Hi8/Digital8 digital conversion options?
 

Hi everyone!

I found this forum via Reddit. I am currently looking for the best ways and services to transfer Hi-8/Digital 8 cassettes recorded on our old Sony camcorder (1998-2004ish) to a digital format. I was initially looking at purchasing another camcorder as we have not found our old one. However, I also wanted to explore digitizing services for this format due to time, convenience, and expertise being an issue. I searched a bit on here, but found only a few recommendations for tape instead of cassettes.

What current recommendations might you all have for a service that would convert our cassettes? Thank you all for bearing with me on this! Thank you.

For 8/Hi-8 use Hi-8 camcorders. With those Digital 8 what can play 8/Hi-8 you will not get the best result. However it depend on your capture system - it should be good. Digital 8 already is digital, so copy it through i.link (firewire). But keep in mind - you should learn how to clean camcorder heads first (remove head cover, remove and throw out original head cleaner and clean heads with paper or lint free swabs soaked with IPA or ethyl alcohol etc.). If you do not want to learn all that and buy necessary (and expensive) devices, better send your tapes to professional. Owner of this forum lordsmurf can help you.

Some members of this site can transfer these tapes for you such as myself, so feel free to reach out.

And we are currently accepting Hi8/Video8/Digital8 projects. :)

I've sent you a PM.

I've heard a variety of things in terms of what is "best" for Hi8. My understanding is that using the S-Video output of Digital 8 cameras that support Hi8 playback are just as good as Hi8 cameras. The D8 cameras being a later feature set would generally have newer technology in them, but it's unclear how important the quality of the analog playback was for those since they were more focused on the D8 technology.

Theres's also the pro Hi8 decks like the EVO-9850 and EV-S9000 which both have built in TBCs. I've restored a few of both, but I haven't really tested if they are superior to a lower hours Hi8/D8 camcorder when it comes to video reproduction. I will say a D8 camcorder is much less complex and likely to work without heavy restoration/recapping like the two models mentioned there almost always need. Pro decks are almost always "more used" than the average consumer camera, but the pro models do have "Hours Counters" within the menus on them so you can actually determine "how used" they are before choosing one.

Other argument for using a "deck" style player would be that it's probably easier to remove a tape if it happens to get stuck. Pro decks can also play back PCM audio whereas the consumer cameras can't, but odds are your content does not have PCM audio recorded to it if it was made by a consumer camera.

Benefit to using a service would be that they probably have multiple playback decks/cameras, some of which may play back better than others tracking/stabilization-wise.

And remember that some of us have decades of experience at these things. :wink2:

If you ever get into the hobby and see some of what goes into losslessly compressed capture and the difference between it and other methods you’ll understand why there prices are higher. Especially the ones with a ton of experience and gear.

You might as well start by getting a Digital8 camcorder that supports analog Video8/Hi8 playback (not all do -- see the list on Wikipedia). Then Digital8 playback will be as simple (or not!) as DV capture via FireWire, and you can decide for yourself whether the quality of the camcorder's own built-in analog to DV conversion is good enough for you.

I did not tried to capture 8/Hi8 trough Digital8 camcorder s-video output, because I have good Hi8 camcorders with built in TBC what I use for this purpose. But I tried capture through firewire simply to compare how it looks. And difference is huge. So firewire is for Digital 8 and nothing else. Difference is larger than between composite and s-video outputs, what sometimes, with good system, is minimal (however always noticeable). First what can see anyone (if compare) it is colour problem - it looks like dawn in the kingdom of Hades. Colorblind people, on the other hand, will notice the difference in noise level. ;)

The difference has little to do with the DV codec itself and a lot to do with the ancient ADC resides in those digital camcorders, So yes, a modern analog capture device from the S-Video output should produce better results than the internal conversion of the camcorder, its digital processing and the final DV encoding.

Most modern analog video capture devices are junk, using Chinese knockoffs of video ADC chips from the '90s and early 2000s. For example if you buy a brand new Dazzle USB video capture device, it uses a Chinese knockoff of the genuine Philips chip they were using 20+ years ago.

So, yes, I'd trust a Handycam from 2004 more than I'd trust most stuff made in 2024.

It seems to be from a religious rather than a technical point of view.
D8 camcorder 8/Hi8 conversion to DV quality is very bad, but you trust :)

By modern I don't mean chinese junk, decent capture devices from 2010's and onward with good reputation.

It is, I personally had to recapture my entire family tapes that I had previously done with a DV workflow, and it wasn't a camcorder, it was the Edirol VMC-1 which is far superior than a consumer camcorder.

