Quantcast
Questions about Setting Preamps for a classical recording - Page 4 - Gearspace.com
The No.1 Website for Pro Audio
Questions about Setting Preamps for a classical recording
Old 2 weeks ago | Show parent
  #91
Gear Guru
 
1 Review written
🎧 5 years
Quote:
Originally Posted by kludgeaudio ➡️
I have done live concert feeds with an rnc and....it's by no means a high end broadcast processor and it sure would be nice to have a digital limiter after it... but... it allowed us to meet the specs and sound decent without actually spending money.
--scott
a valid point...
what i wanted to get across is that in a situation with a lot of mics, i like to leave some of the work to the equipment so that i can concentrate on the mix - and if i didn't own the equipment, i would rent it for such situations.
Old 2 weeks ago | Show parent
  #92
Gear Maniac
 
🎧 5 years
Quote:
Originally Posted by kludgeaudio ➡️
I have done live concert feeds with an rnc and....it's by no means a high end broadcast processor and it sure would be nice to have a digital limiter after it... but... it allowed us to meet the specs and sound decent without actually spending money.
--scott
I sold mine a long time ago and sometimes wish I hadn't. Nifty little box.
Old 2 weeks ago | Show parent
  #93
Lives for gear
 
🎧 15 years
Just a reminder, that in 24 bit encoding there are 256 additional amplitude steps in between each 16 bit amplitude step. Surely you want to exploit that increased amplitude accuracy during encoding.
Old 2 weeks ago
  #94
Lives for gear
 
🎧 10 years
For those who only record two stereo tracks, you can borrow the old tape recorder safety strategy of double recording the same input signal onto two different tracks, with a 6-12 dB of recording level difference. Of course, in digital recording, this would mean using two ADCs and recording 4 tracks.

This is a very effective trick that is widely used in digital imaging world, as well as in just about every digital microphone.
Old 2 weeks ago | Show parent
  #95
Gear Addict
 
🎧 10 years
Quote:
Originally Posted by David Spearritt ➡️
Just a reminder, that in 24 bit encoding there are 256 additional amplitude steps in between each 16 bit amplitude step. Surely you want to exploit that increased amplitude accuracy during encoding.
And recording in DSD provides an additional +3dB of above 0dB headroom, with an additional +3dB cushion above that with significantly increased distortion

Since DSD is analog, it behaves like tape where distortion increases with signal amplitude, the opposite of PCM. So for acoustic music recording, there's little to be gained by pushing close to 0dB as a practice.
Old 2 weeks ago | Show parent
  #96
Lives for gear
 
Plush's Avatar
 
5 Reviews written
🎧 15 years
Post #95--total nonsense. What you wrote is grossly wrong.

Consult Lip****z and Vanderkooy:

Why 1-Bit Sigma-Delta Conversion is Unsuitable
for High-Quality Applications

by

Stanley P. Lip****z and John Vanderkooy
Audio Research Group, University of Waterloo
Waterloo, Ontario N2L 3G1, Canada



https://timbreluces.com/assets/sacd.pdf
Old 2 weeks ago | Show parent
  #97
Gear Maniac
 
Tsk, tsk - there goes Plush confusing the conversation with facts.
Old 2 weeks ago | Show parent
  #98
Lives for gear
 
David Rick's Avatar
 
🎧 15 years
Quote:
Originally Posted by tailspn ➡️
And recording in DSD provides an additional +3dB of above 0dB headroom, with an additional +3dB cushion above that with significantly increased distortion

Since DSD is analog, it behaves like tape where distortion increases with signal amplitude, the opposite of PCM. So for acoustic music recording, there's little to be gained by pushing close to 0dB as a practice.
Plush is spot on. You don't ever want to overload a one-bit delta-sigma modulator or, by extension, a DSD encoder. That's why the Sony mastering standards for this format require such generous headroom.

I don't know where @ tailspn got this "DSD is analog" nonsense. DSD is about as digital as you can get. It has only two possible digital levels, making it an extreme corner case of conventional PCM. The reason it works is an extreme amount of oversampling plus high-order noise shaping.

