Why do you continually make up strawmen? Gibbs has been around longer than we are alive, we all know it.
As I pointed out, using Scott's difference technique allows us to see exactly where the filtering occurs in time.
As I said, the envelope modulation was not considered well in the choice of sampling frequency, and sidebands cut out will impact envelope.
Again with the strawman argument. As I said, any attempt to test for the effect requires knowing what to test for. Scott's diff test is one significant measure to determine if and when a sideband is being truncated.
If he started with low bandwidth recordings, what would be expected?
Sure, no problem..I'll just take content off of one of my CD's..
It's not a case of liking or disliking your sims. It's a case of understanding what one is looking for. If you (Hans) choose (as you did) a modulation which does not take the sideband over Fs, you can't state that "there's nothing to see". I cannot tell if you chose a modulation which would not show the effect, thinking nobody would be the wiser and you'll look real smart...or if you really didn't realize the need to go with a steeper slope. The same with the raised sine modulation choice error..
I would prefer to think you simply chose things incorrectly but without malice, and I truly hope that is the case. I note that it would not be the first time on this forum where someone chose to do that.
Also, for the sim, the abrupt start will easily confound the sideband truncation as well, since the gibbs occurs over a few cycles, so I mentioned the time mirrored exponential into flattop as a better test option.
Oh, btw..in the past, I mentioned enjoying sparring with you, and someone had a real difficult time with that. A sparring partner is never chosen as an easy opponent to beat. They are always chosen as an equal to improve ones' self.
This is not a contest, never has been..
And, thank you for your effort with the sims.
I note you didn't answer my question regarding the ugly output envelope. Can you determine if it is pixellation?
Cheers, jn
That would have been a decision to hide a result that confirmed an effect and provide a lesser one that did not show it as proof it did not occur.
I would not accuse Hans of that, are you thinking he made a choice like that? If that's the case, I certainly disagree with you.
jn
I thought you meant when the CD sampling rate was chosen. That's what I meant.
Last edited by a moderator:
Ah, my apologies.
At that time, I would have made the exact same decision they did. Or at least, I wouldn't have argued.
jn
At that time, I would have made the exact same decision they did. Or at least, I wouldn't have argued.
jn
Last edited by a moderator:
I have listened to this now and it's wonderful. Yes the fidelity is low, but the emotion of piece is there in spades. Maybe the two men had met? 1947 is a time few of us would now be able to comprehend but at least to me you can feel some of the horrors of 2 wars coming out in the music.
I will have to hunt out more of his recordings. Thank you again.
First there was fighting over Jenkin's ear now are we fighting over Gibb's ears? Neither fights seem productive.
No new light, folks aren't about to change their positions, facts don't count.
No new light, folks aren't about to change their positions, facts don't count.
Elektroj, please note in the digital paper that the Soundstream was only BACKUP for the direct disc. The vinyl that was sold was directly recorded until it ran out. Yes, there WAS a difference! I made the console electronics for Crystal Clear, the microphones were B&K 4133 with 2619 preamps. John Meyer handled the interface between my electronics and the B&K's (power supplies, connectors, etc.) Direct disc to the recording lathe was the best, and what was primarily used. I did not make any analog recorders for Crystal Clear. I made them for a competing company, Mobile Fidelity, and then later, my best, with Wilson Audio.
Last edited:
Well, lets see. I made what appears to be a bold and challenging statement to some here about the sound of CD. In a nut-shell, the sound is much better at low freq than at high freq compared with real/live.
The sound progressively gets worse as the frequency is increased... using cymbals for example is most obvious to me.
The HD or Hi-Rez down loads offer the best performance because of all those things we talked about here... sampling, filters, bits, BW and Masters are often used to get closest to the original recorded sound.
Maybe you dont need it to enjoy Your music but it is more accurate and realistic sounding
IMMHO
THx-RNMarsh
Last edited:
But of course! Sorry, I've got that mixed up.
Did MoFi or Dave Wilson ever rented their machines out to other recordists/labels? It would be interesting to know what else is out there recorded with your electronics.
Regarding that Fiedler session - apparently it was released in CD format too:
Arthur Fiedler And The Boston Pops* - Capriccio Italien - Capriccio Espagnol (CD) | Discogs
Arthur Fiedler & The Boston Pops*, Tchaikovsky*, Rimsky-Korsakov* - Capriccio Italien / Capriccio Espagnol (1990, CD) | Discogs
I'm not going to compete with collectors for vinyl, but I might buy one of those just for kicks. Usually, if the recording is good, it "cuts thru" no matter what the release format is.
Unless you especially like the music, I think that buying the CD would be a waste of time. It is 2 generations away from the original direct disc recording. The recording itself is 'laid-back' and not really exciting, due to the placement distance of the microphones. We almost did not release it, even with direct disc, because of this, BUT it actually sounds about as close to a live performance as someone sitting in the audience would normally get. I don't know if the CD is being taken from the analog recording (the digital would be even worse, because of the state of digital recording in that period), but in any case it would be a lose, lose proposition. Recently, at our SF Audio Society, a lucky customer found one of the original vinyl recordings for $1, and he commented how good it sounded to me just last week.
Last edited:
John, all you said is true. I was only playing with the idea of buying one of those just for the feel of it. It's not a big loss. If it turns out to be to my taste, then I might start looking for original LP.
(Although I'n not sure if I need yet another recording of Cappricio Espagnol, Cappricio Italien or Fiedler 😀)
(Although I'n not sure if I need yet another recording of Cappricio Espagnol, Cappricio Italien or Fiedler 😀)
I was totally surprised by the lack of information about it. All specified the flat frequency response up to 20 kHz (within a fraction of a db) and distortion but considering the importance of the anti-aliasing filter I'd assume that at list the reviews in the professional mags aimed at the recording professionals would examine that, but no.
So presumably you're right about the quite common aliasing products at that time; surprising that the PCM-1630 already had phase compensation and even a switchable dither noise generator but otoh used the mentioned filters with a cutoff frequency of 24 kHz.
Against the old saying sometimes you can build on sand. 🙂
I didn't calculate it but from the graph in the AD713 datasheet the filter has a cutoff frequency around 30-35 kHz, so scaling it down by two when doubling the capacitor values could be a bit too low.
As scott wurcer said, it was intended as an anti-imaging filter in a two times oversampling system, where the digital filter (hopefully) provided sufficient attenuation at and above 22.05 kHz and the anti-imaging only had to deal with the remains around 88.2 kHz and multiples.
Last edited:
Too bad.
Of course you're right, it is mainly because I'm interested in such historical details; IME often things clear up in hindsight (see for example the myth around the "Beethoven 9th" as explanation for the quite late change in diameter) .
Wrt sampling frequency choosen, it seems that it was mainly based on the pcm adapters already in use which were using video tape for storage and had their sampling frequency tied to the two formats (44.1 for PAL, 44.05x kHz for NTSC) although the Soundstream experience with Fs of 50 kHz already existed and had allegedly some benefits.