Hi guys!
i got into a discussion today about what difference it would make between two discmechanisms, say a Sony and a Philips IF you use all the same electronics (same DAC, PSU and analogparts) ???
and another Q:
how important is the cabinet, ordinary sheetmetal case or 1 inch steel, in which way does the chassis affect the soundquality??
-help me out here please and i will be greatful for all times to come
/L
i got into a discussion today about what difference it would make between two discmechanisms, say a Sony and a Philips IF you use all the same electronics (same DAC, PSU and analogparts) ???
and another Q:
how important is the cabinet, ordinary sheetmetal case or 1 inch steel, in which way does the chassis affect the soundquality??
-help me out here please and i will be greatful for all times to come
/L
i belive both questons can be answered with: 'quite alot'
the case and disk transport are designed to mimimise interferance from the outside world (movement ect.) and different designers chose differenct ways of doing this.
dampening the case (adding weight and cutting down noise) is one of the best mods that can be done to a player - i have doen that to a relitivly cheap CD play and it sounds completly different
the case and disk transport are designed to mimimise interferance from the outside world (movement ect.) and different designers chose differenct ways of doing this.
dampening the case (adding weight and cutting down noise) is one of the best mods that can be done to a player - i have doen that to a relitivly cheap CD play and it sounds completly different
CD-Transport Quality
Hi Mr. Triatec,
I changed from Sony to Philips recently.
Sound with my KWAK-DAC was distinctively better with the Philips.
I had a Sony CDPX33ES. I tried Philips CD-380 and CD-650. The CD-650 was slightly better than the CD-380.
I also tried the CDP-XA30 ES of my friend. It sounded about the same as my 12 year old Sony, provided I installed my clock.
I am not sure the difference in sound is due to the laserunit, the chipset or both. Mechanically the Philips is inferior to the Sony. The guys at Philips are thinking if we can make that part also of plastic we can save another 2 cent....
The Philips CD-650 weighs 3.5 kg. The Sony CDP-XA30ES 9 kg!
Also interesting: The sound of my old Sony is very digital unmodified; the Philips sounds OK in original condition.
I did not perform any mechanical mod so far, but I can make the Philips more heavy with "lead bitumen" dampening etc.
Hi Mr. Triatec,
I changed from Sony to Philips recently.
Sound with my KWAK-DAC was distinctively better with the Philips.
I had a Sony CDPX33ES. I tried Philips CD-380 and CD-650. The CD-650 was slightly better than the CD-380.
I also tried the CDP-XA30 ES of my friend. It sounded about the same as my 12 year old Sony, provided I installed my clock.
I am not sure the difference in sound is due to the laserunit, the chipset or both. Mechanically the Philips is inferior to the Sony. The guys at Philips are thinking if we can make that part also of plastic we can save another 2 cent....
The Philips CD-650 weighs 3.5 kg. The Sony CDP-XA30ES 9 kg!
Also interesting: The sound of my old Sony is very digital unmodified; the Philips sounds OK in original condition.
I did not perform any mechanical mod so far, but I can make the Philips more heavy with "lead bitumen" dampening etc.
Based on the experience of both myself and some friends of
mine, we have all experienced that there is a correlation
between good mechanical build and sound quality. This
would suggest that read errors are less frequent with a
good mechanics and a sturdy case. However, it is unclear if
this is the case. It has been discussed in a number of
threads previously, and some people argue that non-correctable
read errors are extremely rare. An alternative explanation
could be that no manufacturer would take the cost of a good
mechanical build unless the electronics is already very good
so the price can be justified. Still, if the players sounded as
good with a weaker build, it is hard to see why they would
bother about the mechanics at all, since this is a very expensive
improvement in manufacturing. Yet a possibility is that it is all
about vibrations interfering with the electronics and that it has
nothing to do with actually reading the disc. Still, for whatever
reason, I would say that yes, generally players with good
mechanical build will sound better than otherwise comparable
players.
mine, we have all experienced that there is a correlation
between good mechanical build and sound quality. This
would suggest that read errors are less frequent with a
good mechanics and a sturdy case. However, it is unclear if
this is the case. It has been discussed in a number of
threads previously, and some people argue that non-correctable
read errors are extremely rare. An alternative explanation
could be that no manufacturer would take the cost of a good
mechanical build unless the electronics is already very good
so the price can be justified. Still, if the players sounded as
good with a weaker build, it is hard to see why they would
bother about the mechanics at all, since this is a very expensive
improvement in manufacturing. Yet a possibility is that it is all
about vibrations interfering with the electronics and that it has
nothing to do with actually reading the disc. Still, for whatever
reason, I would say that yes, generally players with good
mechanical build will sound better than otherwise comparable
players.