To get PCM to behave like analog, it needs to be properly dithered. Unfortunately, as Lip****z and Vanderkooy pointed out the paper cited by Plush, you can't dither a single-bit quantizer.

David L. Rick
Old 2 weeks ago | Show parent
  #99
Gear Addict
 
🎧 10 years
Quote:
Originally Posted by David Rick ➡️
Plush is spot on. You don't ever want to overload a one-bit delta-sigma modulator or, by extension, a DSD encoder. That's why the Sony mastering standards for this format require such generous headroom.

I don't know where @ tailspn got this "DSD is analog" nonsense. DSD is about as digital as you can get. It has only two possible digital levels, making it an extreme corner case of conventional PCM. The reason it works is an extreme amount of oversampling plus high-order noise shaping.

To get PCM to behave like analog, it needs to be properly dithered. Unfortunately, as Lip****z and Vanderkooy pointed out the paper cited by Plush, you can't dither a single-bit quantizer.

David L. Rick
David, please don't think of Pulse Density Modulation, of which DSD is the 1-bit variant, through your digital experience knowledge.

The fundamental difference between PDM and PCM, is that PCM is composed of discrete measured digital samples at some resolution (bit depth) and frequency. They are represented by a binary 2's complement number digital VALUE. On the other hand, PDM, either 1-bit or multibit, contain NO DIGITAL VALUES, only varying percents of modulation. It's simply an analog representation of changing signal levels through a bitstream consisting of changing 1's and 0's bit DENSITIES, proportional to a modulating signal level.

Again, there are no digital VALUES represented. There's only analog bitstream(s), whose average density is proportional to the modulating signal level, expressed by percent of modulation of a bit clock carrier. Since there's no digital VALUE(s) represented, they can only be stored and retrieved in/from a digital computer. It's the reason that although the modulation process is complex, the 1-bit variant (DSD) can be retrieved through simple integration. A plane old Gaussian 6dB/octave integrator would surfice if the modulated bitclock we in the hundreds of Megahertz, but at 11.6MHz, DSD256) the integrator needs to be somewhat steeper.

As a simplistic model, please consider PCM like movie film; where each frame is a stand alone picture VALUE, and PDM/DSD like FM radio, where an audio signal is modulating a carrier, (again, no VALUE represented). The carrier in this case is a square wave bit clock. Alternating 1's and 0's represent zero modulating signal level. All 1's represent 100% modulation of positive signal, and all 0's represent 100% modulation of negative signals. 0dB is specified as 50% modulation, allowing the theoretical +6dB headroom I mentioned in my above post. It's really simple when you eliminate all the digital value complications of PCM.

Also, please remember ALL A/D converters available today are front ended with Sigma-Delta modulators, which produce multibit modulated bitclock bitstreams (6 bits wide typically). To be made processable in a digital enviorment, the bitstreams are converted through the lossy process of decimation filtering and interpolation. A pure DSD256 recording, only edited and not converted to PCM for post processing, sounds quite different and natural when compared to the PCM converted file in low level detail and spaciousness.

And put Lip****z and Vanderkooy in perspective. They've been beating that same DSD64 drum now for 20+ years, and imply all recordings must be converted to PCM to be useful. They are correct for the example they theorize, but no DSD recording today is recorded at 2.8MHz bitrates.

Thanks,
Tom

Last edited by tailspn; 2 weeks ago at 03:30 PM..
Old 2 weeks ago | Show parent
  #100
Lives for gear
 
David Rick's Avatar
 
🎧 15 years
I do arithmetic. Arrest me now!

Quote:
Originally Posted by tailspn ➡️
David, please don't think of Pulse Density Modulation, of which DSD is the 1-bit variant, through your digital experience knowledge.

