Equalisation curves (very good)
EQ OLD RECORDS
Oldie (notice the distortion curves)
http://www.aes.org/aeshc/pdf/frayne_history-of-disk-recording.pdf
Radiotron Designers Handbook 3rd Edition (1941 print) pdf download:
http://www.tubebooks.org/Books/RDH3.pdf
Radiotron Designers Handbook 4rd Edition (1960 print) pdf download:
http://www.tubebooks.org/Books/RDH4.pdf
George
EQ OLD RECORDS
Oldie (notice the distortion curves)
http://www.aes.org/aeshc/pdf/frayne_history-of-disk-recording.pdf
Radiotron Designers Handbook 3rd Edition (1941 print) pdf download:
http://www.tubebooks.org/Books/RDH3.pdf
Radiotron Designers Handbook 4rd Edition (1960 print) pdf download:
http://www.tubebooks.org/Books/RDH4.pdf
George
Now we are getting somewhere. many people are told by well known Mastering engineers to only dither once at the final step then send the file to the mastering lab. But if you are doing EQ to recorded channels and other things... level changes to get a better 'mix' and other editing/processing, you are lossing data thru truncation and limited word size etc. This happens through-out the mix-down processes. It is quit unlike mixing in analog.
However you look at what constitutes 'distortion' changes in the file data is going on. Maybe hundreds or thousands of such affects on the data word. Is it changing the file data enough to end up as an audible delta? But, how would that be a delta when comparing master files and a CD play back? Is there a data change threshold that a single dither cannot cover up?
BTW - I didnt mean for 'distortion' to mean harmonics or thd/Im form. more general at this point -- like freq response deviations is a 'distortion' of the original flat response.
Distortion in the meaning of thd/IM is something else which i want to see when I go thru both the ADC to DAC at different input levels. Just using test tones - no DAW software.... though, that might be another test (with a popular DAW).
THx-RNMarsh
Last edited:
agreed, distortion is a change in the signal, not defined how.
as a thought exercise, it can get quite nebulous. Mixing today is from essentially a playback of the individual tracks or a stem of several tracks to a mixing console or a passive mixing buss or in the box, or .....
Most projects are a composite of several of those methods.
I believe PT has moved to a "larger" mix bus architecture in the recent releases, much has changed, and mixing and such can happen at the higher precision word and be kept there til rendered out, at which point one dither pass can be applied from 48 to 24. (I think I remember 48, could be wrong).
Once you get to the actual 2 track mix, the referenced dither advice is relevant. but so much happens before that. I, and I know others, feel that is one of the larger sources of the oft mentioned coldness of digital, truncation.
as a side note Richard, for your thoughts re: why 9624 sounds so different, I would consider the filters involved in the ADC to be a good subject for examination as well.
Cheers
Alan
as a thought exercise, it can get quite nebulous. Mixing today is from essentially a playback of the individual tracks or a stem of several tracks to a mixing console or a passive mixing buss or in the box, or .....
Most projects are a composite of several of those methods.
I believe PT has moved to a "larger" mix bus architecture in the recent releases, much has changed, and mixing and such can happen at the higher precision word and be kept there til rendered out, at which point one dither pass can be applied from 48 to 24. (I think I remember 48, could be wrong).
Once you get to the actual 2 track mix, the referenced dither advice is relevant. but so much happens before that. I, and I know others, feel that is one of the larger sources of the oft mentioned coldness of digital, truncation.
as a side note Richard, for your thoughts re: why 9624 sounds so different, I would consider the filters involved in the ADC to be a good subject for examination as well.
Cheers
Alan
Last edited:
All good and relevant points. I'll be doing many tests once I get all set up... still waiting for the new BenchMark ADC to show up. meanwhile, i have been going back and forth on two other projects waiting for use of the limited test bench space.
Regarding the filters..... I have long wondered if we dont do something like LP EQ from the early recordings...... not using the exact same filter characteristics on record/playback.
THx-RNMarsh
Last edited:
Interesting that the tools that are relied upon by many, likely are making their
changes to the original content that are not always known and would be almost
impossible to quantify, maybe during as well as after the fact.
The distortion....as defined as a deviation from the original signal...is not limited to
audio, I would also say to all media in general. It happens with Pictures and Video
also. I think we all would agree that we all have to start the process off with the
highest quality, deepest resolution, highest bandwidth, et al possible to get the best possible output.
It is pretty easy to see this in pictures or images as we have lots if pixels in our eyes
and as our brains developed this further we have been quite successful as a species.
We might ask ourselves does that one bit, one byte really matter?
You bet it does. Think about each of our own personal and collective histories...
...the only reason we are here, is because each of our bits has been a success since
the beginning of time. Had one bit been OFF, neither you would be reading this
nor would I be typing this.
Of the sub-modalities of thought most normally organized people are in the image
mode...when we get to auditory and kinethetic (sp)...of those auditory has what I
would call the most play the most room for error or deviation in it. I would also
say harder to pin things down when they aren't just right.
Like caps or resistors had to measure the true sonic effect some claim there are
none, but some with big brains seem to be able to get things right...
just as Richard observed earlier, "However, over time, JC designs have been
consistent in their subjective ranking by a wide variety of people and groups.
Could it be then, JC actually does know what else is needed but we arent listening
or asking the right questions?"
