CD Japan to server

[jtwrace]

If you load your server with a higher quality disk to start, does that mean that it will still sound better when played through your DAC? I would think that the better the quality is on the music going in, the better it will be coming out? Anyone try?

 

These are the formats I'm speaking of:

http://www.cdjapan.co.jp/popular_formats/index.html

 

Thanks-

 

[flat6]

i hv quite a number of SHM-CD rips. one is able to hear the diff after they've been ripped.

 

as for the "quality" of SHM-CDs, i've found some SHM-CDs to be significant better, while others hv only marginal improvements (compared to their redbook counterparts). they differ title to title IMO.

 

[cbarber]

If you do a secure rip that reports 0 errors using a REGULAR CD, how could it sound any different from a secure rip with 0 errors on an SHM CD?

 

[solidsqual]

Some disks like the Mobile Fidelity Gold disks are better engineered sonically. IT's the same songs but standard discs of the same album often sound far worse whether played on a CDP or computer to DAC.

 

[cbarber]

I can definitely understand that if played on a CDP, but if you take the data off with a 100% perfect rip, would not the data be exactly the same and therefore the sound quality exactly the same?

 

[sandyk]

The Blu-spec sampler discs from CD Japan,"Feel The Difference Of The Blu-spec CD- The Rock Selection" and also the Jazz Selection, come as a set of 2 CDs with identical content. One CD was mastered from a stamper created using a Blue Laser, and the other using a stamper created using the normal "red" laser. With my equipment, the Blue Laser created version sounds better when played via my Oppo 981HD SACD/DVD-A player, as well as ALSO sounding better when played back from the HDD after ripping using E.A.C. (!) This is despite both lots of ripped .wav files having identical check sums.

.wav files created by ripping the same CDs, sound even better when played back from my HDD after being ripped by my LG BluRay writer,in comparison with those ripped by my Pioneer DVD writer. and the CD created by the Blue Laser stamper, still sounds even better. Once again, the checksums are identical. (Both writers were calibrated using the EAC Server) The quality of the player used via SPDIF IN, or directly from the CD/DVD writer, does have quite a bearing on the sound quality of the ripped .wav files, just as it does when played directly from SPDIF OUT of CD players into a high quality DAC.

Please see my post , and a later post by Silverlight in :

http://www.computeraudiophile.com/content/PC-Performanceupgrades-impact-sound-Honest-Canvass-Opinion

 

SandyK

 

[Audio_ELF]

Sandy. I'm pretty sure that you are kidding yourself about differences. IF the checksum of the file is the same, then there is no way that the files can be difference. It's like saying - after copying the file to a Super Acme USB memory stick and back again the sound quality improved - it just is not possible. If the files are read correctly off the CD, an the WAV files are identical, then the files MUST be identical and CANNOT sound different.

 

There maybe differences in the mastering of "blu-spec" CDs, but then the checksum would be different. There is certainly no way, unless one had errors, that two rips from two different drives would sound different IF THE CHECKSUM WAS THE SAME.

 

Eloise

 

[sandyk]

AUDIO_ELF

Did you read the other post, and the confirmation from Silverlight that the LG BR writer does make better sounding rips than the other optical drive he had been using ?

As I stated in the other post, I am not prepared to argue any longer with people who claim that just because 2 .wav files have identical check sums, means that the files MUST sound the same.

I have posted numerous files for comparison by Rock Grotto Forum members, and the majority of those who downloaded the files were able to hear the differences, and pick which file was which , without being prompted. I also posted an invitation in the other thread for members to email me for links to a couple of uploaded high resolution files for comparison. These had identical checksums. I only received 3 replies asking for the links. To date, only one member has replied with what he found, and subsequently, several other files were uploaded and compared by him. We also compared rips of a Norah Jones track using EAC,which he uses as a reference, and we were both able to pick audible differences between them.(Another RG member was also involved in the comparison of these uploaded files.) Now that he is using the same LG BR writer in an external case, we are also able to obtain identical check sums from both machines using the same CD that we have in common. This further aids our comparisons.