The fundamental difference between PDM and PCM, is that PCM is composed of discrete measured digital samples at some resolution (bit depth) and frequency. They are represented by a 2's complement digital VALUES. PDM, either 1-bit, or multibit, contain NO DIGITAL VALUES, only varying percents of modulation. It's simply an analog representation of changing signal levels (represented by varying percent of modulation) through bitstreams consisting of changing 1's and 0's bit DENSITIES, proportional to a modulating signal level through percentage of modulation.
This is a fair interpretation of an abstract bitstream signal, though it ignores how it came into existance. But it's equally valid to view it as a sequence of numbers in which the mantisa is always unity, so only the sign need be transmitted.

Quote:
A plane old Gaussian 6dB/octave integrator would surfice if the modulated bitclock we in the hundreds of Megahertz, but at 11.6MHz, DSD256) the integrator needs to be somewhat steeper.
So we're agreed that in all practical cases, we're dealing with multi-pole integrator. In fact, an integrator of high enough order to compete with current performance benchmarks is too difficult to stabilize, so modern audio converters always employ multi-bit modulators (with heroic measures to improve their linearity). Nonetheless, it is always necessary that the input signal be restricted to some fraction of "full scale" to prevent loop instability. The higher-order the noise shaping, the smaller this fraction must be.

Quote:
As a simplistic model, please consider PCM like movie film; where each frame is a stand alone picture VALUE, and PDM/DSD like FM radio, where an audio signal is modulating a carrier, (again, no VALUE represented). The carrier in this case is a square wave bit clock. Alternating 1's and 0's represent zero modulating signal level. All 1's represent 100% modulation of positive signal, and all 0's represent 100% modulation of negative signals. 0dB is specified as 50% modulation, allowing the theoretical +6dB headroom I mentioned in my above post. It's really simple when you eliminate all the digital value complications of PCM.
This is confusing because 50% duty cycle implies 0 Volts, not 0 dB. But if you have a signal, that swings between 75% duty cycle and 25% duty cycle, then I imagine that's what you mean by 50% modulation. You could define that as 0 dB for a simple modulator that's stable at 0.5*(full scale). Higher-order ones are not however, so the convention doesn't fit current technology.

Quote:
Also, please remember ALL A/D converters available today are front ended with Sigma-Delta modulators, which produce multibit modulated bitclock bitstreams (6 bits wide typically). To be made processable in a digital enviorment, the bitstreams are converted through the lossy process of decimation filtering and interpolation. A pure DSD256 recording, only edited and not converted to PCM for post processing, sounds quite different and natural when compared to the PCM converted file in low level detail and spaciousness.
Tom, in my production world, even simple editing requires crossfades. Those can't be done in one-bit math, it has to be some kind of longer-word PCM unless you up-sample it further. The only way to maintain "purity" is to mix it analog, but even that involves math because modern DAC's all have a multi-bit core.

I simply don't understand why people think doing arithmetic is a crime.

Quote:
And put Lip****z and Vanderkooy in perspective. They've been beating that same DSD64 drum now for 20+ years, and imply all recordings must be converted to PCM to be useful. They are correct for the example they theorize, but no DSD recording today is recorded at 2.8MHz bitrates.
Pick whatever rate you want, the bitsteam format implies a one-bit modulator which is fundamentally flawed. In my view, there should only be one of those in the production chain. I want it to be in the final encoding step at the back end. Everything else can be high-rate multi-bit, and better for it.

David

PS: Apologies for dragging this thread so far off topic. Don't overload your converters folks, you'll be sorry!
Old 2 weeks ago | Show parent
  #101
Lives for gear
 
kludgeaudio's Avatar
 
Quote:
Originally Posted by tailspn ➡️
And recording in DSD provides an additional +3dB of above 0dB headroom, with an additional +3dB cushion above that with significantly increased distortion
DSD buys you plenty of headroom at low frequencies, not so much headroom at high frequencies. It behaves as a slew-limited system and peak-reading meters no longer tell you everything you need to know about levels in the DSD world.

Quote:
Originally Posted by tailspn ➡️
Since DSD is analog, it behaves like tape where distortion increases with signal amplitude, the opposite of PCM. So for acoustic music recording, there's little to be gained by pushing close to 0dB as a practice.
DSD is analogue? It's quantized in time, it's quantized in amplitude.

