Hi all, I just wanted to share my experience with everyone as to a problem, and solution, to a project I worked on recently. A client of mine had a DIT issue where their raw rushes had been shot as frames alternatively to two different cards on an external recorder and unfortunately the part time DIT mis-diagnosed the situation as duplicate recordings and consequently only backed up half of each of the rushes... Every other frame! You could normally get around this by using the camera proxies, that were recorded, however in this situation the delivery was 4K UHDTV, and was to be QC'd the hell out of by the eventual end client's lab. To make it clear, quality of the deliverable was of the highest priority. To get around the 'problem' I used Bulk Rename Utility on Windows to put the alternative frames back into sequential order. At this point I loaded in the footage into Digital Vision Nucoda and calculated the first frame timecode of each shot and cut the shots in. I then used DVO Twister to retime the footage from 25fps to 50fps, essentially changing the playback speed to 50% using motion compensation to fill in every other frame. What I like about DVO Twister was the amount of options for motion compensation to help sort out issues with fast movement and blurring. Considering the material contained everything from plaid shirts to highly detailed textures, all with movement, I was amazed at how well the software worked first time, and ultimately how well the problem shots improved by tweaking the settings. Finally for any of the more detailed issues, I used DVO Dropout + Fix to either paint the whole frame or paint individual sections of the image. This works by using the previous and next frames to create an interpolated fill. My client was thrilled and naturally it passed it's grueling 4K tech review in the lab. To be fair, when we QC'd it ourselves, we were struggling to work out which were the original frames and which were made up!