Sorry, no, the analyses took me a few hours. The third wasn't up at that time. Should you or anyone wish to do this and more, the only programs I used were Audacity and Open Office, both free. I imported the files into Audacity for the visual and autocorrelation. The latter is a standard analysis function and provides a data export function. I did this and imported the resulting text files into an O-O spreadsheet for the correlations. I'd do more but I have to close up shop here soon. The contractor is coming to start work on my house's fender dent. Someone ran their car into it. Nothing normal ever happens to me.
But you cannot be absolutely sure that there is no change in the rest of the chain. It's like wanting to measure how strong chink # 6 in a chain is by measuring the pull when the chain breaks and claiming that you stand by the fact that all chinks are equally strong. They may be; then again, they may not.
Jan
Last edited:
No worries over the other files. I'm very grateful that you did what you did and showed how similar the first two realtime files were. Errors such as those would surely cover any disc in normal use.
Hope you get the house sorted OK.
Both men are great sources for jitter...... My family in Nepal are my first priority. There were heading towards where the ipicenter of the quake was for a wedding. :-( I have to get to the Western Union and get money to them..... no one will be able to work for awhile for income.
That method sounds very interesting and maybe easy to do... it is a steep learning curve to do setups and tests with the A-P and hopefully some canned software. I am only going to do some tests once or twice to learn the magnitude of the issue. But, also need to do the ADC-DAC thru test at various input levels for distortion. Again a lot to do for a once or twice test. Esp if I dont even plan on using a CD in the future. but, first things first.
Thanks for the info.
-Richard
Last edited:
Well the analysis by drmcclainphd has shown just how similar the first set of files are. That can't be a fluke. The system has worked exactly as it should given errors that were (as it turned out) fully correctable.
Hope you have good news Richard, I've just been watching the evening TV news here. Lots of coverage about it as you would expect..
That result is meaningless. The test method is flawed so you can't attach any meaning to an outcome.
Actually, what you say here is: the outcome fits my expectations, so even when the test is flawed, I accept the result because it fits my expectations. Not good.
Jan
I don't agree. Knowing now that the manufactured disc errors were fully correctable, a null result is to be expected. The two discs would give bit for bit identical data to the D/A convertor in the player, and consequently the audio output should be identical too. Thus far and it appears it was.
We'll have to agree to disagree for the moment then 😉
We'll have to agree to disagree for the moment then 😉
Does anybody remember any of this?
Block codes? Data in blocks. For correction of burst errors.
Generally speaking, a technique for re-ordering a bitstream so that normally contiguous bits are no longer in close physical proximity.
Then redundancy is added to permit a degree of error correction and detection. Then in this case 2 different coding schemes are interleaved, the second more robust. This traps errors flagged by the first block coder.
This was one of the most boring things I ever studied, especially since nobody bothered to explain what it was really for and since it's already been invented nobody really has to think about it, 'cause it just carries on doing its thing there in the background, and it's only when some flake brings up yet another infinitely improbable variation on the audiophile thesis that we all have to remember those boring hours writing out matrices of 1's and 0's to prove to the lecturer's satisfaction that we knew how it worked.
From what I recall, CDs will correct burst errors in the thousands of bits.
Block codes? Data in blocks. For correction of burst errors.
Generally speaking, a technique for re-ordering a bitstream so that normally contiguous bits are no longer in close physical proximity.
Then redundancy is added to permit a degree of error correction and detection. Then in this case 2 different coding schemes are interleaved, the second more robust. This traps errors flagged by the first block coder.
This was one of the most boring things I ever studied, especially since nobody bothered to explain what it was really for and since it's already been invented nobody really has to think about it, 'cause it just carries on doing its thing there in the background, and it's only when some flake brings up yet another infinitely improbable variation on the audiophile thesis that we all have to remember those boring hours writing out matrices of 1's and 0's to prove to the lecturer's satisfaction that we knew how it worked.
From what I recall, CDs will correct burst errors in the thousands of bits.
Last edited:
Torture CD
This is what I've used as a torture disc for the last 27 years. Black stripes would have been better I imagine but I only had white pin striping tape.
Track 1 is clear, track 2 has 2 stripes, track 3 has 4 stripes.... add 2 additional stripes per track up to 16 stripes for track 9 and 10.
Most players don't get to track 3 and found that the players from the 80s generally performed better than the newer players I tried.
The best performer was a 24 year old Sony CDP-X55ES which even attempts track 10 (last track). These days it starts playing up from track 5 but still attempts up to track 9. My Yamaha CD-S700 falls over at track 8 and the computer DVD drive at track 5.
What is surprising is how good the error correction works as it's hard to pick the operation in the sound until the stutter and skips start. There may be some harshness and grain in the later tracks but still remarkable.
BTW.... don't give me any grief about the green pen treatment.
This is what I've used as a torture disc for the last 27 years. Black stripes would have been better I imagine but I only had white pin striping tape.
Track 1 is clear, track 2 has 2 stripes, track 3 has 4 stripes.... add 2 additional stripes per track up to 16 stripes for track 9 and 10.
Most players don't get to track 3 and found that the players from the 80s generally performed better than the newer players I tried.
The best performer was a 24 year old Sony CDP-X55ES which even attempts track 10 (last track). These days it starts playing up from track 5 but still attempts up to track 9. My Yamaha CD-S700 falls over at track 8 and the computer DVD drive at track 5.
What is surprising is how good the error correction works as it's hard to pick the operation in the sound until the stutter and skips start. There may be some harshness and grain in the later tracks but still remarkable.
BTW.... don't give me any grief about the green pen treatment.
Attachments
Last edited:
I just got an email from a relative who says they are all OK. 🙂
But the house was completely leveled.
Thx
-Richard
Here is another example of theory working perfectly on paper. In the field it varies by cost and related factors. IMO it is the actual operation of a given player that I will be listening to which matters and how well it behaves affects my listening. It is interesting that the best transports and the best dsp et al happened a long time ago and has actually gone downhill afterwards.
-RNM
-RNM
Last edited:
Yes, yes and yes. (I think 😀) It was a very long time ago sitting in a classroom at Sony with yard after yard of paper tape with representative 1's and 0's and showing how missing bits could be recalculated from the parity bits and check sums.
Strangely, when faced with a skipping customers unit (not a skipping customer, we didn't get many of those) full of dog hair, or discs smeared with goodness knows what, none of that seemed very relevant. The customer knew you should be able to do all those things and it still play perfectly... 'cos it was on TV that you could.
Hi
I think the amount of error correction data recorded for a data disk is greater than for a CD. The amount of such data for a CD was restricted to allow enough music data on
the given disk size. The strength of error correction is directly related to how much extra data can be stored.
I think the amount of error correction data recorded for a data disk is greater than for a CD. The amount of such data for a CD was restricted to allow enough music data on
the given disk size. The strength of error correction is directly related to how much extra data can be stored.
- Status
- Not open for further replies.