ArVid: Russians squeezed 4 hard drives into one VHS tape in the 90s
jacobfilipp.com
jacobfilipp.com
44.1kHz lets you store all frequencies from 0 to 22.05kHz, which ought to be more than enough for most humans.
Filters with a gradual roll-off are much smoother, cleaner and cheaper. 44 kHz allows a filter with only 20dB roll off starting at ~16kHz to be used instead.
Filters are necessary because if you digitize something at 44kHz, any noise above 22kHz becomes noise at a frequency < 22kHz, aka audible.
If you actually weren't able to hear past 16kHz you wouldn't be able to hear the ringing caused by a good filter above 16kHz, and you wouldn't be able to tell a difference in the transients either.
Alas I can still hear 1khz.
Handy resource:
https://www.szynalski.com/tone-generator/
Note that higher frequencies are very directional and you may have to move your head around a bit to catch the tone, this works best by finding something you just can't hear and then to move your head around to see if you really can't hear it or if you only can't hear it from that particular direction.
Less so now, which is too bad. I recall there being specific TV's which I found the whine really soothing while watching stuff. Computer monitors I never heard the whine which is up around 31khz according to some guys on Reddit.
Which itself is derived form the mains frequency of 50 or 60 Hz depending whether you're from the metric or the eagles/freedom land.
So the digital sample rate is related to how fast the power pant spins it's generator turbine.
Sorry for nitpicking, but the huge country I live in uses metric but has 60 Hz for the mains frequency (and our analog TV standard is PAL-M).
https://en.m.wikipedia.org/wiki/PAL-M
> [PAL M] is unique among analogue TV systems in that it combines the 525-line 30 frames-per-second System M with the PAL colour encoding system (using very nearly the NTSC colour subcarrier frequency), unlike all other countries which pair PAL with 625-line systems and NTSC with 525-line systems.
So presumably PAL M runs at 60 Hz just like NTSC, which again comes back to the domestic AC freq
Also based on how incandescent bulbs will appear to visibly flicker at lower frequencies.
I wonder whether it had to anything with Volkov Commander which was a COM-sized much faster alternative to Norton Commander.
Fun memories: I uploaded it to SIMTEL in 1993 whose operator promptly complained to the operators of the university mainframe where I had an account for pirating Norton Commander and in turn I was booted from that VAX. Being 18, I ... didn't react the best however this incident started me on very interesting paths. One, I have social engineered my way back to another account (it didn't last long but still) and started using Unix and soon Linux because of this... way more useful knowledge it turned out than VAX/VMS.
[1] https://en.wikipedia.org/wiki/DOS_Navigator [2] https://github.com/maximmasiutin/Dos-Navigator/blob/master/A...
https://en.wikipedia.org/wiki/File_manager#Orthodox_file_man...
This is technically incorrect. The BBSes were part of the "official" FidoNet [1] network, Zone 2. Fido7 was mostly post-Soviet (and not just "Russian") project to allow migration from PSTN to InterNet as a carrier. Also many "official" FidoNet nodes had additional InterNet channels which could even prevail traffic-wise over PSTN lines, one couldn't have a purely InterNet-connected FidoNet node without any PSTN presence, as that would violate FidoNet Policy (specifically, adherence to Zone Mail Hour (ZMH) [1]). ZMH was considered a core requirement for a FidoNet node (pretty much everything else was optional) and there was no consensus on dropping this "tradition".
Fido7 project was created with a goal of resolving the issue of ZMH by establishing an additional Zone 7 as an overlay to the existing 6 Zones of FidoNet, that would allow for existence of InterNet-only nodes without modifying Policy for the existing Zones. The project was never accepted as part of the "official" FidoNet structure, but for technological reasons it outlived the "official" FidoNet.
The author is referring to Google mirror of "fido7.su.hardw.support.arvid" "newsgroup". In reality, this was a Fido7-provided mirror of the original "SU.HARDW.SUPPORT.ARVID" ("echomail conference-group" names were traditionally capitalized) where "SU" stands for "Soviet Union". In reality, it was a post-soviet conference-group hosted by backbones of FidoNet Regions that belonged to post-soviet countries. Such mirrors were created by Fido7 project in an effort to promote FidoNet beyond "old school" PSTN-bound community. These newsgroups (with double prefix, such as "fido7.su.") did not belong to Fido7 itself but were forwarded from the "official" FidoNet to Fido7 and then presented as UseNet newsgroups and archived by Google which resulted in confusion that led author to think this was Fido7 content.
This was unique at a time when backup vendors preferred to pretend that errors didn’t exist. And it was a source of pride for the creators. This decision gave users the feedback they needed to improve their use of the system..."
It would be great if in the future all Flash RAM/non-volatile/persistent memory storage devices -- would give users easy access to what their hardware knows about its own internal errors...
For example, in a Flash RAM device, most notably a USB thumb drive -- tell me everything you know about the blocks/regions of memory that have went bad and had to be remapped, when a remap occurred, and what new physical blocks/region was used to remap the data.
Basically I as a user -- want to know anything and everything about what the hardware knows about the state of the underlying storage medium.
Companies that are transparent about this in the future -- will succeed brilliantly!
Flash drive manufacturers whose devices sell at your local department store for under $15 probably don't want to give people a way to check what's really going on and make a case to return them.
I would love an open-source standard storage device platform, though, with a standard firmware update process over USB or whichever interconnect. Then open-source storage device firmware could be developed which could have any other extra functions you could dream of.
