Why? Sit in front of 2 speakers with a laptop and some software. Pan a mono signal using intensity only, delay only or both combined. You'll quickly get a feel for what each of those does and its subjective effect. Not hard to do and easy to interpret.
I'd love to Mark, but I'm a long way from Oregon now, and don't have the equipment.
If you can borrow a spectrum analyzer, even one of the USB based models like a Picoscope, you can see a lot. I would recommend it, if you are looking for the lowest noise layout and filtering.
It is all from long memory so if you can find his work, please share a link. But I remember it was a generalized solution so for all F, steady state indeed. It is quite an extensive derivation.
About steady state or not, I see your point here, but wonder to what extent it is relevant. For clicks, tests have been done and ability to discriminate is about three times worse than for steady state (either noise or periodic).
See below
The math does not support the generalized statement. At any one frequency, it is possible to find a ratio which will correctly support both IID and ITD (within a frequency range). However, what it cannot do is support one pan ratio for a wide range of frequencies.
As such, the pan ratio for a specific location will spread other frequencies to either side of the sweet frequency.
Griesinger even did a three speaker pan law test, which supports my claim. It was either fig 5 or 6 in one of his papers that demonstrated the deviation from the ideal pan law. Of significance is the fact that there were different curves for different frequencies. Haven't read the paper in five years or so, but remember the graph quite vividly.
John
Yeah, OK. That's interesting and relevant, for sure. But again, not hard to test with 2 speakers and software. If you find this to be true, the math would be great to see.
Well... I'm not sure.
I guess I was just thinking that the amplitude ratio would be more of a constant, reproducible effect from one typical stereo playback system to the next, whereas any delay effects would interact with / be altered somewhat by the distance between the playback speakers, wouldn't they?I suppose I'll just have to try your suggestion, and mess around a bit with different pans & delays. I did an experiment similar to this once, but conceptually it was sort of coming from the other direction, and it may be skewing my understanding of the current discussion a bit (this is a copy & paste from an old post, please forgive my laziness):
----------------
Many many years ago (like early 90s?) the seating arrangement in my tiny apartment was such that I couldn't get to a nice sweet spot between the stereo speakers, so had to compensate with the balance control. At some point I acquired a little Sony Dolby Surround decoder box, which had among its features for some reason the ability to set it up as a simple full-bandwidth (well, 16/44 anyway) stereo delay. Cool, I thought - now I can properly address my seating offset. So I set about adjusting the delay on the near channel with a mono source until the direct sound from both speakers was once again correlated at my noggin. But now something else sounded wrong! I quickly realized that the reflected sound in the room was now decorrelated, which sounded noticeably not-right in its own unique way. Worse yet, this new problem of course existed at any listening location. (sigh) So back to balance control for me, and back to the closet for the Sony gizmo.
Since then I've never understood all the emphasis on exact interchannel delay settings with surround systems. I don't use it myself - it seems to solve one problem only to create a worse one.
----------------
- Jim
OK Jim, I see you've done some of this. Thanks. In a couple of my old systems, I had the easy ability to add delay to one channel or both. I did a bit of that to center the phantom image, along with a small amount of amplitude adjustment. A delay and amplitude balance control. It was the combination of the two that gave me the best results, the best center image, rather than just one or the other. Perhaps I didn't need enough delay to muck up the reflections - of which there were not a lot. It wasn't much of a shift to center things.
I have thread somewhere about Fixing the Phantom image that has nice details about discussions about what happens with 2 speakers and two ears. Decorrelation is important there.
As to the space between the speakers making a difference, I don't see that. At least in anything near a normal set up. A delay in one channel is going to move the sound away from that channel by a certain amount that is dependent on your head and ears, not the speakers themselves. How much of a playing field you may have could depend on the angle of the speakers, but I don't think that X delay would simulate a different angle just because the speakers are placed differently. It's an HRTF effect, no a speaker effect. Could be wrong, but I don't think it depends on the speaker angle.
I have thread somewhere about Fixing the Phantom image that has nice details about discussions about what happens with 2 speakers and two ears. Decorrelation is important there.
As to the space between the speakers making a difference, I don't see that. At least in anything near a normal set up. A delay in one channel is going to move the sound away from that channel by a certain amount that is dependent on your head and ears, not the speakers themselves. How much of a playing field you may have could depend on the angle of the speakers, but I don't think that X delay would simulate a different angle just because the speakers are placed differently. It's an HRTF effect, no a speaker effect. Could be wrong, but I don't think it depends on the speaker angle.
Your question is indicative of a very good understanding of the problem despite your statement.
When the images are set in the studio, it is done for those speakers at that distance and that angle. How in the world can that be correct for every setup out there? It can't.
John
Course, I'm probably wrong....any math that requires I take my shoes off for is probably over my head.
Last edited:
Yeah, I'm honestly hung up on the same thing that Jim seems to be, in contrast with acoustic path length to my ear (even if we go with directly incident sound), ns-µs level delays due to propagation seem inconsequential. Heck, even the wave launch from the respective speaker (i.e. their acoustic centers) move with frequency. How that doesn't completely swamp these much smaller differences, I'm unsure.
That's certainly not to say that they're directly correlatable or that I'm not completely missing the point, but it's a head scratcher.
