canopus v other devices

[edDV]

Yes, that is why the EIA resolution chart has similar wedges within in a circle so that horizontal and vertical resolution measuerments are normalized. 16x9 wide video in the same bandwidth would cost proportional horizontal resolution but that goes beyond VHS.

[lordsmurf]

DV is not an optimal method, no. The colorspace compression can and often does muck up VHS input, although it's mostly a problem for cheap boxes, including the ones from Canopus (yes, that includes the ADVC-300, too).

While the "problem" of NTSC DV is well known, I don't see any problems with PAL. I've thrashed the inputs of my ADVC110 with some extreme colour test charts, and it behaves impeccably. This includes levels and frequencies way beyond anything you could get from VHS.
What problems are you seeing with VHS?
Cheers,
David.

PAL is fine, correct.

[lordsmurf]

I bet (S-)VHS captured in 2001 to lossless, filtered as best as was possible back then, and encoded to DVD as best as was possible back then, will look worse than my (S-)VHS > DV > DVD efforts now.

This is true for me, but it's more an issue of skill growth over a period of 8-9 years than anything else. It's not hardware related. There have been SOME advances in software, but not a lot -- not like most people would imagine.

Plus in a few years time the processing will be even faster and easier (and maybe slightly better too).

I doubt it. Every time faster hardware comes along, more gluttonous software comes with it.

DV or lossless - it still looks like (S-)VHS. With a decent deck+TBC, decent de-noising, and gentle sharpening, it'll look like surprisingly stable, clean, and sharp (S-)VHS - but it'll still look like (S-)VHS. Using lossless in place of DV won't help that at all.
Obviously lossless makes perfect sense if you capture, filter, encode, and burn tape-by-tape. The only advantage to DV then is that, in general, people seem to make it work with less hassle. However, if you intend to capture, and then store all the files, working through them at your leisure, it has the obvious advantage of size.

There's something to be said for high bitrate MPEG-2, or even H.264 with moderate bitrates.

Also, if you really care about what you've captured, it has the advantage of being able to dump it to DV tape if you really want, which (should your HDD backup strategy fail) you (or someone) should be able to play in 10-20 years time with slightly less difficulty than procuring a new decent transfer of VHS at that time.

DV tape has the same physical problems of VHS tape --- it's still a tape. More on longevity of tape at http://www.digitalfaq.com/guides/media/longevity.htm ... consumer tape, be it DV or VHS, is still made like crap. It's never been a long-term archival solution.

[danno78]

More reviews about DV converters. Are in japanese, use google translate or babelfish.
http://power-m-web.hp.infoseek.co.jp/index.html

Canopus ADVC300 review:
http://babelfish.yahoo.com/translate_url?doit=done&tt=url&intl=1&fr=bf-home&trurl=http...rUrl=Translate

Canopus ADVC100, more pages:
http://babelfish.yahoo.com/translate_url?doit=done&tt=url&intl=1&fr=bf-home&trurl=http...rUrl=Translate

[lordsmurf]

Did I read that correctly? Those Canopus tests also showed and discussed colors/hues being altered? This is what I've been telling people for years now -- the colors don't necessarily stay true when processed by a Canopus box. It doesn't matter what DV codec is in use, either.

[edDV]

Did I read that correctly? Those Canopus tests also showed and discussed colors/hues being altered? This is what I've been telling people for years now -- the colors don't necessarily stay true when processed by a Canopus box. It doesn't matter what DV codec is in use, either.

That needs explanation. My ADVC boxes have never failed to place recorded color bars dead center on the V-Scope boxes. I mean dead center on the 2% boxes. The color behavior of DV is exactly like uncompressed 4:2:2 or MPeg 4:2:0.

In my experience most issues of "color shift" (99% plus) turn out to be luminance errors. Remember all the time wasted on out of spec DVD recorder black levels?

This can all be shown with rational experiments. Show me an example where the DV format causes "color shift".

And as far as the ADVC DV codec, it tracks the Sony VX2000/PD-150 exactly.

[jagabo]

Show me an example where the DV format causes "color shift".

Of course, it wouldn't be because of the DV format, but rather particular DV devices who's A/D's aren't functioning properly.

[edDV]

Show me an example where the DV format causes "color shift".

Of course, it wouldn't be because of the DV format, but rather particular DV devices who's A/D's aren't functioning properly.

Yes difficult to transfer an analog original. Maybe FedEX?

[lordsmurf]

Show me an example where the DV format causes "color shift".

Of course, it wouldn't be because of the DV format, but rather particular DV devices who's A/D's aren't functioning properly.

Yes. Although the DV Storm cards apparently have the same Canopus DV encoding chips, those cards never get the same observations. That too, was in the above (Japanese?) article. Like we've seen with other devices, the Canopus ADVC boxes may be all over the map. It's either design flaw or bad QC. Another example is the DataVideo TBC-1000's, where there have been noise bar and softness issues reported, but it varies owner to owner.

Higher end DV conversion is okay. It's the video camera passthrough and lower-end boxes like Canopus ADVC that appear to be craptastic at quality.

[edDV]

Show me an example where the DV format causes "color shift".

Of course, it wouldn't be because of the DV format, but rather particular DV devices who's A/D's aren't functioning properly.

Yes. Although the DV Storm cards apparently have the same Canopus DV encoding chips, those cards never get the same observations. That too, was in the above (Japanese?) article. Like we've seen with other devices, the Canopus ADVC boxes may be all over the map. It's either design flaw or bad QC. Another example is the DataVideo TBC-1000's, where there have been noise bar and softness issues reported, but it varies owner to owner.

Higher end DV conversion is okay. It's the video camera passthrough and lower-end boxes like Canopus ADVC that appear to be craptastic at quality.

At least bound the issue. The Japanese article is showing errors well within 5% and only in the mid luminance levels at certain luminance frequencies. The "errors" are due to flatness of luminance frequency response as a function of luminance levels. Their test methodology and results are difficult to understand from that translation but may be caused by the Philips A/D running in 8 bit mode rather than 16 bit. These kinds of errors would be in the subjective ranges you are color correcting anyway and would have little impact on a VHS-> DV capture since the VCR and TBC would introduce far greater errors than those shown in this test.

Then it seems the ADVC-300 in DNR3 mode is somehow significant. I'd like to see similar comparison tests for typical consumer Conexant based capture and hardware encoding cards. These Japanese guys tested only about a dozen devices with their fancy sweep generator back in 2000-2004 then seem to have lost interest.

[juhok]

edit: Deprecated content removed.