Editing in low res and rendering in high res - how does it work?

[LittleGreyCat]

Having read various threads about editing (I am working up to editing a lot of AVCHD video) one strategy seems to be to convert the original into a lower resolution, edit the lower resolution, then use the results to render the original data to produce the finished product.

Simple editing strategy seems to be dragging small clips from the original data onto a new timeline and optionally inserting transitions between clips.

So the question is - how does the software keep track of where in the original timeline the clips came from so that it can match the low res edited version exactly to the high res original version come rendering time?

Is this all done by recording stuff in a description somewhere in the bowels of the software, is it perhaps in metadata, or is the original timeline data included in the new clip in some other way and then stripped out during rendering?

Just trying to get my head round editing one set of video data and having the results applied to a different (original) set of data.

Cheers

LGC

 

[noiseboy72]

In the old days, offline and online editing was done via EDLs - Edit Decision Lists. These used the time code on the tapes to track where all the clips came from and to stick them back together in the same order. These days XML files and similar are used to do a similar job. (I use Premiere Pro, and this allows me to transfer projects to Final Cut Pro and Avid Express)

If you want to transcode footage to a lower res to edit, you will need to ensure you use the same frame rate and timecode settings, so the software can track your changes. If not, when you point the package at the new footage, the results will be all over the place. Depending on your clips and edit package, you may also need to match audio sample rates etc.

On most machines with decent power, you can now edit the original footage and use low quality preview settings. This saves having to transcode footage, but still able to see final cut. Rendering at high quality on the output can then done at leisure.

 

[PhilipL]

Hi

Having read various threads about editing (I am working up to editing a lot of AVCHD video) one strategy seems to be to convert the original into a lower resolution, edit the lower resolution, then use the results to render the original data to produce the finished product.

Simple editing strategy seems to be dragging small clips from the original data onto a new timeline and optionally inserting transitions between clips.

So the question is - how does the software keep track of where in the original timeline the clips came from so that it can match the low res edited version exactly to the high res original version come rendering time?

Is this all done by recording stuff in a description somewhere in the bowels of the software, is it perhaps in metadata, or is the original timeline data included in the new clip in some other way and then stripped out during rendering?

Just trying to get my head round editing one set of video data and having the results applied to a different (original) set of data.

Cheers

LGC

Yes the editor just records the cuts and transitions to apply by using the frame number. It doesn't care if the file changes as long as the replacement carries the same number of frames.

Software like Corel VideoStudio will create low resolution files and use these automatically (Smart Proxy they call it), then when you render out it switches to using the original files.

Other editors might need a manual step to switch to the original files, usually by renaming the folder/moving the lower resolution files will prompt the editor when you open the project to ask you to pick the files again, you then pick the high resolution ones.

Obviously have a try out with your chosen editor before embarking on a big edit just to make sure it works.

Regards

Phil

 

[DocJackal]

Having read various threads about editing (I am working up to editing a lot of AVCHD video) one strategy seems to be to convert the original into a lower resolution, edit the lower resolution, then use the results to render the original data to produce the finished product.

It's called Conforming.

 

[12harry]

Editing software has a lot to do, esp it it's to render in a reasonable time on existing PC's.

Whilst it's true you "need" a powerful PC, it isn't quite the point - it's a question of "TIME". If your PC is a quarter of the rec "spec" the it will take at least 4x longer, so 1hr become 4.

I don't think the speed of a PC actually affect the final output - although I'm having severe "Render" issues myself....I suspect I have a setting wrong - or maybe several.

It makes no sense to edit the full-size clips when a less clip will work just as well. Provided the computer keeps track the final Render should be just fine.
Ext HDD
These are great places to store video-files. However, to keep Render-times down, it "probably" makes sense to have a portion of the C-drive (or any internal drive), dedicated to accepting the Clips/media and (perhaps) another area dedicated to receiving the Rendered output.
USB-2 (or -3) are quite fast, so this issue is somewhat less, but USB-1 was quite slow which could delay the Render process, with each movement of files. Since this is the "bottleneck" it will be good sense to eliminate it, with a simple "fix".

 

[KelvinS1965]

I'm not sure if this is somehow my problem today:

I editted some AVCHD footage into a 1 minute clip. I rendered it in various outputs and they all have nasty blocking and other effects that aren't on the original footage. I checked the original files and they play back fine using DVB Viewer I happen to have on the same PC. The source is 1080/50i AVCHD files from a Panasonic HDC-SD9.

I deleted the proxy files and reloaded the project and re-rendered to make sure that it 100% must be using the HD files rather than somehow the proxy ones and it still has these issues. I couldn't do any other editting as my PC is quite low spec (currently CoreDuo @ 1.8Ghz though I usually overclock to 3Ghz before I start editting), so this proved it was using the HD files as any further editting was really slow and jerky.

Maybe some other issue, but seemed one area to start. Really annoying as I haven't done any editting for ages and when I finally get round to some it plays up.

 

[12harry]

Need more info - like yr processor speed, RAM etc . .
The real issue though is what happens when you create a DVD (say) and watch that on a TV via a dedicated hardware "player"? Is this also "wrong" as described?

I've had issues with loss of "sync" after rendering and creating a DVD - woes!
Then someone (here) suggested I check the sync back in Vegas....the result was it was OK - so the only place for an error is my PC . . . it's just not up to the job, although being Dual-core 2.8GHz and 4G RAM. (32-bit).... I though I had sufficient power!

This threas is touching (as I read it) on the concept of swapping HD media for low-res for the Edit, then swap back for the Render. A process that has attractions, yet it's another process that takes TIME and if you (or I) foul up, it's futile. I can't help think a more-powerful PC is a lot easier.
I shall go for 6-core more RAM and maybe 64-bit, although Vegas Studio does warn users there are some Plug-Ins that will revert to 32-bit - but by then, maybe these plug-ins will be written for 64-bit operation which can make use of extra memory.

Hope that helps...

 

[sterankin]

I think this is proxy editing and is something I had to do because my PC wasn't powerful enough.

Basically had the original footage (avchd) and then converted it all to DV-AVI.
Using premiere pro I created a new project that matched the AVCHD footage (1080-50i I think).
Import all the DV-AVI into this project (it will looks smaller in the preview pane so just zoom it to fit).

Then edit as normal.

Once you are happy, right click on each file individually and choose "replace footage..". Replace it with the corresponding AVCHD file. Takes a while if you have a lot of footage and this has to be done one at a time. The edits will remain in place and you can then export to your desire output.

Another way to replace the footage (without having to replace all one at a time) is after having editing the DV-AVI, save and close the project, then open the .pproj file in notepad, then do a "Find-replace all" (i.e. replace '.avi' with '.m2ts' or '.mts' depending on what you avchd file extension is). Then save the file and then reopen the pproj file in premierepro.

It *should* now load all the AVCHD footage instead of the DV AVI. But I haven't had as much success with this method, so I suggest copying your PPROJ file before trying it. Test it out and see.

Note this method only applies to PPro - for other NLE's I am sure it may be different.