Incidentally, there was also an article in HIFI Choice ,April 2009, (Ripping Yarns) where 2 respected audio (and computer) journalists came to the sudden realisation that computer generated rips were light years behind the Naim HDX, a machine that was designed and built by audiophiles to rip CDs.

Obviously, there are other factors involved in the playback quality, otherwise why have several members, including the founder of this forum reported, and been able to demonstrate improved sound quality when the files are stored and played back from a SSD ?

I understand that even Peter St. has heard differences when using a SSD, but is so far unable to explain why.Give him time, though !

I am prepared to upload further tracks for comparison by members who don't have a closed mind, as well as a minimum of a reasonable soundcard and a high quality DAC, preferably via SPDIF, although having recently had the opportunity to compare a new Benchmark USB DAC with my own heavily modified X-DAC V3, it does also appear quite suitable via USB, due to it's jitter reduction regime. BTW, the differences between high resolution rips of 24/96 and 24/192 from DVD-A , are even more pronounced than 44.1 16 bit tracks.

SandyK

 

[Audio_ELF]

Yes I read the previous comments. I still stand by my statement / belief that if the checksum is the same, the files mst be identical. I understand you desire not to argue and that you stand behind you belief, but to me it's just rubish and there is no objective evidence that there can be a difference, but plenty of subjective evidence that there can't be. Once the checksum is the same, then it's just like copying from one device to another.

 

Please note there is evidence that PLAYBACK can be affected by device (SSD vs HDD) but this is a different issue. I just cannot find any reasonable evidence or explanation that there is a difference between ripping on a LG Bluray vs a Pioneer DVD-RW when the resultant files checksum identically. If indeed the WAV files do cecksum identical, then they are the same file whatever the source of the file.

 

Eloise

 

PS: does the term "checksum" refer to doing an MD5 checksum of the WAV file, or the checksum given by EAC after ripping? Not that it makes a difference (IMO) just to clarify.

 

[souptin]

Maybe this is a stupid question to ask of all you big brains, but does an identical checksum absolutely guarantee sample for sample identical data? I can understand that if the checksums don't match then it follows that the data doesn't match, but not the reverse.

 

If the reverse was true, then shirley (forgive my crass oversimplifications, and the Airplane joke) the 10 to 20 character checksum (not even a kilobyte in data size) could somehow be decompressed into the original music file of 20MB or more. Which would be pretty cool, come to think of it.

 

[Audio_ELF]

Two complete different samples can give a different checksum, but two samples that very a small amount will not.

 

A simple checksum is to check a series of numbers 4, 7, 9, 4

to get the checksum digit we're going to add the four numbers (total 24) and just take the smallest digit (4).

 

Now in this simple exampl it's obvious that a lot of combinations will give same checksum. However it's unlikely that all 4 digits will change -maybe sequence will be read as 4, 7, 8, 4.

 

When the checksum is created we'll see that you get 3 not the 4 that we know it should be, so we'll know there is an error.

 

Obviously the MD5 checksum is a lot more complicated, but that's the basic idea. Two completely different files will create the same checksum, but two files that vary by a few bits here and there will not.

 

Eloise

 

[sandyk]

AUDIO-ELF

The checksums are verified using ExactFile. These days I only worry about the MD5 checksum, although originally I had ExactFile compare using several other different types, and ALL checksums were identical.

I find it interesting that in all cases, the files ripped using the BluRay writer, especially after further sound and vibration dampening in the PC, appear to have what appears to be an improvement in signal to noise ratio. i.e. very low level detail is more easily discerned, and fading voices appear to last just that little bit longer.

I have seem similar replies to yours many times over, and going into greater depth as to why this is impossible. Your reply is at least not sarcastic to the extreme, or plain nasty like some.

