Home |
Search |
Today's Posts |
#321
|
|||
|
|||
Ron Capik wrote:
Isn't imaging, sound stage, (etc.) something contrived by the recordist, mix engineer, and mastering engineer. Don't they all have an impact on the spatial image? They do. But I know what concert hall X sounds like, at least from the seat I like. When I put in this recording made in concert hall X, I want it to sound like that. Does it? If you weren't the engineer how could you possibly map that vision to one's own listening space? Heck, with multi-tracking the entire space is contrived. Who's to say what's ~more~ natural. That's where the problems start coming in. I can say that I have a good solid image that extends beyond the speakers and has a sense of depth to it... but maybe that's not at all what they wanted when they created that recording. An A/B test may identify a difference but can it provide any insight into which is closer to the producer's/engineer's vision (or should I say auralization... ) ? Has anyone calibrated their listening room with respect to the mastering studio? Just what was that target average listening room anyway... ? I don't know, but I know what the recordings I made are supposed to sound like, so I can use them to judge equipment. I know what recordings made in certain halls should sound like too. But if things don't sound like that, is it the fault of the recording or equipment? How can anyone tell? [We seem to have way more questions than answers.] That's what makes audio fun. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#322
|
|||
|
|||
|
#323
|
|||
|
|||
Mike Rivers wrote:
I've never tried to listen for the effect of jitter since I don't have a way to control it, but I recall some of Bob Katz' writings of half a dozen or so years back. He was introducing what, by today's standards, would be considered massive amounts of FM on a data clock and listening to what happened when he turned it on and off. One effect that he observed was that things that were clearly located in the stereo field were no longer as clearly located. This same sort of thing can be a result of crosstalk, phase shift in the signal chain, maybe even listening position. But when he heard a change that he described in this way by doing nothing but adding jitter to the clock, it's not unreasonable to suggest that this is an effect of jitter. The key words I see a "by today's standards, would be considered massive amounts of FM on a data clock and listening to what happened when he turned it on and off" This grants that today we'd only rarely if at all, experience these levels of jitter. IOW, Katz's experiences of years and years ago are almost entirely irrelevant to our situations, today. |
#324
|
|||
|
|||
|
#326
|
|||
|
|||
|
#328
|
|||
|
|||
|
#329
|
|||
|
|||
Sure, but remember that Edison did plenty of public demonstrations with his cylinder machines, where many folks claimed to be unable to distinguish the live and recorded sound. Those folks were probably shareholders. ;-) Same things go on today. |
#331
|
|||
|
|||
|
#332
|
|||
|
|||
In article znr1119975982k@trad, Mike Rivers wrote:
In article writes: The answer is that if any of the sidebands created by jitter are much below the -80 dB level, people generally can't reliably discern any difference at all. Back to "distortion" again, I see. This is mostly correct, BUT audibility depends a lot on how far the sideband is from the main lobe. You can take the attitude that there is a point below which everything is inaudible no matter how far out it is (which is what Arny does with distortion), but if you do that you have to be damn sure that you picked the a point low enough. Since I keep hearing weird things I cannot explain, I sometimes wonder if the points are picked low enough. Looking at the spectrum sometimes helps us explain things. The overarching irony of this discussion is that as a rule, digital recording systems have far lower jitter than analog recording systems. In analog systems we call jitter stuff like scrape flutter and general flutter and wow. Yes, but those have definite sonic artifacts. The "imaging" thing isn't so well defined. I've never heard a digital system that has so much clock instability as to sound like flutter. Nor have I heard flutter in an analog recorder influence the location of a reproduced sound (but I've never tried). The thing is that the jitter spectrum is usually very different than the modulation domain spectrum of scrape flutter. You can use an old Minicom test set into a spectrum analyzer to see the spectrum of the phase noise of a tape machine and most of what you see is usually one big rotational peak followed by a couple smaller ones. In the case of a 440 most of what you see is at the frequency that the scrape flutter idlers rotate at, or the frequency that the supply idler rotates, or the capstan rotation frequency. One of the neat things about the 440 design is that all of these were carefully picked to be different frequencies that weren't harmonically related. On the ATR-100 there's really only one rotational peak but it's a lot smaller. Anyway, if