Sam Houchins II
May 11th, 2019, 08:01 AM
Too many generations of rendered versions leading to loss of resolution / sharpness / detail.
Source is 4k XAVC-S. Output to be 1080 available online.
2-hour session, 3 cams x 2 recorded sessions, + 120 ppt slides
Tasks to accomplish = color correcting, synchronize, choose multicamera cuts, crop and motion track, apply overlays, mix in previously edited takes from the 1st session.
In the past, I'd edit an entire project left to right on the timeline, with the intent of rendering the whole once the whole was finished. Consequently there was no generational degradation, but I was tortured with unnoticed and unintended glitches in distant surrounding edits due to user error with ripple effect or track adjustments that suited one clip while mangling another.
Plan to mitigate my bumbling was to do more renders of "completed" sections to lock those edits down while I proceeded down the timeline to other sections.
Glitch in that plan was that those renders were still more or less only drafts in the process, that then received other layers of edits, which were then locked down with another render, sometimes cumulatively combined with newly compiled sections.
1. Must I avoid multiple renders at all cost, because there is no way to avoid generational loss? Early on I used the Sony XAVC / XAVC S, XAVC S Long 3840x2160-29.97p template for rendering sectional drafts.
2. Once a portion of 4k footage has had cropping or motion tracking applied so as to uses less than 4k of that portion of the clip (as little but no less than 1080), should that portion thereafter only be rendered as 1080?
3. At one point midway, I switched to Sony AVC/MVC Internet 1920x1080 template, but ended up with still more generations of edits and renders that followed. Was that fatal?
4. I've not yet delved into proxies(?). Dunno if there's a helpful use down that path.
Thanks in advance.
Source is 4k XAVC-S. Output to be 1080 available online.
2-hour session, 3 cams x 2 recorded sessions, + 120 ppt slides
Tasks to accomplish = color correcting, synchronize, choose multicamera cuts, crop and motion track, apply overlays, mix in previously edited takes from the 1st session.
In the past, I'd edit an entire project left to right on the timeline, with the intent of rendering the whole once the whole was finished. Consequently there was no generational degradation, but I was tortured with unnoticed and unintended glitches in distant surrounding edits due to user error with ripple effect or track adjustments that suited one clip while mangling another.
Plan to mitigate my bumbling was to do more renders of "completed" sections to lock those edits down while I proceeded down the timeline to other sections.
Glitch in that plan was that those renders were still more or less only drafts in the process, that then received other layers of edits, which were then locked down with another render, sometimes cumulatively combined with newly compiled sections.
1. Must I avoid multiple renders at all cost, because there is no way to avoid generational loss? Early on I used the Sony XAVC / XAVC S, XAVC S Long 3840x2160-29.97p template for rendering sectional drafts.
2. Once a portion of 4k footage has had cropping or motion tracking applied so as to uses less than 4k of that portion of the clip (as little but no less than 1080), should that portion thereafter only be rendered as 1080?
3. At one point midway, I switched to Sony AVC/MVC Internet 1920x1080 template, but ended up with still more generations of edits and renders that followed. Was that fatal?
4. I've not yet delved into proxies(?). Dunno if there's a helpful use down that path.
Thanks in advance.