And in the modern PCM world, distortion doesn't increase with decreasing amplitude any longer. It's not 1985 back when converters all had massive crossover distortion... we don't live in that world today, thank God.
--scott
Old 1 week ago | Show parent
  #102
Lives for gear
 
🎧 15 years
Quote:
Originally Posted by tailspn ➡️
Also, please remember ALL A/D converters available today are front ended with Sigma-Delta modulators, which produce multibit modulated bitclock bitstreams (6 bits wide typically). To be made processable in a digital enviorment, the bitstreams are converted through the lossy process of decimation filtering and interpolation. A pure DSD256 recording, only edited and not converted to PCM for post processing, sounds quite different and natural when compared to the PCM converted file in low level detail and spaciousness.
Most, but not all. The Lavry Gold ADs are most likely based on the SAR (successive approximation) principle only.

And today SAR chips suitable for audio are availible with specs surpassing the latest DS chips.

Pure PCM converters can be build today that sound closer to straight wire than anything else, that transmit the essential qualities of audio, the effortless power, movement, lieveliness, timbre, pinpoint directivity and immediacy.

The sooner we get rid of the hazy, muffled, artificial sound delta sigma conversion has introduced, the better, IMHO.

And DSD is a dead end, and not needed, IMO.
Old 1 week ago | Show parent
  #103
Lives for gear
 
David Rick's Avatar
 
🎧 15 years
Noise shaping is your friend

Quote:
Originally Posted by living sounds ➡️
Most, but not all. The Lavry Gold ADs are most likely based on the SAR (successive approximation) principle only.
Those may have been flash converters, not SAR's; either way they were still lower resolution converters with significant oversampling and subsequent decimation to get to the desired output rate and bit depth. The difference is, Dan did his own digital decimation filters with proper arithmetic precision and quite likely some moderate noise-shaping. And since he was starting with a multi-bit front end, he had the option to do proper dithering. (I don't know if he did, or simply used a lot of guard bits.)

I suspect the reason Dan was initially resistant to higher sample rates was that he was already using the best flash converters he could find and knew that reducing the decimation ratio would just put more quantization noise in the passband. His later stuff may possibly use noise shaping with some non-DC zeros, so the trade-off isn't quite as stark.

Quote:
And today SAR chips suitable for audio are availible with specs surpassing the latest DS chips.
Data sheet or it didn't happen.

Quote:
Pure PCM converters can be build today that sound closer to straight wire than anything else, that transmit the essential qualities of audio, the effortless power, movement, lieveliness, timbre, pinpoint directivity and immediacy.
Historically, "pure" PCM converters were anything but. For instance, the DAC in my first Sony CD player (which still works!) uses a PCM56 DAC chip. As I recall, those were segmented DACs in which low-order bits were done using a resistive ladder and higher-order with an R2R network. That meant there were a number of boundary transitions where the differential linearity went to hell. The most important "major bit" transition could be manually trimmed using external circuitry, but that only worked if you could hold the temperature constant.

Quote:
The sooner we get rid of the hazy, muffled, artificial sound delta sigma conversion has introduced, the better, IMHO.
I eventually found that playback quality of my PCM-based Sony player could be signficantly improved by using an outboard DAC based on more modern delta-sigma converters. Apparently SFDR matters to me. But one can still buy PCM56 and PCM58 chips and boards today, and some audiophiles are a bit obsessed with them.

Quote:
And DSD is a dead end, and not needed, IMO.
Well, at least we agree on one thing!

David L. Rick
Old 1 week ago | Show parent
  #104
Lives for gear
 
🎧 15 years
Quote:
Originally Posted by David Rick ➡️
Data sheet or it didn't happen.
Lot's of measurements can be found in this thread:

https://www.diyaudio.com/forums/equi...2380-24-a.html



Quote:
Historically, "pure" PCM converters were anything but. For instance, the DAC in my first Sony CD player (which still works!) uses a PCM56 DAC chip. As I recall, those were segmented DACs in which low-order bits were done using a resistive ladder and higher-order with an R2R network. That meant there were a number of boundary transitions where the differential linearity went to hell. The most important "major bit" transition could be manually trimmed using external circuitry, but that only worked if you could hold the temperature constant.
Historically, yes. But a modern discrete R2R DAC based on the sign-magnitude principle has excellent linearity, better than delta sigma chips.

