VHS-C picture quality compared to VHS

[artemisc]

I'm trying to digitize the VHS home videos my parents recorded over the years. My understanding is that their camera recorded onto VHS-C tapes, then they copied the footage to regular VHS. Many of the tapes have multiple copies. I'm recording every copy of every tape and only saving the recordings with the highest quality (because some of the copies are not in good condition).

I thought, since the VHS-C tapes are the source, they would be the highest quality. But I was surprised to find that all the VHS-C tapes are in some ways inferior to their copies. Their video quality is blurry and less detailed and the audio is balanced better in the VHS copies (speaking is significantly louder than background hiss in the VHS recordings). The only thing that really makes the VHS-C tapes better is that sometimes they have segments at the beginning that were cut out of the VHS copies.

I've attached an example of the difference in picture quality (not the best example of the audio difference, it sounds good here and other tapes are much worse, but I'm mostly asking about the picture quality right now).

See how the detail of his beard isn't as clear in the source tape? The VHS is grainier, but more detailed. I've tried to recreate the detail by sharpening the VHS-C recording, but it ends up looking worse than the VHS tape and still with less detail.

My questions are...
1) Is the extra detail in the VHS tape just an illusion caused by the change in format and/or enhancement settings on the VCR that was used to copy? If it's not an illusion, how could the copy possibly have more detail than the source?
2) Which one of these recordings should I save? The VHS-C has a segment at the beginning that the VHS lacks, but if I append that segment to the VHS recording, the sudden jump from blurry to more detailed quality looks insane.
3) Is there any way I can just make the VHS-C recording look more like the VHS recording using VirtualDub? I've only tried sharpen and that didn't do it, but I'm not sure what to try next.

[Alwyn]

[Attachment 83452 - Click to enlarge]

Definitely some image "modification" occurred when copied to the VHS. Look at the noise on the walls and ceiling. Also, the edges are much more defined; the experts will probably expand on that but I'd say the VHS VCR was set to sharpen and other enhancements to the image.

I think, from a "raw" or "baseline" point of view, your VHS-C capture looks better and would come up better. On the other hand, some areas (eg the bloke's beard) could be slightly lacking in detail in the VHS-C video; the result of old tape age (although I assume the full-size VHS is about the same age...)?

[timtape]

Yes so long as the original tape isnt damaged, it should have the best picture and sound. Analog copies are always inferior to whatever they are copying but the difference can vary. Yes it's not possible to extract more picture or sound information from a recording than is actually there. But it's very possible to extract less.

There does seem more detail (and more noise) in the VHS copy picture. The extra noise is to be expected but not the extra detail. Are you sure you dont have the captures reverse labelled?

Increasing contrast gives the impression of more detail (and more noise) and sometimes that's necessary but it's not really increasing detail.

Perhaps the VHS C transfers were softened at capture. Is it possible that between capturing the VHS-C and the VHS copies the front panel "sharpness" setting on your VCR was changed? Or settings on a proc amp or TBC in the capture chain?

[lollo]

An analog copy cannot be better than the original. That's a fact.

In your captures, the vhs copy is better in term of "details", while the vhs-c is blurred.
Here a comparison: https://imgsli.com/MzE3NzY0
Another one: https://imgsli.com/MzE3NzY4

This depends in part on the over sharpening performed by the player in the vhs copy capture, you can see a clear oversharping halo here:



but that does not explain all. Even with advanced post processing is impossible to remove the blur and match the details of the two captures. There is also somehow a "distorsion" in the vhs-c capture.

So my impression is that something just went wrong with the vhs-c capture, i.e. bad adapter, different hardware (BTW, what is your workflow?), different day of the week (just joking), ...

[dellsam34]

I see two compounded scenarios/issues here: First, the player used to playback the VHS-C tapes back then obviously was better than your consumer camcorder, it had better details to begin with. Second, the VHS copy obviously undergone some analog processing using probably some pro or semi pro processing/TBC unit, while sharpening brought some details it created unnecessary white edges or halos from over sharpening.

So what you can do? In the first scenario you would need a S-VHS VCR with line TBC and outputs separate Y and C. In the second scenario, there are a lot of post capture tools that can do better than that analog-digital-analog processing unit used to create those VHS tapes.

[timtape]

Re the audio, the copy is about 2db average louder than the original. In blind testing people tend to subjectively prefer the louder source.
After that 2db difference is corrected, there's not much in it IMO. Whenever subjectively comparing two audio sources, it's best to try to match the levels first, preferably to within say 0.5db.

It's harder to compare the baseline noise levels. We would need a sample quiet passage.

