|
The difference in sound quality between a good CD player and a really expensive CD player is not the mechanism. I will state again, that unless it has significant scratches, error corrected CDs are bit accurate representations of what is put on them with the odd unrecoverable error. That is true with cheap mechanisms and has been for a very long time. Quite easy to rip a CD at single or 2x speed and see what the different error rates are, and that CD in your computer was even cheaper.
If you buffer and reclock, (assuming you have good power supplies), then you have isolated any impact of the mechanism on the output. It comes down to a good DAC and mixed signal design at that point.
That Bryston has a low(er) cost mechanism does not have to have any impact on the resultant sound if they design the rest of it properly.
|
This is what you said "sunshine". You started by dissing something for no reason other than uninformed bias. Who is trying to look good? Sorry, but that thing just uses a cheap $20 slot load media disk spinner, like this
Try going back to a real CD transport, with some cred. |
This is not remotely how error correction works. There are many errors correction levels and it is most certainly not a "guess". The only time data is interpolated is after all the error correction fails which is very rare if the disk does not have significant scratches. With a well aligned transport, which most are even inexpensive, and well aligned writer, BER for a good quality CDR media can be lower than a manufactured disk. No wonder the burnt cd’s always sound brighter to many, there’s many
more errors being fixed, and that’s just a 50% chance to get a 1 or a 0
correct, because an error is replaced by what was read before, and
that’s a 50% chance to get it right. This is why many CD players won’t
even play burnt cd’s as they can’t even read the TOC (table of contents)
|
If only there was a way of showing what the correctable and uncorrectable error rate coming off a CD rip or similar was available .... oh right, there is. Fancy words, easily proven correct. So .... let’s have it. I am sure you have lots of data to support this hypothesis given how easy it would be to collect that data? What, you don’t have any data to support your hypothesis. Color me shocked! Shocked I say!!! The Reed Solomon error detection and correction algorithms are much more effective for predictable errors like radial scratches and fingerprints than for unpredictable errors like scattered CD laser light and fluttering of the CD whilst playing, for which the laser servo feedback mechanism is not 100% effective. |
So let me get this right. You think that an optical output using AT&T high speed glass fiber somehow has an impact on the error rate reading off the mechanism. Well then, there is probably not much use of further discussion. Posting pictures of the Wadia does not make your statement any more correct. You can’t measure error rate at the output of a CD player, or even a transport for that matter, you have to have access to raw data in the transport (or player) in order to determine error rate. I think there have been some players that showed it over the years, but not a popular features. "Not an audiophile". Is this your poor attempt at "no true Scottsman" or simply an ad-hom? If you let me know, I can better address your attempted but poorly executed insult. Some of us actually care enough about acoustics and audio to have made both an academic and working career out of it. p.s. I can’t remember anything from Arnie and Paul w.r.t. error rate testing at the output of a transport (since you can’t really do it, it is already corrected at that point), but I do remember a somewhat questionable "report" on jitter. Arnie really was into this concept of black CDs for a while too. Iaf you were an audiophile which your clearly not, you would probably know Arnie Nudell (rip) and Paul McGowan did such a thing (you find out what) and showed it to you in a numerical error counter on the display.
Frightening seeing the differences between good and bad transports, that the error correction didn’t get right.
The lowest count I saw was the Wadia T2000 transport using it’s AT&T High Speed"Glass Fiber" optical output connections using expensive indexing fluid on both ends.
https://ibb.co/PYGZd7t
https://ibb.co/yNmgR1W |
For those that are interested when there is a bit-error, or even multiple bit errors, I do suggest reading about how it works, not the inaccurate guess stated below. Interpolation only occurs when errors cannot be corrected. There are multiple levels of error correction and fully 25% of the data on a CD is redundant. If you don’t scratch your CDs, then you will have almost no unfixable errors on your whole CD. Contrary to what many believe, you don’t have data bits and error correction bits all closely packed physically, they are spaced around the CD to reduce the impact of scratches. You don’t have a "good byte" on either side that is used to guess the one in the middle. That is not remotely what happens. That only happens if the scratches / defects are so severe that error correction cannot correct them. The methodology used for CDs allow full correction of up to 4000 bits, or about 2.5mm: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.462.3524&rep=rep1&type=pdfFor those interested in what happens when a byte can’t be read.
