Yes John,I was describing the worst I had heard and of course one cannot know all the details of the transferring process for reissues.If you like this kind of thing then I am sure you will agree with me that the early SRO Ansermet CD transfers were awful.
I liked the Speakers Corner LPs more than the RCA or Testament issues but I do not remember a single case that I listened to where one of these reissues matched either originals or early cheap reissues, ace of diamonds, vics or concert classics(early EMI reissues over here) Given the fairly high price of these reissues I did not persist with them.This was probably 10 years ago now and I did not have the equiptment at my disposal that I have now. I am always prepared to have my mind changed!
Regards,Nicholas
I liked the Speakers Corner LPs more than the RCA or Testament issues but I do not remember a single case that I listened to where one of these reissues matched either originals or early cheap reissues, ace of diamonds, vics or concert classics(early EMI reissues over here) Given the fairly high price of these reissues I did not persist with them.This was probably 10 years ago now and I did not have the equiptment at my disposal that I have now. I am always prepared to have my mind changed!
Regards,Nicholas
At the time, [UNDERSTATEMENT]I was not wealthy[/UNDERSTATEMENT]. I had a Magnavox CDB473 that I thought sounded pretty good for cheap. Turns out it had the infamous TDA1541 in its guts. I'm not sure if I had my Sony SCD-CE775 SACD player yet. Both eventually broke down and are now history. (Who knew that both those cheapies would become regarded as keepers?) I also played the CDR master and commercial dupe on my computer setup (Plextor SCSI CDR into RME Digi96-PAD w/ 20-bit DAC's) and on the producer's CD player, which was his computer with a Plextor SCSI CDR drive and the above-mentioned Mixtreme hardware, into Mackie HR824 active monitors. The differences were apparent on both, as I recall. I remember he thought the CDR master sounded a little fuller. We just accepted it and moved on, as the final product does sound pretty good (if I do say so myself...). At the time, I expected the commercial dupes to sound a little different than the master CDR. I'd read that that was likely, I think on Bob Katz's website (Audio CD Mastering, Mixing & Replication), but I'm not sure. He used to have these long discussions of digital audio issues up there, but they seem to be gone now. Replaced by FAQ's.
Basically, we used the same CD player for each comparison, so how would it have mattered had we used better CD players or DAC's?
I bet if I uploaded the EAC ripped WAV files from the CDR master and the CD dupe, nobody would hear a difference, because you'd be listening to the same media (the WAV files played through whatever). I think (as in "believe") the difference is in the media used, Mitsui Gold CDR vs. whatever blanks, process, etc. the duping house uses to make commercial discs. But I'm curious to test this now, too.
ADDENDUM: I must apologize for not having time to look for the disc backups last night. I haven't thought about them for years. The producer has a CDR backup as well (at least I hope he still has it). It may take me a little while to scrounge these up.
--
Thanks Ron for doing this. I will start us down an interesting path.
Off the top of my head I would say with a better DAC you might hear more of a difference, with a better CD transport, less of a difference. But that's a big guess.
Off the top of my head I would say with a better DAC you might hear more of a difference, with a better CD transport, less of a difference. But that's a big guess.
Good point, Werner!
L2 does do SRC. You play a 24/96 file in, do your thing in L2, and select the bit depth/sample rate for its output, all within the plugin. It has noise shaping and dithering options too. I won't claim it's the best mastering limiter in existence. It's just the best one I had access to (the producer had bought it).
I think I had a Steinberg SRC plugin at the time that featured POW-r noise shaping, and I got that all fouled up with POW-r doing SRC. The Steinberg plugin probably came with the noise reduction set I got. I have no idea if it was any different than WaveLab's internal SRC. That was probably WaveLab 4, by the way.
That leaves WaveLab's internal SRC, and according to http://src.infinitewave.ca versions 5 and 6 perform somewhere between attrocious and sloppy.
Many mastering engineers claim tangible benefits from 88.2k and 96k, but about just as many mastering engineers have been using sample rate convertors of a blatantly incompetent design. Until recently it was almost like the more expensive the tool, the worse the SRC. And this is surprising, as the recipe for correct SRC has been known for decades.
At the time, it was recommended that files be edited at as high a bit-depth/sample rate as possible, then down-sampled to CD in one fell swoop at the very end of the process. That might have been to compensate for the lackluster software based SRC available at the time. As always, results are influenced by how careful and attentive the operator is at all points in the process. One tries one's best with the tools one has available.
I should point out that this was a completely "desktop mixing" and "desktop mastering" endeavor. Very, extremely low budget. We blew the bulk of the budget on the studio time for recording basic tracks (er, files). Recorded live to 8 tracks, using all good Neumann condensers (U87, XY pair KM84 on the bass viol, with a U67 on the voice, I think) with an EV dynamic on the one and only guitar amp (a little MusicMan hybrid tube/ss jobbie). The talent was a lone singer/guitarist with acoustic bass accompanying. I think we had two room mics that we didn't use in the end, so six tracks total, live to digital. Then we took the tracks to our home studio setups... Which were not top-end, I assure you. But again, we did our best with what we had available.
