Does anyone have strong thoughts on rendering in LRT vs AE? I personally think LRT works well because of the multi-threading which makes rendering the intermediate sequence faster.

Some of my friends use AE to render directly from the RAW files, claiming better quality, but I don’t think that’s the case because in LRT you could render TIFF (lossless) files directly and then use that intermediate sequence for video rendering. Behind the scenes it’s my understanding AE also creates a lossless bitmap type of image for the video encoder but it’s so much slower than LRT given lack of multi-threading.

Curious to hear from others what the appeal of using AE vs LRT is for rendering final timelapse videos, assuming you’re not applying any other effects. I know in AE you can submit batch…but given how much slower it is are you really benefiting? Thanks.

Thx for posting this here also so that anyone take profit. I'm adding my answer that I already wrote to you via email here now.
--
It's true that you can bring a RAW+XMP sequence to After Effects and if you have exactly the same versions, AE will apply the XMP edits and render a video. But you are right, it will also develop a Bitmap (like TIFF) in background and use that to assemble the video. The whole process is very slow because AE doesn't do any multithreading.

Basically LRTimelapse does exactly the same, only that it gives you access to the intermediary tiff sequence (which is nice, because you can easily rerender with different settings etc...). That's why I chose to have an "open" intermediary sequence. If I'd hidden that as AE does, you would have it in one step but much less flexibility.

This means: there is no loss in quality since TIFF is lossless. Only if you use a JPG intermediary sequence there might be a theoretical loss.

In reality the quality that you achieve with LRT is even better because:

• LRT uses the ffmpeg renderer which is know for delivering better results than the adobe renderer in many cases
  
• LRT internally uses a full Rec.2020 Workflow which will take preserve that huge color space until the end. To do that in AE, you'd really need to take care of every step and you'd have no control about it.
  
• LRT is much faster since the Lightroom developing with Lightroom Classic will do nice multithreading so will LRTimelapse.
  

Bottomline: there might be reasons to use after effects, for example for stamping etc - but I would not prefer it for rendering. The LRTimelapse workflow is streamlined for Efficiency and Quality and definitely has no disadvantages in these terms

Thanks as always Gunther! I will post this in the forum today, curious to get a discussion going on this. I think there’s quite a bit of confusion or misinformation.

Do you think most professional timelapsers are using jpgs for the LRT workflow, rather than TIFF? Do you use jpgs normally? I personally find the TIFF files huge and and the resulting prores file which although theoretically better is not very practical or efficient.

I think AE must be using a compressed bmp intermediate image, because the resulting prores video file isn’t too big. For example 10 secs is only 5GB. If I create tiffs and then a prores render the file is over 100GB! Given this there must be compression applied to the bmp file.

Usually I create the intermediaries in 8bit tiff. This is a very good compromise between quality and space. Harddrives are cheap and huge today, and for the intermediaries a Harddrive is sufficient. Personally if you need to worry about storage space, I guess you should just invest in some bigger storage.
The size of the resulting ProRes doesn't have anything to do with the format of the intermediaries.
if the AE ProRes is much smaller, the quality setting is inferior. Usually the LRTimelapse quality settings are on the high end. I usually don't go higher than "High".

Thanks Gunther!