This is one of the issues I have been interested in for a long time, but I didn't find time to research it well. I wonder if read errors are common or if the differences are only because of jitter. I would like to make a test cd with some good wave files and then analyse the sp/dif datastream with some sort of a digital recorder. Then I could compare the original with the generated data stream to check for errors. I could try if things like "De Mat" on top of the cd will increase/decrease the amount of errors.
I think that the differences in sound quality from damping etc. are in the amount of jitter. If the laser assembly is unstable, more corrections of the servo are necessary to follow the holes of the cd. These servo corrections very likely influence the digital supply (give spikes on the supply) which will increase the amount of jitter. Regrettably, I do not have the equipment to check these thingies
Fedde
ps. in the case of cd-players instead of transports, sound changes will also occur due to IC vibrations and microphony of caps.
I think that the differences in sound quality from damping etc. are in the amount of jitter. If the laser assembly is unstable, more corrections of the servo are necessary to follow the holes of the cd. These servo corrections very likely influence the digital supply (give spikes on the supply) which will increase the amount of jitter. Regrettably, I do not have the equipment to check these thingies
Fedde
ps. in the case of cd-players instead of transports, sound changes will also occur due to IC vibrations and microphony of caps.
They *all* benefit from it, just because of cost reducing measures manufacturers leave it out. What's wrong with adding damping material ? The mechanism benefits from it and it is not expensive so what is the problem ?
I have to admit that I prefer electronic modifications first, later followed by mechanical mods.
I've thought the same thing many a time myself, mainly in regard to claims made here about the supeiority of one particular CD transport over another, or the boost in sound quality than an unintuitive modification made, etc. I mean, the truth is, digital is digital, and it's not all that complex. It's 1s and 0s and a timing component, and unlike analog audio where you can always invoke that tired claim that there's some mysterious, unmeasurable quality to your sound, every aspect of digital audio is, by definition, measurable. And it doesn't seem like it would be particularly difficult to actually go about measuring it--indeed, it seems like such a simple thing that you'd think some manufacturer would have done so already and would be trumpeting their results, but I've not heard of one doing so.fedde said:
fedde said:
No need to go through all that trouble. Most every decoder chipset out there has error detect pins on it (usually flagging for both C1 and C2 errors) that can be monitored to see how many errors there are.
As Christer has already stated, this has been done and it's been found that uncorrectable errors are extremely rare.
I think that the differences in sound quality from damping etc. are in the amount of jitter. If the laser assembly is unstable, more corrections of the servo are necessary to follow the holes of the cd. These servo corrections very likely influence the digital supply (give spikes on the supply) which will increase the amount of jitter. Regrettably, I do not have the equipment to check these thingies
Yes. John Atkinson at Stereophile uses the Miller jitter analyzer for testing digital components but I don't recall his ever using it to check for differences in jitter due to various CD tweaks.
se
I agree that digital data can be transferred without errors theoretically. I don't know how many data errors are made by standard transports. The clocking of data is certainly measurable. Though it is not a matter of absolute jitter, but also of the jitter spectrum. Also, the jitter experienced at the DAC is relevant. So for example the shape of the waveforms also matters. Finally, what is important is the correlation of jitter properties with the experienced sound.
I really wonder why you would need a Pitracer (EUR 29000,-) to have a really good transport. I'd guess that a DAC with an extremely good clock and digital supply (fed back to the transport) should give similar results. If not better. I don't understand why you make an extremely expensive transport and then transfer the data in S/P-DIF format and use 50 ohm tulip connectors...
Fedde
I really wonder why you would need a Pitracer (EUR 29000,-) to have a really good transport. I'd guess that a DAC with an extremely good clock and digital supply (fed back to the transport) should give similar results. If not better. I don't understand why you make an extremely expensive transport and then transfer the data in S/P-DIF format and use 50 ohm tulip connectors...
Fedde
Steve Eddy said:
If true, that is reassuring. Though I could also imagine a category of non-detectable errors. Also, an interesting matter is if LSB's are changed during the buring of CD-R's. Some people say that happens. If not and in the case of perfect reclocking in the DAC, CD-R's should sound the same as the original CD's.
Fedde
fedde said:I agree that digital data can be transferred without errors theoretically. I don't know how many data errors are made by standard transports. The clocking of data is certainly measurable. Though it is not a matter of absolute jitter, but also of the jitter spectrum. Also, the jitter experienced at the DAC is relevant. So for example the shape of the waveforms also matters. Finally, what is important is the correlation of jitter properties with the experienced sound.
I really wonder why you would need a Pitracer (EUR 29000,-) to have a really good transport. I'd guess that a DAC with an extremely good clock and digital supply (fed back to the transport) should give similar results. If not better. I don't understand why you make an extremely expensive transport and then transfer the data in S/P-DIF format and use 50 ohm tulip connectors...
