Interfaces and Cables: Sound podcast transcript

This is the transcript of the eleventh release (July 3, 2014) of the “Sound” podcast. In it, Dmitry Kabanov and Timofey Shikolenkov talk about interfaces, cables, and related myths.

[ Listen to this release ]
')
[ More on the podcast ]



Dmitry Kabanov : Today we continue to discuss digital sound. There is a proposal to briefly talk about digital interfaces, understand what these interfaces are, how to work with them, what the components and components are, and what you will have to deal with when dealing with this topic.

Timofey Shikolenkov : The topic is large, the topic is heavy. People have the misconception that if we are talking about something digital, it means that everything is simple there - zeroes, edinichki. But if we understand the nature of the cable, the way information is transmitted, we must understand that the cable itself is not digital. It cannot be digital, it transmits electrical current in one form or another. This raises many questions.

It’s important to start from where the digital interface story came from, and go back to the last century, when Sony and Philips developed CDs and CD players. There was a need for methods of transmitting the digital signal via cable, then certain standards were developed. There were certain bandwidth requirements.

It is very hard sometimes to talk about complicated things with simple words. However, what do we have now? We have two interfaces: S / PDIF coaxial and Toslink optical. In fact, they both solve the same problem. They transmit a digital signal from the source to the receiver. In fact, they have the same specifications at the entrance. When all this was developed, there was still the same CD, or rather it was not there yet, but it should have appeared soon, and there were certain requirements.

If we talk about "coaxial", then there are certain conditions, certain technological limitations for the cable itself is the characteristic impedance of 75 ohms. The specification allows only five percent deviation. An important condition is that [resistance in] 75 ohms must be observed along the entire length of the cable. If some company produces not very high-quality cable, then in some place of this hefty coil, from which pieces for future cables are “biting off”, this figure may not be maintained.

What are the problems? For this you need to understand how the signal is transmitted. The signal in digital audio devices is transmitted in pulses.

Dmitry : It seems to me that it is worth trying to describe everything in simple words. From zeroes and ones to what is really happening in the wire.

Timofey : Okay. Zeros and ones are impulses. At some point, impulse 0 is sent, at some point impulse 1 is sent. As a result, they add up to a more interesting picture, which we talked about in past issues.

Transmission 0 and 1 occurs at some point in time and at a certain duration. On both sides there are devices: a device-transmitter and a device-receiver. When a transmitter device transmits 0 or 1, the receiver device must, at the same time, receive this pulse and realize that it is 0 or 1.

There are two problems. First, you need to understand whether this is 0 or 1. And since the pulses have a certain attenuation period, problems may arise in the cable itself when the signal is attenuated, and in the transmitter they may occur. The impulse must reach [to the receiver].

And here comes the second problem. D / A converters operate at certain frequencies. These frequencies, since the devices are different, are not in any way synchronized. Frequencies with which the transmitter and the receiver work are different. As a result, it may happen that the pulse transmitted from the transmitter will not be perceived by the receiver or will be perceived with an error.

This problem is not new. It originated at the very beginning when these interfaces were developed. Naturally, this was thought about, it was laid into account. Therefore, there are special systems of control, correction. For example, the well-known parity check system is one of the most common ways. Information is transmitted with redundancy, that is, for example, in the case when we are talking about the format of a compact disc, four impulses are transmitted instead of one, that is, four bits, and another fifth additional - parity control.

It would seem that all this allows you to transmit the signal exactly the way it came out of the transmitter. But, unfortunately, there is a phenomenon called jitter. Scary stuff. Some people think that jitter was invented by marketers or engineers so that people buy expensive cables.

If we had a board on which to draw, explain, it would be much easier. Since there is no board, I will try to explain how I would draw it. Imagine a coordinate system, the X axis. The Y axis is not very important in this case. Draw rectangles. One rectangle went up from the X axis, and a little further, another rectangle went down. This is the impulses: one impulse - 1, another impulse - 0.

The beginning of the rectangle is the beginning of the impulse. The problem is that by transmitting this impulse to the cable, the receiver can begin to “listen” to this impulse not from the beginning, but from the middle or closer to the end. From how and in what place he begins to “listen,” the whole problem arises, because he can “hook” a piece of the next impulse. As a result, horror will arise, because he will not be able to understand what kind of impulse it is: 0 or 1.


Differences between the ideal signal and the signal with offset (source: exfo.com )

The signal is long, there are a large number of pulses, the receiver needs to identify each of these pulses correctly. This does not happen in one hundred percent of cases, because the digital interface has redundancy in data transmission. The correction system sometimes copes with this, but, unfortunately, it does not always cope, it is imperfect. And even if it were perfect, it could not be perfect, because it only predicts what impulse was implied.