That's certainly not to say that they're directly correlatable or that I'm not completely missing the point, but it's a head scratcher.
Ah! This illustrates a key difference between your results and mine - a matter of degree. In my tests, I was using quite a bit of delay in an attempt to correct a significant seating offset - we're talking close to 10 feet, so roughly that many milliseconds.
To "hear" this in another way: If we were to use that much delay as a "reel flange" effect (mixing the delayed and undelayed sounds electrically instead of acoustically), it would impart a distinct comb-filtered "pitch" to the composite sound with a "fundamental" frequency of the inverse of the delay time - let's say around 110 Hz in this case. That would be the sound of a lot of decorrelation, and would've had a clearly audible effect on even the already chaotic reflected room sound. It would certainly explain what I experienced all those years ago. In my case, I probably screwed up the reflected sound so bad that it trumped any positive effects of correcting the direct arrival times. Your adjustments are likely much less severe, and so the benefits are the other way round. I need to rethink all this, and run some updated tests with my current living room menagerie.
And of course my verbose sidebar into playback setup is similar but different to the original discussion regarding panning during recording... Sorry about that.
Oh! I'd started reading that but forgot to subscribe. Thanks for the reminder.
I suppose we'd need to start with: What exactly is our objective when we attempt an artificial pan in a recording? If it's to create as natural and realistic a soundstage as possible, then what exactly is involved there, beyond shifting the amplitude ratios between the channels, or even applying some sort of delay?
If we consider a minimalist two-mic live recording, there are not only amplitude and delay differences between the direct sounds of the different instruments, but also the different, complex ways in which the various instrument locations are interacting with the reverberant space they share (back to reflected sound again). In a studio recording with a synthesized stereo field of panned mono sounds, we could surely synthesize these reverberant field locations as well. As Bonsai alluded, there's gotta be plugins for that by now, right? Just not on any of our old favorite records, unfortunately.
- Jim
Last edited:
Blumlein’s discussed work
Sound-transmission, sound-recording, and sound-reproducing system
Main’s filter design
https://www.hpe.ee.ethz.ch/uploads/tx_ethpublications/04909406_01.pdf
Stray Capacitance of Transformers
https://www.hpe.ee.ethz.ch/uploads/tx_ethpublications/biela_IEEETrans_ReviewStrayCap_03.pdf
George
Sound-transmission, sound-recording, and sound-reproducing system
Main’s filter design
https://www.hpe.ee.ethz.ch/uploads/tx_ethpublications/04909406_01.pdf
Stray Capacitance of Transformers
https://www.hpe.ee.ethz.ch/uploads/tx_ethpublications/biela_IEEETrans_ReviewStrayCap_03.pdf
George
Yes, for me it was a delay in the order of 0.5-2 milliseconds, at the most. That's all that was needed to shift the center. You might have been at 8 or 9 ms.
In my former listening room, a lava cave 70' long, I was lucky enough to have over 30' behind my semi-open baffle speakers. The sense of depth in some recordings was astonishing! Almost anything recorded in a large venue sounded big and deep. Even mono recordings could have great depth if recorded in a large space. But most studio recordings stayed flat, right between the two speakers, sometimes going back a foot or two. Some recordings were a combination of the two - think Sinatra right between the speakers with the orchestra deep behind him.
I have heard enough big, high quality systems to understand how much information is (usually) hidden in recordings. Hearing that much depth from my simple system was a shocker to me. Some of the artificial ambience recording tricks worked OK, but nothing beat the sound of a real space. Intensity or delay panning won't get you that, but it can be very effective for other things.
Obviously amplitude and delay panning, or a combination of the two is not going to create a sense of space. But it is handy for placing sounds across the stage where you want them.
In my former listening room, a lava cave 70' long, I was lucky enough to have over 30' behind my semi-open baffle speakers. The sense of depth in some recordings was astonishing! Almost anything recorded in a large venue sounded big and deep. Even mono recordings could have great depth if recorded in a large space. But most studio recordings stayed flat, right between the two speakers, sometimes going back a foot or two. Some recordings were a combination of the two - think Sinatra right between the speakers with the orchestra deep behind him.
I have heard enough big, high quality systems to understand how much information is (usually) hidden in recordings. Hearing that much depth from my simple system was a shocker to me. Some of the artificial ambience recording tricks worked OK, but nothing beat the sound of a real space. Intensity or delay panning won't get you that, but it can be very effective for other things.
Another milestone in the development of the autoranger!
Finalised and tested the cal procedure.
Especially calibrating the freq response for 0.1dB flatness out to 100kHz is a challenge if you don't have a fully equipped lab
I now have a procedure that needs only an AC DMM with reasonable performance and a signal generator that can output 10kHz and 100kHz, preferably up to 10V or more but that's not critical.
The procedure directs you to do some measurements and enter values in a small spreadsheet and the spreadsheet then tells you what to adjust (cap trimmer) for which DMM indication.
For -20dB and -40dB settting, in both SE and BAL mode.
It's done faster than described!
Jan
... and I did a write-up this afternoon, it's here.
The power supply I developed for the AR is really a nifty little stand-alone unit and is described here.
I will add more detail over time.
Jan
- Status
- Not open for further replies.