Here are measurements of an older Soekris:

https://www.audiosciencereview.com/f...ibit-dac.3956/

THD is not nearly as low with these as current DS converters, but thanks to the sign-magnitude principle the "distance" between useful signal and harmonics stays very high, even at much lower amplitudes.

And distortion is still lower than any analog tape machine, most analog consoles, preamps, outboard etc. And speakers, of course. Being correlated with the signal the distortion is masked to the human ear anyway.

After using R2R DACs for a few years now, I definitely don't want to go back to delta sigma. And as a side effect I cannot stand listening to most music recorded digitally after the early 90s now, especially not remasters.
Old 1 week ago | Show parent
  #105
Gear Addict
 
Quote:
Originally Posted by living sounds ➡️

After using R2R DACs for a few years now, I definitely don't want to go back to delta sigma.
Could you pls share which model?
Old 1 week ago | Show parent
  #106
Gear Addict
 
🎧 10 years
Quote:
Originally Posted by living sounds ➡️
And DSD is a dead end, and not needed, IMO.
Ya think?

nativedsd.com

79 labels, 27,832 subscribed customers world wide. Admittedly, the vast majority of these labels record in DSD using Merging hardware and Pyramix DAW. These use the real world ARDA AT1201 and newer AK5394A converter chip A/D's, converting analog signals through multi-bit PDM initial conversion, then interpolated to 1-bit at the same bitrate. No decimation filtering.

All have the resources to choose whatever equipment and recording methodology to attach their product and reputation.

Tom

Last edited by tailspn; 1 week ago at 08:39 PM..
Old 1 week ago | Show parent
  #107
Lives for gear
 
🎧 15 years
Quote:
Originally Posted by mcgilroy ➡️
Could you pls share which model?
Check out soekris.dk
Old 1 week ago | Show parent
  #108
Lives for gear
 
🎧 15 years
Quote:
Originally Posted by tailspn ➡️
Ya think?

nativedsd.com

79 labels, 27,832 subscribed customers world wide. Admittedly, the vast majority of these labels record in DSD using Merging hardware and Pyramix DAW. These use the real world ARDA AT1201 and newer AK5394A converter chip A/D's, converting analog signals through multi-bit PDM initial conversion, then interpolated to 1-bit at the same bitrate. No decimation filtering.

All have the resources to choose whatever equipment and recording methodology to attach their product and reputation.

Tom
Yes, as a recording medium. To make money with, it will probably live on for quite some time.
Old 1 week ago | Show parent
  #109
Lives for gear
 
David Rick's Avatar
 
🎧 15 years
About the LTC2380-24

Quote:
Originally Posted by living sounds ➡️
Lot's of measurements can be found in this thread:

https://www.diyaudio.com/forums/equi...2380-24-a.html
Ah, somebody tried to use the LTC2380-24 seismic ADC for audio! Data sheet here.

A Linear Technology rep (they're now part of ADI -- as is Maxim, starting last week) calls on me regularly. He thought I might like this part for ultrasound or radar, but asked me what I thought of it for audio. The best-case SNR/SINAD specs were impressive, but mostly involved a lot of averaging, resulting in lower output bandwidths than audio folks need. I care much more about inharmonic distortions and the only relevant SFDR plot showed -120 dB at 125 ksps output rate. That's promising, but AKM was already doing better at 1/3 the price, so I understood why LTC had not bothered to do further characterization for audio use.

