Lets colabAnalog Tape Processing: Done cheap at home, but need pro confirmation

[LeeRosario]

I was thinking about the whole "analog tape emulation vs the actual thing" debate and I started doing some experimentation of my own. I took a very scientific approach to the whole thing and want to present my findings with the community. I'm notoriously OCD with this type of stuff, so you can skip to the conclusion if you don't want to read the details.

The process is a pain in the ass since I have no way to sync my VHS to any of my gear. However, I did come up with a way to re-sync my tracks later on in the process.

I'm hoping the professional mastering engineers can help me refine my testing approach and/or technique. I might have missed something or simply lack a better way to do this. I figure this is somewhat incomplete. I can use some other brains on this...

I call it "Hi-Fi/Lo Fi"

My Report on Analog Tape done at Home using cheap materials​

Live Drums Sample:
http://soundclick.com/share?songid=8003433

(First half is unprocessed and second half is processed through the VHS deck. Separated at the blip sound)
Tracked at my place using an 002 rack @ 32-bit, 96khz stereo
File: MP3


Problem and Question: I hate the harshness and "coldness" of projects I must record at home. I'm not satisfied with any tape simulation software I've found out there, so how can I get actual tape processing at home?


Hypothesis: By running my tracks through VHS tape at home, then bouncing it back to my computer, I can get a true to life "analog tape" sound that is much more pleasant to the human ear.

So if research tells me that VHS is about the same as CD quality, why not try it?

Variables that I must take into account:
Quality of the source recordings
The VHS tape Medium
VHS Tape Quality
VHS Deck Quality (specificly the write head quality)
A/D/A Converter Quality of the digital multitrack (in this case the 002 rack)
Jitter
possible issues using an unbalanced signal transfer method (RCA cables)
Bit Depth and sample rate
Sync issues and track realignment
Phase issues (if passing stereo signals as opposed to mono)

Summary:Research on the VHS Tape Medium

Like 2inch tape used in professional studios, the VHS medium is inherently magnetic in nature and uses the same basic principles of operation. The magnetic tape medium is widely regarded as "extremely pleasent to the human ear" by order of harmonics, wow and flutter.

The quality yielded by VHS tape is comparable to CD quality at the very least. It can hold the full range of audible frequencies with a relatively low signal-to-noise ratio, which is dependant on the quality of the tape record head, circuitry and A/D/A quality during the transfer process.

Obviously, a VHS tape deck record head was never intended for professional production applications and is not comparable to a $100k pro quality tape machine that boasts highly expensive record heads. Nor is a VHS tape compared to a 2inch reel that runs into a 3 digit price tag per reel.

Inevitably, compromises must be made.


Tools used:

-Signal generator to test for phase, frequency response, dynamic range, tape compression (if any) and any other changes.
-Modified Digidesign 002 with RCA In/Out.
-Simple DAW software with the ability to test for phase, freq response, dynamic range and any other variables.
-Sony Hi-Fi Stereo VHS recorder with 19 Micron head
-Maxell T-160 Standard Grade VHS Tape
-XDR Pres on Mackie VLX Pro board
-Beyer Dynamic DT990 Pro reference headphones
-Tannoy Reveal (Passive)
-Roger Nichols inspector Metering
-Bob Katz Mastering Audio Book (reference)
-My ears


Testing Process:

Step1: Scientific Testing
Before attempting to run my mixes through VHS tape, I wanted to find out what the deck was scientifically doing to my signal. I ran a few simple tests through my RCA sends and returns.

First, I sent out a stereo @ 32-bit, 96khz pink noise signal at about -3 RMS (my target quality with headroom), recorded it to tape and recorded it back into my DAW. I obviously couldn't do this in real time, since there is no way to sync my tape deck with the multitrack. So certain tests where impossible to conduct.

I was happy to find that the frequency response had not changed except for a high roll off at around 36khz. I assume this has to do with the 002 A/D/A conversion process.

A problem showed up when I tested for stereo balance. The processed signal gave me unequal RMS for Left and Right. Left read around -2.5 while right read around -3.5. That's an entire DB of difference. I can't understand why.