Or maybe it's a musician thing...or and electro-musician or maybe musician-electro
that somehow those musician roots make the design subtleties required to make
the amp, pre amp, come alive--somehow it makes it a bit more real to our ears
and perceptions of hearing along with incorporating those kinesthetic's (sp) that
an engineer alone cannot cope with.
It's part of the soul, the heart, the feeling that someone is putting into it.
It sounds crazy I know....but it's true.
Why else would one producer have a string of platinums and others don't?
Why else would one musician have it while others don't?
How else could the "col" have the "killer", Cash, and Elvis, under contract
and other's not?
If you have enough of IT, you can break through the format's limitations
and touch the soul of another human being...
It's way more than just zeros and once.
As Unkle Ludwig used to say,
"From the heart,
may it go to the heart."
changes to the original content that are not always known and would be almost
impossible to quantify, maybe during as well as after the fact.
The distortion....as defined as a deviation from the original signal...is not limited to
audio, I would also say to all media in general. It happens with Pictures and Video
also. I think we all would agree that we all have to start the process off with the
highest quality, deepest resolution, highest bandwidth, et al possible to get the best possible output.
It is pretty easy to see this in pictures or images as we have lots if pixels in our eyes
and as our brains developed this further we have been quite successful as a species.
We might ask ourselves does that one bit, one byte really matter?
You bet it does. Think about each of our own personal and collective histories...
...the only reason we are here, is because each of our bits has been a success since
the beginning of time. Had one bit been OFF, neither you would be reading this
nor would I be typing this.
Of the sub-modalities of thought most normally organized people are in the image
mode...when we get to auditory and kinethetic (sp)...of those auditory has what I
would call the most play the most room for error or deviation in it. I would also
say harder to pin things down when they aren't just right.
Like caps or resistors had to measure the true sonic effect some claim there are
none, but some with big brains seem to be able to get things right...
just as Richard observed earlier, "However, over time, JC designs have been
consistent in their subjective ranking by a wide variety of people and groups.
Could it be then, JC actually does know what else is needed but we arent listening
or asking the right questions?"
Or maybe it's a musician thing...or and electro-musician or maybe musician-electro
that somehow those musician roots make the design subtleties required to make
the amp, pre amp, come alive--somehow it makes it a bit more real to our ears
and perceptions of hearing along with incorporating those kinesthetic's (sp) that
an engineer alone cannot cope with.
It's part of the soul, the heart, the feeling that someone is putting into it.
It sounds crazy I know....but it's true.
Why else would one producer have a string of platinums and others don't?
Why else would one musician have it while others don't?
How else could the "col" have the "killer", Cash, and Elvis, under contract
and other's not?
If you have enough of IT, you can break through the format's limitations
and touch the soul of another human being...
It's way more than just zeros and once.
As Unkle Ludwig used to say,
"From the heart,
may it go to the heart."
Wow his thread really moves!
One thing that caught me in this paper is the use of a square wave test signal and then looking for even harmonics in the output signal. That seems like a interesting test - has anyone here ever done that or know of a reference where it has been done?
Jan
Mixing in analog brings its own evils (worse on my point of view).
As the bus mixing preamp is, one way or another, an OPA in inverting configuration, its gain increase each time you insert a track. As a consequence, the noise, distortion, bandwidth reduction, losses of details increase too.
It is obvious when you listen, comparing a single track alone with nothing else plugged in, and the same with 31 other tracks inserted (with no modulation or fader at 0) plugged in in the same time. (a 30dB attenuation of its level followed by 30dB increase of gain in the OPA).
Not so much as we suppose, in fact. We have a very little area witch is sharp and detailed in the center of our vision. (Just try to read a text while you focus on the same surface same distance) a little aside. It is our eyes witch constantly move to bring our center of interest in the center, and our brain witch correlate the successive images.
Last edited:
32bit floats have only a 24bit mantissa. I posted an experiment where an exact 1kHz sine wave only exercises a few codes (48) at 96KHz and the truncation harmonics are almost exactly the same as when truncated to 24bit integer. As for dither I recognize theory and practice are very different, and I never meant to comment on processing chains and weather to apply it at every step or at the very end.
Last edited:
I did not know that about the summing amp was so audible when loaded up with signals on every channel........ I wonder now what was the thd/Im level and cross-talk of that increase? More dots to help with my audible threshold numbers.
THx-RNMarsh
Sure, but 32 bit float allows summing of channels without DSP magnitude overload, and also allows gain reduction of channels without loss of their 24 bit resolution.
Pro Tools has taken a very long time to come to the party, Samplitude has incorporated this signal flow since day one.
For some obscure reason, the technically and sonically inferior prior versions of PT have been the defacto standard in pro audio world....those who are more critical have always advocated Samplitude.
What plugins do is important also.
I need to revisit my experiments with dithering depth and spectral types, but as I recall, white noise dithering is readily audible, as are other flavours including the likes of Panasonic MASH and Sony SBM, amongst others.
As I recall, I preferred the 'natural' distortion of non dithered mixdown..... shaped dithering attempts to provide a reduction of Fletcher Munsen weighted artifacts, and flat dithering adds a white noise masking.....both are not natural sounding to my ear.
Dan.
Last edited:
I never measured-it. But each time i was○able to replace original OPAs by CFAs, there, the improvement was obvious, and sound back to life again ;-)
Yes, mostly in the video chapters of catalogs, a specialty of Analog design ;-)
- Status
- Not open for further replies.