A/V The good ol' Digital-to-Digital cable thread

Though I would make a new thread for this topic.
As discussed in the "Good ol' DVI vs VGA" thread
http://forums.bit-tech.net/showthread.php?t=90035

Talking about digital-to-digital cables, not analog. This will probably whip up a hornets nest with some members.

Disclaimer:
I am not ranting on Monster Cables analog stuff...I fully beleave in that, so no Monster fans come attacking me.

Take two of Monster Cables digital cables, and the way they talk about them:
I will attempt to translate some of their marketing lines into what is really happening.
Ultra Series THX® 1000 Digital Coaxial Interconnect

Silver plating copper is great...but does nothing for a digital signal.
Low Energy loss means they saves a few tenths of a volt in power loss, again this does nothing for a digital signal.

Again this is good stuff...for analog...Unless you have a microwave right next to your A/V setup double-shielding won't help all that much, as far as effecting the digital signal.

Good dielectrics will keep impedance in check...great, but still has little effect on a digital signal.

Frequency response, this is a measure of how stable the impedance and reactance of the cable is over a variaty of frequncies...but digital only travels at one frequency (44Khz for digital stereo..higher for surround sound)
This has little effect on digital signals.

This doesn't mean anything for a digital video signal, cable quality won't effect the one's and zero's.

There is no such thing as a more precise digital signal. Sure there are errors, but these errors are often not the result of RF interference. Many DACs included error-correction for a single bit. Fixing and error in a digital signal will not effect the final analog output.

Ultra Series THX® 1000 Fiber Optic Digital Interconnect
Some stuff on this page is just funny.

Hand-polishing can't make a optical fiber transmit a more accurate digital signal. You can't get a clearer '1', or more accurate '0'.

The alignment of the optical fiber has nothing to do with data throughput. If you are mis-aligned you will get serious errors or no signal at all, but this type of misalignment is pretty hard to do on a tos-link connector.

Jitter on a optical fiber is a problem...for telecommunications and long distance optical fibers that transfer millions of phone conversations. It has very little do with with your 7.1-channel audio.
The bandwidth of a optical fiber, even low quality is well above the needs of a home theatre system. Bandwidth has nothing to do with the "precision, clarity and accuracy" of the digital audio/video signal.

...now let the firestorm begin....

 

Surely all the stuff with regards to extra shielding and a 'better' dielectric will help reduce jitter? Or is jitter predominantly to do with the electronics at either end of the cable, and the cable itself has a very limited effect on jitter when compared with other things like the power supply regulation to the master clock and so forth?

Jitter can have an audible effect on the sound quality in my experience - listen to the improvements you gain from reclocking a cd player with an Audiocom superclock or trichord clock 4, which act to reduce jitter significantly.

Personally, and this is going off topic slightly, I can't understand why so few manufacturers don't start to go with 110ohm balanced AES/EBU digital connections rather than the usual co-axial 75ohm connection - by virtue of being balanced it is superior and less succeptible to interference.

 

Yes, all the things they specified do in-fact effect the signal, but in a analog way. The problems is that they are putting analog terms into selling a Digital cable.

Jitter happens when a signal rises or falls in sync with some other signal. (Most of the time 60hz interference from the wall)
Digital signals are not effected too much by this.
Take DDR memory (stuff I am most experieneced with) the voltage has to rise all the way up to 3.6volts before it's considered and error. Even with jitter the "1" is still a "1"..even if it comes in at 2.5volts or 3volts peak.

 

well i think you all know my opinion from the previous thread

i wont paste that all in again.

the only thing i will say is... if its all 0's and 1's then how come other components affect the sound?

surely a dac is a dac as long as it doesnt lose or corrupt the original signal, but different dacs sound different. Also if it is just 0's and 1's then how can a cd transport affect the sound?
Surely all audiophiles would buy a £20 cd transport with a digital out and then put it through a more expensive dac. Yet better transports cost more and sound better. Reviews clearly list the differences between £10k transports and this differences are the same in several independant reviews

I know its sort of off topic but im just trying to show you that its not just 0's and 1's.

Also this thread is making me want to start an analogue cable thread now

 

Monster come up with the biggest PR bullsh!t i have ever seen from a cable company.

trust me when i say "don't trust them". if you even look into some of the **** they say, you'd be rolling in the aisles.

i see you've already debunked most of their guff.

please, eveybody, do yourselves a favour and never buy monster cables. ever. it's money, not just thrown away, but shoved up your arse so hard that it finds its way through your entire digestive tract, flies out of your mouth and THEN into the trash.

