I have several sequences that still have visible bright differences where I have altered the exposure during shooting. The one I am working on now is a night to day shot, 2/3 exposure adjustments made by watching the histogram to avoid clipped highlights. The sun is not in the image area.

I have tried the "Holy Grail" workflow panel, and the 2 and 3 star images are close, but not a match, there is a visible jump. I maunally adjusted the 2 and 3 star images, did the specials and followed the rest of the workflow. I am only altering the exposure slider between the two frames, all other adjustments are synced. I toggle back and forth to make certain the they are a match. I still get a visible exposure jump in the rendered video.

I then went to the "deflicker" workflow panel. After manually matching the 2 and 3 star images, I clicked on the auto transition button, saved, and read the metadata in Lightroom. When I then view the images in Lightroom, some of the 2-3 star jumps nolonger are a match in brightness. The difference is small, 1 or 2 tenths of a stop on the exposure slider, by quite noticable in the rendered video.

The sequences do deflicker fine, this is a separate issue.

I even tried to run them through GBdeflicker in Premiere Pro CC, but that doesn't remove the brightness jumps.

Does Gunther or anyone else have suggestions?

Thanks,

Peter

I maunally adjusted the 2 and 3 star images, did the specials and followed the rest of the workflow. I am only altering the exposure slider between the two frames, all other adjustments are synced.

After doing this you will have to apply the "Auto Transition Special" again, this time for the "2*/3* Holy Grail Exposure to make the transition between your corrected keyframes.

Try to avoid deflicker if editing Holy Grail if possible, deflicker will only happen between the adjustments, they will not make the adjustments "better". Deflicker is only meant for "real" Flickering due to closed aperture, short shutterspeed etc.

I have redone this sequence several times, including following the video tutorial to make certain I did not leave out any steps. What seems to happen is that the Holy Grail wizard does not perfectly match some 2/3 keyframe pairs. (Most pairs are a good match) When I match the problem pairs manually, and then click on the Auto Special keyframe transition, the keyframes in question go back to their slight mismatch of exposure. On the Exposure slider in Lightroom5, they may only be .1 to .15 off, but it is noticable in the rendered video. Perhaps it is a something about the content of the images, where there is dark and light areas and what Lightroom's Match Total Exposure is looking at.

If you use Auto-Transition Special for **/*** only the transition will be calculated again, the 2*/3* keyframes will not be touched.
Make sure to not have deflicker activated.
The matching is not being calculated by analysing the images content, only by Exif-Data. The slight differences are because of rounding/tolerances in the values that the camera writes to exif compared with the actual amount of light that comes in.

If you use Auto-Transition Special for **/*** only the transition will be calculated again, the 2*/3* keyframes will not be touched.
Make sure to not have deflicker activated.
The matching is not being calculated by analysing the images content, only by Exif-Data. The slight differences are because of rounding/tolerances in the values that the camera writes to exif compared with the actual amount of light that comes in.

So, if there is flicker, I ought to use the two pass method? That is, I should go through the normal Holy Grail workflow and then apply deflickering to the jpg or tif files in the LRT_ folder?

Thanks for the help. Other than this one sequence, I have had excellent results from LRT 3. Quite frankly, I wouldn't be integrating so much time lapse into my work without it. I hope it has been financially successful for you so that you can continue to develop it.

Thank you. Yes, the LRT_ intermediate sequences are great for a two step deflicker approach.