Open source VHS: how would you do it?

complete cobblers

NT

Reply to
tabbypurr
Loading thread data ...

I've been reading more and getting to understand the technology and what it can accomplish. As several people have mentioned, it's probably not worth trying to build the tape assembly, especially anything related to helical scanning. Some success may be had using stationary heads, with a scheme similar to compact cassettes, but I doubt one can put enough data onto those things to make it worthwhile.

I made a rough approximation about the amount of raw data that could be put to a VHS tape. I'm basing my calculations on System M (for reasons that will become apparent in a moment), but giving it a lot of slack. Assuming 400 scanlines, with 400 fields per scanline, 25 frames per second and 8 bits per field ("pixel") gives us a bitrate of 32 Mbps. A 2.5 hour tape can then be estimated to hold 33 GB of data. Such a tape would be labeled "DF300", giving an easy approximation of the data you can put on it. Same goes for DF360, DF420 and so on.

Is the estimate valid? D-VHS, the digital variant, puts 31.7 GB onto a DF300, 44 GB onto DF-420 and so on. In addition, comparisons to other formats (LTO, DDS) all feature the same rough numbers and the same rough timeframe (2000-2004). Therefore, I'd say the estimate is roughly valid. In fact, the estimate may even be underestimating the amount of data. I distinctly remember VHS having more than 256 colors, which means it encodes more than 8 bits of data per "pixel", but the big question is encoding and ECC. A few lines going wrong on a video don't mean much, but if every 400th block of data comes out of the tape drive bad, that's a big problem.

With that out of the way, the next big problem is usability and viability. So I asked myself how well does a hypothetical tape system compare to other external storage systems.

A big part of this is the cost of the medium. Where I am (Serbia), I can get the tapes for prices that range from $5/GB to $12/GB, depending on the deal offered. The competition, like external USB HDDs and thumb drives are really stiff. You can get them at about $8/GB, and they pack a theoretical transfer rate of 400Mbps with random access. That's hard to beat. On the other hand, CDs and DVDs go for about $1/GB (and random access) which is pretty much impossible to beat. I would basically need to pack 8 times more per length of tape to get those kinds of prices, which is a steep order. It's basically a three-generational improvement over the "base case". If this were a commercial project, it would be dead at this stage (but only buried after the next few paragraphs).

Fortunately, this is a "let's do something" project, so the insane price competition by the optical storage is not a real issue.

I spent time researching the actual mechanism of storing and retrieving data on magnetic media. I knew the basic principle involved, but not the actual specifics. It seems ultra-precise electromagnetic assemblies are used. I wasn't aware of the dimensions involved, but not that I am, hand making these things is simply out of the question. If I had a precise robot, it could make it for me, but I don't. Interestingly, it seems that when reading, these heads rely on the movement of the magnetic medium to induce the current (and voltage) in them that gets sensed. This poses a whole new set of problems, because not only would I need to precisely control the electronic component of the device, but I would also need to worry about the mechanical component. In a linear scan device (like LTO or compact cassettes), this can even be "faked" while relying on generous tolerances, but a helical scan device may be too complicated.

All in all, if using a helical scan, the best option is probably finding a device with working mechanical components and changing the electronics.

There is also an interesting business aspect of using ever-shrinking magnetic heads, in that at some point they become critical components. While not all of my numbers align, it seems the storage capacity of a tape depends solely on the physical size of the "signal" on the tape, which in turn depends on the size of the magnetic head. Depending on how the business is organized (are you producing your own heads or not), your profit margin could become hostage to the suppliers good will. Again, if this were a commercial project, this is the point at which it would be buried.

As for the rest, I saw an ad for an old VHS for 10 euros. The seller guaranties the device turns on, but says he does not have a TV and can not check if the device actually works. :) It's just a question of whether I want to do it or not.

If I actually do make something usable, I'll post some kind of a guide here. But it won't be a fast project.

Reply to
Aleksandar Kuktin

PAL VHS used 240 lines

I'm not clear what that means, pixels per scanline? I forget what VHS PAL r educed the video channel to, but it would be less than 400 pixels.

The error rate on VHS is very high, so you need some heavy duty error corre ction.

I'm not familiar with D-VHS. I presume that's 31.7G that gets lossily compr essed to rather less then put onto a tape.

IIRC colour level was entirely analogue. And of course of much lower bandwi dth than the luminance signal.

g

The error rate is much higher than that. As a matter of routine you'd have maybe 10 lines or so bad at the bottom of the picture when moving a tape fr om one machine to another. Then you'd get heavy dropouts, the odd second of misalignment now & then & so on.

.
3 hours to read 30G with no ability to do anything but a full tape sequenti al access. So only usable for backups.

I'm surprised by how much you pay for all this stuff.

