Testing audiophile claims and myths

[castleofargh]

 

[bigshot]

He’s degenerated to idle blather.

 

[old tech]

Are you claiming PCM has hidden goodness that comes out when you convert to DSD?

A genie is released from the PCM file.

 

[Slaphead]

I will say that there is only one good reason to convert PCM to DSD, and that is that DSD a very good digital archive/preservation format due to the way it works. It can take quite a battering of flipped bits and still sound good as all bits are treated equal, PCM on the other hand is extremely susceptible to flipped bits as PCM assigns bits in increasing magnitude.

As for an improvement in sound quality - nope not going to happen

 

[redrol]

I will say that there is only one good reason to convert PCM to DSD, and that is that DSD a very good digital archive/preservation format due to the way it works. It can take quite a battering of flipped bits and still sound good as all bits are treated equal, PCM on the other hand is extremely susceptible to flipped bits as PCM assigns bits in increasing magnitude.

As for an improvement in sound quality - nope not going to happen

My computer doesn't flip bits. If it did, it wouldn't work. If you are talking about bit-rot on big HDs.. I don't care this isn't a data center.

 

[71 dB]

My computer doesn't flip bits. If it did, it wouldn't work. If you are talking about bit-rot on big HDs.. I don't care this isn't a data center.

Well, you won't probably use the same computer for decades do you? You buy a new computer before the old one starts flipping bits.

 

[Slaphead]

My computer doesn't flip bits. If it did, it wouldn't work. If you are talking about bit-rot on big HDs.. I don't care this isn't a data center.

Yes it does, all computers flip bits, it's just that through inbuilt error correction you don't notice it, most of the time - That BSOD that you got one time and never again, or that unexpected program crash you got once and never thereafter, well they are mostly as a result of a bit or two flipping randomly in such a way that the error correction thinks it's actually valid data.

However, what I'm referring to is long term storage of music - the archiving and preservation. DSD, through its very nature, provides enough redundancy that when corrupted data enters the stream that cannot be corrected it will have a mostly inaudible effect on the analog signal, and by mostly I mean below the level of human hearing. Flip the most significant bit in a PCM stream and you'll get that loud CD click. Flip a bit in DSD and you won't even notice it because all bits are equally significant.

Anyway that's really the only advantage of DSD over PCM. Overall however PCM has way more advantages than DSD as a working format.

 

[bigshot]

I think you are talking about preserving it aeons after you are dead and buried. Simply maintaining backups will do the same, wouldn't it?

 

[sonitus mirus]

Yes it does, all computers flip bits, it's just that through inbuilt error correction you don't notice it, most of the time - That BSOD that you got one time and never again, or that unexpected program crash you got once and never thereafter, well they are mostly as a result of a bit or two flipping randomly in such a way that the error correction thinks it's actually valid data.

It is absolutely ridiculous to attribute any one-time event of a BSOD to a flipped bit when there are a lot of other more realistic causes for such a predicament. I've had faulty RAM with one bit stuck discovered after testing and analyzing. Even with that obvious situation, my PC did not crash to blue screen on a daily basis before the problem was identified and resolved. The chance of some random bit going wrong here and there causing a BSOD would be exceedingly uncommon. It is far beyond reasonable to be concerned about something like that impacting music playback.

 

[analogsurviver]

I agree too. One of the best recordings I've ever heard was an LP of Fiedler's Gaeitie Parisienne that was recorded in 1952. I think it was the second Living Stereo recording made by RCA. The stuff that matters are the judgements made by the musicians and engineers, not incremental increases in data rates. I think sometimes audiophiles spend more time listening to the formats of their music than they do the music itself.

Hmmm ... careful with that one.

Back in those days, they all strived for the absolute best they could possibly accomplish. Bean counters have not taken over yet. And it was recording on the best medium/format then in existence - not the one that merely/barely yields so-called "audible transparency". Sure the recording mentioned does not stop in frequency response at 22.050,00 Hz - dropping off to nothing.

This recording underwent more format changes and variations than almost any other. I would love to lay my hands on a NOS sealed copy of this LP - the original first pressing. And make a decent digital high sampling frequency recording and FFT analysis of what is actually in the groove - from 1952.
All the digital meant only deciding how much of this original analog sound can be preserved in "digital of the month" while keeping the price low enough to generate sustainable sale figures. It never was about just incremental increases of data rate, it has always been about preserving the quality of the original.

Lacking the ability to grow extra kidneys etc that would allow for the purchase of such so old LP in case it ever again pops up for sale in such condition is making this issue - unfortunately - moot.

 

[bigshot]

The only sound in an LP groove above 20kHz is noise and distortion.

The Fiedler album doesn't sound good because of its specs. Its specs are inferior to a CD. It sounds good because the miking and balances were very well chosen. Engineering matters more than format.

I have a colored vinyl audiophile pressing of this album. I think I paid $15 for it. Sounds great.

 

[Contrails]

Engineering matters more than format

This.

 

[old tech]

Hmmm ... careful with that one.

Back in those days, they all strived for the absolute best they could possibly accomplish. Bean counters have not taken over yet. And it was recording on the best medium/format then in existence - not the one that merely/barely yields so-called "audible transparency". Sure the recording mentioned does not stop in frequency response at 22.050,00 Hz - dropping off to nothing.

This recording underwent more format changes and variations than almost any other. I would love to lay my hands on a NOS sealed copy of this LP - the original first pressing. And make a decent digital high sampling frequency recording and FFT analysis of what is actually in the groove - from 1952.
All the digital meant only deciding how much of this original analog sound can be preserved in "digital of the month" while keeping the price low enough to generate sustainable sale figures. It never was about just incremental increases of data rate, it has always been about preserving the quality of the original.

Lacking the ability to grow extra kidneys etc that would allow for the purchase of such so old LP in case it ever again pops up for sale in such condition is making this issue - unfortunately - moot.

I thought you'd be back at some point.

You know that the state of the art 8 track recorder used at Abbey Road studio back in 1968 (along with the best tape formulations back then) had a SNR of ` 70db and a frequency response of 30 to 15khz at +/- 2db. Given records used a master of at least three generations down from the recorder, apart from noise, how would a 1952 record have musical content anywhere near 22 khz?

Btw, the Beatles White Album sounds great despite the constrained SNR and frequency response of the SOTA equipment of the day.

 

[71 dB]

I thought you'd be back at some point.

Back or not, I'm ignoring him. I have zero interest of what he says, because he has zero interest of learning anything.

 

[analogsurviver]

The only sound in an LP groove above 20kHz is noise and distortion.

The Fiedler album doesn't sound good because of its specs. Its specs are inferior to a CD. It sounds good because the miking and balances were very well chosen. Engineering matters more than format.

I have a colored vinyl audiophile pressing of this album. I think I paid $15 for it. Sounds great.

It is not true that the only sound in an LP groove above 20 kHz is noise and distortion.

It might be true for the Fiedler album of 1951 - because tape recorders back then did not go that high in frequency.

I agree it sounds good for the reasons you cited.

However, there ARE analog records produced later that DO contain SIGNAL above 20 kHz. Mainly in jazz, where percussion can go well past 70 kHz - either in official 192/24 download or in recording from original LP using a cartridge with response extended at least to 50 kHz - where the leve(s) and frequeny limits might be a little less, but definitely "playing with the pulse of music" - and larger in amplitude than random residual noise.
To my surprise, I found clearly defined trumpet harmonics well past 50 kHz - on some digitally ( obviously, NOT at 44.1 kHz sampling ) recorded jazz records from GRP. They have used DMM ( Direct Metal Mastering ), which was advertised to have improved high frequency range compared to normal lacquer cutting - but I never saw any official specs/claims by how much. It clearly can and does perform better.
BTW - if record is mastered at lower than realtime speed, the high frequency limit of the resulting record gets higher inversely to the actual cutting speed used. Over 50 kHz is possible. And it has been used and the records are out there.