Dynamic Range seemed relatively equal with little or no tape noise in the signal.

A 1khz tone confirmed the L/R balance offset. Scientifically, this is no good for stereo mixes, because I didn't get an exact return on L/R balance. It might throw off the entire mix. However, I figured I'd let a subjective test of an actual stereo mix give me a final verdict.


Step2: Practical Testing

After the scientific stuff was done, I took an old unmastered stereo mix I had done and passed it through the tape deck.

I was happy to find that most of what I predicted occurred. I got a softer, rounder, fatter signal with a little grit added. Some tape saturation was also present.

However, the stereo image had degraded substantially: more notably in the center channels. The middle of the stereo image became somewhat cloudy and cluttered with a subjective loss in clarity of around 30%.

100% being an exact copy in transparency and 0% being a completely undefinable copy. Completely mushed sound.

Sending out the signal at a lower RMS (no greater than about -15 RMS) yielded slightly cleaner results, but still not transparent enough for me to consider it as "pre mastering" tape processing. The tone was great, but the image was suffering to much.

Several different premastered mixes confirmed the findings. Whether it was due to the unbalanced connections, the circuitry involved, actual tape compression/saturation or the A/D/A conversion, I don't know.

I wish I had a ultra high quality A/D/A converter to really confirm this.


I then decided to try processing individual tracks through the tape deck, leaving the stereo summing up to a more capable DAW and/or analog desk. Obviously, I don't have an SSL 9000J on hand, so I did the summing "inside the box."

The main issue here was syncing up my tracks. Without a way to sync the tape deck to any studio gear, I developed a very simple method of re-syncing my tracks.

I used a 1khz square wave signal as a marker at the beginning of my tracks. I recorded the processed tracks back into my DAW with the square wave marker. Then by simply cutting at the zero crossing point of the marker (using any simple tab to transient function), I was able to cut and slide my tracks back into sync.

The results yielded very fat, soft, rounded and saturated tracks. I also still have the power to mix and additionally process the individual tracks to my tastes. The harshness was gone and I still had total control over my mix.

Conclusion:


In conclusion, by conducting simple tests with a signal generator and a few standard measuring tools, as well as subjective listening tests to actual mixes, I was able to determine that the VHS tape medium does in fact yield a smoother, warmer, fatter and thicker audio track that it's unprocessed counterpart does not possess. It also yields a slight amount of "musical grit" that adds to a sense of excitement. These of course are all highly debatable and subjective terms.

I felt smoother highs and rounder, thicker lows. I noticed a definable mid range that's not overbearin and that the overall tonal "feel" is more pleasant, even in a digital medium. I observed a more natural stereo feel when leaving the summing outside the VHS tape deck.

The test assumes that you have a capable VHS deck with RCA ins and outs, as well as some form of DAW with RCA ins and outs as well. Theoretically, better A/D/A converters and generally better gear, should improve this process.

By processing premastered and mastered stereo mixes through the VHS tape deck, I did observe a significant loss of transparency that yields a 30% cloudier stereo image. The mid channels suffered most while the side channel material seem least affected. I also determined a Left/Right offset of about 1db that may be unacceptable by professional standards. In some forms of music, particularly grungier music, this may be desired. However, it comes with the risk of unpredictability.

I was unable to determine if it was simply because of the VHS Deck, VHS tape or DAW circuitry itself.

I found better results in processing mono tracks by means of using only the left side (the hotter side) of the VHS deck and DAW, using a simple "tab to transient" signal marker method to realign the tracks. This yielded the benefits of analog tape while still giving me complete control over my mix. This also leaves stereo processing to the mixing platform itself and not to the VHS deck.

So while VHS tape can never be compared to 2inch reel and a VHS deck compared to a professional tape machine, the benefits are clearly present with a bit of work involved.

Thanks for your time, I'm open to advise and help!

