D
Deleted member 24284
I will introduce this subject by offering what I have found so far:
https://drive.google.com/drive/folders/1orfIKUxQK18SwvABKMz07LF0Le8HiHgi?usp=sharing
The file Crown_Mic_Memo_issues_over_22_years.pdf in there is an archive of an informative discussion held over many years on the benefits of PZM mics.
I bought two of these mics new, and I have been testing them both in my home studio https://goo.gl/maps/2C8uA2abQhG3tQ5E6 and in an open field, and also comparing them against the Vanguard v44s gen2.
I will post those test tracks later if there is interest.
I have found that the stereo effect using only interaural time delay by placing the two mics on the floor with capsules 18 - 20cm apart is very good in my very non-reverberant studio. Recording with that mic placement in an open field, as long as it's quiet enough with low wind, yields even better results.
The mic capsule spacing in the ORTF style of microphone placement is usually 18cm. Experimentally, I have found 20cm to be just right when I put the two mics on the floor, both pointing the same direction. The PZM mics are omnidirectional, so my mic placement does not enhance the stereo separation like ORTF does when it aims the two cardioid mics in different directions, and yet, the stereo effect is still convincing and sounds natural.
The mic placement used in https://en.wikipedia.org/wiki/The_Trinity_Session is not the same, as it uses directional mics placed very close together on the floor, so it solely relies on inter-channel amplitude difference to establish the stereo effect like a pan pot on a mixer, and not interaural time delay in what I'm doing. What it does have in common with what I described above of putting two PZMs 20cm apart on the floor is the rejection of floor bounce, hence it picks up only long time delay echoes from the church's walls and ceiling, and avoids short time delay echoes that come from sound bouncing off the floor.
Are these mics used still in the recording industry or small scale non-commercial efforts like mine?
Is there interest from folks in my area in Orange County in southern California in doing further tests with these mics in configurations different from what I have done above?
Is there interest in doing recordings of small ensembles using my mic placement or some other placement using the PZM mics, either in my home studio or in an open field?
https://drive.google.com/drive/folders/1orfIKUxQK18SwvABKMz07LF0Le8HiHgi?usp=sharing
The file Crown_Mic_Memo_issues_over_22_years.pdf in there is an archive of an informative discussion held over many years on the benefits of PZM mics.
I bought two of these mics new, and I have been testing them both in my home studio https://goo.gl/maps/2C8uA2abQhG3tQ5E6 and in an open field, and also comparing them against the Vanguard v44s gen2.
I will post those test tracks later if there is interest.
I have found that the stereo effect using only interaural time delay by placing the two mics on the floor with capsules 18 - 20cm apart is very good in my very non-reverberant studio. Recording with that mic placement in an open field, as long as it's quiet enough with low wind, yields even better results.
The mic capsule spacing in the ORTF style of microphone placement is usually 18cm. Experimentally, I have found 20cm to be just right when I put the two mics on the floor, both pointing the same direction. The PZM mics are omnidirectional, so my mic placement does not enhance the stereo separation like ORTF does when it aims the two cardioid mics in different directions, and yet, the stereo effect is still convincing and sounds natural.
The mic placement used in https://en.wikipedia.org/wiki/The_Trinity_Session is not the same, as it uses directional mics placed very close together on the floor, so it solely relies on inter-channel amplitude difference to establish the stereo effect like a pan pot on a mixer, and not interaural time delay in what I'm doing. What it does have in common with what I described above of putting two PZMs 20cm apart on the floor is the rejection of floor bounce, hence it picks up only long time delay echoes from the church's walls and ceiling, and avoids short time delay echoes that come from sound bouncing off the floor.
Are these mics used still in the recording industry or small scale non-commercial efforts like mine?
Is there interest from folks in my area in Orange County in southern California in doing further tests with these mics in configurations different from what I have done above?
Is there interest in doing recordings of small ensembles using my mic placement or some other placement using the PZM mics, either in my home studio or in an open field?
I think Frank Zappa said they solved a recording issue for him, so they must have a place in the toolbox when nothing else seems to work.
