In my experience (about 18 years of video editing + many years before that as a VCR user), and as owner of quite a few VHS tapes, "color fade" just doesn't happen. It's seems to be a myth, as far as I can tell. I can't think of a logical explanation that would ever allow for "color fade".
While tapes do age, these things are not inkjet prints or paper, it's not something that just "fades" away.
As far I am aware, video aging errors present themselves by way of humidity or heat damage, or use damage, not in a way that would leave a signal intact, but merely lacking in color. This is mostly drop-outs, be it signal dropouts or visible magnetic dropouts (starbursts, sparkles)
This idea of "color fade" is being mentioned by a lot of video companies as a reason to "transfer your tapes now!", but my hypothesis is that it's complete bullshit.
VHS data is color-under and stored as analog data, with color being derived from the chroma, which is played back by a VCR. The playback heads read the color under signal, and a poor relation here (due to head wear/debris or tape debris) also determine the color saturation (amount of chroma). The AGC/ACC/AFC is what keeps it from fluctuating. A messed up head/tape can result in total loss of chroma. Then again, if the tape is this fubar, then you'll notice a heck of a lot more than just some "color loss". Bandpass filtering further limits color interference.
In other words, "color fade" is more a statement about a poor VCR than anything else. Or, the tape is trashed, in which case you'll see ALL KINDS of problems. (And that being the case, I'd still call "color fade" a bad myth, since it can't happen all by itself as a singular error.)
Although I mostly talk of VHS, by extension, I also want to include 8mm, Hi8, Betamax, Betacam and S-VHS, too -- all standard tapes.
I'm hoping davideck and edDV can give some input, not sure who else here may have some good analog video knowledge.
Have I overlooked something?
Note that I am going to be aggressive in this thread. If all you plan to reply is "yes, it happens, I've seen it", then just move along, this thread is not for you. I want some engineer-level discussion, not comment from a casual VCR user that doesn't know anything. Sorry if that sounds mean, but I want a certain level of discussion that is at or above my level.
+ Reply to Thread
Results 1 to 30 of 32
I have come across an interesting situation which I thought might add as a scientific control (if you can call it that), but I think it is relevant to your question, and I am very curious myself.
Some of my tapes are hard to find commercially made tapes featuring music and documentaries from anywhere between 1993 to 1999. Many of the shows were 3 or 4 hours so you would end up with 2 tapes for the whole show or documentary. In any of these situations both tapes look to be made from the same stock, both were produced at the same time, both were stored in the same environment together, and both were played on the same VCR's (all of the tapes are first generation). In some situations Tape 1 will play very well, then I will throw in tape 2 and the chroma is either extremely high or close to non-existent, and vice versa. I have seen that happen in 20 or so cases. The tapes with the bad chroma I will usually pop in all of my decks just to see how they might react in a AG-1980, JVC 9800, a JVC DVHS, and a old Panasonic Omnivision, and I have the same results.
My BVP-4 can usually fix these errors, but I have always been a little curious (especially in the rare cases when I can't match both tapes with the BVP-4) as to the reasons for this phenomenon.
I should also note that when these were originally purchased in the 90's they did not have any color saturation issues, they had the same quality that you would get with any new commercially produced tape. I am curious as to what might have caused this, and if it was something I did years ago caused this.
Last edited by OldMedia; 18th Apr 2010 at 12:58.
(I only want to express my self, and leave in peace.oK?)
Certainly you can keep your tapes safetly inside a lead's box, together with your money.
Tapes that they may be precious for you.
Results of a hall life !!!
Probably an engineer would suggest:
<< Life goes on !
But where and when do you come from?
Get my modern 3D-TV, with my 3D-HD player, and my software which will turn your tape's content into a 3D-HD format.
This is the fashion now, it's more "plaisir", and please let me keep my job.
And remember: having a job i wouldn't be a threat to you.>>
(Probably, he wouldn't mentioned the last.)
But everybody has to live, in this f-----g life!!!
I understand you and i feel the same, my friend.
I have no idea what the last post is about. Anyway...
According to IFLA:
It must be emphasized, however, that, contrary to layman's expectations, the magnetic information on properly stored and handled tape does not fade away.
It's just unfounded myth being repeated by idiots.
Sharing what I find, and the search continues...
This also applies to sharpness, not just color. I've come across a few sites claiming VHS "loses sharpness" as it ages, but again, it looks to be unfounded malarkey.
The truth is more or less in front of everyone's eyes, but it all depends on the eyes eyes.