 

[SouthSIDE Glen]

Unless you are using an increasingly rare VHS-HiFi machine from the mid-80s to early 90s that had a separate manually controllable audio preamp section (Man, I miss my old NEC) or a semi-pro to pro S-VHS machine with similar capabilities, you're probably using a machine that's running the audio through an AGC (Automatic Gain Control) circuit. How does that play into the results?

G.

 

[mshilarious]

Results are going to vary considerably depending on the particular VCR. I tested mine some years ago and posted the results here; I didn't have bandwidth anything near 36kHz; 17kHz is the figure that comes to memory. SNR was better than cassette tape but still horrible compared with 16 bit CD, I think 70dB was the number. There was a significant interference signal somewhere around 15kHz (no, TV wasn't on, it wasn't even connected, but my VCR does have an internal tuner and probably some other internal oscillators, who knows?). THD and intermod were high.

In short, it's an effects box. If you want that instead of a tape sim I suppose it makes sense.

That said, if somebody built a dedicated analog tape deck that didn't use VCR-style rotating heads, that could possibly be a useful medium.

 

[LeeRosario]

you're probably using a machine that's running the audio through an AGC (Automatic Gain Control) circuit. How does that play into the results?

This is why I'm glad I brought this in here. I need all the forward thinkers I can get.

Honestly, I missed that part in this process, so I didn't test for it specifically. And that's a good one too.

Anything circuitry specific, involving opening up the gear I didn't really do. I don't know how.

Unofficially, I'm guessing this would affect the overall RMS of the track? I did observe more leveled off tracks. I actually observed that sharp transients where most affected.

So in other words, if I had a loud snare "crack" at 0 db. It came back at around -3 db peak. But again, that's an "unofficial" answer cause I didn't think to test for AGC affects. I can't tell if that even has to do with an AGC circuit?



I wonder how I can even test for that on a standard VHS deck thats roughly 9 years old?

 

[LeeRosario]

In short, it's an effects box. If you want that instead of a tape sim I suppose it makes sense.

exactly my point. I definitely found that out with this test. I'm absolouty avoiding it as a "mastering" box, but I think it really does great on individual tracks.

Again, with that pain in the ass "realignment" process. It's not too bad, it just takes patience while I record and play it back to my DAW.

That really would be an awesome thing though, a dedicated VHS tape with music production in mind. Definitely it starts with the head (that's what she said), but maybe even more discrete circuitry? Something that still uses all those exsisting unused VHS tapes still floating around out there?


I'm just saying just in case somebody actually does figure out a way to make a very popular and simple deck that people would actually buy. I'd like to dream that we could make money off that idea.

 

[Tim Gillett]

This issue was recently discussed on the analog forum.

VHS hifi doesnt qualify as "analog tape sound" except in the sense that it uses magnetic tape. You may as well say a DAT has "tape warmth".

VHS HIFI uses an FM modulated carrier which is so noisy they had to use inbuilt aggressive double ended NR to bring the noise down. Push the record level over O and it just clips.

 

[LeeRosario]

actually thinking about it now:


some input metering on the VHS deck would be REALLY nice (I heard the old ones used to have them). I have no idea how hot I'm hitting the tape.

 

[SouthSIDE Glen]

I wonder how I can even test for that on a standard VHS deck thats roughly 9 years old?

Well, off the top of my head, I'd start with working out the gain structure. First, I'm assuming that the RCA I/O on the VHS is calibrated for a consumer -10dBV line level. I'd keep that in mind when setting up the audio out to the VHS (as well as monitoring how it's returned.)

I'm not sure of just how the AGCs on VHS machines are spec'd out - e.g. how thei threshold and gain reduction are actually set - and I also assume that it probably does not remain standard from VCR brand-to-brand, if even from model-to-model. But it would be interesting to see how, if any, it changes when you send it audio at, say, -16dBV peak, -10dBV peak, +4dBu peak and +6dBu peak (just for example; I know, that's a lot of testing ) to get some AGC parameters.

This can be particularly important (I would think, anyway) if one is working with pro audio gear calibrated and metered to work at +4dBu and forgets to step down to the consumer level for recording to VHS.