If about modern cards, I have experience with about 3 models of Blackmagic (not loved here) cards and my opinion is - if Blackmagic dealer is your friend, you can buy new Blackmagic Intensity Pro and return for exchange if something is wrong. And it it 50% and even more possibility there will be some problems with analog capture, especially s-video. Problems are minor but different, some may not notice them at all if they don't know where to look. And if Blackmagic dealer is not your friend he will refuse claims, because "card is working" and Blackmagic has no support at all (even for dealers). Their politics is "if it does not work as you wish buy another until you get one that works". Terrible. But if you can test and return card if it is not work as it should you can buy it. Price is good and card is not bad. But they has some problems with new modern computers under Win10 (do not know about 11) - after about 1h capture they loose audio out (I tested 2 cards on 3 PC). Sound is still recorded, bet no monitor output. Maybe some bias settings should be made for that card I do not know. But anyway I use them under Windows 7 and all is fine. Now I have 2 good working cards and 2 are returned to dealer (my friend :)). Avoid Intensity shuttle (USB3) it has a problem with dramatic audio out of sync (no, not because of dropped frames).

How do you know the audio sync issue isn't due to dropped frames? Is the audio ahead of or behind the video? Typically dropped frames will result in audio that is behind the video and it'll get further behind the further you go in the tape.

Audio is behind of video. With analog capture. HDMI is OK. But it start right from beginning. You can eliminate (sometimes) it with PC restart before particular capture, but it is still here. As I wrote above Blackmagic has no support at all, even for dealers. So no answer why. No such problem with Intensity pro PCIe (both - older 2K and present 4K). With Blackmagic is simple - if it works as it should - use it. If not - do not waste time to try to understand why, leave it alone and use another. I do not know, maybe my particular shuttle is defective.

It was dropping frames.

No, total length of audio and video is correct (the same). In theory you can edit it in Adobe Premiere and manually sync audio to video (if you can match :)). But I never did it. When I found (years ago) this problem I simply stopped to use it. Together with Blackmagic dealer we made some tests, did not find an answer why and did not got an answer from Blackmagic support. Then I stopped to waste time to find reason "why" with Blackmagic problems.

That degrades the quality, tossing out ~50% of the color data, and adding blocks. It's far from ideal. DV was never intended to be used as conversion, and the quality of output shows.

Yep.

The choice is that binary.
It's not either
(A) 2020s junk (mostly USB, mostly Chinese)
(B) 1990s junk (ancient compression schemes like DV/MJPEG/etc)

There were several ideal options made between those extremes. Many good capture cards were released in the late 00s especially, during the peak of digital conversion. Namely ATI AIW, but also other ATI, certain Pinnacle (not Dazzle), Matrox, etc.

Video has those. :(

That VMC-1 has interesting uses, but capture isn't one of them.

BM is an HD card that "also does" SD, and quite poorly. It was never made to understand or tolerate lower-quality consumer SD signals, which is a shame. I remember when BM cards were announced (around 2009?), and had high hopes it could be an AIW-killer. It wasn't even close, and was pretty pitiful.

It is, it has to be.

Yep.

That doesn't matter. The audio can just truncate when the video streams ends. Then it would "match". But the video is shorter, for sync to be lose. That means dropped frames, which means lack of frame TBC.

Well, audio delay is about 3-5 sec from start to end within 3h. At start there is only hiss (similar to audio tape hiss), after video start and then audio with delay. And when video stops (after 3h) audio stops the same 3-5 sec later. Hard to imagine this is because of dropping frames.

P.S. It is constant audio delay.

In the analog ingest realm, true constant audio is often still caused by initial frame dropping. Sometimes by divergence in the workflow path (such as not routing audio through ES10/15).

If a mere audio shift exists, it can be adjusted in VirtualDub, by +/-ms. In fact, if this shift fails at sync, it proves in-recording sporatic dropped frames.

Given this discussion is about BM cards, I'm not convinced your experience as simple shift. BM cards dropping frames is a known known.

Audio shift is a real thing (could be looked at as video delay), and basically guaranteed if you're using a frame TBC since it buffers a frame or two which results in at least a 1/30th - 2/30th second of a delay, but it should be fairly constant throughout the capture. A few of the Snell and Wilcox TBCs will actually display what how much the video is being delayed through it live and it doesn't change much.

I've had a theory about the best way to know what your specific audio delay is in a certain chain would be to use a linear time code generator which visually burns in/overlays a generated timecode on each frame and also creates timecode audio. From there, you'd just run a stable source through it like a DVD player or pattern generator and capture both audio and video and then compare the decoded timecode audio to what is seen burned into the frame. The difference between the two at any given point is the video delay of that chain. Feeding it an unstable source shouldn't change the delay, but the benefit of using it here is that no frames should be dropped that would have the visual timecode burnt in.

The key with dropped frames though is that the audio should get further and further out of sync the longer the capture goes and will get more out of sync randomly, so in other words, if the video is 1 second ahead 15 minutes into the capture, it could be 5 seconds ahead at 30 minutes into the same capture if a bunch more frames were dropped between the 15m and 30m mark.

The different capture solutions compensate for dropped frames differently, so you might not get the audio sync problem with certain sets of preferences (even without a frame TBC) with some capture apps/configurations. Frames may still be dropped though.

