Is there a way to get a better transcoded result from PDF files? Right now they are super pixelated, like it’s only using a preview version of the PDF to do the conversion.
I know in the WebUI that it’s possible to grab the full res version, but it would be nice to not always have to do that, plus it’s a bit more complicated to do this in Shotgun Create.
Yikes! Considering what types of imagery and text that are usually in a PDF, rasterizing text to 720p is pretty dismal. I guess I’ll be putting in that request straight away!
Totally agree with Chad, PDFs are usually used to present layout of the printed content or to present detailed storyboards and briefings. Compression to 720p just makes review tool unusable, all the details are lost, pictures look blurry. We have to look into alternative solutions that handle PDFs and high resolutions for the images properly. That forces us to pay twice for the review tool, as well as breaks the solid workflow of reviewing/getting feedback/applying feedback process. We’ve already checked all the shotgun articles for compression, as well as requested the #feature on the roadmap website. We hope it is possible to improve the review experience for PDF and high resolutions pictires for print content.
Hi Patrick,
I have not done it myself but it’s absolutely possible.
I would do it in the following order (in code of course):
convert pdf into single images
convert image sequence in mp4
upload to uploaded_movie_mp4 field
I already use the same approach to review single images in much higher quality than 2k.
The resolution is only limited with the mp4 h264 codec maximum resolution which is 8K UHD (8192x4320)
So you need to have/find a python programmer who can do it for you. Let me know if you need any help
Sorry to revive an old thread, but I’m curious how folks from the 3D animation world are uploading design packets for review in Flow. Let’s say you’re designing a 3D environment and you have a 15 page pdf design packet that needs to be reviewed by your art director - how would something like this be uploaded and reviewed. In our case, the packet applies to a single asset. We’re hesitant to add something like image magick to our configuration just for transcoding pdf files. It seems clunky to publish separate versions for each image since a version in our pipeline is considered an iterative unit of work (each version would imply an additional round of notes).