She does not know one hundred percent what should be the impulse. If she didn’t count him, if she didn’t accept him, even considering the redundancy of the signal that is transmitted, a problem will arise. What is the result? The result is jitter.

Understanding jitter is a separate story. Some of the famous audio gadgets described jitters. In the description of the order of fifty pages about how jitter appears, its causes, its properties and how it is perceived by the listener. As a result, there may be problems in the low-frequency region, that is, problems with the bass - there may be various strange sounds and overtones. There may be clicks in the high-frequency sphere - it all depends on the format in which the signal itself is transmitted ...

Dmitry: Anyway, we are faced with the problem of attenuation. Both cable and connector play a role here.

Timofey: Yes. Whenever they talk about it, they forget one important thing. When we talk about cable, we’re not just talking about wire. We are talking about cable in general. What is a cable and how is it different from a wire? The cable has connectors. It depends a lot on how the connectors are soldered to the cable.

I have had situations several times - you come to a friend, and he says: “This is like a digital cable, but the quality is not very good.” I say, "Let's see what you bought." And he bought something at the market. Unscrew the connectors, you look, and there before the contact of the central core and the sheath of a millimeter. How it works at all is not clear. It cannot work by itself. The wire seems to be normal. It remains only to solder normally.

Since this is a cheap Chinese production, everything can be soldered there somehow. As a result, this worsens the transmission pattern. At some moment we can lose these impulses. What number of pulses we lose is difficult to say. In different situations it can be different. As a result, we will not get what we sent [to the input].

Dmitry: So it is. What solution?

Timofey: The solution is simple. You need to buy a normal cable.

Dmitry: That is not on the market.

Timofey: The problematic itself is not about where to buy, but what to buy. It is likely that you can buy quite a decent cheap cable. There is simply no guarantee. You can learn this only when you connect and hear.

But here comes the second problem. To understand that there is some difference, you need to compare with something. We take a cheap Chinese cable, we don’t know if it’s bad or good, we connect, we listen to music and we think that this is how it should be.

We do not know that [with this] under the cover of our amplifier, the DAC is working and is trying to correct the huge amount of interference it receives. We are not interested in this question. We bought this device, it should work. And it somehow works. In fact, the sound may be quite different. We give the opportunity to try the cable in your home [to compare the sound and, if you like it, buy a cable].

Dmitry: The same as portable players.

Timofey: Yes. People come. We ask them what cable they have, we suggest they try a different one, they say that this is nonsense and there is no need to try. We tell them: “Try it! This is not worth anything. You take the cable, come home, connect, listen. If it seems to you that there is no difference, then what's the point of paying for it? You bring him back. ”

When I tell such stories, skeptics say that people should not hear anything. And if they heard, they persuaded themselves. One can believe that a certain number of people really persuade themselves, but there are practically no exceptions. Virtually every person who wiretaps a cable leaves it. A cable is connected, a person listens to music with this cable for several days, then he says that there is no difference. He connects the old cable, and then his jaw falls off: “How is that? After all, there was no difference? Where did it come from? ”

Dmitry: “How am I going to stay with this?”


In Audiomania, you can not only “listen” to the selected cable at home, but also make a cable to order.

Timofey: And of course, no one is ready to change it. Moreover, when we talk about high-quality cable, it is not some kind of space money. Of course, there are very expensive cables. There are many questions to what extent [pricing] is adequate.

Of course, on more expensive equipment the difference will be more noticeable. But, anyway, saving on the cheapest cable is definitely not worth it. Still [worth paying attention to] some more or less serious brand, not quite for a penny. Roughly speaking, a decent digital cable less than a hundred dollars is not worth it. Just by definition.

I am not saying that there can be no cheaper good cable. Just a guarantee of [quality] no. And there is no guarantee that now you buy this cable, you will be lucky - a neat girl sat and soldered these connectors - and the next time another girl will solder. What is the difference between branded and non-branded items? Brand things are always the same.

Dmitry: Manufacturer's warranty, standards.

Timofey: Of course. These are not only the standards that underlie the standards for the data transfer itself. We are talking about production standards. If this connector is soldered in a certain way, then this is how it is soldered. And everything turns out well. If any wire is used, then only it is used. And if we are talking about some incomprehensible manufacturer, then there may be different connectors all the time, and girls are different with different soldering irons. A good cable might come out, but it's a big risk.

What are we buying equipment for? We buy it to enjoy the music. If we say that we spend on having pleasure and enjoyment, then what's the point of saving enough small money, which are here [talking about]?

Dmitry: Yes. We have not yet disassembled the optical cable. It may be worthwhile to dwell on it, to see the specifics.

Timofey: Yes. Toslink is a fiber that has microlenses on both sides. How is the data transfer? On the source side there is an LED that emits light pulses. And on the receiver side there is a photodiode that catches these pulses.

