[thomas]

Quote:

Bits isn't always bits. Frustrating, but true. Digital cables do make a difference... it's small but audible when listening carefully.

I'm not saying that everything digital sounds the same, or that digital cables can not matter... I'm saying the reason they can sound different is because the timing of the bits has changed, not because the bits have changed.

 

[markjia]

Hirsch,

I would think that the change in sonics would be do to errors/timing resulting in the reading of CDs in poor CD-Roms/CDPs. Plextor makes excellent CD-Roms/CDRs, but if the CDR media used is of too low quality and/or burned faster than it can handle, some players will not be able to read the CD accurately. The change in sonics that Plextor describes maybe the result of differences in the read accuracy of players.

This is the first time I have heard of CDRs being burned at different angles, so I am really just guessing (if someone knows better, please correct me). But consider that CDs (not CDRs) are mass produced using a stamping technique, therefore, they could not have pits at any angle other the 90 degrees. So, the only reason I can think of for burning at any other angle is to allow easier reading in poor drives (and in the process changing the sonics in those, and only those, devices).

 

[dvw]

There are two types of error correction in audio.

1. Reed Soloman. No CD is perfect and there will be error on the disk (defect density is the measurement). What R-S do is that it interleave the data so a single block of errors is spread out to different places in the data stream and that can be corrected. But if the entire frame is wiped out (not likely in disk defect during manufacturing) then the data cannot be repaired. That why you don't want to scratch data in circular motion.
2. Auto correlation, after the data is transmited and when the DAC detects an error, it will guess the value for that error.

In term of digital cable, expensive cable sometimes is worst than cheap cable. The main problem is that some consumers think thicker is better, so the manufacturer makes thick digital IC. However, the line impedance is determine by the ratio of inner diameter and outer diameter of th cable (as well as diaelectric). The thick cable actually has lower impedance. This impedance mismatch causes refection of the signal (and thus bit error). In TV, you'll see that as a ghost image.

Some cable changes the isolation from PVC to teflon, but did not change the physical dimension. The result is a 50 ohm cable instead of a 75 ohm cable. You can see this in some cable manufacturer's web site and they actually advertise this as an advantage.

My point is price and hype of insulation does not determine the quality of a cable. Just use common sense.

 

[grinch]

read about yamaha's new burner:
http://www.cdr-info.com/Sections/Art...ies=0&index=10

it rights the pits larger so the cd has less space (80 minute blank only allows 68 minutes to be burned) but is supposed to raise quality AND reduce jitter.

also, i suggest using eac for EVERYTHING involving audio cds. it has the ability to do write and read offsets which, to my knowledge, no other burning program in the world can do. check it out, it's neat.

ExactAudioCopy
the coaster factory (tutorials and such)

look at those links (more importantly the links on coaster factory about "determining the offsets parts i and ii)

i just set mine up recently (wouldn't believe in the offset making a difference for years) but i haven't a/b'd any copies yet. i hope to soon, and i can't wait for my pre-ordered yamaha burner to come to me.

 

[AssafL]

So as I understand from this thread that a high quality digital cable would be a 75 Ohm (video) cable, and preferably a high quality one at that (perhaps soldering the shield all round the RCA plug to minimize reflections).

What about Toslink fiber optic cables? What are the specs for a good Toslink cable?

 

[markjia]

I don't think that a video cable would be well suited for audio use...I think what is being said is simply that you cannot judge a coax cable by it's price. Instead, you need to do some research into a specific cable, and see what people say about it.

Toslink is generally not considered to be as good as coax. If I remember correctly, it is due to an increased amount of jitter.

 

[dvw]

Video cable can be used as digital cable, but it's not very good. The connector is different and they typically come as a single core (too stiff). Good digital cable however can be used as video cable.

On Toslink, a good Toslink can be better than coax. This is because optical signal does not suffer electrical interference. However, Toslink are usually made out of plastic and the quality varied a lot. For example, uneven thickness, poor termination, improper refractive index etc. My guess is you can have 50/50 chance of getting a good Toslink even from the same manufacturer. I could be wrong.

For those of you that are technically inclined, here's a little more tidbit of information on 1 nad 0.

The digital signal is actually coded with biphase or dibit. What that means is a 1 is coded as 10 and a 0 is coded as 01. The reason for this is to make it easier for the PLL to recover clock. Today's DAC is much more jitter tolerant than the older DAC that is used. You can check out Crystal semiconductor's data sheet. So if you have a newer DAC that is 24/96 capable, chances are the cable requirement will not be as critical.

 

[dariusf]

I have read about this topic quite a bit and accumulated a few interesting articles. I'm my humble opinion after reviewing all I could find on this subject I'm convinced that 'there is a deference'. Even with lower cost equipment you can definitely tell the deference. I have compared some old Technics CDP, JVC XL-V141 CDP, JVC XV-SA70BK DVD Player, NAD 522 and NAD 521. The NAD players have a huge audible deference over the other low end players in this list. I'm also quite sure that my quality amplifier (Proton AA2120 dual mono block) and a good set of speakers (AR S40) made it even more evident.