D
Deleted member 24284
I did a test in a park's baseball field this morning, with a drum moving in a semicircle around the pair of microphones placed on the ground. The angular spacing is only approximate, but the drum to mic distance is mostly constant as I used a string to maintain that.
https://drive.google.com/drive/folders/1n8Y2mNsK1frtylreh8S6uGjk-8Fblry1?usp=sharing
The mic capsules are 19cm apart.
A listening test should reveal that the 18cm recommended capsule spacing for ORTF is correct, as the stereo image moves off center too quickly in my test using 19cm when I listen via headphones.
I have photos and a wave file in the folder above. The wave file has the drum starting straight ahead then moving to one side, and it starts straight ahead again then moves to the other side. The angular span is 90 degrees each side.
My hope is to get a chance to record a string quartet or a small jazz ensemble in that open field environment and the result should be good.
The grass was wet this morning as there was an overnight rain, hence there is minimal ground bounce, even though those mics placed on the ground like that should reject most ground bounce anyway even on a concrete floor. The EXIF data on the photos show you where it is. The wire above each foam piece covering the mic is to keep it from blowing away in the wind.
https://drive.google.com/drive/folders/1n8Y2mNsK1frtylreh8S6uGjk-8Fblry1?usp=sharing
The mic capsules are 19cm apart.
A listening test should reveal that the 18cm recommended capsule spacing for ORTF is correct, as the stereo image moves off center too quickly in my test using 19cm when I listen via headphones.
I have photos and a wave file in the folder above. The wave file has the drum starting straight ahead then moving to one side, and it starts straight ahead again then moves to the other side. The angular span is 90 degrees each side.
My hope is to get a chance to record a string quartet or a small jazz ensemble in that open field environment and the result should be good.
The grass was wet this morning as there was an overnight rain, hence there is minimal ground bounce, even though those mics placed on the ground like that should reject most ground bounce anyway even on a concrete floor. The EXIF data on the photos show you where it is. The wire above each foam piece covering the mic is to keep it from blowing away in the wind.
Interesting recording, thanks for that. Localization with closed spaced omnis, especially in the far field, would be expected to be almost entirely generated by intraural delay. My stereo listening position (coaxial drivers) is back to a window. With the blinds open the recording's imaging falls apart, the drum sounding almost stationary throughout. Closing the blinds restore some element of proper left/right location.
If I'm not mistaken aren't cardiods typically used for ORTF and wide separation with omnis? MA recordings for a commercial example. An exception is when the mic array itself provides acoustic shadowing, a good example Crown's PZM based SASS system.
If I'm not mistaken aren't cardiods typically used for ORTF and wide separation with omnis? MA recordings for a commercial example. An exception is when the mic array itself provides acoustic shadowing, a good example Crown's PZM based SASS system.
Attachments
D
Deleted member 24284
Both the ORTF technique using cardioid mics and the Crown SASS stereo mic make use of the amplitude difference between channels in addition to arrival time difference to establish the stereo image. My recording only uses the arrival time difference, as the PZM mics are omnis.
I hope everyone has heard it on headphones and agree that the stereo image is clear there.
If it's not as clear when listening to it via speakers, that could be due to reflections of the emissions from the speaker from surrounding objects messing up the sound, or the speakers are not driven by the signal at the same time between the two channels. If the drum only goes one way and not the other way, then I think the arrival time difference is mostly at fault, but that's only my guess.
I think using solely arrival time difference to establish stereo image has probably been thought about for decades and even tried many times, but it didn't reach commercial widespread use since the effect is only guaranteed on headphones and is quite hard to get it right on speakers. But nowadays more and more people solely listen on headphones so maybe it can again be a usable technique with commercial viability.
I hope everyone has heard it on headphones and agree that the stereo image is clear there.
If it's not as clear when listening to it via speakers, that could be due to reflections of the emissions from the speaker from surrounding objects messing up the sound, or the speakers are not driven by the signal at the same time between the two channels. If the drum only goes one way and not the other way, then I think the arrival time difference is mostly at fault, but that's only my guess.
I think using solely arrival time difference to establish stereo image has probably been thought about for decades and even tried many times, but it didn't reach commercial widespread use since the effect is only guaranteed on headphones and is quite hard to get it right on speakers. But nowadays more and more people solely listen on headphones so maybe it can again be a usable technique with commercial viability.
