Scott,
I did say MOST.
The U47 has an almost impossible 1" Capsule.
1/4" Recording Mikes are more common, even 1/2" are not that common nowadays.
Sure, at the physics limits that also means the HF goes to hell very early, at a guess 10Khz or less...
Ciao T
I did say MOST.
The U47 has an almost impossible 1" Capsule.
1/4" Recording Mikes are more common, even 1/2" are not that common nowadays.
Sure, at the physics limits that also means the HF goes to hell very early, at a guess 10Khz or less...
Ciao T
I think there's supposed to be a point in there, but it's obscured enough by the underbrush that I don't see it.
Time to get out your drip torch.
An externally hosted image should be here but it was not working when we last tested it.
But don't be surprised if you still don't see anything once the underbrush is cleared.
se
There are no real rooms, with real musicians in them (people are noisy) that can't be fit into 16 bits. Modern recording practice of course requires a lot of headroom around this. And if 0VU is defined as the *very* loud 85dB SPL at listening position that is used in mastering studios, and that peaks reach maybe +20dBVU, there are very very few listening rooms that don't fit into 16 bits.
As a separate issue, our hearing covers a wide dynamic range, but not all at the same time. We hear a smaller range that slides up and down dynamically.
Thanks,
Chris
As a separate issue, our hearing covers a wide dynamic range, but not all at the same time. We hear a smaller range that slides up and down dynamically.
Thanks,
Chris
thanks Ed, I hadn't read that part closely before
My Fasil, Zwicker Psychoacousitcs (3rd ed, 2006) give a summary of critical band measurement techniques – they seem happy with the 24 Bark scale
Another “factoid” would be their ~640 jnd pitch resolution steps from FM modulation measurements - might be another argument against "infinite" human hearing resolution, how Digital Audio "finite" frequency resolution (a flawed interpretation in of itself) could possibly give good results
And pg 155-157 gives a result that should be a strong argument against theories of PIM being particularly audible, offensive, to the point we should follow Otala’s flawed prescription of “flat” loop gain amplifier design
They show one critical band measurement technique using AM vs FM modulation, it is described with graphs showing subjects more sensitive to “AM” modulation for small frequency differences of <64 Hz @ 1 kHz center and “AM”, “FM’ jnd thresholds becoming the same for larger IMD frequency differences
Otala’s prescription of flat loop gain to reduce PIM is about as dead as an intellectual concept can get
– no support for greater sensitivity to “FM” IMD products in the psychoacoustic literature – we see the opposite
– clear demonstrations in theory, simulation and measurement (Cordell) that flat loop gain isn’t a requirement for low PIM in amplifiers
John seems incapable of accepting this information – it doesn’t invalidate his preference for “flat loop gain” designs – but it really should cause him to consider that his continual asserting, in Otala’s original formulation, the cause “must be” PIM isn’t supportable
My Fasil, Zwicker Psychoacousitcs (3rd ed, 2006) give a summary of critical band measurement techniques – they seem happy with the 24 Bark scale
Another “factoid” would be their ~640 jnd pitch resolution steps from FM modulation measurements - might be another argument against "infinite" human hearing resolution, how Digital Audio "finite" frequency resolution (a flawed interpretation in of itself) could possibly give good results
And pg 155-157 gives a result that should be a strong argument against theories of PIM being particularly audible, offensive, to the point we should follow Otala’s flawed prescription of “flat” loop gain amplifier design
They show one critical band measurement technique using AM vs FM modulation, it is described with graphs showing subjects more sensitive to “AM” modulation for small frequency differences of <64 Hz @ 1 kHz center and “AM”, “FM’ jnd thresholds becoming the same for larger IMD frequency differences
Otala’s prescription of flat loop gain to reduce PIM is about as dead as an intellectual concept can get
– no support for greater sensitivity to “FM” IMD products in the psychoacoustic literature – we see the opposite
– clear demonstrations in theory, simulation and measurement (Cordell) that flat loop gain isn’t a requirement for low PIM in amplifiers
John seems incapable of accepting this information – it doesn’t invalidate his preference for “flat loop gain” designs – but it really should cause him to consider that his continual asserting, in Otala’s original formulation, the cause “must be” PIM isn’t supportable
Yes, I am incapable of agreeing with you. I learn from my experience as much as I learn from math. IF I could make a high feedback, low open loop bandwidth design sound as successful as a lower feedback, medium open loop design, I would use it, almost exclusively. AND if I could make a lower feedback, medium open loop design sound as good to ME and my learned associates as an open loop design, I would change the Blowtorch back to the JC-80 topology. (I designed both)
It is a matter of accepting what works and describing what seems to make it that way.
