Adding magnets to the magnetic clamp may in the long run have an effect om the tray loading gear and maybe even motor, as the disengaging of the clamp done purely by brute crude force of the tray has been made harder too.
May be an insignificant comment but beware before you apply I think.
Using a CDM9 Pro, using a hall effect motor made from unobtainium I am not about to stick anything on it to see if it improves or changes anything. If it were a cheesy KSS or VAM12 that is easily available I would do it in an instant.
May be an insignificant comment but beware before you apply I think.
Using a CDM9 Pro, using a hall effect motor made from unobtainium I am not about to stick anything on it to see if it improves or changes anything. If it were a cheesy KSS or VAM12 that is easily available I would do it in an instant.
Hi Phn,
Guaranteed not to misread - now that's a stronger assumption.
A CD or DVD ROM can retry if it failed to read data - it doesn't have to stream data in real time in a read once scenario. And it also has further error correction.
You really cannot compare raw PCM audio data to MPEG data of any form (as on a DVD or bluray). You can lose a lot of data before it is noticeable with MPEG video because of the error correction and compression methods used and the drastic drop in quality compared to what the original would have looked like (which would require about 10x the bitrate). A DVD transport could be very poor and it wouldn't matter for playing a DVD video.
Tube-nut,
I agree, I've done this before on an old CDM12, but am not so keen to do this on my CDM9 as they are rarer.
Cheers,
Phil
Guaranteed not to misread - now that's a stronger assumption.
A CD or DVD ROM can retry if it failed to read data - it doesn't have to stream data in real time in a read once scenario. And it also has further error correction.
You really cannot compare raw PCM audio data to MPEG data of any form (as on a DVD or bluray). You can lose a lot of data before it is noticeable with MPEG video because of the error correction and compression methods used and the drastic drop in quality compared to what the original would have looked like (which would require about 10x the bitrate). A DVD transport could be very poor and it wouldn't matter for playing a DVD video.
Tube-nut,
I agree, I've done this before on an old CDM12, but am not so keen to do this on my CDM9 as they are rarer.
Cheers,
Phil
Andy,
I'd be very interested! I definitely need a spare.
I bought a cheap CD player with a CDM9 in it for spares off of Ebay - only to discover it had a poor laser. Ho hum.
Perhaps I should send you a PM?
phn, you have decided that a transport is perfect and will never misread. I'm sorry you feel frustrated, but I think it was a good debate, even if you might disagree (like with the subject matter
)
Cheers,
Phil
I'd be very interested! I definitely need a spare.
I bought a cheap CD player with a CDM9 in it for spares off of Ebay - only to discover it had a poor laser. Ho hum.
Perhaps I should send you a PM?
phn, you have decided that a transport is perfect and will never misread. I'm sorry you feel frustrated, but I think it was a good debate, even if you might disagree (like with the subject matter
) Cheers,
Phil
poynton said:
Agree with you!!!!!!!!
And for data cd, CRC and other measures are taken to correct the reading error, maybe hardware or software or both in the circuit of CD-ROM.
While for audio cd player, maybe they don't use CRC, interpolation and mute is OK.
I think mute is inaudiable and muting can't be heard. (I don't know why:->. Could poynton explain your reason? thank u
The follow link is a more detail about CD and CD-Player, u can read the last 3 paragraphs of chapter "CD information storage and playback"
enjoy!
tabooes said:
Agree with you!!!!!!!!
And for data cd, CRC and other measures are taken to correct the reading error, maybe hardware or software or both in the circuit of CD-ROM.
While for audio cd player, maybe they don't use CRC. interpolation and mute is OK.
I think mute is inaudiable and muting can't be heard. (I don't know why:->. Could poynton explain your reason? thank u
The follow link is a more detail about CD and CD-Player, u can read the last 3 paragraphs of chapter "CD information storage and playback"
enjoy!
the link:
http://www.repairfaq.org/sam/cdfaq.htm#cdtek
Yesterday i have been fidling with a CDM2. Tapping on a loaded cd sounds not dampened because cd is only in centre supported. With a diy dampening disc extra the sound is very dull en doesn't ring trough in time compared to bare CD. A dampening disc on top of cd helps only to dampen, not to equalise rotational speeds.
Well, the question is still: does less speed fluctuation improve audible quality? Has anybody experimented with this?
