Is the application written to take advantage of multi-core processors?
  • The user's preferences within the software application. Are your display and/or processing settings selected for "speed" or "quality?" Is the initial image you see the embedded jpeg preview image included in the raw file? Or a newly processed version of the image applying default settings (WB, contrast, etc.) at the time the file is opened?

The reason the difference will typically be fairly negligible is because getting the contents of the raw image file into a state where that raw data can be processed by the CPU and/or GPU (again, depending on the hardware configuration and the application, some programs can use the GPU to do a lot of the number crunching) is usually fairly trivial compared to the time it takes to process that uncompressed information, apply the needed operations to the raw data, then send it to your GPU to send to your display.

It's possible that your hardware could have specific limitations that would significantly increase the time needed to decompress a compressed file until it would become noticeable with very large files. Or that your application, such as Adobe Camera Raw that runs under the hood of Lightroom, could be poorly written in such a way that it takes longer than optimal to decompress the contents of the raw file. But if your machine and software version are up to date that should not be the case.

Ultimately which takes longer, reading a larger file or decompressing a smaller one, and how much the difference is depends upon the size and variability of the contents of the file as well as the strengths and weaknesses of the hardware you're using.

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.