I will also state that myself and others, have also been able to notice the deterioration in SQ of .wav files that have been moved between HDDs in an undampened P.C., or to a USB memory stick and back.One Sydney RG member even took some files home from my P.C. and tested them. He verified my claims, as he was also able to hear a deterioration with a file that had been moved between HDDs. Several Rock Grotto and DIYAudio members from Sydney have also readily heard these differences directly from my P.C. and a more recent visitor had great difficulty believing that the compared .wav files came from the same source.

Incidentally, a very respected member of this forum has also said that he believed he was able to hear differences between files that were uploaded, despite not using a DAC at the time.He was rather embarrassed to admit this, as he could not explain why.

The differences between earlier rips and current rips are now more pronounced than they were back then.

 

I repeat my offer to upload a couple of tracks for comparison purposes ONLY, to members with an open mind and suitable high quality PC playback. I do not consider direct playback from ANALOGUE OUT of a soundcard as suitable.

 

SandyK

 

[cbarber]

Re: I will also state that myself and others, have also been able to notice the deterioration in SQ of .wav files that have been moved between HDDs in an undampened P.C., or to a USB memory stick and back.

 

Are you saying that the very act of copying a file from one disk to another can affect the file's sound quality?

 

We keep finding we have to peel back our expectations of physics. We can get to the physics later. But for starters, is this what you're saying?

 

[souptin]

This all sounds very interesting. I'm skeptical but also curious as to your theories about what might be making the difference. For example are you convinced that it's a difference made in the ripping stage and not a universal playback improvement on your own pc as a result of your work on vibration damping?

 

Have you or any of your testers done a full sample comparison between files? No doubt you'll have read my earlier question about the possibility of having different files with the same checksum, which is the reason why I ask. I'm afraid that I don't know if any software allows this sort of comparison to be done easily.

 

Since your testers are reporting differences as a result of copying a file between hard disks and to and from usb sticks, have you been tempted to compare the result of copying the music file directly off the cd instead of going via ripping software? By extension, this makes me curious about whether comparing files transferred across the internet would give much insight, since the same degradation might take place.

 

Incidentally would it be at all possible for you to make these comparison files available via one of the free filehosts like yousendit / rapidshare etc, or is there some reason why it needs to be done via email?

 

[sandyk]

cbarber.

Yes, I am indeed saying that. However, since extensive sound and vibration deadening to all drives, and especially the SMPS, I have not noticed further degradation when moving files between INTERNAL HDDs, although it is still quite noticeable with an external HDD. I suspect that this would be a non issue if all HDDs were SSD. I have also found that saved 24/96 paid .flac downloads from HD tracks, now sound better after reconverting to .wav files , in comparison with files converted from the same downloaded .flac files before the extra sound and vibration dampening was done.

SandyK

 

[tmfidelis]

How about bypassing the glass master and CD by importing the data from a DDP image?

 

Regards,

 

Tim Marutani

Emeryville, CA

 

[sandyk]

 

tmfidelis

Wouldn't the record companies love that !!! NOT.

 

I think the best we can hope for is an expensive Reference Recordings specially prepared DVD ?

SandyK

 

[sandyk]

souptin

I have found as several RG members have found, that the improved SQ of the rip from the LG BR writer is a key factor. As this writer is also internally mounted, it has also further benefitted from the PC sound and vibration dampening. Even files ripped last year by my Pioneer writer now sound better too, but are still quite audibly inferior to those from the BR writer.

I have not attempted to copy directly from the CD in the LG BR writer itself, as although playback SQ is noticeably better than with the Pioneer writer, it doesn't sound nearly as good as the ripped .wav file. It sounds very much like a decent CD /DVD player does when listened to directly.

Much to my surprise, files uploaded to Rapidshare and Filemail, appeared to have suffered very little degradation when I downloaded them again . They still sound better than the same files burned to a 24K Gold Kodak Archival CD, which I find gives better than average playback results.

The files that I have been uploading for comparison purposes using Filemail for recent uploads, are copyrighted material, and I have no desire to post a direct link in a publicly accessible area of a forum, which at the very least, is likely to result in a reprimand from Admin, and removal of the links.

