What are the best s-video cables to get?
 

Hello, everyone! I'm new here. I've done a bit of video capture over the years just to have a handy copy of family videos on the computer, but never for true archival purposes. I've recently realized that I need to get a permanent digital copy of those old VHS tapes sooner rather than later. I re-captured some of the same videos last year with a newer Hauppauge USB-Live2 dongle, and was surprised to find that the quality looked worse than the captures I had done 15 years earlier.

At first, I thought that my VHS tapes had degraded. After a lot of research online -- most of which kept bringing me back to this forum -- I've come to realize that the problem probably isn't my tapes, but my equipment: a consumer grade VCR and a capture device that's optimized for modern signals instead of obsolete ones.

Long story short (too late?), I've now spent more money than I ever thought I would on decades old A/V equipment, and next week I should have a JVC HR-S9900U VCR, AVT-8710 TBC, and ATI All-in-Wonder 9600 capture card arriving next week. Possibly to be followed by a proc amp, if it seems like it will help.

I'm sure I will be on here peppering all of you with questions as I embark on this journey. The first of them is this: What S-Video cables should I get? It seems like a ridiculous thing to worry about compared to the expense of the equipment, but by the same token I don't want to shoot myself in the foot by chaining my new equipment together with cables that degrade the signal I'm trying to clean up!

I know from past experience that cable quality can be all over the spectrum, and that price isn't really an indicator of quality. I'm pretty sure I have some S-Video cables somewhere in my basement that I have collected from various electronics over the years, but I've never used them (I transitioned straight from composite to component, skipping over S-Video entirely), and I tend to doubt the quality of any cable that comes included with a device.

So... what are your suggestions and tips for S-Video? Any particular features I should look for? Any particular brands more trusted than others? I assume I'll want to keep the individual and overall cable lengths as short as possible, and route them away from any potential sources of electrical interference. Any other considerations?

I'm an Amazon Prime member, so for convenience I'm partial to something I can buy from there. But I also have accounts at B&H, Adorama, and Midwest Photo Exchange, so if there's something that a pro house is more likely to have, I can check there.

Thanks in advance!

There are two schools of thought on cables, whether RF, s-video, component. digital coax, analog interconnect, whatever. Those two schools of thought are:

A) All cables are alike. Wire is wire, plugs are plugs. All wires look and sound alike.

B) All cables are not alike. Wire differs by materials used, impedance characteristics, and purpose. All wires do not sound and look alike.

The main problem that people in the (B) group have is that people in the (A) group don't have aural or visual sensitivity to detect a difference, and/or they don't care one way or the other.

The main problem in the (A) group is that no one in the (A) group has a convincing answer for why so many professionals, videophiles, and audiophiles are in the (B) group.

I have tried the cheap cables from Monoprice, Cables To Go, Amazon Basics, Belkin, AR, Monster, and plenty of others. I threw all of them away for unexciting or just plain bad performance except for those that broke down or lost connecting pins or ruined one of my input sockets. Those that didn't break gave me soft pictures or grungy audio.

The cables I have stood by are the Pearstone line from B&H photo (forget about the hype on gold connectors, they don't hurt but they don't matter either) which used to go by another name. They've been durable and gave decent results that I can't complain about. On the other hand my premium cables are the very durable and more expensive s-video cables made from Belden broadcast-grade 75-ohm wire and good Euro connectors, sold by BlueJeansCable.

There is also a "C" group claiming that the more you spend, the better the cable. Nonsense. I've tried s-video cables from the USA and UK that cost up to $600 USD for 3 feet. None of them were worth it, including the cockamamie idea of charging $350 for a piece of cable where two parallel wires were frozen solid until they stuck together (thank heaven I got my money back. They were horrible, and you couldn't bend the damn things). I stuck with the Belden pro wire and more reasonable cost, which has given excellent low-noise performance, good contrast, and no color casts. BlueJeans makes some pretty clean sounding audio stereo and clean looking component/composite solid-core cables, too.

On the other hand if you subscribe to the (A) group, it's easy. Just go to Amazon and look for the many pages that list a lot of s-video cables, stop at a page you like, say eeny-meeny-miny-mo, and get the cable your eyeballs settle on. Or click the button that says something like "list prices low to high", then click on the cheapest cable and order all you want.

I definitely consider myself Group B, or I would bother asking for cable recommendations. :)

As for Group C, I know that isn't true because Monster Cables exist (Monster Markup / Monster Marketing).

What are your thoughts regarding cable length? Thinking about where I'd have everything laid out on the desk, I could have 1.5' cables between the VCR, TBC, (and maybe proc amp), and then a 6' to where the capture PC sits. I just figure there's no sense in getting a bunch of 6' cables and having all that excess dangling when most of the equipment will be in close proximity to each other.