Its all about knowledge and equipment. Say you recorded video to 8mm film some 20 years. Its more than likely that you were an amature taking video. But heck, its 20 years ago and you look like a pro because you are prob the only one with a $2000 camera, right? So you play the video on your tv back then. And, it looks great, right? True-that! But you prob don't know at the time that the video looks poor or that something is wrong because you, the amature, are not yet educated in the things of video. That will come some 20 years later, today.
But how much did you know about that camera and its features. Did you know that it did not use good color space processing? Or, did you know that the film stock you used was prob poor quality (ie, Eastman) and that over time, the video's color levels does fade? Just wait till you transfer it to your computer. But you probably won't notice or realize something is yet, still. So, you come here to read more about video and editing and processing and so on, and then it finally dawns on you that something is wrong--because you read somewhere (and seen) some discussions and demos of before/after examples. Lightbulb!
hmm.. or, how about when someone (could have been anyone, amature, pro, sky's the limit) transfered to vhs tape, but prob did so when the film stock was already loosing its properties, or the person's equipment was not to spec, ie poor color space processing, and so on.
There are so many different reasons for "color fade" or "color washout" is the result when one finally does work on their home video transfers, and these are just some examples or their origin of scope.
Botton line is more than likely that people that came here complaining of color washout or color fade are those who did not know their equipment all that well, and were or read misinformed information, and so on and so forth.
I'm thinking it's a mix of that, as well as the higher quality of modern displays, poor VCRs (either now or then), and simply having a bad memory.
Note how SignVideo mentions other degrade issues, but makes no mention of color loss on a proc amp page: http://www.signvideo.com/single_dual_proc-amp_video-processor.htm
Even if the chroma had reduced in level on the tape, you wouldn't see it as such.
The chroma signal includes the colour burst: a moderate amplitude reference signal, at the mid frequency of the chroma information. It sits between the horizontal sync pulse and the active video (picture information). It's recorded and replayed along with the picture content itself.
The frequency and phase of this colour burst is important, but so is the magnitude. Most displays and capture devices will measure the amplitude of the colour burst, and AGC the entire chroma signal to bring the colour burst (and, proportionally, the chroma picture information) to the correct level.
So if the chroma on-tape was reduced, the colour burst would be reduced too - and the display or capture device would simply boost it - meaning you wouldn't see any difference over all.
(Except it might be a bit noisier - and if it was way down, the colour killer would activate and give you a B&W signal only)
Lots of professional video from 20 years ago seems to have "faded colours" - the conclusion must be that it looked like that when it was first created, but we didn't notice at the time - either because we were used to it, or we had our colour saturation controls set higher to compensate. Or (same difference) video now is typically more saturated, so we have our colour saturation control set lower to compensate.
You can supposedly knock the high frequencies off magnetic tape simply by dropping it hard, repeatedly. This might be an old wives tale - it would be easy enough to check these days.
Any magnet (and magnetic tape is just a bunch of tiny magnetic crystals bound to a tape) can be demagnetized a bit by violent knocks (the magnetic domains become misaligned). They also slowly lose their magnetism over time. I don't know that any of this is of real consequence to VHS tape though.
Another thing to keep in mind is that the signal on the tape isn't simply an analog amplitude. The luma and chroma amplitudes are frequency modulated onto a carrier frequency. My understanding of how that works is shallow so I can't say if you could lose chroma without losing luma and the associated horizontal sync and colorburst references.
Last edited by jagabo; 19th Apr 2010 at 11:29.
Another patent with interesting information related to color and VHS:
There are various manners of filtering a spatial detail in an image area to a degree that depends on the degree of color saturation. The detailed description provides an example in which a spatial detail in an image area is emphasized to a degree that is greater as the degree of color saturation is higher. Emphasizing a spatial detail is tantamount to high- pass filtering. As another example, a spatial detail may undergo a low-pass filtering to a degree that is greater as the degree of color saturation is higher. That is, the spatial detail is deemphasized to degree that is greater as the degree of color saturation is higher. Such a filtering technique may be advantageous for processing, for example, a video signal obtained by reading a videocassette of the VHS type. On such a videocassette, color information is stored by means of a modulation technique that causes relatively strong noise in saturated color planes. This noise can be reduced by applying a degree of low-pass filtering that is greater as the degree of color saturation in the plane concerned is greater. Perceptual image quality will improve. This may also be obtained by applying a degree of high-pass filtering that is greater as the degree of color saturation in the plane concerned is lower. In such an application, details are emphasized only in image areas where color saturation is relatively low, which avoids an emphasis of color noise that would otherwise adversely affect perceptual image quality. There are numerous ways of filtering spatial details in image. The detailed description provides an example in which a difference in value between a pixel of interest and neighboring pixels is amplified on a pixel by pixel basis. As another example, a filter arrangement may detect particular sets of pixels that represent details, such as, for example, a set of pixels around a border of an object. The filter arrangement may then amplify high- frequency components that are present in such a set of pixels. For example, high-pass filtering may selectively be applied to areas in image that comprise relatively many details and in which colors are saturated to relatively high degree. Frequency analysis techniques may be used to identify areas that comprise relatively many details.