In the 1980s, it wasn't really yet feasible to handle digital video in a meaningful way. No true colour framebuffers at 720x480 pixels and no way to move that many bits in and out of a framebuffer fast enough, either. But digital electronics could run > 50 MHz even back then, and there were fast analog-to-digital and digital-to-analog converters. So why not digitize the entire analog video signal, directly? D-1 does just that, sampling the the entire video signal at like 14 MHz, writing hundreds of megabits of data per second to tape uncompressed.
In other words, a perfect digital copy of the analog video waveform. Same signal the camera generated. D-1 was rather widely used in production in the late 80s and 90s, and for broadcast playback. Surviving material on D-1 is quite rare today - it was a very expensive tech and tapes got reused. Here's an example though - Depeche Mode performing Personal Jesus on German TV in 1989 - https://www.youtube.com/watch?v=ELRR7rPDvh0
Looking at that video reminded of what it looked like to watch analog live broadcast from a high quality camera, or high quality film (Hollywood) on an analog TV set through broadcast. It wasn't bad.
Todays cable MPEG-2 SD channels actually have worse quality. (Some cable and satellite HD channels have worse quality today even, when they crank the compression too hard!)
This is an interesting all-in-one unit with the VHS deck built in - https://www.youtube.com/watch?v=WVDCxTtn4OQ - released just a year or two before the CD came out.
In the end I never stored any critical archives on that thing, as I suspected (correctly) that a number of years later I'd have no way to read them back.
I didn't use this either, as I worried about wearing down the camera heads, and DVD-R's were cheaper and more plentiful.
https://manualzz.com/doc/o/9paan/amos-system-commands-refere...
https://members.optusnet.com.au/spacetaxi64/MAIN/VFL-Video-f...
The format wasn't successful though. The DVD dominated the market. HD televisions, which could take advantage of HD movies on D-VHS, weren't really available at the time, and content producers didn't like the recording ability of (D-)VHS anyway.
Interestingly, the German Wikipedia claims (without source) that some devices could even record D-VHS on standard VHS tapes, though with a higher error rate. Which would suggest they squeezed out much more than the roughly 2 GB ArVid did a few years before, albeit with specialized hardware.
Unfortunately this relies on unverified Wikipedia information.
> The Canadian Conservation Institute says that VHS tapes have a 10-30 year lifespan, on average.
https://www.canada.ca/en/conservation-institute/services/con...
That sucks, I just got in VHS for various reasons and bought a 1987 copy of Ghostbusters. It still works despite being 36yrs old. I'm curious what it means for tape to degrade and how to detect it in films.
So, I do vaguely remember seeing VHS-based backup systems, but the price/capacity wasn't something that, at least in the US, was competitive against existing consumer tape systems. I do remember though having a couple big DAT drives at work, which were the data grade versions of VHS, but the drives were a lot more expensive.
There were also similar systems for the Amiga, I think one was called VBS.
I haven't been able to find any ads for it, and I figured "word of mouth among Sysops" was the way it spread. It would be great if you could share your story.
That sounds like a massive waste of space, no? Wouldn't at least using RGB and within that only using three luminosity levels essentially dramatically increased the capacity with little increase in error rate which could have been addressed with a level of redundancy?
Doing something multilevel PAM purely in the luminance signal is probably better approach, but one has to take into account that the overall level of the video signal tends to get "averaged-out" by various ALC circuits in the VCR.
Also the cards appears to be largely software driven (it is too simple to do any real HW acceleration), so the main limitation of the bitrate probably is the performance of CPUs at that time.
Both redundancy encoding and DMA were hardware-accelerated by an onboard FPGA (although DMA had lots of motherboard compatibility issues)
[Edit: alternatively it is possible that the EPROM contains some kind of "microcode" that serves as sequencer for rest of the circuitry, which is probably more likely]
Another interesting things are the 4 wide DIPs at the bottom edge of the card, these are clones of AM29705, which is 16x4bit dual port SRAM, apparently this serves as some kind of minimal buffer between whatever the rest of the card is doing and the DMA activity. As there is no IO except DMA I suspect that the card continually samples input and produces output, probably with the IR input and output being interleaved into the "video" data in some way (as one of the 16 bits?).
Edit: what dfox said.
I wonder what the practical reliability was. For comparison breathing in the same room as my first CD-burner made it fail.
"your 500MB hard drive is overflowing with software, games, and documents."
Mh ye no. Zero people filled up their HDs with documents.
I tried this thing as a curiosity many years later. It sounded nice in theory, but was pretty slow and unreliable compared to the specialized streamers. It required a fairly janky driver, and depended on the quality of the VHS recorder and the tape. Seeking was also a problem. Besides, HDDs out-sized VHS pretty quickly.
You know, it really depends what you use a computer for.
But ye, of course I am exaggerating. Surely there were maybe a handful of persons.
There were some inherent problems associated, but nonetheless, it was quite known back then, at least in ru.
Plus it reminded me that I do need to finally start on my own LTO project, still today nothing can beat tape if you have huge amounts of data to backup and/or move.
It’s basically between cloud like B2 backblaze and LTO tapes for many 100s of TBs. Currently using backblaze B2 and I think it’s time to bite the bullet and get two LTO drives and tapes, at least it will scale up to as big as one can imagine and then have full control and access.
As someone who owned VHS tapes and several 90s-era hard drives, and remembers the sizes of each, the title had me confused for a minute.
Basically like an Iomega Jazz drive.
Realistically it wasn’t what was talked about, but it was a fun idea while it lasted.