G.

 

[LeeRosario]

This issue was recently discussed on the analog forum.

VHS hifi doesnt qualify as "analog tape sound" except in the sense that it uses magnetic tape. You may as well say a DAT has "tape warmth".

VHS HIFI uses an FM modulated carrier which is so noisy they had to use inbuilt aggressive double ended NR to bring the noise down. Push the record level over O and it just clips.

wow, I had no idea it was brought up in here. But I mean isn't that the whole point? To simply get the tape sound?

The only understanding i have with actual tape machines is from using a few different Studer 2inch Tape machines I've used in the past and some research on the subject of why tape sounds good. I've never actually opened one up or thought to ask a qualified studio tech on the extreme specifics at work there. Which really sucks.


I guess what I am asking is, at what point does it stop being about the tape and actually more about the circuitry in the tape machine itself?

 

[LeeRosario]

Well, off the top of my head, I'd start with working out the gain structure. First, I'm assuming that the RCA I/O on the VHS is calibrated for a consumer -10dBV line level. I'd keep that in mind when setting up the audio out to the VHS (as well as monitoring how it's returned.)

I'm not sure of just how the AGCs on VHS machines are spec'd out - e.g. how thei threshold and gain reduction are actually set - and I also assume that it probably does not remain standard from VCR brand-to-brand, if even from model-to-model. But it would be interesting to see how, if any, it changes when you send it audio at, say, -16dBV peak, -10dBV peak, +4dBu peak and +6dBu peak (just for example; I know, that's a lot of testing ) to get some AGC parameters.

This can be particularly important (I would think, anyway) if one is working with pro audio gear calibrated and metered to work at +4dBu and forgets to step down to the consumer level for recording to VHS.

G.

I definitely see what you mean here. I mean I wonder if the AGC (if there is one in here) is one of the biggest factors inside the circutry of a typical VHS deck.


Like for example, as far as I know right now, these are the big factors inside a standard tape deck:

-The record head
-AGC
-The brand specs (how each chooses to calibrate the signal)


I wonder what else is at play inside a VHS deck signal chain? Well that's important anyway...

 

[miroslav]

Off of Wikipedia....

<<<<<<<<<<<<<<<<<
In the original VHS format, audio was recorded as baseband in a single linear track, at the upper edge of the tape, similar to how an audio compact cassette operates. The recorded frequency range was dependent on the movement of the tape past the audio head, which for the VHS SP mode, resulted in a mediocre frequency response of roughly 100 Hz to 10 kHz. The signal-to-noise ratio was an acceptable 42 dB. Both parameters degraded significantly with VHS's longer play modes, with EP frequency response peaking at 4kHz.

More expensive decks offered stereo audio recording and playback. Linear stereo, as it was called, fit two independent channels in the same space as the original mono audiotrack. While this approach preserved acceptable backward compatibility with monoaural audio heads, the splitting of the audio track degraded the signal's SNR to the point that audible tape hiss was objectionable at normal listening volume. To counteract tape hiss, decks applied Dolby B noise reduction for recording and playback. Dolby B dynamically boosts the mid-frequency band of the audio program on the recorded medium, improving its signal strength relative to the tape's background noise floor, then attenuates the mid-band during playback. Dolby B is not a transparent process, and Dolby-encoded program material will exhibit an unnatural mid-range emphasis when played on non-Dolby capable VCRs.

High-end consumer recorders took advantage of the linear nature of the audio track, as the audio track could be erased and recorded without disturbing the video portion of the recorded signal. Hence, "audio dubbing" and "video dubbing", where either the audio or video are re-recorded on tape (without disturbing the other), were supported features on prosumer editing-decks. Without dubbing capability, an audio or video edit could not be done in-place on master cassette, and requires the editing output be captured to another tape, incurring generational loss.

Studio film releases began to emerge with linear stereo audiotracks in 1982. From that point onward nearly every home video releases by Hollywood featured a Dolby-encoded linear stereo audiotrack. However, linear stereo was never popular with equipment makers or consumers.