Fedde
Fedde, others,
We did some tests on various drives, measuring the data and have a PC doing a bit by bit comparison. All data was the same.......
My suggestion is to make the drive as good as possible (lowest achievable jitter) and make the DAC as imune as possible.
Ciao
fedde said:Ah, there we have our jitter expert!
Thanks for your information. I remember that you advise to insert your (fine!) clock into the DAC and feed back the clock to the transport. Is the transport than still critical in that situation? If so, do you have a clue why !?
Ciao,
Fedde
Hi Fedde,
Yes, I advised that (I prefer to send back a PLL control signal and control a VCXO at the drive to prevent sending back a > 10MHz clock)
Yes, the transport is still important in that situation as jitter ripples through untill (I should say in) the DAC chips.
See our DAC design where we applied reclocking. If I would do a new DAC design I'd apply multiple reclocking wherever possible.
Ciao
This is the Chassis I am working on for my CDPRO 2, I have not got into the Electronics yet. The chassis weighs in at 35 pounds, the top that will have the drive mounted inside is 3" thick 12 x 16 Acrylic.
http://24.70.80.252/Photo/view_album.php?set_albumName=album39
Anthony
http://24.70.80.252/Photo/view_album.php?set_albumName=album39
Anthony
I've been stuck with a Cambridge Audio D500 for the least year (spending money elsewhere in the system), but I've taken the isolation and damping seriously. It sounds darn good (it will soon get a DDAC1543 to show what it really can do) at the moment. Just a few days ago it finally got to sit on proper roller bearings (not on the photos linked below yet)
http://didnt.doit.wisc.edu/audio/cdp/cdp.html
Messy looking, but it sounds great
http://didnt.doit.wisc.edu/audio/cdp/cdp.html
Messy looking, but it sounds great
Every time a discussion like this comes up about the influence of CDM's on audio quality I see this wonderful combination of myth and math...
As I worked for the CD-system-evaluation group of Philips I tested an awful lot of discs, drives players under a wide variety of conditions.
There is a CD-standard to make sure that every disc plays on every player. That this is taken for granted is a good thing (we succeeded in our objectives...).
To make a very long story short. A CDM fullfilling the requirements of the CD standard (so in 99% of all players) will give exactly the same data stream. What you do with this data stream after the CDM makes the difference...
If you have an old Philips CD200 series player with a SoPhi board in it you can even measure the errorflags yourself. Probably the most you can see is how much you care about your CD's.
Only a certified disc like the Philips 5A is a stress test. But normally the error correction should be capable to cover for 100%.
Of course a well designed CDM has more headroom for nasty errors like fingerprints, wobbely tracks etc, etc and probably has a longer (within spec) life.
Ward
As I worked for the CD-system-evaluation group of Philips I tested an awful lot of discs, drives players under a wide variety of conditions.
There is a CD-standard to make sure that every disc plays on every player. That this is taken for granted is a good thing (we succeeded in our objectives...).
To make a very long story short. A CDM fullfilling the requirements of the CD standard (so in 99% of all players) will give exactly the same data stream. What you do with this data stream after the CDM makes the difference...
If you have an old Philips CD200 series player with a SoPhi board in it you can even measure the errorflags yourself. Probably the most you can see is how much you care about your CD's.
Only a certified disc like the Philips 5A is a stress test. But normally the error correction should be capable to cover for 100%.
Of course a well designed CDM has more headroom for nasty errors like fingerprints, wobbely tracks etc, etc and probably has a longer (within spec) life.
Ward
EDUM said:Every time a discussion like this comes up about the influence of CDM's on audio quality I see this wonderful combination of myth and math...
As I worked for the CD-system-evaluation group of Philips I tested an awful lot of discs, drives players under a wide variety of conditions.
There is a CD-standard to make sure that every disc plays on every player. That this is taken for granted is a good thing (we succeeded in our objectives...).
To make a very long story short. A CDM fullfilling the requirements of the CD standard (so in 99% of all players) will give exactly the same data stream. What you do with this data stream after the CDM makes the difference...
If you have an old Philips CD200 series player with a SoPhi board in it you can even measure the errorflags yourself. Probably the most you can see is how much you care about your CD's.
Only a certified disc like the Philips 5A is a stress test. But normally the error correction should be capable to cover for 100%.
Of course a well designed CDM has more headroom for nasty errors like fingerprints, wobbely tracks etc, etc and probably has a longer (within spec) life.
Ward
Ward
Most confsuion is not about data but about jitter (or the ignorance of it)
I did some tests, extracted bit wise equal data from different drives, but jitter ade the difference
(I am still within Philips....)
regards
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.