If a scratch has created read errors, you’re not completely hosed., "there is a pretty good chance that an uncorrected byte still has a good byte on either side". If that’s the case, then your CD player will take an average for those two values "and make an "educated guess" about what the missing value should be in between".
If the number of missing bytes gets to be too large, the system will suppress the error by muting the sound for a fraction of a second, which is hopefully too short a period of time to be detected. |
Georgehifi,
You are just digging A bigger and bigger hole. Jitter is not remotely the same as uncorrectable read errors. A buffered and reclocked CD player using a $20 mechanism will be near jitter free within the limits of the interconnect method which has nothing to do with the transport.
I don’t appreciate the constant attempts at personal attacks to cover up for your own lack of knowledge. Based on your own test it should be clear I am the real audiophile and you are not. You called me not a real audiophile based on my perceived lack of knowledge of a test that didn’t even exist. I knew what the tests actually was. Your test for an audiophile not mine.
You are applying 1980s and early 1990s CD architecture operation to newer CD players. As another has noted, where this all started, the Bryston reads into a buffer and reclocks. The mechanism is at that point meaningless to jitter. Your argument is akin to saying that jitter on a streamer must be huge because jitter over an internet connection is enormous. Obviously that is not the case.
|
Ignore me all you want, but that doesn’t change that a buffered and reclocked output negates timing issue in a low cost transport not to mention the way that error correction does work, not the way you claimed it does negates almost all errors except on well scratched CDs.
I am sorry that you think accurate information is twisted.
|
Except the error correction is designed to correct up to 4000 bit burst errors, hence why you have 0 tests or data to support your hypothesis even though this data would be ridiculously easy to obtain for almost free if not free. Without scratches unrecoverable errors are near nil with even a cheap transport. CDRom only adds to the error correction of audio CD. They work, fluttering, scattering light and all, at 52x speed. Imagine that ..... Buffering the data doesn’t prevent or correct the damage to data done
when the laser attempts to read the physical data on the disc due to
fluttering of the disc and scattered laser light getting into the photo
detector. Buffering only stores the damaged data temporarily. The damage
to the data and sound occurs in the first picosecond when the laser
tries to read the data. All the King’s horses and all the King’s men
couldn’t put Humpty Dumpty together again. 🍳 |
I am not the one making the easily proven hypothesis for which you can offer no proof. Don't blame me for that. That is on you.
|
Or you could join the world in 2020, and realize that what was true in 1992, when flash memory was $100/megabyte, ram was $30-40/megabyte, and a good DSP was at least $20-30 is not the same world as today, when I can get all that for a couple dollars in reasonable volume, and it is more than enough to handle enough pre-buffering to never run out due to minor speed fluctuations from the transport, and would do so at far less power (i.e noise), than even the ECC chips in late 80's, early 90's gen processors.
Some people are lost in the forest, and are quite happy to be there. Problem is, they want to drag other people into the forest with them.
Stereo5, most likely is that Esoteric sounds the way you like. Not technically better than another DAC, but sounds the way you like. Playing off a local SSD or memory card offers the potential for far lower noise than the electronics required to run a CD transport and error correction/interpolation. Rip your CDs to storage, and many programs will tell you exactly how many uncorrectable errors there are (if any). Many will even take multiple passes in an attempt to get an error free read. Everything is done "offline" so any transport or other issues disappear. At that point, it really comes down to the DAC, and if integrated, you pretty much eliminate any jitter from an optical or electrical interface.
|
It's one of the reasons I listen to radio during the day. If I stream, I am too easily distracted to change the song.
|