I'll put put Pearstone and BlueJeansCable on my list to check out. Thanks!

Most of my cables (s-video, component, composite, audio, digital coax are 3 feet to 4.5 feet, but I have three 10-foot s-video cables, a 12-foot s-video, and a 35-foot s-video in regular use.

Pearstone was at least as good as the best cheapo Ive ever used, but BlueJeans were best for capture. Among cheaper cables if you can ensure they are made of solid-core wire you'll have fewer impedance problems and can use longer lines.

I've had Monster. Never again, not at any price high or low.

It's rare to have an s-video cable perform poorly, regardless of the brand or cost. Cheap and expensive cables often perform the same. In fact, I find that expensive cables often perform worse. Quality differences in s-video is mostly imagination. Component, composite and HDMI is where you can run into issues, not really s-video.

I recently had to toss out a DataVideo branded cable (came with the TBC-100 cards), because the header was shorting out, and started to attract line noise.

My favorite s-video cables are the ones that came with JVC S-VHS VCRs for free.

I'm not a fan of Monster. You pay for the name brand, not any actual quality. The shielding on their composite and component is good, but at twice the price of other equally good wires.

I forget the brand off-hand, but most of my wires have come from Monoprice in recent years.

It's easy to see line noise, when your TBC, VCR and capture card output clean signals. I test cables this way. I mostly have to throw out composites, because they break down over time, used or not. I toss several per year.

Here is my personal opinion on cables:

For high impedance cables such as interconnects, shielding and length are important because any noise picked up by badly shielded or long interconnects will be further amplified with the signal and reduces signal to noise ratio.

For low impedance cables such as speaker wires, gauge is important especially for long runs.

In either case expensive cables are not the remedy, Keep high impedance cables as short as possible, as shielded as possible. and low impedance cables as thick as they go longer, shielding is not important because the signal has already been amplified.

None of the above apply to digital cables, $5 HDMI cable does the exact same job a $120 monster cable does assuming they are rated for the same data speed.

So to answer your question, A well built S-Video cable with enough shielding is all what you need, Keep the length under 3 feet. Such cables usually cost anywhere from $5 to $15. Some sellers show the structure of the cable layers, otherwise just buy from a reputable vendor like monoprice cables.

Something like this, Or even Monster cable if it's priced right.

This is pretty much representative of the "A" group response, which was described earlier in post #2, above. If you are in the "A" group you are wasting your time reading articles and opinions about wire, since just about anything will work for you. For an "A" individual, monoprice is the primary source for all gear and cheapest is usually best, unless you're swayed by marketing and markup at BestBuy (but I would advise against Monster Cable's widespread use of cheap 50-ohm hookup wire in their premium priced products or MC's false ads that claim HDMI is 75ohm wire).

If you're in the "B" group you'll spend more time (and less money) reading articles that go into more detail about specific wire types and applications. For example, a "B" member might be looking for interconnects or speaker cable at Absolute Sound or HomeTheater.com or forums like AVS, while a "B" person looking for the real thing and descriptive detail about HDMI might be browsing https://www.whathifi.com/products/ac...es/hdmi-cables. To summarize the two points of view from the earlier post (without casting bias toward either group):

I can't agree with some of the foregoing post, for instance the implication that impedance is somehow a factor in speaker wire. Unless a speaker cable's impedance is unusually high (impedance increases with length and diameter), the main factors with speaker wire are inductance, capacitance, construction (stranded or solid core) and amplifier loading characteristics depending on the speakers used. For the "A" group, heavy-gauge A.C. cord from Home Depot is great for long speaker wires, while the typical "A" living room would have lamp cord tucked behind the furniture. "B" users on a budget would gravitate toward Audioquest Type4 while a more obsessive "B" would have Kimber Kable 12TC's soldered directly to their speaker connections.

Debate plus comparison tests and actual use reports are of interest to the "B" group. I don't know why anyone in the "A" group would be interested in this thread or in similar articles.

Signal cables (audio, video, RF, digital information, etc.) and not as simple as the Group "A" would have one believe. Group "B" provides a studied approach, attempting to match the cable performance to the requirement. And depending on the application it entails looking at issues such as losses, impedance, insulation value, shielding, connectors, durability, plenum rating, electrical environment of the application, etc. (Impedance matching of cable, connectors and i/o becomes an issue the cable lengths approach a significant fraction of the wavelength of the highest frequencies carried on the cable to avoid issues with reflections/ghosts. It is also a factor in efficient electrical power delivery.)

Group "C" is into making a statement; e.g., their pigs sport a Rolex and wear Guerlain KissKiss :love:. (In contrast Group A's pigs wear Timex and Rite Aid's Wet n Wild :2cents:, while Group B's pigs generally don't wear a watch or lipstick :congrats:).

I believe HDMI signal cable is based on twister pair and a nominal 100 ohm impedance (+/- 15%). Twister pair provides some noise immunity in balanced configuration. It also has still evolving specification (e.g., HDMI 1.0, 1.1, 1.2, 1.3, 1.4, 2.0, 2.1, etc) for loss, frequency response, phase shift, cable length, signal paths, etc.

For normal audio and SD composite, s-video, and component video in a typical home edit studio environent there is little if any benefit from spending for exotic cables. The bandwidth is limited, the runs are short (much less than a wavelength), and the electrical noise environment is not demanding if not close to a transmitting tower or under an EHV power line.

I'm both amazed and amused we're having an in-depth discussion (argument?) over cables. :)

Cables are one of the easiest aspests of video, one that requires little effort and costs.

Cables either work, or they don't.
Either noisy, or not noisy.
There's not a gradation of quality, in terms of visual performance.

The only aspect of a cable that is graded/variable is physical construction, such as gauge and shielding. But it only matters if problems arise, as caused by the environment, the power, or feedback from other devices in the workflow. I often tell the story of moving from TX to TVA, and back again. I had issues with wires, UPSes, TBCs (DataVideo), and others, on that first move, requiring switching to different products (ie Cypress). And upon moving back, the TBCs (both DataVideo and Cypress) and most wires were fine, but the UPSes were hosed by the power grid. I've seen many cases over the years, of certain VCRs or TBCs or DVD recorders not liking certain cables (or other devices in a workflow), which is why I so often have to test full workflows, not just individual pieces. It's exhausting at times.

But, in the end, visually, it just works and does not. No middle ground.

All capture devices are alike. They just look different and sometimes come in different shapes and use different materials, but they all do the same thing, so they're all alike. Because all wires are all alike, it doesn't make any difference how capture devices are connected to players or other capture components. You don't have to try recommendations by those who claim that some capture devices are better than others, because they're all alike and so the results will always look and sound the same, so why try different products and compare them? After all, light is light, color is color, sound is either there or it isn't. Either the products produce a picture and sound, or they don't. So what difference could one component or wire make over another? They all deal with the same colors and sounds and they all send the same colors and sounds to the same places, and once a video is in digital form they all look alike because all of them are just 0's and 1's.

Comparing the performance of different products along the same product line and publicizing reports and performance experiences and getting into details is a waste of time. It's all in the imagination. Pictures are all alike, sounds are all alike, 0's are 0's and 1's are 1's. So why is this being discussed?

Time for my nightly shiraz.

Audio and video have many variables, and variables within variables, and varying degrees of accuracy and acceptability. So capture devices are rarely alike, each with different ways to interpret that audio and video. (Unless using like components, a clone/rebadge.) Any picture or any sound is not good enough.

It's very different from the binary nature of cabling. Because again, any picture is not good enough. You want no noise, not noise. There really is not middle ground, no semi-noise.

Skip the shiraz, have some champagne, you deserve it. :)

Can't accept that. If you're going to use one form of home-made logic for certain entities that you don't like or don't understand, you have to apply the same logic elsewhere.

Just as captured devices are not all alike (as you say) and don't perform in the same way, so all wire is not alike and does not perform in the same way. To say that all types of the same product are alike is not logical because it is pretty self-evident that all wire components don't perform in the same way. The discussion is not about price. It's not even about numbers. Anyone knows that an under-$100 ATI 600 USB captures cleaner lossless AVI video from VHS than a DV or Black magic device. People don't use "measurements" and "numbers" to critique the differences in quality. Of course, there are those who say there is no difference. What would you say about those who contend that differences don't exist?

Those who look at three different captures from the gear and three methods described above and conclude that all three captures look alike would be in for considerable disagreement with viewers who have better eyesight. By the same token, viewers and listeners who contend that all wire looks and sounds alike perhaps should observe more closely.

My own experience with monoprice s-video is that the images are soft, lack contrast and detail, have unusually murky shadows, and display a lot of fuzzy transmission noise that is similar to "floating tape grunge" especially in large gradient areas. My experience with Monster s-video is that it alters colors visibly (darker flesh tones and shadows look green, most midtone skin colors are too reddish), chroma blotches appear in dark areas (usually cyan), the wire "enhances" edges with noisy oversharpening effects, and accentuates highlights. This is indicative of similar effects with other wire that doesn't meet video impedance standards. On top of that, Monster's connectors are really cheap. My experience with s-video store brands from BestBuy and Amazon is similar to the monoprice stuff. Acoustic Research "blue jacket" s-video cable is well made but is disappointingly over-warm, soft. and noisy. Their Pro Series "gray jacket" is heavier and images look cleaner and color is more accurate, but they do lack a certain "snap" in detail compared to the few best that I've used -- but I'd prefer them to most other s-video at much higher prices.

If they all look the same to you, then just say so. It's not the same thing as saying "they're all alike". What you really mean is that "they all look alike to me". They don't all look alike to many other users. My wife knows zilch about video, but she can certainly see differences between video cables.

I wouldn't be surprised if this continues forever. However, I would like to see more detail from the "A" side. I don't mean just naysayers, they're everywhere. Other than the cheapest piece of junk wire you could find, what else have you actually tried? I've extensively tested and used about 30 s-video products over the years, from dirt cheap (is 57 cents cheap enough?) to $600 USD. How about you, how did you test, and what was your experience?

The electrical properties of cable; i.e., resistance per foot, impedance, capacitance and inductance per foot, conduction chemistry, conductor diameter (think skin effect), shielding effectiveness, thermal properties on the materials used, length, etc. will all have an effect on the ability of the cable to pass a signal. The connectors and assembly also play a significant part. The result is some cable will have lower losses, high frequency roll-off, phase shifts, noise pickup, etc. than others. Longer cable runs will tend to increase these effects, and impedance discontinuities caused by mismatched connectors, cable kinks, and terminal gear can induce reflections and other distortions. Wonder how many recall the good old days to antenna TV reception and the sliding a bit of AL foil wrapped around 300 ohm twin lead to "tune out" interference.

I am not one of the folks who can see the difference between most VCR/DVD player manufacturer-provided s-video cables and some premium brands in my applications, which are mainly short (<3') runs. The other warts of SD and VHS sources pretty effectively mask them. I have encountered issues with pins breaking on the NXG Sapphire s-video cables I tried after only a couple uses. But I am not about to spring for a $600 cable for testing purposes.

Is it time to start a discussion of the effects of storage temperature and humidity on cable performance? :yahoo1:

The properties you mention obviously affect cable performance. I wish we had more definitive data on that, although impedance, capacitance, and inductance are commonly mentioned in many discussions I've seen. For construction properties, I usually detect a difference in effect between solid-core and stranded cables especially with audio wire.

For me as well, most wire looked and sounded generic. A very few had definite "problems" and glitches (such as Monster) but if you were to say that "most of it is alike", I'd agree. As far as "premium brands", I had no luck with them, from Monster to Kardas, and the $600 brand from the UK was no better than the generic monoprice stuff (beautifully made, but so what?). The only s-video "brands" I ended up with were budget Recoton (aka RCA and a few other marketing names) and BlueJeanCables, and I guess I owe a qualified nod to Acoustic Research Pro II. BJC is the pricier of the three, but that's a long way from the fantasy prices some high falootin' brands were asking and was cheaper than Monster's usual retail.

Storage temperature? Now, that might be fruitful.....

Given some of the sloppy work I've seen coming over HDTV cable these days, it does make me nostalgic for the simple life. Our calibrated CRT had a picture so movie like, that.....well, that's another story.

-- merged, 3 months later update --

Now that I've ordered and auditioned a new replacement YC2 s-video cable to replace an older, shorter one, it seems BlueJeans has made some undesirable changes that aren't so great. The new version has a soft, lower contrast image that looks pretty much like generic s-video from Amazon or Monoprice. I have to withdraw my recommendation for that cable. Phooey.

Cables *can* affect signal quality, but whether they do, in what manner, and how much is a very big, "it depends".

In general, any cable, regardless of type, construction, or materials will have the following properties: resistance, inductance, and capacitance. Now, the materials used, and the type and construction of the cable will have an effect on the relative quantity of those properties per cable length. And all of these properties will interact with the electrical properties of both the sending device and the receiving device to create an electrical voltage divider that causes overall signal loss, and filter effects which alter the signal.

Every single cable in existence does this, because physics. This effect cannot be escaped. The important question is: Is the effect large enough relative to the information we were trying to transmit over that cable to begin with, that the effect is worth caring about? And in a cable fit for its stated purpose, the answer is, "no". Otherwise, it wouldn't be fit for purpose, now, would it? But, how much effect is too much effect, and therefore worth worrying about? More than a dB? Maybe. It certainly depends on the size of the effect relative to what else is going on in the signal. But certainly anything less than a dB is utterly undeserving of concern.

For an example, almost a decade ago, some guys on a guitar forum were arguing about kinds of speaker cable; whether certain sizes caused tonal shifts or overall sound losses, and whether premium cables were worth it. At audio frequencies, and output power levels, any filter effects formed from the speaker cable's inductance and capacitance are so small as to be utterly swamped by the amplifier's output transformer or the speakers themselves. The only factor that could possibly be worth worrying about was the resistance of the speaker cable itself. Willing to be That Nerdy Guy, I bust out Ohm's law, which shows that in the worst case scenario of a guitar amp putting 100W across a 4 ohm speaker load, assuming 6 feet of 18AWG between the amp and the speakers, less than a quarter watt is being lost as heat in the speaker wire. At 100W of output power, .25 W loss in the cable isn't even a hundredth of a dB. While perhaps measurable with sufficiently expensive equipment, will certainly not result in a perceptible reduction in volume. And <0.25W of heat being dissipated by the wires themselves isn't going to be enough for them to heat up enough to cause any damage, either. So, any possible electrical effects from 6' of 18 AWG speaker wire aren't worth caring about in this scenario

Coming back to video-land, what do we care about when it comes cables? The same thing: that any amount of signal loss or filter effects are below the "don't care" threshold. Just where is that threshold? That threshold is probably different for each member here, but for me, it's sufficient bandwidth to carry the information contained in the tape. The cable's quality could go to hell just after whatever our chosen bandwith threshhold is. As long as any effects from the cable are below 1 dB within the bandwidth we do care about, any effects outside of that are utterly not worth caring about. There's going to be 10s of dB of noise on the tape itself, swamping any effects from the cable in comparison. You've got bigger fish to fry than worrying about cable effects well outside the bandwidth needed for video.

How much is sufficient bandwidth? From what I've been able to read, when digitizing an SD analog signal, that signal is sampled at 13.5 MHz for luma. Per the Nyquist-Shannon sampling theorm, the highest frequency that can possibly be accurately captured and represented without distortion is half the sample rate, so in this case, 6.75 Mhz. This is the same rule that makes the CD's 44.1 KHz sample rate sufficient for 20kHz audio. The 20kHz limit of human hearing is about 91% of the theoretical limit of the sample rate, which gives us a little bit of wiggle room to account for losses in real-world electrical devices. A sampling rate of 48kHZ for audio gives us even more wiggle room (83.333...% of the limit vs 91%). So applying that same amount of wiggle room to our concern of luma bandwidth, as long as our video signal doesn't contain more than 5.625-6.1425MHz, we can digitally capture it. The luma bandwidth of 3.4MHz for a VHS tape and 5.4MHz for a S-VHS tape are within this limit. Any theoretical information or noise in a video signal above the sampling limit of 6.75 is not worth caring about, because your capture card was never going to capture it, and thus you were never going to see it anyway.

So we have a range of thresholds here for whether or not a cable has an effect worth caring about: 3.4Mhz of bandwidth for "Good Enough", to 6.75Mhz of bandwidth for "Stop wasting your time on money on something that doesn't matter, and spend it instead on something that will actually have a real effect worth caring about, like a TBC".

How to determine if s-video cables have sufficient bandwidth? That, I'm still figuring out. I'll get back to y'all when I think I have something.

Edit: Certainly, anything built from this: https://www.showmecables.com/media/s...ec%20sheet.pdf, is more than sufficient. Note the loss spec on the 2nd page: 1.7dB per 100 feet at 5MHz. At 6', any losses are so small that they probably can't even be measured without all but the most expensive lab equipment. In any cable made from Belden 7700A, the quality of the connectors and termination of the wire thereto will be the only things capable of any level of signal degradation worth caring about. Same goes for Belden 1807A or 1808A. Money spent on trying to find something better than those three cables is wasted money better spent on a TBC, quality S-VHS deck, or quality capture card.

Not sure what you mean by that measure. Keeping in mind that dB is a ratio.

FWIW: 0.25 watts is -26 dB below 100 watts.
VHS signal to noise rations run about -40 dB.

You're right. I phrased that poorly. You phrase it better below.

While true, .25W being lost to heat in the cables out of 100W from the amplifier means that the power reaching the speakers is 99.75W, which is less than a hundredth of a dB below 100W. You're not losing -26dB to the 6 feet of wire connecting the amp to the speakers.

Which reinforces my point. If your cabling degrades SNR by less than 1 dB within the bandwidth of interest, it's not worth caring about. You might be able to measure it, but I highly doubt you're gonna see the difference between an SNR of 40dB and an SNR of 39dB in your capture.

Because science. :D

Some of the above discussion makes loose use of the term "dB." In terms of losses a cable doesn't know signal from noise, so except for noise added by the cable (e.g., stray field pickup and thermal effects) the the signal and noise will be attenuated equally for any given frequency.

Audio cables, for short runs and as in typical in-home applications, can be generally modeled by the DC parameters of the wire. (But for long runs such a phone lines the DC models are not adequate.)

For video applications it becomes complicated at much shorter distances. Cables, such as s-video, are in effect transmissions lines for radio frequencies and modeling them is more complex. Losses are more than simple DC resistance of the wire, and they are frequency and distance dependent.

There has been a lot of marketing for exotic cables, and buzz words such as oxygen free (which may be a 1% impact on bulk resistance of the copper). Oxygen free was invented for use in plating/sputtering applications. Use in audio cables arose as a marketing ploy to separate consumers from their money.

Because signals are often compressed over portions of the brightness range (think gamma curves), a small effect in signal strength can have a larger effect on the displayed image.

Bottom line is the construction of the cable: i.e., durability, quality of shielding and dielectric, and electrical discontinuities at connectors (e.g., changes in characteristic impedance), are what matter. Problems can be minimized by using shortest possible cable runs, routing away from power cables, variable magnetic fields and currents, and sources of high frequency emissions and by care in handling to avoid physical abuse of the cable. And super cheap cables are risky, while expensive cables are no assurance of noticeably superior performance.

My best guess for why it's complicated for video but not audio has to do with the relative wavelengths of the maximum frequency of interest, and the length of the line being relatively close or relatively distant from a significant fraction of that wavelength. Is that the case, or did I just go further out into left field?

Okay, I'm curious what models for video applications are appropriate and where I can read up about such models and how to use them? That's not intended as a challenge. If my mental model for evaluating video cables is inadequate, then I need to remedy that.

I feel like I'm derailing this thread. Perhaps this should be taken to another thread, perhaps in another section of the forum?

On this, we are in complete agreement.

Some leisure reading for a start. (There is significant math involved.)
https://en.wikipedia.org/wiki/Skin_effect
https://en.wikipedia.org/wiki/Telegrapher%27s_equations
https://en.wikipedia.org/wiki/Reflec...nducting_lines
https://en.wikipedia.org/wiki/Oxygen-free_copper
Beyond the above overviews you can delve into the chemistry, physics and related engineering (material science) if you wish more depth.

Note that frequency and wavelength are related by the speed of the electrical signal through the media. For air this is essentially the speed of light in a vacuum (~3x10^8 m/s). For wire/cable it is less; e.g., typically 79% of the speed of light for RG-59/59 coax, different speeds for different cables).

Not a derail. Carry on. :cool:

Okay, you've got me reading and learning things, so thank you. You're right, there is some non-trivial math involved in this. Even though some of the math is over my head, I think I'm starting to have fun.

I did some more reading about transmission lines. You're right. Skin effect, characteristic impedance, reflections, it's all important. It's what I was intuiting and alluding to in my prior statement:
However, reading up more on transmission lines, there's a caveat here, the characteristic impedance of a transmisison line, and its consequent effects (reflections, etc) only matter when the transmission line is longer than the 1/4 of the wavelength of the highest frequency of concern.[1]

Recall that our highest frequency of concern is 6.75 MHz[2]. The formula for wavelength is λ = v/f, where λ is the wavelength, v is the velocity and f is the frequency. Your reference to the speed of light in coax was another clue I needed, apparently electricity in a cable does not propagate at the same speed as light in a vacuum, it can be a good deal less, depending on the properties of the cable. The Wikipedia article on Velocity Factor[3] contains some examples.

Assuming the speed of light in a vacuum, the quarter wavelength of 6.75MHz is 11.1 meters, or 36.4 feet. However, as dpalomaki correctly mentions, our signals are not propagating at the speed of light, but at some fraction thereof. Per [2], data cables can have a velocity factor of as little as 42%. To get an idea of a worse case, let's assume we have cheap, low quality cable with a velocity factor of 40% of the speed of light. 40% of 299792458 m/s is about 119916983. Plugging that into our wavelength formula above, we get a 1/4 wavelength of 4.44 meters, or about 14.5 feet.

Shorter than this distance, it will be the interaction between the the input of the capture card and the output on the S-VHS deck that dominates the quality of the signal your capture card, well, captures. This assumes that there are no appreciable transmission losses, or additional noise coupled in caused by the cable itself.

So as long as our hypothetical cheap cable is shorter than 14.5 feet, has losses less than 1 dB at 6.75Mhz, and doesn't let more than 1 dB of noise to get into the signal, and doesn't have any other problems like breaks in the insulation or shorts between conductors, than this cable is good enough, electrically. It could have other problems, like flimsy connectors at both ends that are going to break the next time you look at them wrong, but for the goal of getting the analog signal from your S-VHS deck into your capture card, you are not going to be able to tell the difference between it and a fancier, more expensive one with a better bandwidth or velocity factor.

I would be willing to pay money to someone who could prove otherwise in a properly done, double-blind ABX test.

Now, that doesn't mean all cheap cables are good enough, nor all expensive ones wasted money. More expensive cables should (hopefully) have quality connectors that are going to survive being connected. And note that 14.5 feet isn't so crazy a distance for a cable. Most consumer video electronics I've worked with have come with cables that range from 3 to 6 feet long, and honestly most of them aren't constructed like ideal transmission lines. Well, that's a reasonable distance for the most common scenario of the Xbox or DVD player being placed on a shelf just below the TV. At that short of a distance, you don't have to necessarily worry about characteristic impedance as long as you're not allowing noise to leak in or causing appreciable signal losses. However, if you need to put your capture computer on the other side of the room as the S-VHS deck, because you live in a small apartment and have to play Tetris with your furniture layout thus, you could find yourself needing a cable longer than 14.5 feet pretty quick.

Well, 14.5' was an estimated worst case, with a cheap, crappy cable. Playing with the numbers from my "definitely good enough" candidates of Belden 7700A, 1807A, or 1808A, we get a lengths of 28.4' (They all list a velocity factor of 78%). That's certainly a lot better, but a still a distance you could conceivably run up against in a practical situation (maybe the S-VHS deck has to sit in the next room). Fortunately, the characteristic impedance of all three of those is 75 Ohms, which happens to be the common standard source and input impedance for input/output ports for video signals.

So you could take an S-video cable made from Belden 7700A, 1807A or 1808A, terminated well to high-quality min-din 4 connectors on either end, and that cable could be almost 100' long before you had to start worrying. If you need your S-VHS deck to be more than 100' away from your capture card, I think the easier and cheaper thing to do would be to find a way to move them closer together, rather than worrying about cable lengths.

Now all we need to do is find some quality mini-din 4 connectors. Terminating the cable to the connector will still be a crapshoot, as not everyone can or cares to learn how to solder well, but there's nothing I can do about that.

1. https://www.allaboutcircuits.com/tex...mission-lines/
2. http://www.digitalfaq.com/forum/vide...html#post62771
3. https://en.wikipedia.org/wiki/Velocity_factor

Just an an interjectory reminder, the cables that came with DataVideo TBCs and JVC S-VHS VCRs are what I use almost exclusively.

- The JVC have worn out in time, especially seeing as most of the came from the late 90s and early 00s.
- But the DataVideo have not. However, some DataVideo were just bad right away, tossed aside.
- Other branded cables have been all over the place, with the main issue being visual signal noise, mostly on runs longer than 3 ft.

I've never really found a digestable/understandable reason for this, so I'm reading your conversation here.

Carry on... :)

Some potential sources of noise with cables:

- Poor shielding allowing stray signals to enter.
- Cold solder joints/connections that might rectify RF that is picked up by the cable.
- Weak connections that generate static as there is microscopic relative movement of contact points, perhaps even microphonics induced by ambient sound, building vibration, or mechanism vibration.
- Ground loops
- failure to have a low impedance common ground for all connected gear.
- fields from nearby power lines, light dimmers, switching power supplies, etc.

https://incompliancemag.com/article/...out-shielding/
provides some interesting reading on cable shielding.

Lots of factors come into play, and a lot depends on the specifics of the installation at hand. Bottom line is go with what works for you. Pricey Monster cables and their ilk do come with bragging rights :beer:.

I suspect the reason you're seeing noise in some cables that are longer than 3', is the top reason that dpalomaki lists there: poor shielding.

Per the pinout of S-video the pins are Y, Y Ground, C, C ground. Both the Y line and the C line are supposed to be individually shielded.

I suspect no small amount of S-video cables are made wrongly. I've also seen the pinout listed as Y+, Y-, C+, C-, which would imply that the Y and C lines are differential instead of single-ended. Someone could see this and think that he can use 2-pair unsheilded twisted pair to make an s-video cable. Unsheilded twisted pair does have noise rejection properties, when the transmitting and receiving ends are wired to use it as a differential transmission line. In such a case, the noise is common mode, and gets cancelled out, as the receiver receives the difference between the voltage on both wires. If you've ever worked in audio and dealt with XLR cables on microphones or other studio gear, those work on the same principle. However, since the Y line and C line are each supposed to have their own grounds, both are supposed to be shielded, twisted-pair isn't going to do much for rejecting noise when the input it's connected to was designed assuming the line would be shielded.

There's another way a cable could end up poorly shielded, and that's by cheaping out on the on materials. Instead of using a pair of shielded wires, one might end up using simple 4-wire cable. Both shielded and unshielded variants of such exist. In an unshielded setup, noise is gonna couple into that line pretty quick. But, even in a shielded 4-conductor cable, since the Y and C lines aren't shielded from each other, you could easily get cross-talk between the two. That'd be the Y signal coupling into the C signal as noise, and the C signal coupling into the Y signal as noise.

I kinda want to take apart a bunch of S-video cables now to see what they do right and what they do wrong.

Another crazy thought: maybe I should pick up a spool (or long length) of 7700A, and make a few cables of varying lengths for people to test.

I will agree with lordsmurfs and other posts regarding this. For S-Video, expensive cables are a waste of money. I find the same with things like HDMI cables. I wouldn't buy the cheapest 99p cables either, mostly for durability issues as opposed to functional issues with composite as an exception.

I paid £2 for my S-Video cable and £2.50 for my component ones, and they feel thick and sturdy and has given me no problems. With S-Video, you are not as likely to have problems as composite, as both Y/C is not mixed together on the same cable.

I would not worry, just use what you can, and only maybe buy one a couple of quid more if you need to, but no need to buy super expensive cables.

I know Sony sold these PS2 RGB scart cables, that people buy on eBay for £100+. I found some at a boot sale and flogged them, I use a £3 RGB cable that does the same job with no difference. Not sure if they buy it for collector's value? I was more than happy to sell it at that price if they wanted it at that price, but I did laugh my head off.

Save your pennies and put it towards better capture equipment, though for obvious reasons avoid the crazy cheap 50 pence cables, mostly cause they will break on you quite quickly.

AKA: Balanced or double-ended. Much of the data networking cable is differential/balanced (e.g., Ethernet) as well as HDMI, USB, Firewire, RS422, (but not RS232), SATA, DVI-D, the old TV 300 ohm twin lead and 450 ohm ladder line to name a few. And of course professional audio connections.

There is a wide variety of shielding on coax cables: foil, foil with a drain wire, braid (with varying degrees of coverage), spiral strands, and combinations (e.g., foil with braid). The low cost s-video cables I've cut into used spiral strand shield around the Y and the C hot lead, the shields being the ground for Y and C. The slightly premium NGX Sapphire s-video cable that has a failed connector used a combination of foil and spiral wrapped strands around 4 wires (red & black hots and white grounds for Y and C). Decent shielding from outside field, but probably not between Y and C. At least it is a pretty color :woot:

The 300 ohn twin lead has much lower losses than coax at TV broadcast frequencies, but suffered from other issues.

Has anyone found a good cable that uses Belden 7700A etc. cable? I just bought a good capture card (AIW 7700), TBC, and VCR, and now I’m trying to get good cables.

These old-fashioned plain vanilla 75-ohm svideo's at https://www.amazon.com/BLUECELL-Gold...7805330&sr=8-4 are probably as good as you'll get these days. I've used these and similar for several years. Meanwhile the Belden line isn't what it used to be: I find their latest svideos to be blurry and low-contrast. The worst are overpriced Monster Cable, nothing more than cheap 50ohm stranded core hookup wire with a lot of chroma noise. You might still be able to find a decent set of Acoustic Research Pro II s-video wires around.

Most people claim that the 99-cent cables they use look like all the others. Since they haven't used anything else, I wouldn't take their word for it.

The cables that once were selling for over $75 are for less than $10 now thanks to the internet:
https://www.ebay.com/sch/i.html?_fro...call_filtering

I see two different models for Accoustic Research II on Amazon. Which is the right one, PR-129 or PR-121?
https://smile.amazon.com/Acoustic-Re...s%2C160&sr=8-1
https://smile.amazon.com/Acoustic-Re...s%2C160&sr=8-4

Came across these S-Video Cables
 

I too was searching for some solid cables since pearstone has discontinued all cables under 10 feet long. I came across these and ordered a bunch of them so I can wire one of my TBC-4000's up to all of my capture VCR's so I no longer have to unplug cables to use different devices.

https://www.cablesforless.com/cables...o-svhs-cables/

These cables are on closeout so I bought a bunch. I got 10 cables (6 x 3' and 4 x 6') for under $20 and that includes shipping. The cables look decent but time will tell. Mine are supposed to arrive on Friday so I'll report back if they end up being junk. The cables I purchased are the closeout performance cables that are blue in color.

P-129 is a camcorder cable. PR-121 is has standard s-video connectors.

Are the 121 cables the ones people are talking about when they talk about getting Accoustic Research II if you can find them?