There are numerous ways of detecting the degree of color saturation. The detailed description provides an example in which an absolute difference is determined between a chrominance value that belongs to a pixel of interest and a central value. As another example, respective chrominance values of respective pixels in a particular image area may be analyzed so as to determine a degree of color saturation. A pair of chrominance values may be expressed in the form of a vector, which has a given length. The degree of saturation may be derived from the length of such a vector. There are numerous ways of generating a control value, which reflects the degree of color saturation, on the basis of one or more chrominance values. First of all, a suitable way of generating a control value will depend on the color format of the color image. Furthermore, a control value may vary with the degree of saturation in a nonlinear fashion. The degree of detail emphasis need not necessarily be proportional with the degree of saturation.
I've been trying to wrap my head around this one, and I tend to buy into with the premise that the playback device is completely rolling off the chroma signal due to lack of sufficient chroma information on the tape. The color fade would not be on the tape itself; but if there were chroma errors, the playback device may be shutting out all but the luminance signal.
Last edited by filmboss80; 20th Apr 2010 at 10:25.
I tend to believe this is a selling point for those involved in transfer services. As after LS mentioned this I did a little surfing to see who's saying what. Most of it was just that, business looking for more business. All mentioned I think has covered the bases(and without going over the same) on possible reasons for "color fade" if you will call it that. Dropout is the only thing I've seen in my many years of working with video. I have tons of 3/4 inch tape that play fine on my VO-5600 or VO-5630 after all these years(stuff from 70's, 80's television). The same with my Beta, Betamax, etc... they play fine, even the stuff I shot back in the mid 80's. But no color fade! (VHS stuff I really never used for storage as the wear on the tape as it moves across the heads, still that's not color fade)) A few have dropout , but this doesn't cause "color fade". It's from a number of things(i.e. wear, cinching in tape, wrinkle, edge curl, etc...). Chroma levels (low or high) may/could give the impression to those less knowledgeable. Duped stuff would have Generational loss, but as mentioned prior this wouldn't be color fade.
Now film would be a different animal, I have many color prints that have color fade . Actually most color film out there will fade in time. Just part of reality.
Again, I think this is just a way to drum up business. No one has shown me otherwise, and I'd be willing to admit I'm wrong if it could be proved.
If the Light ain't Bright, it ain't Right!!
Don't keep the tapes near the TV. When crt TVs are turned on, they produce a strong magnetic field for a short time to demagnetize the crt (and any tapes laying nearby).
For example: http://stason.org/TULARC/pc/video-faq/50-Degaussing-demagnetizing-a-CRT.html
It is unlikely that you could actually affect magnetic media but better safe than sorry.
Warning: Manual degaussers produce magnetic fields that can be damaging to data stored on magnetic media. Keep them well away from floppy disks and hard disks. It is also a good idea to just generally keep magnetic media away from your CRT.
Unlikely = "good idea" in the context above. In both cases, it sounds like guessing or repeating something without further tests to prove or disprove it. I hate that. Neither of those quotes is from an authoritative quality reference, it's just random people writing random things -- about on par with forum posts or sales ads. I don't recall ever coming across a good reference stating "keep VHS tapes away from TV sets because of magnetic fields".
There are some leaky things out there -- I have a really bad clock radio that can interfere with TV sets, as well as blank out audio tapes set on or under it. But you'd have to have a really fubar POS TV set for it to be the same way. I'm about ready to toss out the clock. It had a good 20-year run, but a power surge has one of the numbers perma-stuck at "9".
Thanks to everybody who's input so far.
I sent a few PMs and emails out, asking for input from folks I know "know things".
Maybe a bit of useless oar-sticking here, but... I agree. Nonsense.
It's perfectly possible for the tape media and the magnetic signal to degrade over time (or multiple plays), but that'll manifest as the classic dropouts, smearing, muffling etc that are characteristic of a low-grade or simply well worn video or audio tape.
I've got cassettes around that were bought to entertain me as an infant, when VHS players were still very expensive luxury items (and remote controls were attached by cords, not IR) - so we're talking 25+ years here - and though they show some occasional wear artefacts, the colours are still as vibrant as ever and stand up quite well to recordings made much more recently. Mind you, they haven't seen that much play for the last decade or so!