--
Last edited:
I wouldn't call these devastating monstrosities lacklustre. BTW, the plot I showed was of Wavelab 5. Can you imagine 4's performance?
Files still should be edited at as high a sample rate and word length as possible.
The word length thing is easy to comprehend, as it pushes down rounding errors.
The sample rate is a bit more complex: every non-linear operation executed in the digital domain, be it largely non-linear such as compression/limiting or much more subtle such as rounding/truncation, in-place aliases its distortion components, that spread out to infinity, against the sample rate and thus into the audible band. Operating at an elevated sample rate, regardless the file's original rate, alleviates this somewhat. It does not solve the problem. For this one needs aliasing-aware software which is rather rare. Another testament to the limited knowledge of signal theory some music software developers seem to have.
I did the rip versus read CD experiment I promised earlier.
I took my oldest CD. It was acquired in my teenage years in 1983, long before I got a CD player myself. The disc itself has been cleaned, and damaged, with isopropanol (don't ask!).
I ripped it to wav with EAC, all correction and checkers enabled. EAC reported a total of zero errors for the whole disc.
This was saved as the reference file.
I then put the CD in an old Rega Planet (one of the first samples, end of 1995 I guess), and piped its coax output through 2 meters of analogue interconnected (!) into my PC's Terratec Phase 26 USB interface, itself a total piece of crap.
I used Audition to make a digital recording of the CD's first track. This gave a second file.
Still in Audition I first trimmed the file with single-sample accuracy and then inverted the second track (both channels) and summed it with the reference file.
The result was a total NULL over the full track length.
This means that, compared to the DVD-ROM/EAC/PC rip the CD player did not misread or corrupt a single bit.
I took my oldest CD. It was acquired in my teenage years in 1983, long before I got a CD player myself. The disc itself has been cleaned, and damaged, with isopropanol (don't ask!).
I ripped it to wav with EAC, all correction and checkers enabled. EAC reported a total of zero errors for the whole disc.
This was saved as the reference file.
I then put the CD in an old Rega Planet (one of the first samples, end of 1995 I guess), and piped its coax output through 2 meters of analogue interconnected (!) into my PC's Terratec Phase 26 USB interface, itself a total piece of crap.
I used Audition to make a digital recording of the CD's first track. This gave a second file.
Still in Audition I first trimmed the file with single-sample accuracy and then inverted the second track (both channels) and summed it with the reference file.
The result was a total NULL over the full track length.
This means that, compared to the DVD-ROM/EAC/PC rip the CD player did not misread or corrupt a single bit.
Very well said. That correlates with a lot of things I thought I heard when mangling digital audio files.
Re: Wavelab and its subpar performance...
So even this (however misguidedly) well-regarded tool is to blame for some of the perceived problems with "CD sound"?
I've read that the distortions most likely to occur in digital audio production (and playback) are more readily apparent (and unnatural sounding) than the distortions most likely to occur in analog audio production chain. Now that I've learned that WaveLab and other programs like it are actually horrible distorters of the digital audio data, that makes me think the audio quality to be expected from CD might be doomed by common practice, not by its inherent physical properties.
OK, I'll readily admit that the quality of my mastering job was far short of what it could (or should) have been. As I explained, this wasn't a state of the art process. It was done with commonly available tools by non-experts (you could even say non-engineers, as I'm not one --). We did try to make it "sound good" in the end. My point was that the duped CDs from the duping house sounded every so slightly different than the master CDR I sent them. Your point is...?
What do you suppose the duping house did that might have changed the sound of the final product? Is it possible that it's just the "sound" of the different disc materials? Something that happened in the xfer to glass master?
--
PS - In reading thru M Jones' book last night, I came upon a bit about how he noticed some part or other in a power supply making the amplifier sound different than if using some other part. Sorry, I don't remember the details, but I think it was in the Rectification chapter. He couldn't think of why that could be, so he put it down to his imagination and left it for later. He later thought of a good engineering reason why he might have heard what he thought he heard, and upon further investigation was able to prove to his own satisfaction that he had indeed heard the difference and that he understood why. Just an illustration of how not everything one hears can be explained right away in good engineering terms. Sometimes people hear things and the reason why doesn't come to light for a long time...
--
OK, unless I'm mistaken, that proves that EAC does a really accurate rip, and that Audacity flips polarity without goofing up any bits. That is indeed good to know, as I can download Audacity and use it for free and continue the experiment. (I already have EAC.)
Now if you take those ripped WAV files and burn them to a nice, new CDR (preferably on some really nice media), will the bits be the exact same? If yes (and it would probably pass that test), then send that master CDR out to a duping house. Will the dupes sound the exact same as the master CDR?
--
Last edited:
Earlier in this thread, I mentioned that Winston Ma, the owner of First Impression Music brought in a glass master and a CD made from that master and we played them both for a comparison. The glass master sounded a "lot" better than the CD, although the CD was very, very good by itself. Evidently, as Winston explained, this is fairly normal within the Industry and not an isolated case.
It was a lot like listening to the Master Tape and an LP made from it.
BTW: I think that your last sentence is right on the money!
Best Regards,
TerryO
Some of the perfection of CD is here SY:
1) So nothing wrong with 44.1kHz
2) Nothing wrong with 16 bits
3) And no bit errors.
SY, abraxalito and Werner have all at various points in this thread (regardless of the amnesia that sweeps them now) defended some or all of these aspects of the CD. Abraxalito even avoided the entire logical tenant of perfection being something devoid of error.
I noticed this because I argued that the CD had shortcomings in all 3 of these areas, and was roundly and soundly beaten down by people who consider the CD to be free of imperfection, yet strangely reticent to admit it is perfect.
So yes SY: If you argue that there is nothing wrong with the CD, in your view it is perfect. This is logic of a digital nature..
If you do not consider the CD perfect: please tell me why, I'm curious!
In the meantime: time to put an LP on 🙂
1) So nothing wrong with 44.1kHz
2) Nothing wrong with 16 bits
3) And no bit errors.
SY, abraxalito and Werner have all at various points in this thread (regardless of the amnesia that sweeps them now) defended some or all of these aspects of the CD. Abraxalito even avoided the entire logical tenant of perfection being something devoid of error.
I noticed this because I argued that the CD had shortcomings in all 3 of these areas, and was roundly and soundly beaten down by people who consider the CD to be free of imperfection, yet strangely reticent to admit it is perfect.
So yes SY: If you argue that there is nothing wrong with the CD, in your view it is perfect. This is logic of a digital nature..
If you do not consider the CD perfect: please tell me why, I'm curious!
In the meantime: time to put an LP on 🙂
You're still misquoting me. Sonic transparency of the format (44.1/16 bit) under normal use? Yep, I'm on board with that. To go from that to a blanket "CDs are perfect?" No, never said that. Don't attribute that to me.
It is rapidly becoming tiring.The thread falters.
If the most of us, were reading more carefully the "perfect sound forever",thingy,it would have been obvious,that they were placing the weight of it, in "sound".They never said "perfect music for ever"
B.L
That was a great test and thanks for doing it.
Now we should be curious to see what would happen if Ron does the same with his master CD-R vs the factory made version. Using the program DiffMaker should be even easier.
Lots of things. Bad engineering. Compression. Peaky mikes. EQ for car stereos. Distortion manipulation... If the signal that makes the digits sucks, the audible transparency of the format doesn't mean a damn- it's transparent to a lousy product. And CD is a 2 channel format- 2 channel is FAR from "perfect," even if done well.
Now, a well-recorded CD, that's a whole 'nother animal, though it still suffers from the limitations of a 2 channel medium so can't be "perfect." But saying "CDs are perfect" (what you falsely attributed to me) is utter nonsense as long as record companies and producers keep putting out lousy sounding product, as long as we use imperfect mikes, and as long as we're working with only two channels.
edit: In the list of Bad Things, I should add some editing and mastering software- and the chimps who run it.
Now, a well-recorded CD, that's a whole 'nother animal, though it still suffers from the limitations of a 2 channel medium so can't be "perfect." But saying "CDs are perfect" (what you falsely attributed to me) is utter nonsense as long as record companies and producers keep putting out lousy sounding product, as long as we use imperfect mikes, and as long as we're working with only two channels.
edit: In the list of Bad Things, I should add some editing and mastering software- and the chimps who run it.
Your memory is somewhat poor. I was demonstrating that your claims of CDs flaws were based on false premises. That's not 'defending aspects of CD', rather attacking (not even really attacking - undermining is a better choice of word) erroneous claims and sloppy thinking. If your claims had been accurate ones about CD, then I might have defended CD, or alternatively admitted some shortcomings. However, in this case, your claims missed the mark - so no need for defense. Defense is only called for when there's attack. Any clearer now?
I am not following you here at all. Where did I do what you are claiming? If you're referring to where I declined to 'help you' then you ignored the reason I gave for why not.
Not 'beaten down' - none of my arguments were directed at you as a person, rather your arguments and claims were shown to be mistaken. Note that at no point have you said 'OK, fair point, I didn't realize that before but I'm getting the picture now.' Rather you refer to myself, SY and Werner as 'experts' but don't take our advice. Notice any irony there? 😀
And this is entirely in your imagination. Which is why both SY and myself have asked you to substantiate it. Why not simply admit you cannot substantiate your claims?
Last edited:
Time to back off guys. You are taking this waaaaay to seriously. I'd hate to see you land yourselves in the sin-bin for something so silly.
I wouldn't see doing time as something particularly undesirable - as Werner's already pointed out, I could really make progress with my design for an ultra-advanced anti-aliasing filter...😀
- Status
- Not open for further replies.