Around 1985, JVC added Hi-Fi audio to VHS (in response to Betamax's introduction of Beta Hi-Fi.) Both VHS Hi-Fi and Betamax Hi-Fi delivered flat full-range frequency response (20 Hz to 20 kHz), excellent 70 dB signal-to-noise ratio (in consumer space, second only to the compact disc), dynamic range of 90 dB, and studio-grade channel separation (more than 70dB). VHS Hi-Fi audio is achieved by using audio frequency modulation (AFM), recording each of the 2 stereo channels (L, R) on a frequency-modulated carrier and embedding the 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 under the luminance carrier (below 1.6MHz), and recorded first. Subsequently, the video head erases and re-records the video signal over the same tape surface, but 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 or SP.) Since JVC had gone through the complexity of ensuring Hi-Fi's backward compatibility with non-Hi-Fi VCRs, virtually all studio home video releases contained Hi-Fi audiotracks (in addition to the linear audio track). Under normal circumstances, all Hi-Fi VHS VCRs will record Hi-Fi and linear audio simultaneously to ensure compatibility with VCRs without Hi-Fi playback.

The sound quality of Hi-Fi VHS stereo is comparable to the quality of CD audio, particularly when recordings were made on high-end or professional VHS machines that have a manual audio recording level control. This high quality compared to other consumer audio recording formats such as compact cassette attracted the attention of amateur and hobbyist recording artists. Home recording enthusiasts occasionally recorded high quality stereo mixdowns and master recordings from multitrack audio tape onto consumer-level Hi-Fi VCRs. However, because the VHS Hi-Fi recording process is intertwined with the VCR's video-recording function, advanced editing functions such as audio-only or video-only dubbing are impossible. Some VHS decks also had a "simulcast" switch, allowing users to record an external audio input along with off-air pictures. Some televised concerts offered a stereo simulcast soundtrack on FM radio and as such, events like Live Aid were recorded by thousands of people with a full stereo soundtrack despite the fact that stereo TV broadcasts were some years off (especially in regions that adopted NICAM).

The considerable complexity and additional hardware limited VHS Hi-Fi to high-end decks for many years. While linear stereo all but disappeared from home VHS decks, it was not until the 1990s that Hi-Fi became a more common feature on VHS decks. Even then, most customers were unaware of its significance and merely enjoyed the better audio performance of the newer decks.
>>>>>>>>>>>>>>>>>>>>>>>

 

[Waltz Mastering]

Around 1985, JVC added Hi-Fi audio to VHS (in response to Betamax's introduction of Beta Hi-Fi.) Both VHS Hi-Fi and Betamax Hi-Fi delivered flat full-range frequency response (20 Hz to 20 kHz), excellent 70 dB signal-to-noise ratio (in consumer space, second only to the compact disc), dynamic range of 90 dB, and studio-grade channel separation (more than 70dB).
>>>>>>>>>>>>>>>>>>>>>>>

I used to do mixdowns to this format around 86/87 and then switched to the Sony pcm 601 F1/vhs format until DAT became popular around 89/90

 

[dean1964]

I keep reading this and keep getting confused. It has been my understanding that VHS format audio has been digital since the 80s. So are you 100% positive that your VHS deck is recording an analog signal?

I bought a new Sony VHS recorder in 1998 for $100 and used it as a stereo master recorder for a while because it was digital. ( i had nothing else at the time).

You say you experienced some "cloudiness" of the stereo image once passed through the VHS, and that you had better luck with a mono signal. I'm willing to bet that this might be caused by the phase shift introduced by the anti aliasing filters in the A/D/A converters in your VHS. If the two anti alias filters are not causing the exact same phase shift, this will put the two sides out of phase relative to each other, regardless of how "in phase" the stereo mix was before the VHS.

In mono, only using one side of the VHSs audio channels, this would not be a problem. The phase would shift relative to itself and nothing else. It will sound different but not necessarily any less clear.

 

[mshilarious]

I keep reading this and keep getting confused. It has been my understanding that VHS format audio has been digital since the 80s.

