JVC DR-M10SE performance and recording issues

[jonaszoon]

I spent the last month searching for the best way to digitize my old videotapes. Having read most of the relevant topics here and still not sure which route to take, I had the luck (or so it seemed!) to ran into a JVC DR-M10SE for 20 euros. I bought it, despite knowing of the technical problems surrounding them.

The trouble is that most of the tapes I want to digitize are VCC tapes, an obscure format primary sold in (certain parts of) Europe. The VCC VCRs are old and not that advanced. I believe there were two generations of recorders made, the one I recently bought is a second generation recorder which includes SCART-output, instead of the old DIN AV (or whatever it's called) and apparently even has a noise reduction system. On playing, most of the tapes seemed in fairly good shape (e.g. no tearing), but there was a lot of noise, especially visible in the color red.

So my setup is pretty simple: The VCC VCR (Philips VR2324) connected via SCART with a JVC DR-M10SE. The tape I used for testing is a high grade tape, recorded some 20 years ago (possibly the most recent VCC recording I have). On it, a Rolling Stones concert, which I don't really care about (in case the tape gets worn out).

However, there are two problems. I'm not quite sure which is worst. Problem one is that the JVC simply stops recording after about half an hour. It locks up and the display shows "Loading", just like the buggy units that won't start at all, yet only during recording. Until now I have only used (new) Verbatim 1-2x DVD-RWs. The other problem is that the recording doesn't look that great. Naturally, I've asked myself the classic question: "Am I expecting too much?". But apart from looking roughly the same as the original recording (although no signs of line TBC, I think), it induces severe macroblocks. There's also some sort of chroma noise, which is not directly visible when watching the recording on television. I used the FR180-mode for recording. Could it be that the (subpar?) performance of the JVC is related to excessive noise on the input signal? Is it possible that the signal doesn't get processed (as in: JVC NR magic, etc.) at all? I've attached a still of the recording.

[lordsmurf]

"induces severe macroblocks" = caused by excess noise
"Could it be that the (subpar?) performance of the JVC is related to excessive noise on the input signal?" = yes
"Is it possible that the signal doesn't get processed (as in: JVC NR magic, etc.) at all?" = no
"ran into a JVC DR-M10SE for 20 euros" = nice find
"It locks up and the display shows "Loading"" = read fixing the JVC DVD recorder "LOADING" issue
"Verbatim 1-2x DVD-RWs" = good choice

[aedipuss]

if the unit has a higher bitrate(shorter recording time) mode i would try it and see if the results are better. 3 hours on a sl dvd is too low a bitrate without multipass vbr encoding.

[jjeff]

I agree with above, 3hrs for a noisy source like tape is too much. 3hrs for a clean source like a commercial DVD is doable(although lately I've been more inclined to not go over 2hrs 40 minutes) but with a noisy source much of your bandwidth is wasted trying to preserve all the noise, which leaves less for your actual program.
I believe your JVC is supposed to have some good noise filtering which will help with lower bit rates, but again 3hrs is probably just too slow. For tape sources SP is really your best compromise speed.

[lordsmurf]

That's not true at all.
Read this: http://www.digitalFAQ.com/forum/showthread.php/sp-mode-best-1507.html

3 hours (FR180) should look better than 2 hours (SP), higher allocation. FR180 = XP, for visual quality.

Don't assume 720x480, that's a flaw of Panasonic and others (stupid settings).

[jjeff]

Good point, and since his VCC tapes are probably more like 352x480 no sense tying up bandwidth encoding in full D1.
Exactly what speed/setting does his JVC switch to 1/2 D1?
As you probably know on a Panasonic one must set FR for more than 4hrs(even 4hrs/1min) to get it to switch to 1/2 D1. In this case you really gain no bandwidth by playing with the speed. I believe Pioneers are the same but drop resolution at a faster speed than Panasonic, so you may be able to gain bandwidth by selecting the speed right after the drop.

[lordsmurf]

FR175 = 352x480
I don't really know what the hell Panasonic and Pioneer were thinking with 720x480 @ 3 hours.

[jjeff]

What about Panasonics 720x480 @ 4hrs
I really wish DVDRs would have a simple setting allowing the user to set exactly where they want the switch to be. With a clean source and very little fast movement I think 4hrs at full D1 can work, but other times such as noisy tapes or analog cable, as you said even 3hrs(or less) turns to a macroblocking mess.

"I don't really know what the hell Panasonic and Pioneer were thinking with 720x480 @ 3 hours."

Marketing, most people just read the box and if it says 500 lines of resolution at 4hrs they don't realize their is a downside to fitting so much time with such little bandwidth. Same thing with some DVDRs that advertise 10 or even 12 hrs/DVD If you've ever seen 12hrs on a SL DVD it sure ain't pretty, probably VCD quality if that, but most people don't think of the downside. Even Panasonics brag about 8hrs/DVD which look like crap. 6hrs is OK for low quality talking heads but please, not 8 hrs! At least with Panasonics you can tell the DVDR(in the setup) to not allow the 8hr setting and max out at 6hrs, which is bad enough.