Headphones worked very clearly here. It's the first test done to confirm the localization performance over my speakers was due to the highly reflective surface behind my listening position. The amplitude difference between closely spaced omnis is minimal for any source in the far field so I agree that the strong room reflections probably played havoc with an image created primarily from intraural delay. But you're already aware of it and the potential trade offs. Still makes a useful demo to root out room issues.
While ORTF does leverage spacing amplitude differences it's greatly augmented by the falling off axis amplitude of a cardiod, not available with omnis. The Crown SASS gets around this by using omni elements in a housing that modifies their directivity as frequency rises to partially emulate a human head as per spec sheet below.
It will be interesting to see where this experiment goes.
D
Deleted member 24284
The file Crown_Mic_Memo_issues_over_22_years.pdf in https://drive.google.com/drive/folders/1orfIKUxQK18SwvABKMz07LF0Le8HiHgi?usp=sharing that I mentioned in the first post has a lot of descriptions of usages of PZMs to mic various instruments, and these read like serious and detailed reports of what a knowledgeable recording engineer has tried instead of casual shooting-the-breeze types of discussions. But upon close reading, no one in there has described the simple usage I have tried here, of not using any amplitude difference between channels. One so-so recommendation is to record a choir that insists on using music stands: put one PZM on each music stand, each stand spaced 1 meter apart approximately. They never considered the different arrival times of the same singer's voice at several mics and how that spoils the stereo image.
A rental shop in Hollywood north of Los Angeles has a Crown P-SASS for rent. I have offered to buy it from them, but they said its place in their vintage mic cabinet is already reserved and they will not sell it.
In my highly damped home studio / listening room the JansZen A8 can only show the drum going to the left and not to the right, but curiously the center position always shows correctly in the middle, so maybe there is more than one thing wrong here, since if there is a delay of the left speaker's signal compared to the right, then the whole image will shift. I am feeding it with SPDIF now, and the left speaker's SPDIF input is driven by the right speaker's dedicated SPDIF output per manufacturer's recommendation, so maybe that introduces a delay which is clearly unacceptable, although that can be easily fixed by modifying the software running on its integrated DSP crossover and class-D amps. I will next try an analog feed going directly to each speaker to see if that fixes the problem.
I have been in contact with local big band jazz types of orchestras to let me record a brief segment of their rehearsals, especially if it is held in a very reverberant space like a high school gym, so that I can use a 4 channel digital recorder to capture the sound from two PZMs and also the one Vanguard v44s gen2 positioned directly above it and compare the results. The v44s will reject sound from the rear, but it'll catch the floor bounce. The PZMs will reject the floor bounce but will pick up the unwanted sound from the rear unless I put in a foam right behind it to block that.
A rental shop in Hollywood north of Los Angeles has a Crown P-SASS for rent. I have offered to buy it from them, but they said its place in their vintage mic cabinet is already reserved and they will not sell it.
In my highly damped home studio / listening room the JansZen A8 can only show the drum going to the left and not to the right, but curiously the center position always shows correctly in the middle, so maybe there is more than one thing wrong here, since if there is a delay of the left speaker's signal compared to the right, then the whole image will shift. I am feeding it with SPDIF now, and the left speaker's SPDIF input is driven by the right speaker's dedicated SPDIF output per manufacturer's recommendation, so maybe that introduces a delay which is clearly unacceptable, although that can be easily fixed by modifying the software running on its integrated DSP crossover and class-D amps. I will next try an analog feed going directly to each speaker to see if that fixes the problem.
I have been in contact with local big band jazz types of orchestras to let me record a brief segment of their rehearsals, especially if it is held in a very reverberant space like a high school gym, so that I can use a 4 channel digital recorder to capture the sound from two PZMs and also the one Vanguard v44s gen2 positioned directly above it and compare the results. The v44s will reject sound from the rear, but it'll catch the floor bounce. The PZMs will reject the floor bounce but will pick up the unwanted sound from the rear unless I put in a foam right behind it to block that.
D
Deleted member 24284
I do not have a new test to share this week, as I am still debugging my speakers' refusal to symmetrically move the drum image to the right in the same manner as it readily move the drum image to the left. It does move to the right, but not nearly enough, and it gets stuck about 1/4 of the way there and does not go any further. I have tried an analog feed into each speaker, and that does not help. I have reconfigured the Auralex foam, and that does not help.