No. You are thinking about S-VHS, but not all S-VHS recorded digital audio, only some professional models.

So are you 100% positive that your VHS deck is recording an analog signal?

Almost certainly. OP noted a bandwidth of 36kHz; that would be impossible with S-VHS's digital audio rate of 48kHz.

I bought a new Sony VHS recorder in 1998 for $100 and used it as a stereo master recorder for a while because it was digital. ( i had nothing else at the time).

No. The Alesis ADAT used a VHS cassette to record a digital signal, but plain ol' VCRs do not, unless you specifically bought an S-VHS deck.

You say you experienced some "cloudiness" of the stereo image once passed through the VHS, and that you had better luck with a mono signal. I'm willing to bet that this might be caused by the phase shift introduced by the anti aliasing filters in the A/D/A converters in your VHS. If the two anti alias filters are not causing the exact same phase shift, this will put the two sides out of phase relative to each other, regardless of how "in phase" the stereo mix was before the VHS.

In mono, only using one side of the VHSs audio channels, this would not be a problem. The phase would shift relative to itself and nothing else. It will sound different but not necessarily any less clear.

I don't even know where to start with that one. Modern (>1995, at least) ADCs do not impart a phase shift, but even if they did, it's a digital algorithm that is part of a decimation filter. It's impossible for it to have a different frequency corner on different channels, and the phase shift (if there was one) would be a function of the rolloff frequency. The entire signal isn't phase shifted.

Instead, modern ADCs have a very simple, often first-order filter at say 600kHz. The tolerance in capacitor selection could cause that to drift a bit, but it's hardly important since the bandwidth limitation at the data rate will be done by an algorithm and not a physical filter.

None of that actually matters for an analog VCR, but anyway just had to correct some of these misconceptions about digital audio that are hard to stamp out . . .

 

[SouthSIDE Glen]

I keep reading this and keep getting confused. It has been my understanding that VHS format audio has been digital since the 80s. So are you 100% positive that your VHS deck is recording an analog signal?

Only a guess, here, Dean, but I think you may also be confusing the actual VHS-HiFi recording format with ADAT recording which used VHS-style tape cartirdges. Standard VHS-HiFi has always been analog, but ADAT used the VHS tape format to record digital. Same tape and head scanning structure, but entirely incompatable formats.

@Lee:
No mention of pre-recording circuitry in the form of gain control in that Wiki report. Tim G. mentions the possibility that they just let the signal clip when they reach the end of the FM bandwidth, which is it's own form of hard gain control, I suppose , and maybe that's what I had always misunderstood to be an actual AGC circuit. But I'm not convinced that they didn't include AGC in at least some units that did not have their own manual input preamp control; just letting the signal clip sounds awfully sloppy to me, especially with the inexpensive nature of a simple AGC.

Either way, it'd be an interesting test to see what happens in that regard.

Back in the 80s I used to use an NEC N965U VHS HiFi* deck for poor-man's mixdown/pre-mastering. This deck had separate L/R input gain faders w/metering, output gain control and manual tracking adjustment by meter, etc. And I gotta say, I was very pleased with the results, as far as it making a reliable and quality-sounding medium for these purposes.

It did NOT impart any kind of "analog sound" worth mentioning, which is not surprising since the audio is not amplitude modulated to tape, but rather frequency modulated. But it did give a clean, virtually hiss-free, high-dynamic range and frequency-accurate reproduction of whatever I fed to it.

But would I use your standard don't-worry-about-anything-but-the-play-and-record-button VHS-HiFi for the same purpose? Almost certainly not.

And it's not just a matter of having the manual controls, but also the hardware behind it, AGC or not. Like any type of hardware, quality is going to vary from model-to-model, especially on the recording side. The quality of heads varies and the quality of circuitry varies. And on your average $69 grocery store VHS HiFi machine today, they are not really worried about audiophile *recording* quality, because Joe Netflix was not worried about that either. He just wanted to watch his movies with quality *playback* capability.