 

If you are refereing to, say getting a noise when you move your USB mouse passed to the sound card, this noise would be induced at or after the DAC.
This could be due to bad capacitors (not filtering voltage flucuations).
Bad Inductors (picking up RF noise rather then filtering it)
If the 1's and 0's were effected the signal would be corrupt and not be generated into sound as the sound card would not know what to do with the data. In theory yes, if there is enough interferance before the DAC, but after the sound processor, you could get noise effecting the 1's and 0's but that interference would be so great you would hear it on the analog side as well.

A DAC is not a DAC, that is like saying that any monitor that can display an image can display the exact same image. (same quality, color, brightness)
DACs come in varring levels of quality, precision, frequency response, basicly anything that can be tied to analog. The incoming digital is the same, the outgoing analog can varry. There are 2900 DAC's listed on Digi-key.com alone.

This is because a CD-Rom is a medium involing a focused/aimed laser. A bump or mis-alignment can cause and error, but most likey due to the high speed of the disk the error won't be single-bit. CD-Audio has only a single-bit error correction, so an error reading the medium will result in an error in the sound. There are circuits that can buffer, detect errors or no-reads, and re-read the same spot, these are common and used as shock protection.
http://electronics.howstuffworks.com/cd5.htm

A higher quality transport may involve more processing, a more precise, less error prone laser, but to say that it got more precise digital data off of a CD isn't true. The data is there, unchanged, there isn't anything more to read. You can process it, convert it, resample it, interpolate it; but in truth the data coming off the disk is the same in a $10 sony CD walkman, or a $10,000 high end device.

I don't know if this is a really good CD transport, I just googled it:
http://www.marklevinson.com/products/overview.asp?cat=cd&prod=no390s

That is very true. But with microprocessors and buffers the signals can arrive late and still be put into their proper place. A high quality CD transport may spin the CD at a more precise speed, read it off the disk with a more precise laser, resulting in perfect timing. This perfect timing is more elegant, harder to design and achive, but it doesn't change the digital data.

Exactly. A high quality transport may be more elegant, but not nessicarily more precise.

This is the job of the DAC. DACs have small buffers, the same as most any digital circuit, it is the job of these buffers to take in the data and put it out at the right timing. A more precise DAC does a good job of this, but this happens at the actual conversion, it is not relivant to the reading of the CD, or the transport of the digital data.

 

I didn't want to start a monster cable rant...I was using them as an example.
They do make good analog cables, if you need them.
I have a $1000 HDTV.. I know that no monstor cable is going to improve my TV's picture...analog or digital..the TV isn't good enough to need it.

But later in life when I buy a DLP, or other higher quality HDTV, I will probably use DVI/HDMI and Digital audio to connect 100% of my devices, and therefor I won't need any monster cables anyways.

 

you missed my point a little. you still say its 0's and 1's yet hundreds of reviews and reviewers cant be wrong in talking about the different sound quality from different transports them. Surely if they dont have error problems they would sound the same, yet they dont.

(however... the best cd players at the moment have onboard dacs and external power supplies) (yes i know thats a generalisation)

Also as for digital all the way, i would rather take my audio from the cd player out in analogue form (providing it has a good dac) and taking it into the power amp the pre then off to the speakers if your putting a digital signal into an amp your using its own dac for the sound quality.
No amps i know have good enough on board dacs, infact i know of no amps i would use that have onboard dacs.

Also pure digital amps are not even close to the refinement in sound compared to analogue ones.

This method is proven to be better for sound quality. Ask any reviewer of a major hifi mag and they will tell you keeping it analogue as much as possible gives a warmer more natural sound compared to digital which is clean and clinical.

 

I'm glad my ears ain't perfect, I can't really tell the difference between (say) a 192 kbps mp3 and a CD.
So I can buy mid-range stuff and it sounds just as good as a £10,000+ HiFi to me

 

i dont know, most people i know can tell the difference between 192 and cd on my system and its not upto 10k value yet.

see personally for me its about re-creating the audio as if the people are playing live infront of me. i dont want to be able to tell its playing off a cd or lp etc.