I hope the audience can suggest additional steps to debug this problem.
I can think of taking these three steps next:
Check the output amplitude from each speaker.
Aim the tilt of the speaker by white noise and mic on stand mounted at listening position as these speakers' electrostatic tweeters have a very narrow vertical dispersion.
Aim the rotation of the speaker by mirror placed on the speaker cabinet's face and a flashlight shining onto it held next to my ears, as I look for the reflection of the light to return to my eyes.
https://photos.app.goo.gl/omnh61j1WUpPn1UA7
I hope the audience can suggest additional steps to debug this problem.
I can think of taking these three steps next:
Check the output amplitude from each speaker.
Aim the tilt of the speaker by white noise and mic on stand mounted at listening position as these speakers' electrostatic tweeters have a very narrow vertical dispersion.
Aim the rotation of the speaker by mirror placed on the speaker cabinet's face and a flashlight shining onto it held next to my ears, as I look for the reflection of the light to return to my eyes.
https://photos.app.goo.gl/omnh61j1WUpPn1UA7
D
Deleted member 24284
Thanks for showing us that the amplitude between the two channels in the test clip are very close.
But I was referring to my own speakers having different acoustic output when driven by the same signal. I can go back to using the analog feed into them and drive that with a preamp that has a balance control.
I can also bring the speakers outdoors and play the same clip. If that still doesn't work, then it'll be an interesting thing to report for sure.
But I was referring to my own speakers having different acoustic output when driven by the same signal. I can go back to using the analog feed into them and drive that with a preamp that has a balance control.
I can also bring the speakers outdoors and play the same clip. If that still doesn't work, then it'll be an interesting thing to report for sure.
D
Deleted member 24284
I can hear the drum going both ways on my speakers now after I aimed them better.
The horizontal sweet spot is very small, around 1cm.
I think the mics placed 19cm apart instead of the 18cm recommended for ORTF also made this test more demanding than it should.
In commercial recordings, especially the multi-mic technique used for full-sized symphony orchestras, they rely only on amplitude difference for establishing the stereo image, with incorrect and multiple (too many) arrival times spoiling the image.
Gustavo Dudamel, Music & Artistic Director of the Los Angeles Philharmonic, Walt and Lilly Disney Chair, recently noted that his most recent effort leading an orchestra in a recording used over 40 mics. Recording this way lets you fix a lot of mistakes in post-production, although the stereo image is spoiled irreparably.
Michael Troke, of Micworks Inc. In Fountain Valley, CA, told me five years ago that the best recording he has ever heard was recorded outdoors at a quiet location deep in a forest in Santa Barbara using only two mics. Sadly he had to close his shop recently due to poor health.
The horizontal sweet spot is very small, around 1cm.
I think the mics placed 19cm apart instead of the 18cm recommended for ORTF also made this test more demanding than it should.
In commercial recordings, especially the multi-mic technique used for full-sized symphony orchestras, they rely only on amplitude difference for establishing the stereo image, with incorrect and multiple (too many) arrival times spoiling the image.
Gustavo Dudamel, Music & Artistic Director of the Los Angeles Philharmonic, Walt and Lilly Disney Chair, recently noted that his most recent effort leading an orchestra in a recording used over 40 mics. Recording this way lets you fix a lot of mistakes in post-production, although the stereo image is spoiled irreparably.
Michael Troke, of Micworks Inc. In Fountain Valley, CA, told me five years ago that the best recording he has ever heard was recorded outdoors at a quiet location deep in a forest in Santa Barbara using only two mics. Sadly he had to close his shop recently due to poor health.
I agree regarding the Crown PZMs being great mics for recording large acoustical sources. Their behavior mimics the human hearing analog closer than most other common mic types. Heard and made some great recordings with them, specifically large jazz orchestral material and choirs.
The Jecklin disc is another great setup. Even a cheap pair of Rode M5s in this configuration blows away many other higher priced options. It sounds better to me than the Neumann dummy head or most other ORTF schemes.
The Jecklin disc is another great setup. Even a cheap pair of Rode M5s in this configuration blows away many other higher priced options. It sounds better to me than the Neumann dummy head or most other ORTF schemes.