You say the samples already uploaded dont really show the audio differences well. Feel free to upload samples of the worst VHS C audio compared to the copy.

[artemisc]

An analog copy cannot be better than the original. That's a fact.

In your captures, the vhs copy is better in term of "details", while the vhs-c is blurred.
Here a comparison: https://imgsli.com/MzE3NzY0
Another one: https://imgsli.com/MzE3NzY4

This depends in part on the over sharpening performed by the player in the vhs copy capture, you can see a clear oversharping halo here:


[Attachment 83456 - Click to enlarge]

but that does not explain all. Even with advanced post processing is impossible to remove the blur and match the details of the two captures. There is also somehow a "distorsion" in the vhs-c capture.

So my impression is that something just went wrong with the vhs-c capture, i.e. bad adapter, different hardware (BTW, what is your workflow?), different day of the week (just joking), ...

Yep, you're right that something went wrong with the VHS-C capture. The whole time I thought the VCR's picture settings were set to "EDIT"... turns out it was set to "AUTO", so I guess the VCR decided the VHS-C picture needed to be softened for some reason, and it probably sharpened the VHS recordings. I re-recorded the VHS-C on EDIT and all the detail I would expect to see is there Silly me.

On the other hand, that audio issue I mentioned is still there. I boosted the volume on these two samples for a more fair comparison. The VHS-C audio sounds worse to me, there's a louder hiss than the VHS version. Any ideas why that might be? I looked in the VCR settings and don't see any audio-related settings that would affect this.

I'm using a JVC HR-S3800U, Panasonic DMR-ES10, Hauppauge USB-Live 2, recording with AmaRecTV, and editing with VirtualDub... if any of that matters.

[artemisc]

Re the audio, the copy is about 2db average louder than the original. In blind testing people tend to subjectively prefer the louder source.
After that 2db difference is corrected, there's not much in it IMO. Whenever subjectively comparing two audio sources, it's best to try to match the levels first, preferably to within say 0.5db.

It's harder to compare the baseline noise levels. We would need a sample quiet passage.

You say the samples already uploaded dont really show the audio differences well. Feel free to upload samples of the worst VHS C audio compared to the copy.

Here's a good example of the audio difference. I tried to make the volume levels equal.

[dellsam34]

Some JVC VCRs are not great reading the linear track, The dubbed VHS version sounds better because the player used to playback the VHC-C tape was probably better, You can re-align the AC head for better audio if you know how to put it back to the original position, Also make sure the AC head is clean.

[timtape]

Here's a good example of the audio difference. I tried to make the volume levels equal.

Thanks for the audio examples. The VHS-C example only extends up to about 5 kHz. It has a big hole in the speech clarity part of the spectrum.

The VHS copy is better in that speech clarity range and goes up to about 7 kHz.

But SP VHS and VHS-C can go to 10 kHz. I suspect both audio captures fall well short of what's on the tapes, both VHS-C originals and VHS copies. Tomorrow I'll try and upload some sample linear audio from an old VHS-C camcorder tape for comparison.

Linear audio on slow speed tapes like this has been one of my special interests because extracting the best from such tapes is not straightforward and I've been trying first to educate people about how much better audio fidelity can lie on these tapes, even when shot on modest VHS-C cameras.

Unfortunately unlike the great declicker software tool I demonstrated on your other thread, getting the linear audio sounding right cant be fixed with a software tool post capture. It has to be done at the playback machine.

[dellsam34]

Don't forget that there are other factors that can affect the audio quality during recording, the quality of the microphone, The distance from the subject, the microphone isolation from vibration and noise, the state of cleanliness of the heads during the shoot, The quality of the audio preamp and the recording circuit ...etc. Not related to the subject of this thread but my point is that specs are not rules.

[lollo]

Yep, you're right that something went wrong with the VHS-C capture. The whole time I thought the VCR's picture settings were set to "EDIT"... turns out it was set to "AUTO", so I guess the VCR decided the VHS-C picture needed to be softened for some reason, and it probably sharpened the VHS recordings. I re-recorded the VHS-C on EDIT and all the detail I would expect to see is there Silly me.

It makes sense, thanks for the reply. Use edit mode for both captures from now.

I'm using a JVC HR-S3800U, Panasonic DMR-ES10, Hauppauge USB-Live 2, recording with AmaRecTV, and editing with VirtualDub... if any of that matters.

OK, thanks, this will provide nice captures. I did not recognize the histograms of the USB-Live 2, that's why I asked.

Concerning the audio issues, with timtape and dellsam34 you're in good hands!

