Is anyone using a converter to connect a Toslink source output to a Coax input? Are there any signal loss/degradaton issues in doing so??

[This message has been edited by drhb (edited June 23, 2004).]

no, and yes there is always going to be some loss on a conversion similar to that. however i doubt it is noticeable, and the ability to connect the device how you want/need to would surely outweigh the small loss of insertion/conversion.

------------------
This post has been brought to you by curegeorg, thanks for reading.

For my Sony mega-changer, which only has toslink digital output, I use a Monarchy DIP Upsampler. The DIP allows toslink input with coaxial output. It also reduces jitter and upsamples to 48k or 96k. The change in sound quality was quite noticeable.

Exactly what loss would you always have with that type conversion?

Edited to correct spelling error.



[This message has been edited by Cliff Watson (edited June 23, 2004).]

no "lose", but the loss would be converting light to electronic waves. or vice versa. CONVERSION OF ENERGY.

------------------
This post has been brought to you by curegeorg, thanks for reading.

More accurately, conversion of signal format from an electrical form to an optical form (or vice versa). With a digital signal, it would be possible for a few bits to get lost along the way. Those lost bits should be effectively inaudible (after all, there's inevitably a little data loss anyway, but it is handled by error correction during D/A conversion). Unless the optical-to-coax converter is very poorly implemented, there should not be any measureable degradation of the analog signal produced at the receiver or pre/pro.

------------------
gonk -- 950 Review | LFM-1 Review | Pre/Pro Comparison Chart | Saloon Links

Jamingc, can most users justify a $250 expense just to make the conversion...assuming that the difference is indeed real (wonder if any double blind testing of this converter vs. a standared converter, was ever made...)

This type of conversion is relatively easy since in reality the Toslink signal is converted to an electrical signal anyway inside the 950 and other components so the circuits can process them. There are various freestanding boxes that perform this conversion too.

It is absolutely lossless in any but the poorest engineered components.

A double blind test is not needed when many tests have been done comparing the original bits to the bit stream after conversion. The error rate was nonexistent.

Placing an optical converter in the path is the same thing as connecting it to the optical input on a receiver/preamp.

if one had to output their digital signal as coax, then convert it to optical for connection. that would be a lot of unnecessary converions... ironic that i recall someone complaining about component video switching in a processor (due to loss?) and yet defending this. no switching/conversion of a signal is ideal, everytime it happens there can be losses. better converters/switchers limit this to low levels.

everyone knows that ideally there would be only one conversion in a digital audio signal, that from digital to analog with all the processing done digitally.

as i said in my first post, if you have no other choice but convert the signal to make it connect, then the point is moot and connecting is better than worrrying about bits of data being lost. if however you can leave the signal in one form, then do so. it would be better to buy a new cable than to buy a converter to utilize the cable that you already have...

------------------
This post has been brought to you by curegeorg, thanks for reading.

Obviously, the KISS engineering principle remains fundamentally sound, but comparing component video switching (involving a high bandwidth analog signal) to a format switch from electrical to optical with a digital signal is stretching beyond reasonableness.

Here's another way to look at it: any time you have an optical digital audio cable, there has to be a conversion to optical signal and back at either end of the optical cable. This conversion is one of the most common reasons that some people prefer coaxial digital audio, although many people also make the (very valid) claim that the conversion process is for all reasonable intents and purposes lossless. Going to an optical/coaxial converter is (aside from the coaxial digital audio cable) no different than using an optical cable to begin with.

------------------
gonk -- 950 Review | LFM-1 Review | Pre/Pro Comparison Chart | Saloon Links

>>>reasonableness<<<<

Hmmmmm......

Anyway, there is no real "conversion" of the data, only of the data format in the optical to electrical translation process. As Gonk said, the video issue does not apply here since in that case we would be dealing with analog signals of very high frequency which need a very clean and controlled signal path. Digital video data would behave like digital audio data and it would be extremely immune to degredation. Data formats are converted constantly - take the internet for instance, and yet you don't see massive errors for the same reason you don't see massive errors in digital audio - error correction schemes.

To drhb- for me the DIP upsamlper was a cheap tweek, I have spent that much or more on some cables. I didn't do any blind testing, just my ears. In my rig, the change in sound was very noticeable. The DIP is not just a converter, it lowers jitter and upsamples. But you will have to make the upgrade decision for yourself.

I would have to agree with Gonk, in that on our field installs of a/v systems, nothing in the way of signal loss was noticed, (obviously not via double blind testing...), and moreover, logic suggests that conversion loss is a moot point since, as noted by Gonk, the conversion would occur anyway, at both ends of the toslink connection. Just wanted to get a consensus of what others thought....

”Data formats are converted constantly - take the internet for instance, and yet you don't see massive errors for the same reason you don't see massive errors in digital audio - error correction schemes.”

Digital audio (S/PDIF) does not use error correction. The S/PDIF receiver/decoder uses error concealment by repeating the last known good bit from a small buffer (up to 20 bits before dropping a frame) or in the case of a bad CRC a dropped 32 mSec audio frame.

Point taken, but I wasn't talking specifically about S/PDIF, but just in general terms that errors get corrected (or concealed, nullified, adjusted, un-modified, made good or whatever).

Hey, this ain't the IEEE bulletin board and some shooting from the hip is allowed in the process of making a general point for a general audience.

dig. coax would avoid the 2 conversions that optical creates...

------------------
This post has been brought to you by curegeorg, thanks for reading.