* No, that is not my eBay ad, nor am I recommending it; I'm just using it as the only place where I could quickly find a picture of of one of them.

G.

 

[Tim Gillett]

@Lee:
No mention of pre-recording circuitry in the form of gain control in that Wiki report. Tim G. mentions the possibility that they just let the signal clip when they reach the end of the FM bandwidth, which is it's own form of hard gain control, I suppose , and maybe that's what I had always misunderstood to be an actual AGC circuit. But I'm not convinced that they didn't include AGC in at least some units that did not have their own manual input preamp control; just letting the signal clip sounds awfully sloppy to me, especially with the inexpensive nature of a simple AGC.

AFAIK there were models with record gain controls and others without. The earlier ones tended to have gain controls and stereo metering but the makers seemed to drop them later on, at least the consumer models, I guess for cost reasons and because most people at home were just recording off air from the inbuilt tuner and so all the gain structure could be built in and preset anyway, including the fact TV transmitters dont go over a certain predetermined level. So strictly no need for compression or limiting.

I just said that above a certain level (maybe 0 on the meter) the audio clipped. That was my experience.

I understand AGC as a sort of limiter (high or infinite comp ratio) but with a very slow release time so that it doesnt "pump" in the presence of loud program. It takes up to a minute or so for it to "release" back to normal, hence no pumping. It would have been used in VCR's on the linear audio tracks which had poor S/N, and so some dynamic range compression was afforded. Not strictly needed on the HIFI tracks because of their far better S/N.
But again, as you say Glen, HIFI VHS was not analog in the normal sense and so didnt at all have the "analog tape" sound.

Cheers Tim

 

[dean1964]

No. You are thinking about S-VHS, but not all S-VHS recorded digital audio, only some professional models.

Cool. I was wondering where the difference was. I had a stack of ADATS in the 90s and only once did I put the tape in my "S-VHS" to see what would happen. =)

I don't even know where to start with that one. Modern (>1995, at least) ADCs do not impart a phase shift, but even if they did, it's a digital algorithm that is part of a decimation filter. It's impossible for it to have a different frequency corner on different channels, and the phase shift (if there was one) would be a function of the rolloff frequency. The entire signal isn't phase shifted....

None of that actually matters for an analog VCR, but anyway just had to correct some of these misconceptions about digital audio that are hard to stamp out . . .

One place you could start is with Harry Nyquist. I do not want to re-hash or explain the fundamental process that all PCM ADCs use, but we can still remember the rule. Anyone one here working in the digital domain NOT using PCM?

There is only one complex signal at the input of a single channel. If you distort the signal at any point you have distorted the whole signal. (This is more of an ideal than a fact. Adopting it will get you thinking about the over all quality of your signal as you record.)

"The highest frequency a digital audio system can record and reproduce is one have the sampling frequency." Anything above this results in distortions called aliasing. To prevent this from occurring designers must implement a low pass filter before the P.A.M. stage of PCM. This filter is analog, not an algorithm, and ALL analog filters impart phase shift to one degree or another. Cheep ones will cause greater distortions - kinda like the ones in the consumer S-VHS machines.

This is not a misconception, rather a textbook fact.

Peace ( seriously )

 

[SouthSIDE Glen]

AFAIK there were models with record gain controls and others without. The earlier ones tended to have gain controls and stereo metering but the makers seemed to drop them later on at least the consumer models,

More or less true. The earliest machines in the early 80s usually were pretty basic. It really wasn't until the market matured a little with first 4-head technology and then HiFi that you started to see the more feature-laden, "high end" machines, and even then those were usually reserved for the higher ned of a label's product line, with the cheaper models still being pretty basic in features and quality. Sony was always a bit of an exception in that they tended to offer feature-laden, higher end models at the top of their product line pretty early on. But then again, they had a head start and were Betamax, not VHS.

I guess for cost reasons and because most people at home were just recording off air from the inbuilt tuner and so all the gain structure could be built in and preset anyway, including the fact TV transmitters dont go over a certain predetermined level.