[timtape]

Don't forget that there are other factors that can affect the audio quality during recording, the quality of the microphone, The distance from the subject, the microphone isolation from vibration and noise, the state of cleanliness of the heads during the shoot, The quality of the audio preamp and the recording circuit ...etc.

Of course. All those factors and more can affect the recording. So we are careful to not make the sound any worse in our playback. For that we can control.

[timtape]

Some JVC VCRs are not great reading the linear track, The dubbed VHS version sounds better because the player used to playback the VHC-C tape was probably better,

The player may even have been the VHS-C camcorder which originally recorded the tape.

You can re-align the AC head for better audio if you know how to put it back to the original position...

Yes it seems very likely the main problem here is AC head azimuth misalignment.The black band in the upper trace, (circled in green) represents a loss of signal. We can see it intact in the copy transfer below it. This is classic audio comb filtering. I cant think of any other reason for the particular audio loss as shown in the spectrum trace below. One band of frequencies.

When there is such a loss in such a band , (3 to 4 kHz) we can expect there to also be losses in higher frequencies, which we can also hear in the audio and see in the trace. The highest captured frequency extends to about 5 kHz. Whereas the VHS copy extends to about 7 kHz.

Again, likely the copy is also somewhat azimuth misaligned, either in the playback on the OP's JVC, or at time of copying VHS-C to the VHS, or a bit of both, just not as badly in sum. Unfortunately any error at time of copying from VHS-C to VHS is now 'baked in' to the copy and cannot be removed. Whereas the original VHS-C tape will have no such error. Of course we could recapture the VHS copy with linear audio azimuth aligned but there would be no point as best picture and sound will be in the original. So long as the azimuth error re the VHS-C tape is adjusted out at the JVC at time of digitising, the full audio spectrum should be extracted.

And as already covered, any other audio 'generational losses' will be bypassed when capturing from the original tape.

Important: aligning linear audio azimuth on one tape recorded on one camcorder or VCR, should make it automatically aligned to all tapes recorded on that same camcorder or deck, or close to it.

[timtape]

Sorry, duplicate post.

[lordsmurf]

Remember, there's no rule that you must use the same VCR for both audio and video. I have often taken JVC video, and merged with Panasonic/others linear audio.

But when merging like this, TBCs are extremely important, so the timings match.

It's also not a given that JVC is always bad at linear. In many cases, it's not, and can be made better with audio head tweaks. Sometimes Panasonic others is worst, not the JVC. Lots of factors go into it.

[timtape]

This thread is one of those cases where the OP had a copy tape which had better sound than the original tape, or appeared to have better sound. If he didnt have that copy tape he may never have suspected there was anything wrong with the original tape or his transfer of it. Hence the OP's thread here, as it pertains to both picture and sound quality.

While azimuth is only one factor it's usually significant, sometimes crucial. Adjusting azimuth is interesting. We can never be 100% sure whether the audio head azimuth is correctly aligned to the tape being played for best sound unless we check it. Which means adjusting it. Adjusting it into alignment, past alignment on the other side, then rocking it a little until we're finally sure it is at that true peak. There is only one adjustment point which is correct. It's not a matter of opinion. It's like manually adjusting the tracking control on some pro and semi pro VCR's for strongest video head tracking.

Trying the tape on a different deck, or a few different decks and choosing the one which sounds the best is unlikely to yield a top result. Unless we always adjust azimuth on each tape, or at minimum on each tape recorded on a different deck or camcorder, we will never know if it could or couldnt have sounded better.. with regard to azimuth.

That's not to say everyone can or should use this method but IMO those of us claiming to do top quality transfers should be checking azimuth objectively even if the customer never knows we have. Just part of a premium service.

[lordsmurf]

but IMO those of us claiming to do top quality transfers should be checking azimuth objectively even if the customer never knows we have. Just part of a premium service.

But that's too involved of a process to do with every tape, without adding significant costs to the customers. It's only problem tapes where I pull out decks owned (and customized) solely for tweaking alignments.

[timtape]

but IMO those of us claiming to do top quality transfers should be checking azimuth objectively even if the customer never knows we have. Just part of a premium service.

But that's too involved of a process to do with every tape, without adding significant costs to the customers. It's only problem tapes where I pull out decks owned (and customized) solely for tweaking alignments.

The simple custom mods I made for my own machines make adjusting azimuth as easy as adjusting the factory fitted picture tracking control. No need to open up the deck and poke around with a screwdriver! Safe and quick. Done in seconds.
I knocked up the external adjuster in an hour.