 

SandyK

 

P.S.

I have found that the SQ differences also vary greatly according to which software player is used. Several people report that the much maligned Creative media Source player gives better transparency and detail than Foobar. A friend from Brisbane gets excellent results with a version of Peter St's XXHE.

We have also found that EAC gives a higher quality, more transparent rip than dbpoweramp, although the latter is much faster and has some great features such as album information etc.

 

[souptin]

Fair enough - I see your point about your comparison files. I'd hoped that the CD Japan folks might have had some free sample material that was ok to distribute. Would have been interesting to see the results of, for example, a poll with a larger number of people taking part.

 

To clarify on my other question about copying instead of ripping, I was thinking if you mounted the disk, then copied the tracks you wanted onto the hard disk. Obviously you get files with names like "15 Audio Track" - no niceties like artwork and tags - but they can be played back from hard disk or ssd instead of via the cd drive. My theory being that even if a two files are similar enough to pass a checksum test they may not be 100% identical.

 

Of course if the files are the same down to the very last byte and still sound different then my theory is blown out of the water, but I've experienced enough glitchy files to suspect that even simple data transfer isn't always as error free as it might be. Sometimes software has the ability to repair damaged files without user intervention and (presumably) some file formats contain redundancy to help with this. DDP sounds like an example if my google research skills are up to scratch.

 

I have to say, interesting as this discussion is, it's probably way above my pay grade when it comes to listening skills. For example I still think Rod Stewart's Forever Young is the exact same as the Bob Dylan song, despite having different lyrics, chords, and teams of lawyers to assert the opposite.

 

[silverlight]

jeez, I'm late to the party, totally missed this thread. anyway, I do want to back SandyK up and provide one additional feedback that I DEFINITELY hear differences from the same CD rips I did in the past with a rip done now with the LG Blu Ray drive. While on my speaker system the changes are more subtle as I'm still awaiting more resolving speakers to be delivered, with headphones using the Berkeley DAC, the Weiss DAC2 and the EMU 0404 all with Amarra, it's definitely there. The more I listened the more I started to pick out the differences. Mind you this is with some _seriously_ transparent headphones - the Sennheiser HD800's which I'm demo'ing along with a few other sets in the comparable category. The 3M tape (which I now have a roll of) is also making a difference, as I've now dampened one of the drives and going to A/B rips with another drive after application just for verification.

As an example of the different sound, same checksum. SandyK and I swapped rips of Feelin' the Same Way, Norah Jones (we both have the Blue Note Hybrid SACD pressing). Man were they noticeably different before I got started on making these tweaks. I still haven't reached the SQ of his rip, but getting closer.... (and.... don't ask my why, the logic side of my brain of course is with Eloise on this one)

 

 

 

[The Computer Audiophile]

I can certainly see/hear two rips from different CD Drives sounding different. The drives may not be equal in ripping quality and one may have needed to interpolate the data if a secure rip could not be done.

 

I think the real test is to see if the files match. Some work using the drives' offsets and reripping etc... may need to be done.

 

[sandyk]

Hi Chris

All the .wav files that I upload normally have identical checksums, as verified by ExactFile. The Norah Jones tracks that Silverlight and myself have compared together ,also had matching checksums at both ends. These were both from LG BR writers, and we did compare our respective checksums. Presently, I am uploading a couple of 44.1 16 .wav files for comparison purposes ONLY.

They had identical check sums at the time of attaching, and should still have identical check sums on receipt.

I will happily provide download links on receipt of an email, but I do recommend a minimum of a good DAC,preferably via SPDIF so as to be sure that the differences are clearly heard.

Both files have this checksum. :

5e4af86de7079b5173ff70d2327bf054

 

SandyK (Alex)

P.S. I think downloading them as individual .wav files , instead of a zip file would be a good idea. There is a maximum of 10 downloads, and the downloads will expire in 5 days time. Please use this comparison for personal evaluation only.

