The VHS Challenge!

#1

Hey, I have a challenge if anyone fancies it, it's called "The VHS Challenge" :heart_eyes:

It would be cool to see "VHS" in the library. The VHS audio system adds a unique, solid character to the sound being recorded/played to/from tape (an interesting system).

As a quick read-up, here's a quote from Wikipedia:

VHS Hi-Fi audio is achieved by using audio frequency modulation (AFM), modulating the two stereo channels (L, R) on two different frequency-modulated carriers and embedding the combined modulated audio signal pair into the video signal. To avoid crosstalk and interference from the primary video carrier, VHS's implementation of AFM relied on a form of magnetic recording called depth multiplexing. The modulated audio carrier pair was placed in the hitherto-unused frequency range between the luminance and the color carrier (below 1.6 MHz), and recorded first. Subsequently, the video head erases and re-records the video signal (combined luminance and color signal) over the same tape surface, but the video signal's higher center frequency results in a shallower magnetization of the tape, allowing both the video and residual AFM audio signal to coexist on tape. (PAL versions of Beta Hi-Fi use this same technique). During playback, VHS Hi-Fi recovers the depth-recorded AFM signal by subtracting the audio head's signal (which contains the AFM signal contaminated by a weak image of the video signal) from the video head's signal (which contains only the video signal), then demodulates the left and right audio channels from their respective frequency carriers. The end result of the complex process was audio of outstanding fidelity, which was uniformly solid across all tape-speeds (EP, LP or SP.)

With that, you have an explanation of the process that would effectively need to be modeled in Axoloti. And there's a twist, because this challenge takes you out of your coding zone and drags you back down to the same level as us mere cable-connecting types. Yup, you have to model this technique using the objects already available so that others can learn from it. Coding objects is cool, but completely useless for non-coders who wish to see how it's made, what makes it tick etc, so you must do this by connecting objects that are already available in the factory set.

So there you have it, a fun challenge if anyone is interested, and looking at the Wiki description it's an interesting system in it's own right. No pressure, no deadlines, no prizes, but you do get bragging rights if you produce a VHS patch that is genuinely convicing, and is based on the same process that VHS audio uses.

Basically a stereo audio-in, followed by your VHS patch, followed by a stereo audio-out :grin:

Simple delay object
#2

that would be super nice to have this indeed!

#3

It would, I think it's a difficult one though, cause I suppose you'd have to model tape emulation in there before you could even think about modeling the VHS process itself.

Haha, devilishly mean challenge I suppose, but there's no denying that if whatever goes in one end comes out sounding like a genuine, VHS Hi-Fi recording on the other, that would be very nice!

LoFi Warped Vinyl + Generation Loss
#4

Can someone test my first attempt at this?
I'm trying to add some dirt (aging part of the patch) with the slew limiter but it doesn't behave as I expected.

VHS-woble.axs (32.6 KB)

link to VHS sounds, which should be put in SDcard /samples/

main patch (the uploaded one does't have the eq after the input

Capture d’écran 2017-04-18 à 11.56.15.png1269x749 70.8 KB

woble

Capture d’écran 2017-04-18 à 11.58.01.png1368x881 104 KB

global woble

Capture d’écran 2017-04-18 à 11.58.42.png1529x794 116 KB

#5

Hey, sorry for the late reply, krikor, I had no idea anyone had posted. I can't test at the moment, but will as soon as I can, and thanks for taking the challenge :sunglasses:

#6

Finally been able to see what you made here!

It brought a smile to my face, but for all the wrong reasons, lol. You've really gone for a dodgy tape transport effect with it, and I like the way it does that, sort of glitches in time with the speed instability. I was really thinking more the rubber-band bassy-warmth kinda thing it does to the sound, not so much the instability stuff, but it's interesting to look at the way it works, so thanks for being a sport and creating it :sunglasses:

#7

the "aging" part together with the filtering was intended to do this but I could not get it to work

#8

I'm pretty new here and this is a great idea!

I see it's a patch though...is it possible to like... add this patch into an existing patch as an effect with a couple key parameters available to be adjusted live? (I literally started a couple days ago and have no idea if this is possible)

#9

Sure, just copy the patch then paste it into your own design, and connect it all up :sunglasses:

#10

Yeeee. But is there any way to do that without it taking up so much space visually in the gui? I'm trying to keep everything simple lookin...

