GH13 GH1 Final Cut Pro Claification

Vence Vida

Vence Vida

Active member
I will be getting a GH1 tomorrow, and it will be my first HDSLR (I know it's not technically a DSLR, but semantics...). I plan to use the newest hack to shoot in 24p at a decent bit rate.

I currently work in FCP 6 and I also have Adobe Master Collection CS5.

I have done my homework and spent the evening reading the various threads in this forum for dealing with the files in FCP. I understand that I have to wrap them or transcode them and I understand the software and processes involved.

So here's what's not quite clear to me: What is the advantage to one over the other (transcoding vs. wrapping)? It seems to me that transcoding would have to result in some sort of file degradation, so I would think Clipwrap would be the better choice, but I see others say they much prefer transcoding via Log & Transfer or Streamclip or Toast 10, etc because of something about codecs with I-frames rather than long GOP -- which I have to confess, I don't know much about.

Can someone help clairfy why transcode rather than wrap or vice-versa, and what is the best software for the process - I'm not worried about the cost of the software, I just want to know what's best and why.

Thanks!
 
Mpeg StreamClip does a decent job- it's free and speedy.
JES- Deinterlacer
ClipWrap
Compressor

I hear Avi-Synth is pretty good for resolving the most/best image data from footage--but i believe that's a Windows OS thing.
 
Thanks Zack. I'm pretty excited to finally join the HDSLR revolution.

I'm pretty sure that transcoding and wrapping are not the same. You can wrap a clip in minutes (or even seconds) that would take much longer to actually transcode to another format like ProRes. Divergent Media even advertises that Clipwrap doesn't transcode, so no loss in quality.

I have Streamclip (which actually transcodes) which I've used on m2t files that I get off my DataVideo DN60, and I've also used a wrapper program on the same footage. There is a definite difference in what it's doing. I always assumed that wrapping was better, since you aren't actually changing the files, but after reading a lot of threads here, tonight, it seems some people advise to convert to ProRes. Just hoping someone might know why that could be better.
 
Vence Vida said:
I will be getting a GH1 tomorrow, and it will be my first HDSLR (I know it's not technically a DSLR, but semantics...). I plan to use the newest hack to shoot in 24p at a decent bit rate.

I currently work in FCP 6 and I also have Adobe Master Collection CS5.
Click to expand...

I'll not go in to the details but how many compression schemes work is by only sending 'differences' between frames, and the receiver then 'adds' the difference frames to form the frame to be presented. Think of it this way, a scene has mostly 'static' elements... we aren't talking 'shaky cam'... So, one need only send the portion of the image that changes, because of 'motion' of objects, or lighting, etc...

A 'group of pictures' is such a compressed sequence. The 'decompression' involves CPU time to 'reconstruct' the sequences of full frame images. In 'I-frame' only codecs, there are no 'difference frames', just full 'frames', perhaps using compression on the frame like JPeg, which does spatial and 'color/value' compression, but otherwise one 'frame' is independent from the other. This does take CPU power as well, but quite a bit less...

The recommended work flow is if one is going to do a lot of processing, for example edit in Premiere, send the results to After Effects, bring back into Premiere, the recommendation is to use a codec that is 'easy' for the CPU to process. Apple ProRes and the Avid DNxHD codecs are intended for this 'intermediate' processing use.

This is transcoding... going from one codec to another.

Wrapping is where media is placed in a 'container' format, and tools like Quicktime (the underlying libraries which are used by many tools to 'understand' MOV files), may not recognize that container format. Hence 'wrapping' would be to extract the media that is encoded using a codec that Quicktime would understand, and placing the media in a container file it supports, like the MOV format. Wrapping media should not result in 'compression loss', because nothing should be decompressed, and recompressed. Since almost all popular video codecs are lossy, this will result in loss and is to be avoided...

On the topic of Final Cut... If you do have the full CS5, I'd recommend you learn Premiere, After Effects, and dispatch Final Cut into the dustbin of computer history. The introduction of FCP X, has pretty much clearly indicated where 'professional video/film editing' stands with Apple... that is... nowhere. It is clear that Adobe has moved on, and while it still may not be absolutely equivalent to FCP of the previous generation, it is clearly superior to anything in the near future coming from FCP X.
 