 

[email protected]

 

[Audio_ELF]

You have me believing it is possible, at "different drives create different sounding files" and "anti-vibration helps SQ" but I become totally disbelieving when you assert the files have exactly the same checksum, but sound different. To my mind this is impossible as if the files are the same at a checksum level then HOW does the computer know that one is from the LG and one from the Pioneer to be able to play it different?

 

Eloise

 

[sandyk]

AUDIO_ELF

 

"HOW does the computer know that one is from the LG and one from the Pioneer to be able to play it different?"

Or as silverlight posted, between a Teac and a LG BR writer ?

My guess is that the .wav files are most likely identical, but it is the actual system noise level that is different at the time they were ripped and saved . Perhaps something like an AM radio signal modulating a noisy carrier wave ? TBH, I have no real idea of the mechanism involved, but in the earlier uploads where the differences were far more subtle, most people made a comment about what appeared to be a lower system noise floor, permitting fine detail to be more readily resolved, and natural decay of voices and instruments to be more apparent. With lower level voice in a Dire Straits recording, the comment was made about then being able to hear a subtle echo behind the voice, whereas even the voice was indistinct on an earlier rip. It would appear that jitter plays an important part, as the higher resolution 24/192 and 24/96 .wav files appear to show the greatest differences. At least with the limited number of DVD-A discs that I own. If you have a suitably high resolution PC playback sytem, why not see if your sytem is able to reveal any audible difference between the 2 .wav files that I uploaded several hours ago ? Incidentally, I even included an Electronics Engineer from Queensland in emails with download links to try, as well as offer to send him sufficient 3M tape free of charge, to try with his Mac. However, both offers were declined.

I would hope that not all EEs are that closed minded ?

SandyK

 

[silverlight]

I agree conceptually that it makes no sense that two files with the same checksum would sound different. I think I recall a discussion on AC around ripping software and drive selection (maybe in the Empirical Audio forum) and potentially on the AA forum that discussed this at length, that folks were getting better sounding rips from one drive and software over another (in those discussions dbpoweramp was being highlighted and the DW-224SL TEAC drive), while the files had the same checksum. the attribution was being given to drive offsets at the time as the most likely possibility.

 

Well, I taped up my DW-224SL drive last night top and bottom with the 3M tape, and the rip exceeded what I got in terms of SQ than what I got from the full-size Blu-Ray Drive (LG GGW-H20L), which leads me to conclude there are two factors affecting the sound quality - power and vibration. In order of best SQ to worst SQ: (1) TEAC + 3M, (2) LG Blu-ray + 3M, (3) LG Blu-Ray, (4) TEAC. The big difference in power is that the TEAC drive can run off USB bus power, and if you're using a notebook and pull the cord and run it off battery you're not plugged into the grid or using a switching power supply for the drive. The Blu-Ray (full-size) is a power hog, using a dual voltage el cheapo switching power supply built into the back of the enclosure (5V/1.9A + 12V/3A --- the 5V is for the logic board and 12V for the drive servo). Well I researched and spoke to a couple suppliers on getting a good linear PSU for the BluRay drive, and it was going to cost between $350 - $400 for a two-box solution (because no dual-voltage linear PSU's come with 3A on the 12V side that I could find in speaking with Acopian, Bolder, etc.). A better enclosure with a large external switching PSU (vs the tiny one in the enclosure I used) might help the Blu-Ray drive along with an enclosure that doesn't use an intermittent fan and flashing lights, maybe FW would best USB, dunno (or internally mounted in a tower case like a PC or MacPro, well dampened). Anyway, since I'm into silly experimenting while I'm waiting around for my system to be complete, I ordered the latest slim version of LG's blu-ray drive from HK that runs off of a single 5V/2A bus power or external PSU. Will post results once I get this drive and treat with 3M tape to see what happens.

Last aspect of the configuration which may add to the quality of the rip was that I'm using an internal SSD on the MBP13 (which means no moving parts / no vibration in the computer where the file is stored). the MBP also sits on 3 vibrapod feet, as does the drive when I do the rips.