Consequently, I'm very pleased that @ living sounds posted a link to a thread on the subject at diyadio.com. A member of that forum took the trouble to hack up the LTC demo board and wrap a bunch of additional circuitry around it to make it suitable for audio use. He seems to have worked on it very hard in 2016 and 2017 before eventually stalling out. The last set of measurements show SFDR in the -126 to -128 dB range, which is genuinely admirable. The tested system is not a commercial product, but there's certainly some potential there for a nice product if a boutique pro audio maker wanted to pursue this avenue. It's an open question whether there would be a big enough demand to justify the effort, given how much the product would need to cost and knowing that the delta-sigma folks are already breathing down one's neck.

Quote:
Historically, yes. But a modern discrete R2R DAC based on the sign-magnitude principle has excellent linearity, better than delta sigma chips.

Here are measurements of an older Soekris:

https://www.audiosciencereview.com/f...ibit-dac.3956/

THD is not nearly as low with these as current DS converters, but thanks to the sign-magnitude principle the "distance" between useful signal and harmonics stays very high, even at much lower amplitudes.
I think you're reading too much into the posted low-level linearity test, which filters out everything but the fundamental. I'll concede that 0.02% THD+N is probably "low enough" in practice but a the twin-tone intermod distortion number of -50 dB won't win any ribbons at the fair. The tests do show that it's better (in most respects) than those vastly over-rated Schitt's that some audiophiles were gaga over for several years.

David
Old 1 week ago | Show parent
  #110
Lives for gear
 
kludgeaudio's Avatar
 
Quote:
Originally Posted by David Rick ➡️
Ah, somebody tried to use the LTC2380-24 seismic ADC for audio! Data sheet here.

A Linear Technology rep (they're now part of ADI -- as is Maxim, starting last week) calls on me regularly. He thought I might like this part for ultrasound or radar, but asked me what I thought of it for audio. The best-case SNR/SINAD specs were impressive, but mostly involved a lot of averaging, resulting in lower output bandwidths than audio folks need. I care much more about inharmonic distortions and the only relevant SFDR plot showed -120 dB at 125 ksps output rate. That's promising, but AKM was already doing better at 1/3 the price, so I understood why LTC had not bothered to do further characterization for audio use.
I've used some instrumentation stuff which uses that for a front end and it's definitely interesting. The problem with those SAR things is getting them running fast enough to avoid crazy antialiasing filters, but they do have the stability of a sigma-delta system even if you're still back to the same filter issues you had in the ladder era. I'd be really interested to see how such a system would sound (our system does not have proper antialiasing for audio work unfortunately).

Doesn't that chip have some craziness to reduce power demands, though? I would worry that some of that may also be problematic.

Personally I am a big fan of sigma-delta systems... no need to leave the converters powered up in the hall a day before the session and then spend an hour fiddling with trimpots in the morning...
--scott
Old 1 week ago | Show parent
  #111
Lives for gear
 
🎧 15 years
Quote:
Originally Posted by David Rick ➡️
Consequently, I'm very pleased that @ living sounds posted a link to a thread on the subject at diyadio.com. A member of that forum took the trouble to hack up the LTC demo board and wrap a bunch of additional circuitry around it to make it suitable for audio use. He seems to have worked on it very hard in 2016 and 2017 before eventually stalling out. The last set of measurements show SFDR in the -126 to -128 dB range, which is genuinely admirable. The tested system is not a commercial product, but there's certainly some potential there for a nice product if a boutique pro audio maker wanted to pursue this avenue. Its an open question whether there would be a big enough demand to justify the effort, given how much the product would need to cost and knowing that the delta-sigma folks are already breathing down one's neck.
He's still at it:

https://www.diyaudio.com/forums/equi...nalyzer-8.html

0.000024% THD at 1 khz in the latest incarnation of his own (more optimized) design. I don't think there is a pro audio ADC that good.


Quote:

I think you're reading too much into the posted low-level linearity test, which filters out everything but the fundamental. I'll concede that 0.02% THD+N is probably "low enough" in practice but a the twin-tone intermod distortion number of -50 dB won't win any ribbons at the fair. The tests do show that it's better (in most respects) than those vastly over-rated Schitt's that some audiophiles were gaga over for several years.
I did my own amplitude linearity test and the R2R DAC beat a Lynx converter handsomely.

