Hello,

I spent the last couple of days speaking with Adobe support because of serious speed issue while exporting a batch of files.
After hours of tests and errors I found the on the XMP files that were saved from LRT.

Just to give you an idea about the time took 4h30 to export 400 images on a formatted laptop, with brand new catalog with only 400 images.
Laptop is I7 6core MacBook Pro from 2018. And I'm running latest version of Mojave and Lightroom. My LRT is the trial version.

When exporting Lightroom takes time to read the file and the .xmp file. the bigger the .xmp file the longer it takes to export the file. I did all sort of test.

And the issue happens only with files that contain the .xmp files imported/re-written from LRT.

Images that are not edited are faster
Images that are edited in LR are slightly slower
Images of which the metadata was re-written from LRT takes forever.

Hope you can help me as I really enjoy your software but I cannot use it if brings my working station into crisis.
P.s. attached a copy of the metadata

Thanks for you help
Marco

This post was last modified: 2020-02-08, 23:40 by Gunther.

Hi Marco,
are you sure you are using the latest Lightroom Classic? Adobe significantly improved export speed with the latest updates.
Of course, LRTimelapse XMP Data carries more information, than unedited images, but your times are definitely not normal, I would think.

I've just run a test, on my laptop (PC, 8 core) and 400 images from LRT, quite heavily edited needed 17 minutes to export from Lightroom Classic (latest version).

Of course, it depends which tools you use. Noise reduction, Sharpening, Perspective Correction etc. can be quite intensive, but definitely you shouldn't have to spend hours on a modern PC.

Yes I've checked and is definitely the latest version.

Haven't used none of Noise reduction, Sharpening, Perspective Correction etc.
Could it be that radial filters are generating the issue?

I've added 3 radial filters when editing with Lightroom and I've seen that the voice radial filer is not one of the voices appearing in the columns on LRT.
As I remember in one video you saying for example not to add any new ND filers but only use those provided by LRT. Could that be the problem?

Thanks!

You cannot add additional Linear or Circular gradients, when working with LRT. Use the ones that LRT creates and edit them. Don't add new ones, don't remove existing ones.
But I don't think this will be causing the long export times.

What kind of images are you processing? From which camera? Resolution? Raw / Dng?

Hey, sorry for the delay getting back to you. And thanks for the effort in helping me out.

So that is what I did, I've used a couple of gradient and 3 circolar. Maybe that is what was causing the issue?
Also in both filers I've applied applied the new feature called "Range Mask" (from 2018 or 2019 LR update) that allows to target just highlights or just shadows. Is that supported by LRT?

The images are taken on a Sony a7rIII around 42MP files, and raw format.

Right now after trying everything I've tried to cancel all the metadata from the sequence, export without adjustments and was fast. I tried to export also with the metadata (from the first try) and was incredibly slow.
Last try was to export by following the correct process so without adding any filters (apart from those provided by LRT) and has been pretty fast. Took me about 20 minutes instead of 4h+ (and 4 hours was not a case, did it 3 times).

Can it be because I don't have the full license version? I'm of course planning on buying it but I wanted to try the free version before doing the purchase.

Hi Marco,In every instruction, tutorial and the faq I explain, that you cannot add or remove gradients when working with LRTimelapse. You can only use the ones provided by LRTimelapse. There are 4 Linear Gradients and 2 Circular ones. Please use only those.
Yes, you can use Range Masks. But of course, the more complex the editing, the longer the export will take.
Exporting from Lightroom will be the same speed in the free version as with a license.
LRTimelapse itself might be faster, because it supports more than 4 parallel threads in the licensed versions.