It is a matter of accepting what works and describing what seems to make it that way.
Otala's PIM /= explaination of "flat gain" amps "sound better"
but you keep making the technical claim of causation, in terms susceptible to EE analysis, using circuit analysis, feedback, signal theory, using Otala's formulation of PIM, flat open loop gain
this is testable, was thoroughly examined in theory (JAES), designs, measurements in Otala's own units (Cordell)
if you have satisfied yourself that your experiences, buddies listening comments, means flat open loop gain "does something special" that is intellectually possible
claiming The Cause is PIM doesn't agree with any of the later results, mathematical analysis, no supporting DBT psychoacoustic evidence has been cited
the audio engineering community has examined Otala's ideas, some issues were pointed up that informed designers now address - but his formulation of requiring flat loop gain to reduce PIM is simply incorrect - there are other options, high loop gain feedback amplifiers can be designed with low PIM
you can still assert that you, your friends, reviewers hear something different between the design choices but the accepted ways EE, Psychoacoustics “construct” knowledge Otala's PIM, flat gain formulation is now understood to have been based on incorrect/incomplete models, explains no accepted “evidence” that PIM is a audible problem requiring any different consideration than any other IMD
if you want to assert that there is a frontier out there beyond current EE signal theory, analysis capabilities I’m sure many will be happy for the new intellectual challenges
but when claims are made that are testable well within those understandings, theoretical framework you have to accept the results, or present your new formulation, with supporting analysis, evidence
but you keep making the technical claim of causation, in terms susceptible to EE analysis, using circuit analysis, feedback, signal theory, using Otala's formulation of PIM, flat open loop gain
this is testable, was thoroughly examined in theory (JAES), designs, measurements in Otala's own units (Cordell)
if you have satisfied yourself that your experiences, buddies listening comments, means flat open loop gain "does something special" that is intellectually possible
claiming The Cause is PIM doesn't agree with any of the later results, mathematical analysis, no supporting DBT psychoacoustic evidence has been cited
the audio engineering community has examined Otala's ideas, some issues were pointed up that informed designers now address - but his formulation of requiring flat loop gain to reduce PIM is simply incorrect - there are other options, high loop gain feedback amplifiers can be designed with low PIM
you can still assert that you, your friends, reviewers hear something different between the design choices but the accepted ways EE, Psychoacoustics “construct” knowledge Otala's PIM, flat gain formulation is now understood to have been based on incorrect/incomplete models, explains no accepted “evidence” that PIM is a audible problem requiring any different consideration than any other IMD
if you want to assert that there is a frontier out there beyond current EE signal theory, analysis capabilities I’m sure many will be happy for the new intellectual challenges
but when claims are made that are testable well within those understandings, theoretical framework you have to accept the results, or present your new formulation, with supporting analysis, evidence
Last edited:
by all means lets see new measurements
Ron Quan didn't present measurements of anything more recent than a TL061 - not even the ~30 yr old industry standards TL074, NE5534
understandable since he needed actual differences in his numbers to compare - but not too useful to today’s designers
If he has new measurements, please share - I personally doubt there will be anything above the noise floor for recent LME4562, ADA4898, OPA827 class op amps
do we need to figure out how to get Bob Cordell's error corrected Mosfet amp together with Ron's hardware? maybe Ron can quickly convert Bob's sub 100 ps PIM measure to his test conditions?