Then the same thing happens. Impulses turn into 0 and 1. On the one hand, they turned and passed on, on the other hand - they started. In fact, we have the same problems, except for the fact that affects the cable.

If we are talking about coaxial cable, there can be a huge number of reasons for interference. For example, high frequency interference that may come from a variety of equipment. The same old fridge turned on, and the high-frequency noise slipped through. Air conditioning, included in the same network, some kind of industrial equipment. Weather conditions, the same lightning.

There are no such situations with optics. Optics have little effect at all. Here, unfortunately, there are other problems. For some reason, in low-cost equipment, the whole binding around the optics is of poor quality. I often hear the opinion that in low-cost components, equipped with both optical and coaxial interfaces, the coaxial interface sounds for some reason better in the end. It is difficult to say how it happened and why.

In the expensive equipment, the difference is no longer audible, and you can switch back and forth and not hear any difference. And on cheap everything is different. I can not say that it happens always and everywhere, but it happens.

What could be the trouble? Here, firstly, the trouble can be in the quality of the microlenses themselves, if we are talking about the cable. Optics itself can be of different quality, there also have their own specific standards. I have not studied this problem deeply, but the cables are different, and, as a result, the resulting sound can also be different.

When we connect, we can compare. If I have a bunch of cables, this difference is really audible. Well, the very strapping: you need to understand that in addition to LEDs and photodiodes there is some kind of electrical circuit. The quality of this electrical circuit, the quality of the conversion of these signals into something tangible, audible, different.

That is, the harness that exists around S / PDIF is very different from the harness that exists around optics. And what components are used, which radio components are installed in each particular case, which chips, which manufacturer, greatly influences the result.

By the way, we have a very unusual player in the Audiomania showroom at Electrozavodskaya. Several DACs are built into it, and you can switch them with a button. It would seem that if we are dealing with a number, then what's the difference, and why build some DACs at all? Strange nonsense, it would seem. But it's amazing how the character of the sound changes. We listen to the same composition on the same equipment, replacing only the used DAC inside by clicking on one button. It's amazing how the sound can change, how the impression of the music can change, based on the DAC used.

When we talk about a digit, the main misconception is that the digit is 0 and 1, and everything is the same everywhere. But in reality there are a lot of subtleties, we cannot touch them now, and, in fact, I do not know all of them. The bottom line is that, despite the fact that the figure is 0 and 1, the sound is [all the same] analog.

At that moment, when 0 and 1 turn into analog sound, many questions arise. The issue of data transmission, which we have just mentioned, is one of the big problems. Then the question of transformation arises, which we have already spoken about before, and we will probably talk a lot more. I always propose to compare the sound, because very often there are people who say: “This can not be, because this can not be. I remember from a high school physics course that this cannot be. ”

Here you can recall the earlier school course, when they teach the basics of arithmetic and say that there are no negative numbers, and it is impossible to subtract three from two. And in later classes they begin to tell that it is still possible. Putting this question in the plane of physics and those processes that occur in the equipment and in the cable, you can learn a lot of new things, unfortunately. Unfortunately for the person who believes that many things are impossible.

Now there are few such resources where you can read a truly authoritative opinion. You come to the forum, and there authority is measured by the number of written messages. And it turns out that a person with ten thousand of them is an academician. Of course, he nafludil half the forum, but this did not become smarter. He pours out what he has in his head. But the number of messages information in the head does not occur.

I always recommend listening and comparing. If you have any conviction, let it be yours. Healthy skepticism in this matter is helpful. You will sit in the listening room, for example, in ours, and compare the sound of your cheap cable and cable, which costs more ... I always like to look at people's impressions, because, as a rule, this is an extension of the eyes, involuntary opening of the mouth.

Indeed, people hear this difference, they understand what they are giving up for the sake of their beliefs. Always worth a try. Moreover, in the case of Audiomania, wire tapping, however strange it may sound, does not cost anything. You just have to go to the store and take a chance. The world can roll over. This is a big problem: people are afraid that something in their lives may change. People love constancy.

Dmitry: The old saying: "The best is the enemy of the good."

Timofey: Yes, it seems that everything is fine, and why is there something better? But it is worth trying once, and the world will not be the same. And the quality of life will increase.Recently, when people come to me who have never heard of high-quality equipment, I turn on some arbitrary composition and ask if there were goosebumps. If it wasn’t, then either I didn’t guess the composition, or I don’t like the equipment.

As soon as the person says that the goosebumps were there, it means it worked out. When a person realizes that he has goosebumps from playing music, but he has never had it before (because goosebumps do not arise on cheap equipment, emotionality is not saved), then I understand that I changed the future for a person, changed life for a person. I do not know whether this is good or bad, but it is interesting.