I have however gone and backed them up onto DVDR with a borrowed set-top recorder - both so a young cousin can watch them (without risk of his curiosity ruining these increasingly precious objects - much as my brother did with a different title and a wax Crayola at the same age), and because of paranoia at their future longevity. It is after all just rust laminated onto celluloid, and that's known not to be the most stable of materials in non-optimal storage conditions.
Mind you, I wouldn't bet against the cassettes STILL outlasting the DVDRs I copied them on to! It's really not a robust format even compared to VHS. (And for similar reasons I'm happy to still have an audio cassette player in the car... my home deck is worn enough that I can actually tell the difference between that and a digital source whilst the vehicle is in motion, but I have some that have been kicking around in such environments for the best part of two decades ... doubt CD would stand up to that treatment, even though there's far less risk of the media becoming entangled in the drive mechanism should it hit a random grease/moisture spot whilst playing...)
So ... yeah. I'd expect some kind of wear and signal degradation - but specifically the colour dropping out? Nonsense. It's a composite video signal at heart. It's hard enough pulling the colour out of that accurately when you WANT to. I'd even have fair confidence that this holds true for (AM colour) NTSC as well as (FM) PAL.-= She sez there's ants in the carpet, dirty little monsters! =-
Back after a long time away, mainly because I now need to start making up vidcapped DVDRs for work and I haven't a clue where to start any more!
just rust laminated onto celluloid
have fair confidence that this holds true for (AM colour) NTSC as well as (FM) PAL
Thanks for the quality input.
It's SECAM that's FM chroma. Both NTSC and PAL are Quadrature Amplitude Modulated (suppressed carrier). As LS says, apart from the intentional differences (phase alternating by line (PAL!), different carrier frequencies chosen), they're basically the same.
It's not a composite signal on tape. The chroma and luma are very much separate! That's not a reason why the chroma can fade though.
Regarding the issue of the degaussing coil in a TV set erasing your tapes: This is one of those things where it really, really depends on the exact circumstances in question. (And I've worked on quite a few CRT monitors over the years, so I do have some understanding of what's going on.)
For one thing, we have to keep in mind that both monitor technology and tape formulas have evolved over time, so what may well have been true 20-30 years ago may have gradually ceased to apply to newer equipment, but the "conventional wisdom" still gets repeated as fact because people don't realize the circumstances have changed.
Older color TV sets did put out a pretty significant EM pulse, relatively speaking, when they degaussed on power-up, especially when they hadn't been turned on for several hours and the simple PTC thermistor used to gradually reduce the current (and thus slowly collapse the magnetic field strength to zero) was at its lowest resistance. In addition, the larger the TV, the stronger the degaussing coil, so the coil surrounding the picture tube in a 27" or 32" console TV would throw off considerably more EM than the one in, say, a 15" portable.
At the same time, tape formulas have evolved considerably, with more exotic oxide formulas with more stable magnetic properties replacing the plain old iron oxide. Even tapes sold as "Type-I" today, or in the last few years, aren't the same as the old stuff from the '60s and '70s.
So, I think it's quite likely that back in the "old days", you could indeed degrade or erase the signal off a plain old type-I open-reel tape or cassette that got left left on top of a big 1970's Curtis Mathes console TV for several days, or weeks, of the TV being turned on and off... but that doesn't mean that a modern CrO2 or cobalt-gamma-ferric-oxide formula tape left on top of a 27" TV set made in 2005 will necessarily suffer the same fate.
I used to think that color fade / vhs tapes getting suddenly more blurry over time was real as well.
I had this home movie VHS that I remembered as looking MUCH better in the past and it was one of the reasons I was galvanized into this hobby of making VHS to DVD conversions. I gathered all of my tapes up to start making conversions -- and what do you know I found a dub (a Polaroid dub of a Scotch tape) of the tape I was complaining about before that probably hadn't even been played once since it was made 20 years ago.
Upon watching it I realized it looked exactly like the other tape did -- so either #1) it degraded in exactly the same way the other tape did, despite the fact taht it was stored in a different location and was likely never played or #2) My memory of the original tape looking much better in the past was invented in my head
I stumbled across a document by a guy how seems to understand how VHS works:
Unfortunately, English doesn't seem to be the guys native language. And I still don't grok whether or not VHS chroma could fade (at least theoretically) over time...