Again, mostly true. When recording, most folks were recording off of analog broadcast TV (cable wasn't anywhere near as ubiquitous as it is today, and what cable there was usually suffered greatly in distribution quality control) or playing back rental movies from Blockbuster, neither media of which was ever known for high audio quality . And remember, it wasn't until 1984 (if I remember right) that Miami Vice came along with an actual stereo broadcast track.

There were a few who like me who liked the high-end machines for mixdown purposes, but definitely not enough to keep the market for them alive. Remember, back then there were still plenty of consumer/prosumer Teac, Pioneer, Technics and Akai stereo open-reel decks available brand new in the showrooms that tended to be the favorite of those who would normally otherwise be in the VHS Hi-Fi for audio market. And for video, the real prosumers either already had their Betamax machines or jumped right to Betacam. After all that, by the time S-VHS came along, it drove a coffin nail in what little market for "botique" VHS Hi-Fi machines there was.

But during that short window of opportunity in the early-mid 80s, there were a few VHS HiFi machines by NEC, JVC and (to a slightly lesser degree, IMHO) Matsushita that were actually fairly respectable in their quality.

G.

 

[LeeRosario]

this is really an interesting look into the whole thing.


I'm actually glad I brought it up, mainly because I really think taking the time to understand the physics behind tape and the circuits is monumental.


The AGC thing really does through a wrench into my thoughts...only cause I wonder if it's really an AGC circuit that's affecting my signal and not the actual tape response itself.

I mean I did do some basic testing with square waves that came back rounded off (very indicative of tape) but then again, an attack and release time set a certain way on any limiting circuit could have a similar affect.

I'm really starting to think the benefits come with those special pro VHS decks you guys have been bringining up. About a decade before my time.

I definitely do understand that difference between ADAT Digital and Analog tape itself.

I mean I wonder, cause when I did the sine wave test with plenty of headroom (no clipping), I still did get back a sine wave with even order harmonics...also very indicative of tape.

It's almost one of those "I think I hear the difference, but what if my mind is tricking me?" kind of things

 

[LeeRosario]

I keep reading this and keep getting confused. It has been my understanding that VHS format audio has been digital since the 80s. So are you 100% positive that your VHS deck is recording an analog signal?

I bought a new Sony VHS recorder in 1998 for $100 and used it as a stereo master recorder for a while because it was digital. ( i had nothing else at the time).

You say you experienced some "cloudiness" of the stereo image once passed through the VHS, and that you had better luck with a mono signal. I'm willing to bet that this might be caused by the phase shift introduced by the anti aliasing filters in the A/D/A converters in your VHS. If the two anti alias filters are not causing the exact same phase shift, this will put the two sides out of phase relative to each other, regardless of how "in phase" the stereo mix was before the VHS.

In mono, only using one side of the VHSs audio channels, this would not be a problem. The phase would shift relative to itself and nothing else. It will sound different but not necessarily any less clear.

Well I guess in "scientific terms", the variable test was done against a non-VHS processed signal. In other words, I did test where I sent a out a signal from the 002 (RCA) and brought it right back into the 002 (RCA). I left out the VHS deck entirely. It's definitely all analog at that point.


That came back a perfect copy. Other than some small level differences, it was a perfect match. So at that point, whatever was happening had to be inside the deck itself.

You might be thinking about digital SPDIF which is the same RCA connection, but actually is a digital signal.

I mean, I could imagine the aliasing interacting with VHS may have had something to do with it, but it sounded more of a like a bad limiter. It almost sounded like I ran my mix through the standard shitty limiter plug-in that comes with pro tools.

A shift in phase (assuming it's really bad) would definitely screw with tracks that where recorded as a group. Mono overdubs not so much (like you mentioned)

At that point, drums would probably suffer most since the entire image depends on the tracks being in phase.

Me testing for that would actually help me decide if it's worth processing group tracks for, or just saying "don't run group tracks through tape, just tracks that are not phase dependent to other tracks."

Vocals would be a good one, assuming they are tracked in mono.