I measured the binary data that is written in the buffer RAM with an oscilloscope, and it has much jitter. Of course, the data stream coming out from the RAM is stable. So theoretically mechanical things should not matter. But this is only theory.
I measured the binary data that is written in the buffer RAM with an oscilloscope, and it has much jitter. Of course, the data stream coming out from the RAM is stable. So theoretically mechanical things should not matter. But this is only theory.
philpoole said:Andy,
I'd be very interested! I definitely need a spare.
I bought a cheap CD player with a CDM9 in it for spares off of Ebay - only to discover it had a poor laser. Ho hum.
Perhaps I should send you a PM?
phn, you have decided that a transport is perfect and will never misread. I'm sorry you feel frustrated, but I think it was a good debate, even if you might disagree (like with the subject matter
Cheers,
Phil
A friend of my bought a Volvo. He crashed and died. Hence Volvos are unsafe cars.
I haven't decided anything. I have seen no proof of any errors. But I'm sure you have. I mean, it can't be that you are just full of it?
poynton said:
Please look up logical fallacy.
I have seen a transport read something that it simply couldn't
before. Why wouldn't it read it? I think it could be argued (although obviously a ridiculous assumption) that it was in error.
Obviously, this is evidence that MY transport benefited, not everyones.
What is wrong with these facts?
A CD transport is not faultless. Its a mass produced item. But it can be helped, if its environment is improved (i.e. do the opposite of mounting it on a washing mashing machine or diesel powered generator).
Even if you have seen no proof of errors, can you honestly believe that all of the 671088640 bits have been read correctly? I think you can assume that maybe one or two might have been read in error (shocking as it may seem). Of course, depending on the actual error rate, what we can hear is a totally different argument.
before. Why wouldn't it read it? I think it could be argued (although obviously a ridiculous assumption) that it was in error.
Obviously, this is evidence that MY transport benefited, not everyones.
What is wrong with these facts?
A CD transport is not faultless. Its a mass produced item. But it can be helped, if its environment is improved (i.e. do the opposite of mounting it on a washing mashing machine or diesel powered generator).
Even if you have seen no proof of errors, can you honestly believe that all of the 671088640 bits have been read correctly? I think you can assume that maybe one or two might have been read in error (shocking as it may seem). Of course, depending on the actual error rate, what we can hear is a totally different argument.
phn said:
I think that it has been established that transports misread. The format itself caters for this, Error correction is built in to deal with it. Where data needs to be bit perfect ie computers, The error correction is more sophisticated. This would not be necessary if
If you can supply any evidence to the contrary, then that would be good
Meanwhile, here is a description on how the error correction works on a cd.
Reed solomon error correction
This is the section that interests me. The article basicly indicates its not missing bytes that cause playback errors. Doing more searches on reed solomon gives even more information. particularly that not all drives are equal on the error correction. Some implementation of the code are better than others.
Also when errors become too great, the CD player will interpolate samples to get a reasonable value. This way you don't get nasty clicks and pops in your music, even if the CD is dirty and the errors are uncorrectable. Interpolating adjacent data bytes on a CD-ROM wouldn't work very well, so the data is returned without the interpolation. The second level of ECC and EDC (Error Detection Codes) works to make sure your CD-ROM stays readable with even more errors.
Of course none of this proves if mass loading or clamping a CD makes a difference to sound quality. But hopefully its enough to dispel the "perfect" cd concept.
oshifis said:
I am interested in the quantity of data in the ram, before and after a dampened cd or weighty cd puck. This should differ imo. But the hardware can self dicide how the ram is filled, so not measurable i think.
If the ram is of good proportion, there are allmost no reading errors possible. Errors will be filled in out of the ram. But i don't know how exactly the ram is working, have to read that part over in the datasheet/serv. manual of dcp with SAA7210: Is the ram used continously, so as a buffer, or is it used when the error flag out of SAA7210 decides there is an reading error or datastream error ,and then takes the "spare" bits or words out of ram?
And there is something else: a datastream ON CD can be as:
10101101100111010100111010101001
But could be read by mechanism as:
101111X1100111110100X11010101001
The X'ses are drop outs, and there are some bits misread as 1 while they had to be 0 and vice versa.