#11

I can't remember the proper term for it off-hand, but you can collect everything and store it in a new object. You basically add inlets and outlets to it so that the other parts can access whats inside it. I can't help with that though, so hopefully someone else will step in to explain that one :blush:

#12

Ah, that's it, I think it's called "Subpatch".

#13

add a patcher/patch object
click edit
copy the bits of the patch you want into the editor window
and inlets and outlets
save as a file, close editor
delete patch object and reload it as an independent subpatch from the file
to do that you open the object list and type ./(name_of_subpatch_file) and hit enter.

subpatches are easy, learn how to use them, it can save a lot of clutter in a big patch.

#14

Not really sure how to do any of that yet haha are there any tutorials for how to do these things step by step? It feels like I'm missing a huge piece of the puzzle here..I went through the tutorials within the Axoloti Gui but don't remember doing any of those things.

Once I figure out how to do these things and have a weekend off I'm going to make some tutorial videos for others...

#15
#16

Documentation is pretty thin and there isn't much in the way of step by step tutorials, read the user guide and try a few things out, it's pretty simple to understand how to put things together but without some knowledge of synthesis it's hard to know what goes with what to get the sound you are after I guess.

#17

Yeah! It seems like it should be super straight forward but I'm finding a lot of extremely important stuff is virtually undocumented...but from trial and error and a lot of dumb lucky I've managed to get things working.

Im definitely going to make some tutorial videos soon when I have some time (my work weeks are hilariously long but I wanna help as much as possible!) My friends here in Toronto just about lot their minds when I explained to them what this is but without some sort of video showing how easy it is to get things up and running, they're all assuming you have to be some sort of matrix-level hacker dude to use this magical thingamajig

#18

I'm back on this, it seems my approach to the slew limiter is wrong, anyone know what could be a good treatment for the aging part of the VHS?

#19

So did you get any further, krikor?

Just wanted to point out, I think the defining character of VHS audio must be down to that 'Depth-Multiplexing' technique it uses (see Wiki quote in first post). I suspect that due to the way it's done, storing two signals at the same tape surface position, that what we hear is being contaminated in some non-obvious (but pleasing) way. At the end of the day they're still storing an analogue audio signal to tape, so I think it's really the Depth-Multiplexing part that makes VHS audio have a different characteristic to that of standard tape recording techniques.

So what I'm getting at is, I doubt ageing and glitches would get you that VHS audio characteristic. I do think tape recordings sound better after they've aged, so there is something of merit in the ageing process of a tape recording, but it's not what gives VHS audio it's unique characteristic.

Although I wouldn't know how to do all of this, I'm thinking that taking the audio signal, processing it with an appropriate form of distortion or 'drive', and then mixing a slightly offset copy of it with the original signal, might reasonably replicate the process where, with VHS, it gets recorded, erased, and re-recorded to the same part of the tape surface. What would need to happen to it in a modulation sense, I'm not entirely sure (perhaps an ultra-high frequncy amlplitude modulation to the cloned signal), but once I get a better understanding of some of the objects I'd definitely be exploring routes like that.

#20

Just came across a clip that I believe demonstrates the magic that Depth-Multiplexed AFM (as used in Hi-Fi VHS) adds to audio recordings. The first is a clip where I suspect the magic can clearly be heard, and the second is one where Depth-Multiplexed AFM definitely not involved (and I can't imagine anyonwe would prefer CLIP 2 over CLIP 1). The technology clearly adds a very pleasing form of dynamics and excitation to the audio, resulting in a more powerful, lush, and crisp sound.

CLIP 1 - Suspected WITH Depth-Multiplexed AFM
CLIP 2 - Suspected WITHOUT Depth-Multiplexed AFM

This comparison might be of interest to those among us who are perhaps too young to have enjoyed the pleasure of Hi-Fi Stereo VHS decks. And yes indeed, just in case it has the cogs in your mind churning away in overdrive, people have already discovered that such decks are the ideal master recorder for their albums. It's exactly as the Wikipedia quote in the original post stated;

"The result of the complex process was audio of outstanding fidelity" :wink:

Yup, so we need to capture that in an axo object ... still waiting :yum: :stuck_out_tongue_winking_eye: