A short fiber optic cable was included with my Adopt fiber optic set. Since I wanted to change the switch, I needed a longer cable. The Adopt fiber optic cable would have cost almost €300 in length. I found a fiber optic cable with the same specification on Amazon. I thought to myself, okay, I’ll just buy 30 meters. In terms of sound, the cheapest 30 meter fiber optic cable is better than the expensive 1 meter Adopt fiber optic cable. In any case, I didn’t notice any disadvantages. I had previously read that you should use at least 3 meters of fiber optic cable. If the fiber optic cable is too short, there may be more reflections that shine back onto the transmitter lens. Greetings Andreas
1 Like
The sound is the DA and AD sections that detrmins how the analog signal represents an analog wave form after deconnstruction and reconstruction. One it is constructed as 1’s and 0’s Ethernet can send it all over indefinitily with no errors. The error correction is really good, 10E-12 or 13. That is a small number of errors and there is error correction to fix even those. Check out the “noise” Ethernet can jumble through!
Anything that allows an error to squeek through to the DA “could” cause an infintesimal error on a single bit, but can we hear such a small total error? Gross errors surely can be heard (those POPS we hear in digital playback are packets that are corrupted when we encode AD). Those stay messed up.
There is the distortion on the analog side that can change how the DA works called jitter.
This is the distortion FILTERS have to deal with and how what is called ringing impacts the interpolation of the filters. The firmware adjusts the filters to move the ringing frequency and jitter location around to tune the sound. This is how the PS Audio and other DAC’s change their spots so to speak. This is a simplified explaination for sure, but the filter once designed is consistent as a phono cartridge’s sound based on the design trade-offs. Each filter sounds “different” maybe better or worse depending on your other stuff.
There are no incoming digital “errors” that can consistently “tune” an analog waveform that we’d ever like to hear. The sound is the consistency of the filters approximations through the entire chain of AD and DA. That just adds up same as analog. Digital has the advantage that the SOURCE stream once in place as 1’s and 0’s can be used over and over with no changes where analog gets worse and worse over time. What we feed that source digital stream into is what changes it! Do we use DSD or PAM filters as an example and at what encoding level?
I’ve done Ethernet cable design for like twenty years and it is REALLY, REALLY good at near zero errors. The filters are all about approximations and timing (jitter).
Galen
6 Likes
Thank you for your quick and comprehensive explanations. Is there an explanation for simple people like me? I’ve heard that there is such a thing as characteristic impedance, or that a certain length of cable is more or less susceptible to failure. Not that I’m messing anything up. I’ve heard from some that, for example, 1.5 meters is considered the optimal length for power cables. Is this a myth, or is there an optimal length for power cables, connection cables or Ethernet cables that causes less interference or is less prone to interference? With fiber optic cables, I had only heard once that a certain minimum length is advantageous in terms of back reflection and possible attenuation. Greetings Andreas
Yes, at RF cables have an impedance that needs to be optimally matched to the load. Most loads are a resistor that reflects (pun there) what the cable impedance should be. Why a resistor? Because the cable at RF is a resistive vector, the capacitance and inductance phase cancel. This is why a coaxial cable can have the impedance calculated form the L and C; Impedance equals SQRT divided by (L/C) where there is no “phase” in the equation because the phase angles cancel so at RF a cable is a resistor. Thus the desired load “impedance” is just a resistor. Simple!
But we’re still not through. A cable has what is called SRL, or STRUCTURAL return loss and RL, Return Loss. In the real world we use RL into a fixed 75-ohm load. The load is supposed to be 75-ohm and so is the cable. We can make pretty precisie resistoirs but cable can be less precise depending on the design (foams are harder to be geometrically precise than solid dielectrics).
What happens if we decide to match the load to the CABLE’s structure? We can do that if we use a varying load resistor and tweak it to match the cable that might be above or below 75-ohms in our example. Move the load value resistor until the reflections are at a minimum. In practice, we use a FIXED resistor, though, so we have what is called RL.
Optics is similar but based on matched refractive indexes of materials. Index matching gels are used to ease the transitions in mechanical connections between links but not the load where we have a different aspect to consider. The LOAD (optical receiver) end is easy as it is just an optical lens with a wide acceptance cone, so a match isn’t used like an RF cable to a resistor that matches the load. The receiver has to be placed such that the acceptance cone absorbs all the light.
Ok, now we can see what RF cable (needs a precise load match to stop reflections at the load) and fiber (needs a cone of acceptance to receive all the light) need and thus we can now reflect (ha! another pun) on cable length. Both have what is called ACR and optical budgets, attenuation to noise factors or minimum light quantity. The received signal has to be above the noise floor enough to be recognized but what happens when we have HUGE signals? In super short lengths, the signal can have reflections that don’t attenuate out going to the far end of the cable and bouncing back to the load end. This can confuse the detector as to what is the signal. Opps, problems. We need less signal.
The attenuating of relfection such that reflections are now in the “noise” and not so big to look like a signal is the argument for minimum lengths. We would like to MATCH the wavelength such that it fits perfectly into the cable between the transmitter and the load, making it a more perfect system to minimize reflections. Mismatched wavelengths reflect more. Usually the LOWER frequency range is more critical as it is, weirdly, a stronger and less attenuated RL reflection. The signal is too good! The weaker amplitude higher frequencies are more attenuated and interfers with the load less as a result. We need to manage the higher energy reflections in passive systems.
The way we do that is to make the cable LONGER so we attenuate the RL more and/or match the wavelength to the length of the cable to a full wavelength multiple. This is where the recommended cable length minimum comes from.
For optic is it easier as we generally don’t couple cable lengths together that need index matching, but to make sure we don’t swamp the load with too much light. A consumer grade optical input has an optical attenuator built-in to make sure this doesn’t happen. They kind of know the signal can be too good.
In either case we can be too good for the load. Digital RF can see the equivalent of multipath on a TV (two signals delayed in time by reflections) and optical can swamp the load receiver with too much light and that swamos the detectors rise and fall snsitivity to generate varying amplitude pulses. Same as your eyes go BLING after a flash bulb goes off.
The above is all DIGITAL RF stuff or light. Analog signals need not apply as the wavelength is many, many time longer than the cable length even at 20 kHz, and we terminate into basically infinity loads with I/O cable so the load is way, way higher impedance than the cable’s and we use what is called a voltage transfer function, the voltage drops across the higher impedance, that would be the load (47k-ohm or so) and not the cable (around 2k-ohm or so).
Galen
5 Likes
I just tried a BAV RCA audio cable between my cd transport and DAC and am getting surprisingly good performance.
Would you expect better performance from the 4694P?
If you are using the BAV RCA for a digital output, it is a ~ 110 ohm cable designed for analog. The 4694P is a true 75-ohm RF serial digital cable and is better matched to the expected load that is a 75-ohm resistor. The cable to load has lower RL reflections than the BAV. Technically yes, the 4694P is a better cable for DIGITAL where the BAV is the better cable for ANALOG. We don’t try to mix the two applications but optimize to the expected load.
The proper resistive load for RF is matchiung to the cable impedance =SQT(L/C). There is no phase angle at RF as it cancels out. Thus the cable is a pure resistive vector at RF. We can use a high tolerance resistor matching the cable impedance. Analog isn’t like this…read on.
On shorter digital runs, where the RL reflections can be high, the best resistive load match is important. Longer improper cable impedance (too low 50-ohm or too high 110 ohm) cable lengths attenuate the RL reflections more for moderate distance runs but as you get longer, those RL reflections are a form of attenuation and will limit the maximum reach (less signal voltage) to a lesser distance. On a shorter run with improper cable (110-ohm analog), RL reflections are pretty severe and can swamp the error correction.
Analog cable does not need the extreme geometry as digital and we can use AIR TUBE design (the BAV) to lower capacitance as analog IC cable is a voltage transfer function into a high impedance 47-kohm load (not a 75-ohm resistor like RF, big difference) with analog wavelengths that are far, far in excess of the cable length so the RL reflections aren’t a thing for analog cable.
Using the proper 75-ohm cable will allow optimum use from the shortest to the longest runs because the RL reflection losses are removed and the signal strength is maintained over longer runs. Shorter 75-ohm runs also see far less RL signal bounce and the error correction can be less impacted.
Hope the above helps! Don’t foget, we have the ICONOCLAST BLOG also running so stop by and prime the pump, everyone benefits with more interaction.
Talk to Bob and get the right cable for your DIGITAL.
Best,
Galen
3 Likes
Hi Rob, the 4694P is the best sounding SPDIF cable I’ve tried at any price. Something about the FEP dielectric just seems to do magic on the cable. We are hoping that members will help us to get our website forum running. It’s lonely over there.
Hello Fellow Iconoclasties,
For those who have not checked out the Iconoclast forums, Galen has posted some really interesting/different stuff in the different subject headings. There’s insights for all of us. Iconoclasties should go look and register. No reason for Bob to be lonely.
3 Likes
Thank you Jeffrey! Encouragement appreciated. Bad weather crossing the country. I hope everyone has a safe and enjoyable weekend. Get rested up for Axpona!
1 Like
Google audiophile wire fraud and you will land on a recent 52 minute YouTube video hosted by a prominent figure in the world of audio reviewers with an equally prominent guest discussing, as you might guess, the contribution speaker cables make to system performance. I only watched the first 30 minutes, so I may have missed something, but I was surprised that the conversation never deviated from frequency response.
I get how measurement wars can be waged when talking about data but to ignore data is surprising, especially since I generally respect their technical chops and background. My understanding is our hearing is much more sensitive to timing than amplitude so why ignore or discount related data?
Thank you for sharing Allan. I’m seldom surprised by this type of coverage. Sadly, I find that there are very, very few “white hats/good guys” in the community were advertising dollars and other perks all too often come into play. How many reviewers hold an EE degree with decades of experience in analog and digital signal transmission? We offer our cables for review and just can’t seem to get takers with the really well known publications. Go figure?
Allen,
A different reactive “impedance” caused by changing cable will impact an amplifier’s transient frequency response and damping factor, some desings more than others. We don’t test transient accuracy and as far as I know no real repeatable test exists thus we test into a resistive, not what the amplifier is used into, load. We make ICONOCLAST speaker cable a verified load to known measurements and calculations to mitigate this issue. I don’t see a better way to do it than address all the knowns and apply an even and balanced hand across them all.
We all know different amplifier’s sound different into “loads” and some aren’t recommended for those loads, or are if you like the anomoly. The cable is always a part of the amplifier’s load. We address the speaker, but seldom include the cable into those comments.
Frequency response with no PHASE response isn’t the entire picture, either. A resistor has no PHASE, but a speaker and cable do add reactive phase and that impacts what we hear. This also is what impacts an amplifier’s sound. Looking at a speaker load as a resistor is just wrong, or assuming an amplifier drives dynamic reactance the same as a fixed resistance. A speaker’s messy swept reactance is frequency dependent, a cable also changes with frequency but to a smoother graphed curve.
A speaker’s reactance changes a lot more than the equivalent resistive dissipated power calculation that we see in the test reports. That’s important, but the signal’s phase changes the sound too.
Galen
2 Likes
I Googled it. I saw one of his videos long ago. He seems to have a hero complex as do Amir and many others. Life’s too short for those characters.
3 Likes
Great heads up, THX.
1 Like
Yup, he’s saving the world 1 eye-blink at a time.
2 Likes