Prefered Capture Software?

[jriker1]

I am dubbing many many VHS tapes to DVD eventually however probably Divx or something similar before that to do some cleanup on the audio and video created. I did this a few times with my ADVC100 but now using the 110 after many years of not needing to dub things. I received a copy of Edius Neo with my unit. Is this good software for this purpose or can someones suggest something else? Cost is not a major issue when it comes to suggestions. Want to see what is recommended, then worry about cost.

Thanks.

JR

Edit: This is on a Windows based system by the way.

[donp326]

Scenalyzer is my choice. I have been using it for several years and it has performed well for me.

[GrassValley_CP]

Give EDIUS Neo a try. Though it is far more than just a capture program, it should do you fine for capture and editing. You can read about EDIUS Neo on its product page here.

[Meow]

I actually prefer capturing with VirtualDub. It's free, there's even a x64 for 64 bit Windows versions, and it captures the DV of your ADVC with the colorspace you want, and raw without dropping any frame.

Which brings me to a quick question;
What colorspace has the DV Format going through the firewire when used for 720x576 PAL on an ADVC300 ?

[GrassValley_BH]

DV is YCbCr 4:2:0 for PAL and YCbCr 4:1:1 for NTSC.

It's not very useful to capture the stream to any format other than DV, as it is already compressed as DV when it reaches the computer.

[Meow]

DV is YCbCr 4:2:0 for PAL and YCbCr 4:1:1 for NTSC.

It's not very useful to capture the stream to any format other than DV, as it is already compressed as DV when it reaches the computer.

I beg to differ. As DV is not a lossless compression format, I never do any filtering in it. So unpacking it as early as possible is actually a good idea if you need to work on the material a lot. That's also why I asked, since I then require to know what color-space to translate from.

Thanks for that info.

[THoff]

Even if you take the output of the ADVC an store it as an uncompressed file, you are not going to gain any quality -- you cannot create more detail than what is already there in the DV stream.

Furthermore, any transcoding to another lossy format will result in a degradation of the image and loss of detail, though it may not be visually obvious.

[GrassValley_BH]

Of course you don't want to recompress to DV after filtering, but there's really no reason to decompress the DV to an uncompressed file.

The end result is the same reading an uncompressed file generated from DV to RAM versus directly decompressing a DV file to RAM.

[Meow]

Exactly.
And as good as the built-in filtering of this unit might be, it still can't crop, do advanced de-interlacing and chroma-corrections etc. I have rarely, if at all, used the actual output DV material of the ADVC300 for anything other than "keeping the original somewhere just in case".
If you handle old VHS and Hi8 tapes, downsizing usually is a good thing for overall quality, especially when combined with filtering and such.

So yeah, I always recompress for distribution or DVD. Of course, I limit that to 1 time in the chain of things. A decent x264 encode is absolutely brilliant for ADVC300 captured material.

[GrassValley_BH]

So you're capturing the ADVC's output as analog?

[Meow]

So you're capturing the ADVC's output as analog?

Of course not. Why?

[GrassValley_BH]

Maybe I'm getting a little confused here so correct me if/where I'm wrong

1. You're capturing via FireWire from the ADVC, which means your system is receiving the compressed DV data from the ADVC.
2. You're decompressing that DV data and storing it uncompressed to your hard drive during the capture process.
3. You're filtering and working with that uncompressed data and storing the result uncompressed.
4. And finally you take that uncompressed data and compress it to whatever your distribution format requires (MPEG-2 for DVD, etc)

If I'm correct in the above, then there's a number of things to look out for in the process:

• In step #2, there's really no need to decompress and store the stream from the ADVC as an uncompressed file.
  If you're doing the decompression and saving during the capture process you're also putting extra load on your CPU and risking frame drops and/or synchronization problems without any reason.
• You should verify that your filtering/editing software can process video in YCbCr space, and not RGB. If it's processing in RGB space, then there's a colorspace conversion involved that can cause shifting.
  This is especially important for output to YCbCr distribution formats like DVD.
• You should also make sure that your filtering/editing software is decompressing (loading the case of uncompressed) the video into a working space that is the same or larger than the DV source. This means an YCbCr 8-bit 4:2:0 space for PAL DV, or an YCbCr 8-bit 4:1:1 space for NTSC DV. Ideally, you should use an 8-bit YCbCr 4:2:2 space if you can, or if you have an insane system (or tons of time) a 16-bit YCbCr 4:4:4 space.
  The larger space ensures the precision of calculations is maintained through the filtering/processing, even if the end result is a smaller sample space.
• And finally, make sure your filtering/editing software is aware that DV is not square pixel, especially if there are CG-type effects (circle wipes, titles, etc)

If all of those concerns are met, then you should be in good shape, but you still don't need to capture DV to uncompressed.

[jriker1]

Not sure if this was directed to me, however since I am the OP and this is similar to what I'm doing thought I would respond.

[LIST=1][*]You're capturing via FireWire from the ADVC, which means your system is receiving the compressed DV data from the ADVC.

Correct

[*]You're decompressing that DV data and storing it uncompressed to your hard drive during the capture process.

It is saved initially as whatever Sony Vegas 8.0 or the Canopus ADVC110 does as DV content. Don't really have a choice. When I do a cleanup routine with NeatVideo I try and save it uncompressed. Also playing with Lagarith.

[*]You're filtering and working with that uncompressed data and storing the result uncompressed.

Only saving it uncompressed once after the initial DV import for a single edit cycle. Will then be going to MPEG2.

[*]You should verify that your filtering/editing software can process video in YCbCr space, and not RGB. If it's processing in RGB space, then there's a colorspace conversion involved that can cause shifting.

Don't know on the color space since I'm not sure how Canopus ADVC110 or Vegas is creating it. Not sure if the software or hardware is initially responsible for the color space on the first DV import.

Thanks for your comments and look forward to any response to this.

Thanks.

JR

[Meow]

It is saved initially as whatever Sony Vegas 8.0 or the Canopus ADVC110 does as DV content. Don't really have a choice. When I do a cleanup routine with NeatVideo I try and save it uncompressed. Also playing with Lagarith.

Only saving it uncompressed once after the initial DV import for a single edit cycle. Will then be going to MPEG2.

Don't know on the color space since I'm not sure how Canopus ADVC110 or Vegas is creating it. Not sure if the software or hardware is initially responsible for the color space on the first DV import.

I've found that Vegas seems to indeed do a bitwise store of the stream from the ADVC300, use the correct colorspace etc., but you do need to have it configured so that it doesn't do any filtering right away.

Also, if you later on cut and crop the captured material, and then render it to a final stage, it will also internally be re-compressed by most video-editing software out there (even if you render/save it as DV again). You're not actually cutting compressed DV, you're cutting the unpacked data. That's why rendering takes so long. There are exceptions to that, for example when you only trim it, in which case you can do direct-stream copying in most video-editing software out there.