[jonaszoon]

Thanks for your replies so far. I looked further into the noise problem. For instance, I opened the VCR (which was not the most easy task, given the lack of information on the 'net) and checked out the heads. They looked clean and in good shape (no rust). Nevertheless, I tried cleaning them, however nothing serious came off. So I suppose they weren't dirty to begin with, although the VCR itself definately was. I begin to fear that the problem is purely the degradation of the tapes. Or could there be some other factor? And in case the noise is permanent, would it be better to use something else than the DR-M10SE for digitizing the VCC tapes?

[lordsmurf]

I doubt "something else" would help. Probably just look worse.

What you have there isn't awful, it can be further processed in VirtualDub and/or Avisynth.
For example, Fixing Chroma Noise -- http://www.digitalfaq.com/forum/showthread.php/vhs-chroma-flaws-565.html
Same techniques would fix it.

[jonaszoon]

I doubt "something else" would help. Probably just look worse.

What you have there isn't awful, it can be further processed in VirtualDub and/or Avisynth.
For example, Fixing Chroma Noise -- http://www.digitalfaq.com/forum/showthread.php/vhs-chroma-flaws-565.html
Same techniques would fix it.

I agree that it's not awful. When playing it on my television (old CRT) the difference between the tape and dvd is indiscernible, certainly not worse. It's just that I'm a bit hesitant when it comes to artifacts induced by lossy compression, especially in case of future post-processing. I'll check out the techniques you mentioned.

Also, I checked your thread about the "loading" issue, quite an interesting read. The thing in my case is that the error only occurs during recording and always after about half an hour. Same thing is mentioned in this thread (listed in your thread): http://www.digitalfaq.com/forum/showthread.php/dvd-recorders-jvc-2000.html But the unit there ends up wrecked. It doesn't seem to be related to the power supply problem. Possibly a drive issue?

[lordsmurf]

Drive issue could be the case, as could inferior blank media.

[jonaszoon]

It seems the plot thickens. I hooked up the JVC to my DVB-T decoder and started recording. No errors at all, more than three hours of video on a disc. Same recording mode, same SCART cable, same discs, only different input. Then I tried it with the VCC again, yet after 40 minutes the JVC crapped out. So what could be the problem here?

[lordsmurf]

You may need a TBC between the VCR and DVD recorder, as the signal is "unclean".

[Bix]

FR155 (2:35) is the highest bitrate half D1 setting on that recorder, not FR175.

[lordsmurf]

My DR-M10SL is 720x480 all the way to FR165 for sure.
I don't recall FR170.
But I know FR175 is Half D1, so that's why I said "FR175"

Just check your own machines, to be safe.

[Bix]

Was there more than one DR-M10S version in the US? The one I used briefly was definitely Half D1 at 2:35/FR155. I even used it for at least one VHS conversion that was just over 2 hours, since the bitrate allocation was so much better than SP.

[lordsmurf]

Yeah, there were some conflicting findings between my Japan unit, and my friend's China unit, when it came to the Half D1 switch. That was like 5-6 years ago when we did the comparisons, I barely remembered it until this thread. Nothing else appeared to be different. I can always re-test mine, in case mistakes were made, but I'm pretty sure it was still 720 @ FR165, on this machine I have here.

Will be a while, I'm too busy for testing right now.

[jonaszoon]

In the meantime I bought a Datavideo TBC-1000. I also bought a Philips VR1100 VHS (30 euros this time!), which seems to be a JVC clone of some sort: built-in TBC/DR, etc. Sadly, the latter can't be used as a passthrough (how I tried!), but works great for my VHS tapes.

Anyway, the TBC was said to fix the JVC/VCC recording issues and I'm pretty glad to say it does! The DR-M10SE records without any interruption now and the "digital static" before the recording is gone too. Thanks Lordsmurf!

Some findings along the way:

• The VCC VCR (without external TBC) does never ever generate a copy-protection signal, and I came across some really horrible tapes. Could it be it has some sort of built-in TBC too? I spoke to some guy, who claimed that "back in the days", one could throw a VCC recorder in the chain to beat copy-protection. But I don't take his word for it.

• The VHS VCR (without external TBC) often causes M10 copy-protection "blockage" between tapings / on unused portions of the tape.

• The VCC noise suppression I mentioned in my first post has to do with audio, not video.

• On another forum, they told me that the (great amount of) noise is mostly inherent to the VCC system. Tapes can be flipped, so only half of the tape is scanned during each side pass. Less space for data, faster degradation, more noise; something in the lines of that. One also said replacing the capacitors of the VCR power supply could help, but I'm not going there yet.
• I'm in PAL-land, so half D1 is 352x576. Is this a significant difference in regard to recording modes / time?

• The M10 also craps out when recording off my Videopac, a 70's console sold in the US as Magnavox Odyssey2.

• A somewhat remarkable side-effect of the TBC is that the chroma-noise seems less worse. How could this be? Also, it's clear that it darkens the video and renders it a bit less sharp. Other than the TBC, I am also using different cables and a SCART-to-composite converter, since the TBC doesn't have a SCART input/output.