Ron Quan didn't present measurements of anything more recent than a TL061 - not even the ~30 yr old industry standards TL074, NE5534
understandable since he needed actual differences in his numbers to compare - but not too useful to today’s designers
If he has new measurements, please share - I personally doubt there will be anything above the noise floor for recent LME4562, ADA4898, OPA827 class op amps
do we need to figure out how to get Bob Cordell's error corrected Mosfet amp together with Ron's hardware? maybe Ron can quickly convert Bob's sub 100 ps PIM measure to his test conditions?
Last edited:
I was just pointing out specs on a real product, give or take a little they are right on with your observation. I would disagree, Pearl for one is well above 1/4" and their recordings are first rate to say the least (Waterlily (Kavi Alexander's lable), etc). I'm not an expert in recording I would refer you to Martin Kantola of Nordic Labs and consultant to Pearl, 1" capsules, NOS VF14's, etc. nothing else will do.
http://www.panphonic.com/info.html
EDIT - BTW the 1/2" capsules we played with in Austin were 12dBA self noise. Circuit sophistry wins all the time.
Last edited:
VERY well-stated, jcx.
PIM was virtually unmeasurable in my MOSFET power amplifier with error correction, even with my specially-designed coherent IM analyzer, even though that amplifier had very low open-loop bandwidth due to its very high open-loop gain at low frequencies.
Ron Quan is a nice guy, and I had some good conversations with him last year at AES, but his paper did not add anything to the PIM body of knowledge, and did not provide any evidence to counter the previous work that showed that high open-loop bandwidth does not exacerbate PIM.
John brings this beaten, bloodied dead horse of PIM back up about every six months, hoping that the people here do not remember the last time that the open-loop bandwidth claims for PIM dependence were thoroughly dicussed.
Cheers,
Bob
It puzzles me that Tom Lee of Stanford endorsed this. My only conclusion is that main stream EE's are totally detatched from the audio community. The absence of data from even one modern amplifier is baffling.
One possible conclusion is that there is an agenda afoot. The opportunity to show that yes circa 1969 amplifiers had problems but now this issue is irrelevant is just to obvious. This would require a degree of cynicism.
Last edited:
Scott, I believe the Rode NT1-A is the quietest condenser on the market ATM.
RDE Microphones - NT1-A
WRT ribbons noise, you should be able to calculate it, but obviously the
step up transformers DC R's are the unknown factor - and probably the
ultimate bottle neck.
Hi,
I played with those mikes a few years back. Quiet, nice midrange, poor treble. IT's all a tradeoff. I have been thinking for a while about "2-Way" microphone arrangements to overcome this, with a big quiet 1" and a 1/4" for the top... Then again, I'm also thinking about microphone arrangements for stereo recordings that go past the common configs for stereo space rendering. Very early trials are promising, but I really lack time to work on this more...
The ribbon itself is << 1 Ohm... The transformers often need as much as 1:300 Stepup to manage to get usable levels from Ribbon Mikes and thus accumulate loads of DCR and usually have problems with frequency response.
Ciao T
I played with those mikes a few years back. Quiet, nice midrange, poor treble. IT's all a tradeoff. I have been thinking for a while about "2-Way" microphone arrangements to overcome this, with a big quiet 1" and a 1/4" for the top... Then again, I'm also thinking about microphone arrangements for stereo recordings that go past the common configs for stereo space rendering. Very early trials are promising, but I really lack time to work on this more...
The ribbon itself is << 1 Ohm... The transformers often need as much as 1:300 Stepup to manage to get usable levels from Ribbon Mikes and thus accumulate loads of DCR and usually have problems with frequency response.
Ciao T
Yes I have one that's where I got that number, I didn't mention it because it usually is not considered first rate as a performance recording mike, for the reasons Thorsten mentions. The 1" capsule has problems with the highs. A typical mod replaces it with a good Neumann clone which probably brings up the noise a little. Pearl uses square capsules in some cases, even though I don't typically like the music you should listen to Waterlily's Ry Cooder recording that won a Grammy.
You would also need to know how many Teslas the ribbon is suspended in. As a physics problem the thermal noise of the ribbon compared to the signal of the Brownian rms pressure on the ribbon would be your absolute limit. As Thorsten points out a transformer that does not contribute would be very hard to make.
- Status
- Not open for further replies.