you're curious about how flutter influences imaging, check out some of Jamie Howarth's demos on the Plangent Technologies website, which I think is www.plangent.com. It's a magic box that removes flutter from old analogue recordings by using residual bias as a pilot reference tone. It's kind of weird hearing what happens when the flutter is removed... things become more solid and defined and there is less of a sense of instruments blending together. When is the last time you saw an analog tape or vinyl-based system as good as 0.01% or less flutter and wow of *any* kind, unweighted? Different issue. Same with scrape flutter, which introduces sidebands in the audible range. The last time I used an ATR-100. Did I mention that I really like the ATR-100? Anyway, I think that jitter and flutter are two sides of the same coin, but the end results are a little different because the spectra are different. And I think eliminating BOTH of them will benefit the world. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#333
|
|||
|
|||
In article znr1119976280k@trad, Mike Rivers wrote:
In article writes: Nobody has done a good study on perceived effects vs. jitter levels at various frequencies, though. I think that a more interesting study would be about the effects of random jitter. Jitter at a fixed frequency is probably caused by something that you can actually do something about. Not sure about random jitter, though. Well, we _do_ know that it's less audible when it's more random, and more audible when it's all at one frequency. We've known that since the efforts to deal with tape flutter in the sixties. That is why folks designing PLLs for high-end DACs spend a lot of time looking at plots and trying to get 60 Hz components and residual clock components out of their clock error spectrum. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#334
|
|||
|
|||
|
#335
|
|||
|
|||
Scott Dorsey wrote:
In article znr1119975982k@trad, Mike Rivers wrote: In article writes: The answer is that if any of the sidebands created by jitter are much below the -80 dB level, people generally can't reliably discern any difference at all. Back to "distortion" again, I see. This is mostly correct, BUT audibility depends a lot on how far the sideband is from the main lobe. Agreed. When the FM sidebands are within the critical band of the carrier, the ear perceives the modulation as roughness. If they are outside the critical band, the ear tends to perceive them as separate (usually aharmonic) tones. You can take the attitude that there is a point below which everything is inaudible no matter how far out it is (which is what Arny does with distortion), but if you do that you have to be damn sure that you picked the a point low enough. Agreed. Since I keep hearing weird things I cannot explain, I sometimes wonder if the points are picked low enough. Looking at the spectrum sometimes helps us explain things. Agreed. The overarching irony of this discussion is that as a rule, digital recording systems have far lower jitter than analog recording systems. In analog systems we call jitter stuff like scrape flutter and general flutter and wow. Yes, but those have definite sonic artifacts. As does the most common or one of the most common forms of jitter in poor systems, which is 60 Hz based in the US and 50 Hz based in most of the rest of the world. The "imaging" thing isn't so well defined. I've never heard a digital system that has so much clock instability as to sound like flutter. Nor have I heard flutter in an analog recorder influence the location of a reproduced sound (but I've never tried). The thing is that the jitter spectrum is usually very different than the modulation domain spectrum of scrape flutter. Agreed. You can use an old Minicom test set into a spectrum analyzer to see the spectrum of the phase noise of a tape machine and most of what you see is usually one big rotational peak followed by a couple smaller ones. Vinyl tends to follow the same pattern. However there are often two rotation frequencies, one of the whole disc, and one related to the rotation of the drive motor. In the case of a 440 most of what you see is at the frequency that the scrape flutter idlers rotate at, or the frequency that the supply idler rotates, or the capstan rotation frequency. One of the neat things about the 440 design is that all of these were carefully picked to be different frequencies that weren't harmonically related. On the ATR-100 there's really only one rotational peak but it's a lot smaller. I regret that I did not have good facilities to look at the jitter artifacts in vinyl tape in the days when I was highly comitted to it. Anyway, if you're curious about how flutter influences imaging, check out some of Jamie Howarth's demos on the Plangent Technologies website, which I think is www.plangent.com. It's a magic box that removes flutter from old analogue recordings by using residual bias as a pilot reference tone. It's kind of weird hearing what happens when the flutter is removed... things become more solid and defined and there is less of a sense of instruments