 

Many, even high-end, audio review sites are VERY used to analog terms. Applying analog problems to a digital signal, thinking that it will effect the end resulting analog signal. I would argue that any transport that has a digital output is manipulating the digital signal in some way, either through processing, or re-sampling. They can't cost $10,000 to read a CD and put the signal out raw.
Re-sampling can make a audio CD sound better...this can be done two ways:
Digitaly applying math to the signal, or converting to analog, adding analog filters and such, and sampling it at a higher rate back into digital.

MP3's are compressed digital, that opens up a whole new bag of discussion, we are talking here about full-speed pure digital.

 

DACs can affect signal quality because each has a slightly different idea about which level to register the next number up, or down at. Most of them use resistor ladders in something called a Flash ADC which consists of over 65,000 comparators and even more resistors. Every resistor has a tolerance, usually 1% in these things, so it can get it a little different between DACs.

Cable however does jack buggery squat unless you're moving from a cable which is so poor the interference or losses (particularly in the case of optical with bits missing) is causing phantom 1s and 0s to appear. Otherwise the 1's and 0's are just 1's and 0's there is no way the audio can be affected by replacing a cable that transports non analogue data with 0% loss/error with one that costs 200 times the price. Any 'sound difference' is a placebo effect.

 

You forgot to mention that analog audio equipment is more expensive than digital audio equipment of the same quality. It doesn't cost much to make a digital audio system that sounds very good, but it costs significantly more to make an analog audio system that is as good or better.
In fact, digital audio is so popular that semiconductor companies like Crystal and TI are making highly integrated ASICs just for that purpose.

There is absolutely no reason why digital is not as good as analog. Remember that D/A and A/D converters are now available in the GHz range, well above the bandwidth required for audio.

This reminds me of my math teacher asking, "What gives a more accurate answer: Graphing and measuring or doing it all mathematically?" I answered "doing it mathematically", which turns out to be correct.

As for the cable, it shouldn't matter as long as it's not too bad. With digital, you only need to tell the difference between 0 and 1. If that works, it can't get any better. It's like reading a (text) book. If you can correctly read the words, it can't be any more accurate. Noise caused by the cable is like dirt on the pages. It won't have any effect on accuracy as long as it's not too bad.

With my configuration, the audio remains digital until the very last moment. The audio amp is basically just a digital signal processor and a power booster. The speakers themselves contain only inductors and capacitors to perform the D/A conversion.

Also don't forget that EMI caused by power supplies more easily affects analog circuitry than digital circuitry. Any electrical engineer will tell you that it's a bad idea to put sensitive analog circuitry next to a power supply.

BTW, I pay attention to ZapWizard because he understands electrical engineering very well and is, in fact, a technician at AMD.

 

star882 dont make this personal, i know digital equipment as expensive as analogue.

no reason why anaolgue and digital amps arnt as agood as each other? well ones been around lots longer and had much more r&d put into it.

with your configuration from memory... it really wouldnt matter who long you keep it digital, sorry but you really havnt got the sort of system that is hi-fi.

And i dont ignore zapwizard, i read his posts, reply etc. i am merely having a discussion. and how does working at amd make him good at audio? perhaps he makes the tea

(yes i know he doesnt, its a joke)

 

Although I agree with most of you analogies, capacitors and inductors don't perform any DA conversion - I think you are refering to the smoothing out of the stepped waveform produced by a DA conversion.

 

holy arse - yo must have sucky sound then - why the hell would you ever want to smooth a signal from your amp.

get a better DAC!!!

 

Nope, the amplifier simply switches the output voltage between positive and negative at a very high frequency (~1MHz). Digital logic is used to control this, so the signal remains digital throughout. At the speakers, the inductors and capacitors filter out the high frequency carrier and pass through the audio frequencies.
This explains how it works:
http://www.cirrus.com/en/products/pro/detail/P1070.html

 

you dont have digital speakers though.. so how can the digital signal go all the way to the speakers?

the only digital speakers i know of have an inbuilt amp and cost more than most people spend on cars (its made by meridian)

 

The output is a modulated square wave with only two active voltage levels: V+ and V- (excluding powersave which is 0v). That's a digital signal, right? The resultant signal looks like an audio signal with a 1MHz carrier added to it when viewed on a spectrum analyser. The filter only has to block the 1MHz carrier and let the audio through - a pretty easy task.

BTW, I used to use some 2.1 Philips USB speakers (and still do for semi-mobile applications) which do have integrated digital amplifiers. But like I said before, I upgraded to my current 5.1 setup, which has a separate amplifier.

 

Yes but im talking hifi, not intergrated low quality mobile audio.