DxO caches: why not the same structure as my datasets

Because of ordering a new PC to inventerise the cache files on my hardisks, I fond many: Photolab 2, 3, 4, 5 and 6, of IMatch, Nikon NX, etc.
Why not a cache with the same structure, same directories as my datasets. Or even two…


  1. the original dataset in RAW in my case + the DOT-files (ca 2 Tb)
  2. a cache with jpegs in the original 'preview (will be about 100 Gb?)
  3. a cache with the jpegs (or another) of the last version, to be updated regular (size depending on the settings.

The last one can be used by other programs, like IMatch. But also Wordpress, Word, etc.

The cache will not more than 2 Tb, is even affordable as SSD,

Who can explain to me that thsi is a stupid idea?


Ruud Kampf

A cache is an internal dataset intended to speed the operation of the program.

You are looking for a separate output directory to store the latest output images?

What happens if you select 100 images and make an adjustment. Then what you’re asking is DXO automatically export those 100 images into a folder which is like a view of the edited files? What if you quit PL before 100 exports can be completed?

Yeah I don’t know that I get the real purpose of this. What would you use it for?

It’s an interesting idea but the implementation is potentially problematic (and it would suck battery life from my laptop). I probably wouldn’t use it.

you know, that you can point the cache to another location, resize and also clear it, when you don’t need it (anymore)?

Hi Wolfgang,

Yes, I know. However, I do have the idea that a cache can be more productive.

A cache is indeed normally a separate output, hidden somewhere deep, with a typical programmers structure.

Indeed, not on a laptop indeed. But on my desktop I would love to have it. A 6 Tb Storage cost is a a fraction of what it used to be. Hard disk is 170 Euro nowadays, a 1 Tb SSD less then 100 Euro.

Adjustment could be either automatically, or one command for a directory. Or just switched off.

You basically want to replicate the photo archive with an identical structure containing jpeg (or other) files?