blending together. Here's the web site: http://www.plangentprocesses.com/ Some nice audio samples there. When is the last time you saw an analog tape or vinyl-based system as good as 0.01% or less flutter and wow of *any* kind, unweighted? Different issue. Same with scrape flutter, which introduces sidebands in the audible range. The last time I used an ATR-100. Did I mention that I really like the ATR-100? Anyway, I think that jitter and flutter are two sides of the same coin, but the end results are a little different because the spectra are different. Let's say that they can and often are different, but there's nothing that makes them always be different. And I think eliminating BOTH of them will benefit the world. --scott Agreed. Digital was the practical solution to wow and flutter. Plangent Processes technology is a highly complex but effective after-the-fact solution. |
#336
|
|||
|
|||
In article writes: Anyway, if you're curious about how flutter influences imaging, check out some of Jamie Howarth's demos on the Plangent Technologies website, which I think is www.plangent.com. It's a magic box that removes flutter from old analogue recordings by using residual bias as a pilot reference tone. It's kind of weird hearing what happens when the flutter is removed... things become more solid and defined and there is less of a sense of instruments blending together. Yes, I remember hearing his AES demos, and last year he was pointing this improved imaging out, but I couldn't really tell on the show floor. Anyway, I think that jitter and flutter are two sides of the same coin, but the end results are a little different because the spectra are different. And I think eliminating BOTH of them will benefit the world. I had a chat with one of the engineers at AES about why their Big Ben clock generator got so many rave reviews about making anything that it clocked sound better. The amplitude of the jitter in their spec sheet isn't really that much smaller than other good devices. He said that it had to do with the spectrum (or probably it's really the probability distribution) of the jitter. But he really didn't sound very convincing. -- I'm really Mike Rivers ) However, until the spam goes away or Hell freezes over, lots of IP addresses are blocked from this system. If you e-mail me and it bounces, use your secret decoder ring and reach me he double-m-eleven-double-zero at yahoo |
#337
|
|||
|
|||
|
#338
|
|||
|
|||
Arny Krueger wrote:
I regret that I did not have good facilities to look at the jitter artifacts in vinyl tape in the days when I was highly comitted to it. That's okay, the guys at 3M did, and they wrote a bunch of papers in the AES about it. Send e-mail to Bob Groschen here because I am sure he has some citations. It's actually not a very big deal with vinyl, at least not with a good lathe, because you have a huge amount of mass on the platter to even it out. You still have issues with the main bearing and with the drive mechanism. On a Scully, which has multiple pulleys and belts, I'd bet that a flutter spectrum will show a peak for each one, but I also bet the platter mass makes them all pretty minimal. There are some playback systems, most notably direct drive turntables, that have massive amounts of cogging. This results in one huge peak in the phase noise spectrum. This is a known problem and easily avoided by not using crappy DJ turntables. Agreed. Digital was the practical solution to wow and flutter. Plangent Processes technology is a highly complex but effective after-the-fact solution. Now you have to convince the digital guys to use decent clocking, though. The word is getting around, but it's still not out enough. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#340
|
|||
|
|||
Mike Rivers wrote:
In article writes: Thing is, jitter doesn't change the location where sound comes from. No (so I'm told - I'm NOT speaking from listening experience here) it doesn't change the position. Turning the jitter on isn't going to make the listener point the laser in a different position. The effect, as it's been described, is that it makes a well defined location less well defined. For example, without introducing jitter, the subject pointing at the clarinet ten times might have all of his sightings within, say a 15 degree angular range. Turn on the jitter and he'll still be pointing in pretty much the same direction, but his aim might be spread over a 30 degree range. Average out both sets of measurements and you'll probably come out to pretty close to the same angle, but for any single measurement, the chance of being further off the actual position is greater with jitter. But its not like the image becomes wider, it just becomes more vague. Vagueness is hard to pinpoint. This is why it's difficult to test. ...snip.. Another aspect to consider is the relative spatial impact on harmonics. By that I mean will all the harmonics of said clarinet point to the same point(s) in space. Will (can, do,) the highs rotate about a the locus of the fundamental or randomly jump about or some such. I would expect that spatial variance would be frequency dependent. Are left and right track jitter highly correlated? Later... Ron Capik -- |
#341
|
|||
|
|||
"Ron Capik" wrote in message
Mike Rivers wrote: In article writes: Thing is, jitter doesn't change the location where sound comes from. No (so I'm told - I'm NOT speaking from listening experience here) it doesn't change the position. Turning the jitter on isn't going to make the listener point the laser in a different position. The effect, as it's been described, is that it makes a well defined location less well defined. For example, without introducing jitter, the subject pointing at the clarinet ten times might have all of his sightings within, say a 15 degree angular range. Turn on the jitter and he'll still be pointing in pretty much the same direction, but his aim might be spread over a 30 degree range. Average out both sets of measurements and you'll probably come out to pretty close to the same angle, but for any single measurement, the chance of being further off the actual position is greater with jitter. But its not like the image becomes wider, it just becomes more vague. Vagueness is hard to pinpoint. This is why it's difficult to test. ...snip.. Another aspect to consider is the relative spatial impact on harmonics. By that I mean will all the harmonics of said clarinet point to the same point(s) in space. All the harmonics of a source don't always point to the same point(s) in space, even without jitter. Reflections at least partially localize to the reflective surface, not the source. Reflected sounds may or may not have timbre shifts. |
#342
|
|||
|
|||
Arny Krueger wrote:
"Ron Capik" wrote in message ...snip.. Another aspect to consider is the relative spatial impact on harmonics. By that I mean will all the harmonics of said clarinet point to the same point(s) in space. All the harmonics of a source don't always point to the same point(s) in space, even without jitter. Reflections at least partially localize to the reflective surface, not the source. Reflected sounds may or may not have timbre shifts. Yes, the spatial character of the real world is dynamic and very complex. However, note that I said ~relative~ spatial impact. Thus I'm asking if disproportionate shifting of the spatial distribution, what ever that natural spatial distribution might be, may be taking place. Maybe a bit like chromatic aberration on an undulating screen. Also note that both tape and vinyl have the right and left tracks mechanically linked. Later... Ron Capik -- |
#343
|
|||
|
|||
On Thu, 30 Jun 2005 15:40:12 GMT, Ron Capik
wrote: Arny Krueger wrote: All the harmonics of a source don't always point to the same point(s) in space, even without jitter. Reflections at least partially localize to the reflective surface, not the source. Reflected sounds may or may not have timbre shifts. Yes, the spatial character of the real world is dynamic and very complex. However, note that I said ~relative~ spatial impact. Thus I'm asking if disproportionate shifting of the spatial distribution, what ever that natural spatial distribution might be, may be taking place. Maybe a bit like chromatic aberration on an undulating screen. Also note that both tape and vinyl have the right and left tracks mechanically linked. 've been trying to think about y'all's question without much progress. Maybe the problem could be simplified to just considering the first-arrival sounds (first), and then working on the general case. Very interesting; good fortune, Chris Hornbeck "Betty, is that Jimmy's ring you're wearing?" 'um huh." "Gee, it must be great riding with him. Is he picking you up after school today?" "unh uh" "By the way, where'd you meet him?" |
#344
|
|||
|
|||
Chris Hornbeck wrote:
On Thu, 30 Jun 2005 15:40:12 GMT, Ron Capik wrote: ...snip.. Yes, the spatial character of the real world is dynamic and very complex. However, note that I said ~relative~ spatial impact. Thus I'm asking if disproportionate shifting of the spatial distribution, what ever that natural spatial distribution might be, may be taking place. Maybe a bit like chromatic aberration on an undulating screen. Also note that both tape and vinyl have the right and left tracks mechanically linked. 've been trying to think about y'all's question without much progress. Maybe the problem could be simplified to just considering the first-arrival sounds (first), and then working on the general case. Very interesting; good fortune, Chris Hornbeck For what it's worth, my conjecture would work just fine addressing direct (first arrival) effects. 'twas Arny that added reflective surfaces. In my conjecture said source has an image, as complex or simple an image as you care to envision. Does jitter have a frequency dependent or frequency neutral impact on that image? Later... Ron -- " ...y' get the picture? " "Yes, we see... " |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
loud headphone amp | Pro Audio | |||
loud headphone amp | Pro Audio | |||
Can ears literally bleed from loud noise? | Pro Audio | |||
How loud is loud? | Pro Audio | |||
hearing loss info | Car Audio |