Alos, more recent designs have lower IMD:

https://www.superbestaudiofriends.or...rements.10561/



But again, all these measurements aside, with the right reconstruction filters (which one is free to load or even design oneself) I prefer the sound of these to any DS converters. And this is comparing to analog audio.

Last edited by living sounds; 1 week ago at 12:20 PM.. Reason: ADC, not DAC
Old 1 week ago | Show parent
  #112
Lives for gear
 
tourtelot's Avatar
 
1 Review written
🎧 10 years
Okay guys, we are talking about working solutions to what might present itself on a typical location recording.

0.000024% THD??!! WTF?

Sheesh.

Set your best preamp to whatever assures you that you won't get clipping and push the freakin' red button.

Done now!

D.
Old 1 week ago | Show parent
  #113
Lives for gear
 
James Lehmann's Avatar
 
🎧 15 years
Quote:
Originally Posted by tourtelot ➡️
0.000024% THD??!! WTF?

Sheesh.
But if it turns out that you can indeed reduce your preamp gains by 0.000024% when paired with converter X as opposed to converter Y and achieve a lower overall system noise floor, then this surely counts as a bona fide answer to the OP's original question, no?
Old 1 week ago | Show parent
  #114
Lives for gear
 
🎧 15 years
I'm with you, I believe THD is a good measurement to take to find (potential) problems in electronics, but getting absurdly low numbers doesn't make any difference to the human ear.
Old 1 week ago | Show parent
  #115
Lives for gear
 
kludgeaudio's Avatar
 
Quote:
Originally Posted by tourtelot ➡️
Okay guys, we are talking about working solutions to what might present itself on a typical location recording.

0.000024% THD??!! WTF?
This just goes to show you that THD isn't a useful tool for this. Although I remember in the ladder era when it was not unusual to get full-scale THD numbers lower than .001%, but when you dropped the signal level down by 60dB it turned into more like 6%... that was definitely eye-opening. And that's why folks were obsessive about pushing levels up for years.
--scott
Old 1 week ago | Show parent
  #116
Lives for gear
 
🎧 15 years
Quote:
Originally Posted by kludgeaudio ➡️
This just goes to show you that THD isn't a useful tool for this. Although I remember in the ladder era when it was not unusual to get full-scale THD numbers lower than .001%, but when you dropped the signal level down by 60dB it turned into more like 6%... that was definitely eye-opening. And that's why folks were obsessive about pushing levels up for years.
--scott
Yet everyone is looking at THD to judge something over the internet...

A modern sign magnitude ladder DAC can keep THD low even at very low amplitudes.
Old 20 hours ago | Show parent
  #117
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by Plush ➡️
One would have to turn down a 24 bit signal by 48dB to reach a 16 bit quality level.

On a separate topic, it is the use of waaaay too many mics that presents a situation where one gets screwed over by noise. This film score type 55 mics recording style also ruins depth and perspective.

It’s a piece of crap.

Not allowed here.
Perfect!!
Old 17 hours ago | Show parent
  #118
Gear Guru
 
Brent Hahn's Avatar
 
1 Review written
🎧 15 years
Quote:
Originally Posted by Geoff Poulton ➡️
Perfect!!
Film score recording is a completely different profession.
Old 16 hours ago | Show parent
  #119
Lives for gear
 
tourtelot's Avatar
 
1 Review written
🎧 10 years
Film scoring compared to minimal mic classical recording. Kidding right?

Apples and oranges.

D.
Old 16 hours ago | Show parent
  #120
Gear Maniac
 
Plush said "film score type", which I took to mean applying that type of dozens of mics approach to non film score orchestral recordings; not criticizing how film scores are recorded.
📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 59 views: 5689
Avatar for SurveillanceP
SurveillanceP 24th March 2010
replies: 1856 views: 185437
Avatar for Paul_G
Paul_G 4 hours ago
replies: 3306 views: 61141
Avatar for Synth Guru
Synth Guru 4 weeks ago
Topic:
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump