A few questions regarding DV in general (and Canopus, specifically)...

[Joel Cairo]

I've done a fair amount of work with analog video capture and editing tools, but I am looking into the possiblity of purchasing a Canopus ADVC 100 to do some DV work. Previously, my workflow went something like this:

1. Capture video to 704 x 480, using AVI-IO
2. Do trimming work in MSP 6.0 (though I now have MS Director's Cut 6.5, it's the same thing, essentially.)
3. Reassemble the pieces of the finished project using the "file append" feature of VirtualDub (which gave me the capability to rip what would be the complete soundtrack for the finished project, and work on it independently.)
4. Then, using the "frameserve" feature of VirtualDub, send the files to TMPGenc for encoding.

My question is, will this procedure still be possible to use for DV files coming from the Canopus ADVC-100? I understand that there are two different "Types" of DV files (1 and 2), and that there are some pre-existing "issues" with both VDub and AVI-IO in working with DV in general (I believe, if I've read correctly, that they only work with Type 2 files). MSP says it supports both types (though it's opening splash page in the video editor project module only has setups for working with Type 1 files...); but I've additionally read something about a proprietary Canopus codec (in the ADVC-100 case, a **hardware** codec, no less) that has me leery about the compatability of the files that the unit generates. If the Canopus files are only useful with the box in the loop, I'm not very keen on the whole idea.

So, can someone verify what form the Canopus generated DV files are in, once they're captured to a PC? Are they merely standard DV-type files, once they're on my machine (no Canopus codec needed to work with them)? Is the DV file type decided by the Canopus box, is it dependent on the capture codec (actually, a "filter", I know) that I use to bring it in? Would using the MainConcept codec help?

For effects, would I still be able to use DynaPel's Motion Perfect software (as I sometimes do for my analog AVI's)? Or do they only support one or the other (if any) of the DV file types?

And will TMPGenc work with DV files at all?

**Whew**

Sorry to ask everything at once, but I thought I'd just get all of this out of the way, so I can try to make a decision one way or the other. Thanks for any guidance that you folks can give!

-Joel Cairo

[Dr Mordrid]

The only advantage of the EZDV over a cheap Pyro, SIIG or DV-Wonder is the software bundle. I think you could save a buck or two by just upgrading to MSPro 6.5. It's a nice editor for DV, having a ton of features geard just to it that 6.0 doesn't have at all, and is hell and gone more powerful than EZ Edit.

TMPGEnc is one of the best encoders to have for DV files. It has some features specifically for the problems DV presents under certain conditions of lighting and scene content.

Dr. Mordrid

[AndrewDV]

Hi,

The hardware codec in the ADVC-100 exists outside of the machine. You do not install a Canopus codec on the box.

Once captured, you can use the generated files on any machine with a DV codec. I have played the captured video through a G400-TV, a Hollywood DV-Bridge, and a GeForce2 on separate machines with no problems.

Using different DV codecs is up to you. I have not seen anything to sway me into purchasing one.

If you play them (or ANY DV files) through the ADVC, you do get hardware playback. The benefit is that quality goes up quite a bit.

MSP6.5 really works the best. The new capture utility is a big improvement. I'm also finding that MSP6.5 works better with DV files than other formats (oddly enough, performance is nicer, files generate really quick).

As far as frameserving, Ulead needs to add this feature. Sadly, they do not support it. However, TMPGEnc works great with the files.

[Dr Mordrid]

While Ulead doesn't have frameserving itself one can, however, frameserve its segmented captures, or separate files generated by MSPro, to TMPGEnc from the disk using AVISynth.

Dr. Mordrid

[Joel Cairo]

Thanks for the responses-- they **do** help!

One thing I **am** curious about, though, (and maybe this should be directed to Andrew, who's used the Canopus box)-- what format are the files that come from the ADVC-100... DV Type 1 or Type 2? Or is that dependent on what codec (filter) that I use to capture them to my hard drive?

Again, thanks for the assist, folks!

-Joel Cairo

[AndrewDV]

Hi,

From my understanding, the DV type does depend on what codec you install on your machine. For me, I am using type1 captures since they are OpenDML compliant. Captures can be as large as necessary since I am using NTFS in Win2K.

[Dr Mordrid]

Most devices use type1 now days, but many offer segmented captures & output for FAT32 systems. Matrox's RT-2x00 cards are one example.

Dr. Mordrid

[Phil Pugh]

Don't CANOPUS DV files have their audio encoded differently to DV files 'captured' through OHCI cards? This is one reason that when using a RAPTOR (as I used to do), MSP 6.0.2 and 6.5 needed to render everything if you forgot to select the CANOPUS DV project settings and chose a 'standard' DV project. It was a minor pain as it knew the video was OK - but the audio needed rendering - which didn't take too long.

[Jerry Jones]

DV "type" is determined by your capture software.

It is not determined by your capture device.

Ulead capture software captures to DV Type 1.
MGI capture software captures to DV Type 1.

(Ulead's timeline can handle *both* DV Type 1 and 2. But the Ulead capture software cannot capture to DV Type 2.)

Adobe Premiere 6.0 capture software captures to DV Type 2.

What's the difference?

Microsoft says Type 1 has the advantage where system overhead is concerned. (Type 1 taxes the system less.)

DV Type 2 carries redundant audio and is better for certain "real time" operations. It requires more system overhead.

http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dx8_c/directx_cpp/htm/type1vstype2dvfiles.asp

The Ulead and the Adobe applications use Microsoft's "Direct Show" for video capture.

The Canopus capture software - on the other hand - uses the older Microsoft "Video for Windows" architecture for video capture.

There is:

1. Direct Show DV Type 1 (Ulead, MGI)
2. Direct Show DV Type 2 (Premiere 6.0)
3. Video for Windows DV Type 2 (Canopus)

DV codecs have absolutely nothing to do with capture.

The codec is not even involved in the process as Firewire transfer is simply a transfer of digital data. It's not the same as analog capture.

DV codecs only become involved when rendering occurs.

With Canopus DV Raptor...

...even when one *does* use the Canopus project settings, one will have to re-render audio when editing.

One can observe this to be true by placing Canopus-captured clips into the MediaStudio Pro timeline.

Whenever one sees the red (audio) *and* green (video) lines at the top of the timeline it means that portion of the timeline does not require rendering.

Each time a new Canopus-captured clip is added to the Ulead timeline, one will visually see the red line disappear. That means the audio has to be rendered. This does not occur with OHCI DV boards and Type 1 DV.

Canopus-captured DV files use PCM audio project settings instead of DV audio project settings.

[AndrewDV]

Hi,

I've read similar discussions about this on Canopus' support forum. Files from the Raptor have to go through an extra set of processing for DVD mastering.

The ADVC-100 does not use the same codec as the Raptor. It is jointly developed between Canopus and NEC.

Using MSP6.5, the files have been captured, edited, and mastered using the standard MSP6.5 settings. The same can be said as the DV-Bridge (nothing odd playing back on one or the other).