Last edited:
j1clark@ucsd.edu said:
The recommended work flow is if one is going to do a lot of processing, for example edit in Premiere, send the results to After Effects, bring back into Premiere, the recommendation is to use a codec that is 'easy' for the CPU to process. Apple ProRes and the Avid DNxHD codecs are intended for this 'intermediate' processing use.

Wrapping is where media is placed in a 'container' format, and tools like Quicktime (the underlying libraries which are used by many tools to 'understand' MOV files), may not recognize that container format. Hence 'wrapping' would be to extract the media that is encoded using a codec that Quicktime would understand, and placing the media in a container file it supports, like the MOV format. Wrapping media should not result in 'compression loss', because nothing should be decompressed, and recompressed. Since almost all popular video codecs are lossy, this will result in loss and is to be avoided...

On the topic of Final Cut... If you do have the full CS5, I'd recommend you learn Premiere, After Effects, and dispatch Final Cut into the dustbin of computer history. The introduction of FCP X, has pretty much clearly indicated where 'professional video/film editing' stands with Apple... that is... nowhere. It is clear that Adobe has moved on, and while it still may not be absolutely equivalent to FCP of the previous generation, it is clearly superior to anything in the near future coming from FCP X.
Click to expand...


Thank you for the clarification on the I-frame format. That helps me understand what that means and how it works. I do understand codecs and the whole concept behind only sending the differences within a GOP. I'm just not sure what effect editing with one codec versus another has on the final footage, and I didn't know what I-frame format meant. I can definitely see how that would be better for the process, although would create much larger files.

What I still don't feel confident about is when transcoding from a GOP codec to an I-frame codec is beneficial. I am taking a guess from what you've said and from other threads I've read that if you are doing basic cutting and pasting type of editing, leaving the footage in it's original codec (using a wrapper) would result in less loss in the end, but if I plan to do color grading or other filter-effects type treatments that transcoding to ProRes before I do those things would result in less loss on the final output. Does that sound right? Am I getting the premise, here?

Also, regarding FCP X, I'm not in a position where I need OMF, AAF or EDL outputs, so that doesn't impact me. The only tape work I do can be done via Firewire, and I'm moving rapidly to a tapeless workflow. I believe them when they say multi-camera and XML is coming ASAP in a soon-to-be-released update, and for now, FCP 6 is working just fine for me. I almost never have to go back and re-edit an old project that is more than a few months old so backward compatibility is not a big deal to me. And for accurate live monitoring and HDSDI I/O, I have a Decklink Studio card and the drivers on that have either already been updated or will be shortly to work with FCP X, so I'm in pretty good shape. I think FCP X will actually be a good upgrade for me and the type of editing I do, so I'm not so much on the "FCP is dead" bandwagon.

And I've heard that FCP X works with AVCHD files, so wondering if it will actually allow me to deal with files from my GH1 natively, at some point. Does anyone know if that could be true?
 
Vence Vida said:
What I still don't feel confident about is when transcoding from a GOP codec to an I-frame codec is beneficial. I am taking a guess from what you've said and from other threads I've read that if you are doing basic cutting and pasting type of editing, leaving the footage in it's original codec (using a wrapper) would result in less loss in the end, but if I plan to do color grading or other filter-effects type treatments that transcoding to ProRes before I do those things would result in less loss on the final output.
Click to expand...
Using an I-frame codec can improve motion tracking during video capture. Once the file is recorded, however, there is no quality advantage in transcoding to an I-frame format - if anything the transcoding process will degrade the image. In Adobe CS5 there is no advantage to transcoding at all - a modern PC with a fast video card and plenty of RAM can edit the original video files directly without any need for transcoding. It is only when rendering a composition back to disk that it's preferable to transcode to a higher-quality format than the original. The main advantage to transcoding to ProRes was to facilitate file sharing with Final Cut Studio, which may continue to be a viable option for some period of time.

In FCPX you have no choice as the files are automatically transcoded to ProRes in the background.
 
You must log in or register to reply here.
Back
Top