Are the misreads filled in by the CIRC error correction? The dropouts will be corrected for sure. When the misreaded "1" is a "0" precisely on an MSB spot, the whole word cannot be decoded anymore, and the whole audioword must be interpolated, or mostly done is taking of the previous word and re-use it. Thats a major reading error if you ask me, even if it cannot be heard in the music. Also when a drop-out is right on MSB the whole word must be reconstructed from another word.
A long time ago in a dutch audio magazine Audio & Techniek, they tried and tested a CD polishing fluid. The manufacturer claimed better sound with cleaned CD's. What they did in the test is, besides listening tests, read out reading errors and counted them on a pc or something.
They used 3 cd's:
*one new and non polished.
*a cd polished with the fluid
*a cd "treated" with a kitchen cleaning fluid, known here in NL as Jiff, it has a sort of fine sand in it!
The results where : new CD few reading errors, the polished cd a lot more, and the "grinded" CD showed a huge amount of errors. Philips CDM's didn't have problems with the discs.
After the tests they admired the error correction, with the grinded CD it had to do a lot of work, and was allmost not hearable.
oshifis said:
Not to discourage you, but i had a long time ago sort of same idea:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=54637
philpoole said:
Nobody has questioned anything of that. Only the conclusions, which I tried to illustrate with the Volvo example.
philpoole said:
Why should I not believe? Is there some natural law involved that makes it impossible to read 671088640 bits without errors?
And I don't think what we can hear is irrelevant. I think it's very much relevant. I don't think one missed bit out of 671088640 matters.
justblair said:
Nobody has questioned that transports can misread. Anything can go wrong. People die driving Volvos.
That CDMs can misread was never in question.
The issues are:
1. Misreading CDMs are a problem.
2. Plastic CDMs make it worse. (Hence Meridian CD players must misread CDs.)
3. Inertia reduces the number of errors. (Hence the error prune PiTracer must read more exact than any other transport.)
justblair said:
That's dishonest, not to mention a logical error. That's like a Christian asking me to disprove the existence of God. If I can't, God must then exist.
It's you who assume, not I. I assume nothing.
This is my last post in this pseudo-debate.
---I measured the binary data that is written in the buffer RAM with an oscilloscope, and it has much jitter. Of course, the data stream coming out from the RAM is stable. So theoretically mechanical things should not matter. But this is only theory.---
I have been reading in a serv. manual of a CD640. (CDM2, SAA7210, 7220, TDA1541) It uses a quite sophisticated CIRC error correction. Data is buffered in a 16K X 4 DRAM stored as 4 bit words. The errors can be corrected in different levels (of priority):
C1 is for small random errors.
C2 to correct long burst-errors
And even those errors are adaptive weighted with flags to SAA7220:
as hard error (most reliable), medium, and soft error (least reliable)
The MC control (motor control) to spindle motor can be varied in 62 steps from 1.6% (FIFO full) to 98.4% (FIFO empty) The FIFO-DRAM is about 50% filled, the speed of spindle varies to achieve the 50%.
I am not surprised the data to the RAM has jitter, because its regulated by a MC feedback system, and not with a clock. A heavy puck only gives a large load for the motor, and has no influence on data jitter ingoing the DRAM. Out is regulated by clock.
So i must conclude now too with other members here, a heavy puck is pointless to increase inertia, but imo could benefit some in dampening vibrations.
To reduce jitter a good master clock to the dac, decoder and digital filter helps.
I have been reading in a serv. manual of a CD640. (CDM2, SAA7210, 7220, TDA1541) It uses a quite sophisticated CIRC error correction. Data is buffered in a 16K X 4 DRAM stored as 4 bit words. The errors can be corrected in different levels (of priority):
C1 is for small random errors.
C2 to correct long burst-errors
And even those errors are adaptive weighted with flags to SAA7220:
as hard error (most reliable), medium, and soft error (least reliable)
The MC control (motor control) to spindle motor can be varied in 62 steps from 1.6% (FIFO full) to 98.4% (FIFO empty) The FIFO-DRAM is about 50% filled, the speed of spindle varies to achieve the 50%.
I am not surprised the data to the RAM has jitter, because its regulated by a MC feedback system, and not with a clock. A heavy puck only gives a large load for the motor, and has no influence on data jitter ingoing the DRAM. Out is regulated by clock.
So i must conclude now too with other members here, a heavy puck is pointless to increase inertia, but imo could benefit some in dampening vibrations.
To reduce jitter a good master clock to the dac, decoder and digital filter helps.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.