You could have 50 here for nothing.

A 31.7G HDD costs what, 5 or 10 bucks?

NT

Reply to
tabbypurr

Industrial suppliers sell 'Flat o-rings' that are available in many different materials. They also sell flat and round material to custom make belts.

--
Subject: Spelling Lesson 

The last four letters in American.........I Can 
The last four letters in Republican.......I Can 
The last four letters in Democrats.........Rats 

End of lesson.  Test to follow in November, 2016 

Remember, November is to be set aside as rodent extermination month.
Reply to
Michael A. Terrell

HA, you got that right. ABC. Shame our choices are limited so, but I unders tand the smart ones do not want to run until after the crash. (and the libe rals think they're stupid LOL)

To the subject.

It is a totally fruitless exercise to attempt to use tape as a storage medi um today. It wears out when you play it, and it erases itself even if you d on't, more if you do. Access is clumsy at best, what if you need a file tha t is at the end of the tape ? It is extremely vulnerable to magnetic fields , even putting it near a loudspeaker.

I don't want to be a prick about this but the bottom line is if anyone want s to pursue this I have a bunch of these CD rewinders cheap, and they work on DVDs just as well.

Reply to
jurb6006

Oh dear. I forgot to convert currencies. To get the actual price for tapes, USB HDD and thumb drives, divide the quoted price by 100. CDs and DVDs are quoted in the converted value.

Really changes the price dynamic...

Reply to
Aleksandar Kuktin

I recently went through some of the backup I made in 2000 with regular burned CDR's. There was not a single unreadable file, in >10 disk I had to inspect to find the right backup file. All backups were zipped, so sensitive to single byte failure.

Groetjes Albert

--
Albert van der Horst, UTRECHT,THE NETHERLANDS 
Economic growth -- being exponential -- ultimately falters. 
albert@spe&ar&c.xs4all.nl &=n http://home.hccnet.nl/a.w.m.van.der.horst
Reply to
Albert van der Horst

Thanks very much for the report! It is good to know that these media are holding up.

Although, I did recently read some backups I made onto 9-track 6250 BPI tapes in 1993, and they came back perfectly. So, these tapes had been sitting in my basement for 23 years without loss. I did have to clean the tape heads after every tape, however, so they are shedding a LITTLE oxide.

Jon

Reply to
Jon Elson

Many people would love to know what discs you used then.

NT

Reply to
tabbypurr

partially unreadable zip files can still list contents.

--
This email has not been checked by half-arsed antivirus software
Reply to
Jasen Betts

The problem is to get access to a suitable tape drive... We already had that issue when trying to restore an old DLT.

Reply to
Rob

Back in the late 90s I had an interest in CDR longevity.

I eventually found a NIST(?) investigation that concluded that there were significant differences between all of the discs/chemicals.

Unfortunately manufacturers made it very difficult for a customer to specify the appropriate discs/chemicals, and their part numbers changed so rapidly that you couldn't get a repeat order of a known-good batch.

At that point I gave up trying, and relied on diversity, crossed fingers, and rotating rust.

Reply to
Tom Gardner

and one didn't know what was in the discs in the first place, unless they were blue.

I've found home burnt CDs & DVDs hopeless long term. Proper archival grade ones are of course a very different story.

Media longevity is less of a problem now that we all use multiple backups.

NT

Reply to
tabbypurr

Won't do any good. The manufacturers have been improving or consolidating for years. A bunch of (presumably very reliable) blanks I got about that era, are writable only at a '2x' speed, and modern drives don't SUPPORT that low a speed.

So, if my old burning station ever dies, I can't write any of that batch of disks.

Reply to
whit3rd

From what I remember, the colour didn't give you sufficient information.

Maybe, maybe not. But it critically depends on timescales.

It is already getting less easy to recover data safely stored on rust with an IDE interface. What about in 20 years time?

Reply to
Tom Gardner

To get a ballpark figure, remember that forty years ago a stereo digital audio signal (14 or 16 bit) could be recorded at 44.056 kHz or

44.1 kHz sampling rate on a video tape with a PCM adaptor (like Sony PCM-1).
Reply to
upsidedown

Really? The last tape drive we used was LTO which for the more recent models has media prices like that per TB rather than per GB as you state.

Reply to
Rob

Blue meant the dreaded blue dye, anything else who knows.

What's hard about that now?

NT

Reply to
tabbypurr

Aren't they being phased out in favour of serial interfaces. (I haven't bought a computer in 5 years, so I'm out of date)

Reply to
Tom Gardner

SATA has taken over but I've never found any difficulty in using PATA HDDs. New & old PCs with PATA are around, as are adaptors. I've still got PATA discs in use.

NT

Reply to
tabbypurr

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.