Never had such a problem with PCI express BM cards, only particular USB. All was tested on the same PC with the same VCR and the same video tape. For capture and edit I use Adobe Premiere 6 where it is really not a problem to sync audio to video after capture. When I noticed USB version does not work as it should I simply stopped to use it. I have no experience with dropping frames by BM. At least Premiere capture log do not show something extreme. Yes, there are dropped frames sometimes, with damaged tape especially, but this is understandable.

P.S. I still have this card somewhere (I never try to sell someone something what is not working for me) so it would be possible to make some tests again if I will find it.

No, not that long. The frame delay is one field to one frame, so 1/59.94 to 1/29.97 (or PAL 1/50 to 1/25).

For 29.97fps, 1 frame equates to about 35ms. (Or PAL 40ms.)

Human perception doesn't start to notice audio divergence until you exceed 100ms, approaching 200ms. (FYI, AI chatbots give lower numbers, but AI chatbot are "artificial idiots". I sometimes run facts by chatbots, to see if they're accurate or crap.)

In effect, offset at this tiny ms is imperceptible.

This also assumes the source is 100% correct, which it never is. There will always be some % of lag, from the beginning, simply because of physics. Light travels faster than sound.

Do you simply drag the audio timeline over a by a few frames? Or do you do more? If more, then it's not a static/constant delay.

A good TBC generally won't allow drops here either. All tapes have some % of slop, as consumer formats were chaos on mylar. That is where dropped frames come from.

Video and audio was (1) the same length and (2) perfectly "in sync" if synchronized manually in Premiere after capture.
So it is simply audio delay, not "out of sync" completely.
But audio delay was 3-5 seconds for particular record. So if it was 3sec, then 3sec from beginning to the end. If 5 - the same. I didn't have the health to look into any reason for this, especially when Blackmagic itself couldn't even tell its dealer (my friend) anything relevant, except "try to restart the computer."

:laugh:

Old version of same advice = "Did you try to hit it or kick it?"

Aha! I remember those tube TV when this method worked (sometimes, until a certain point) :)

I thought that the card might be cursed, and I thought that I should call a priest (it was not clear which one, there are many of them and you never know or it will turn out worse).

But then my friend suggested to leave it alone and use the cards that work, so as not to accidentally release a demon from this one as a result of exorcism, not understanding where it will end up! ;)

Chemical degradation of tape with age is one thing. But there are still lot of people who found somewhere in attic their marriage tapes from 80s together with VCR. And what they are doing? You know it, I am sure! :) They connect this VCR to something, turn on and try to play tape. And it works! And it depends if suspicions something is wrong appears within 1 minute, 3 minutes or later, in result tape is damaged 1, 3 or more minutes from beginning. And in this case there are no TBC what could help so dropped frames are common.

"Works' is no the word I'd use. The tape gives an output, but the output looks like ass. :sick:

It doesn't have to be that way. This is why TBC exist.

This is the thing I should learn more about. In my experience, if tape is seriously mechanically damaged by transport (because of tape skew etc.) some dropped frames (at least in capture log) exist along with, of course, distorted image.

I went from a shuttle and a composite VCR to a 1980p ag, Pinnacle card and a frame TBC. When I used a regular VCR with the shuttle it dropped a bunch of frames. When I replaced the regular VCR with the 1980p ag it dropped way less but still dropped.

When I got my Pinnacle card I hooked it up to a regular VCR with no TBC and noticed it handled the signal way better than the shuttle. It dropped too but not as bad. With my frame TBC in the workflow I get no drops.

Yesterday I had a VHS-C SLP tape that squealed when I played it and the tracking was awful. I captured once then thought maybe a transplant would help so I transplanted it to a full size cassette and captured again then as a last resort I put it in my composite cheap deck and messed with the guides and it didn’t track well no matter what but I had no dropped frames and no bouncing around when capturing. I never did capture with my composite deck I just saw how it wasn’t helping with the tracking in my CRT and decided not to do that capture. Only when it was on my CRT with the composite deck did the picture bounce around. I’ll peace together the two captures because that is one of the more important tapes I have. Or maybe I’ll just use the one after the transplant. It has like 66 lines at the bottom and top I can crop to hide the bad tracking lines because it was shot with a VHS widescreen look. It’s not perfect but I was happy with it because I feel like a lot of transfer methods couldn’t of got that footage with no drops and the picture being really steady like that or maybe couldn’t of got it at all.

In this case is one question about VHS tapes (if they squealed). Analog audio tape squealing is commonly caused by one of 2 factors - SSS (sticky shed syndrome) or lubricant dried out (mostly cassette tapes, this is why sometimes Studer A80QC is the only way to reproduce cassette tapes). SSS can be cured by baking or in dehydrator (food dehydrator works :). But some cassette tapes (mostly Ampex) sometimes can be played back only after additional lubrication with silicone oil (and after wind out on spool and on Studer A80QC). Terrible precedure, but if it is unique recording, no way. So what kind of degradation cause squealing of video tapes?

Maybe squealing was the wrong word. It made this noise.

Squealing is usually sticky shed on U-Matic/VHS/Betamax also, but I've yet to encounter it in person on anything other than U-Matic though.