Yeah, that's Engrish + jargon,
but it does essentially state that there is error correction to prevent changes (i.e., fade and more)
The recovery of the original signal color conversion is done through a reverse way as in the recording. The imprecision of the mechanical system and the elasticity of the tape, resulting in reproduction as frequency variations. If we call the signal fsc printing, in reproduction we have (± f'sc f) where f error of reproduction (according to the values given above is (629 KHz ± f)). By means of a converter that is applied to the same RF as that used in the recording, it reconstructs the color information in its original value of sub carrier. For the error disappears, a comparator circuit is used, this device compares the output of the mixer with a standard oscillator, the comparator circuit produces a control voltage which changes the oscillator frequency used for conversion, so that the signal has no error output. (4.211 MHz ± f) - (0.629 MHz ± f) = 3.582 MHz
I believe that "signal has no error" refers to a specific aspect of the signal, not a general "the chroma signal that comes out is exactly the same as the chroma signal that went in". This is plainly obvious when you see that the chroma channels coming off VHS tape are much noisier than the signals that went in -- ie, it's full of errors.
It's talking about the frequency + phase.
If the high frequency composite chroma signal were recorded to tape as-is, the amount of jitter on VHS (even at its best) would destroy the colour information - you'd simply get the wrong colours. The only thing that would survive would be the saturation.
In the colour under system of VHS, the chroma is stored at a low frequency. This already makes it less sensitive to jitter. But there's more...
To move it from the high (original) frequency, to the target (low) frequency, you multiply the chroma signal by a sine wave. If you have chroma at 4.43MHz, and want it at 0.626MHz, you multiple it by a sine wave of 5.056MHz, giving you a signal at 0.626MHz which you record to tape (and another at 9.486MHz which you dump).
When you play the tape, you get back a 0.626MHz signal. To move it back up to 4.43MHz where it started, you again multiple it by a 5.056MHz sine wave, giving you a signal at 4.43MHz (the chroma), and another at 5.682MHz (which you dump).
Now, here's the trick. If, on playback, you used a clean 5.056MHz sine wave, then all the jitter from the tape would be present in the final output. However, if you generate the 5.056MHz sine wave based on the sync signals (15625 of them per second for PAL) coming off tape, so that those sync signals, with all their jitter, give a 5.056MHz sine wave with the same jitter, then the final 4.43MHz chroma output is jitter free - because the jitter in the on-tape 0.626MHz colour-under chroma signal, plus the matching jitter in the 5.056MHz sine wave, cancel each other out.
That's what the guy is trying to say. His numbers are for NTSC. My numbers are for PAL.
In practice, it's more complicated. There are 2 (special) modulator stages, giving two sets of intermediate frequencies - but the principle is roughly as outlined above. And it works!
(There's an even clever trick used to minimise crosstalk in filtering, but I'll leave that for another day).
None of this is at all relevant to colour fade.
I heard this used in a reference to s-vhs recording.
S-VHS tapes are a higher / quality grade etc as far a coatings are concerned.
S-VHS tapes cost more than standard lower quality VHS tapes.
But some cheapskates have worked out that the mechanical difference between the two tape types is a little identification hole drilled on the under side of the cartridge.
Therefore many people who used these machines and did not want to fork out for the more expensive s-vhs tapes have modified standard VHS tapes using this method.
So a standard VHS tape can be used to fool a S-VHS machine that it contains a S-VHS tape and it will record to it in the same S-VHS higher bandwidth way.
The problem is while initially this will work and your recordings will be fine for a while. Due to the type of material used on the lower quality VHS tapes the recordings made to these modified tapes wont last as long and magnetic strength will fade very quickly over time. Possibly leaving you with an unplayable or badly degraded recording.
magnetic strength will fade very quickly over time
Time doesn't much affect this.
"Fade" is also a crappy word to use. The dictionary definition of this word does not apply to anything in the realm of analog VHS or S-VHS recordings (or most other tape formats). See http://dictionary.reference.com/browse/fade -- and no, #4 does NOT apply, as it still implies visual loss (disappearing) or life essence (plants/animals dying).
Note that I did this for about a decade. On a good VHS tape, you could save 50-75% of your funds, and still have excellent quality. This method was mostly applied to hobby uses, of course, and not pro use. (That would have been stupid, although I saw it done by others.)
When you cheated and used VHS tapes in S-VHS mode by tricking the machine (i.e. not S-VHS-ET), the problems were as lordsmurf said - but these applied only to luma - more noise, and poorer bandwidth compared with real S-VHS. It didn't make any difference at all to the chroma, because VHS and S-VHS treat the chroma in exactly the same way. That was one of the weaknesses of S-VHS - the chroma is still lousy!