I have only used the lower end Radio Shack gold plated cables. Which are nothing special but a regular solid cable. Until I do some experimentation with some high quality or should I say expensive cables I guess I will not know if it makes any real audible deference in my setups.

Check out these articles:

Bits is Bits, Right? Home mastering can effect the recording:
http://www.stereophile.com/showarchives.cgi?372

Jitter in, jitter out...
http://www.stereophile.com/shownews.cgi?70

A Transport of Delight: CD Transport Jitter
http://www.stereophile.com/showarchives.cgi?368

Good error correction info:
http://www.emedialive.com/EM2000/starrett5.html

Some more good dox:
http://www.dcsltd.co.uk/papers.htm

 

[dvw]

What dariusf posted are all very accurate. However, it also shows if bits are bits with no error or alteration, they also sound similar.

However, jitter is a complete different issue. There three primary cause of jitter, transmission, receiver and the media (cable). Jitter is caused by the turning on and off of the transitors in the circuit (in the case of circuit. In the cable, it's the charging and discharging of the cable. A more accurate term is dispersion. This also happens in optic due to the reflection when the light hits the fiber. That is why jitter is more difficult to control in TOSlink. If the fiber is glass based with a super thin diameter, the refection will be less and the jitter will be lower, but that will be very expensive (commonly used in telecom equipment).

Designer are faced with a tradeoff in clock recovery (at the receiver), a lower jitter generation will implied lower jitter tolerance (i.e. the digital stream is difficult to lock). So it came down to pscho-acoustic on how much jitter we can tolerate.

Modern design (because of IC technology progress) makes retiming a lot cheaper. Some DAC now have retiming built in (you can also buy a retiming device), so jitter becomes less of an issue in this type of device.

The overall jitter contribution of a cable is very little if the cable is short. Majority of the jitter is from the transmitter.

Jitter is too complex to explain in a single post. In relation to sound, any digital defect or improvement, is random (as opposed to analog IC). A poor cable will sound bad but it will not have any charcteristic of its own. Two good cable will sound exactly the same and there will not a warmer sound or bright sound.

So digital cable is more or less a 1 or 0. it will either sound good or it doesn't.

DAC are more interesting because different DAC have different error correction scheme. I burned a CDR with lots of error (not by intention) and it sounds vastly different on different machines. IMO, if you hear a difference, then you most likely have data error in one of the digital chain.

 

[AssafL]

Regarding Toslink: Interesting thought, as most good fiber optics (glass fibers) goes to telephone and data trunks. Perhaps it is possible to salvage surplus and create good Toslink cables from those (I guess it is not a single mode fiber).

As for the SPDIF interface - I would also guess that some manufacturers skimp on the interface network between the decoder and cable.
Some use impulse transformers that removes common mode noise if exists both on the transmitter and receiver - I guess that most don't.
If the circuits are grossly mismatched, you'll have reflections, which may substantially compromise the signal. That would also explain why some lo end components don't really like to drive high-end components. A case in mind is the 100 Ohm resistor in the DI/O (instead of 75 Ohm).

 

[dvw]

It is a good idea to use glass as Toslink. Unfortunately, it's not a simple DIY job. Splicing the fiber requires special tool. The glass have to be polished at the end and gluing the fiber to the connector takes a lot of skill. If you have a friend that have the know how, the glass toslink will be the ultimate digital interconnect. A glass fiber is actually cheaper than copper. It's the interconnect part that's expensive (because of labor). I think that 's why toslink are plastic.

The draw back of fiber is that it has a small bent radius and can be crushed easily. If you are vacuuming and the vacuum machine runs over the fiber, it can get crushed.

If you going to DIY, I would recommend the 3M fiber. It has a new formulation (or so they claimed) that makes the fiber more crush resistance.

 

[Dusty Chalk]

Quote:

Originally posted by markjia Jitter is a matter for the DAC (assuming it reclocks) correct?

No, incorrect, that's too big an assumption. It's a matter for the entire system. If it's small, it can be dealt with in an advanced DAC, but it usually isn't, hence the move back towards one-box CD players. Quote:

Originally posted by AssafL What about Toslink fiber optic cables? What are the specs for a good Toslink cable?

Glass. As has already been mentioned, most of them are plastic. A good glass one should do the trick. I know some audiophile one just came out with one, but do not remember the name.

One extra-special-plus-good thing about optical digital is that you don't have to worry about ground loops. This was important to me a while back, because I had separate players for my DVD and CD, plus the receiver for my HT setup, so I had so many things to plug in, that I ended up having to use two completely different electrical outlets that were on different breakers. I kept these separated by the optical digital cable, and no ground loops.

My ears aren't good enough to be able to tell the difference between digital cables...yet...give Hirsch some time to educate me...

 

[dvw]

I didn't mention one tweak that you can do with digital cable. If the source and receiver do not have the proper impedance matching, you can add a 10 ohm restistor in series with the cable.

This tweak will not eliminate reflection, but it wil "reduce" amplititude of the reflection. This is more a when all else has failed band aid.

 

[AssafL]

The 10Ohm (or higher) would probably do the trick...

I wonder if anyone manufactures a Redbook SPDIF compliance monitoring equipment.
I would guess if HP (or whatever their test equipment name is now) made one it would have a very audiphile compliant price...