I believe you could do-it for some MP3, not when it is your own heritage preservation.
I wonder why you could think to compress a 24/96 re-mastering, when you know your audience will be audiophiles, looking for the max dynamic they can find and that you work for future.
Dunno about Stalin, but for old 78s a carefully restored version is to me preferable to hearing it as it was heard straight out the horn. Others may disagree.
or Plangent Plangent - A Better Way to Transfer Analog Tape - Audiophile Review where you remove tape induced artifacts with some clever DSP. I would take restored by that over the raw tape. But I accept some may want it as it was before they needed to shave.
or Plangent Plangent - A Better Way to Transfer Analog Tape - Audiophile Review where you remove tape induced artifacts with some clever DSP. I would take restored by that over the raw tape. But I accept some may want it as it was before they needed to shave.
OK, so I 'work for future' and nobody buys it, except 23 audiophiles in the USA. Bad deal if you ask me.
. Not only that, but I will be out of a job faster than you can say dynamic range.(Hint: the world of audio does NOT revolve around a couple of dreaming audiophiles. Sorry about that...).
jan
Stupid. The people who loved those records will want to have the best copy of the original record, as close as possible from the vinyl they probably have in their disk collection. And the kids will want to have the thing with the original perfume too.
It is not at all the same market than the last rap tube.
You can make a colorized version of an old black and white movie for a special occasion, like a tv show, but, for sure, you will make for your archives a black and white restored version as close as the original as possible, removing the scratches and dusts from a used copy if you don't have better, trying to recover the original contrast of the film.
I have some fabulous 1950's and early 1960's recordings on CD. I don't think they were remastered for digital - just a straight transfer and maybe de-wowed and de- fluttered. You can still hear the tape hiss but they are nevertheless masterful. I have jazz (Dave Brubeck), Julian Bream, and some classical orchestral pieces. Wonderful stuff
I do question this statement, as you've quoted what seems to be a reasonable statement to me..
I've been following the whole discussion, and have a question.
If I assume a digital system with x percent distortion during full scale reproduction due to lsb size, if I have a track which is really quiet, like 1% of full scale, won't the lsb distortion be 100 times higher?
I would expect the cleanest sine output to be while using the full 16 bits, but not if the sine is sufficiently low that it uses say, 8 bits.
jn
You are dreaming. The only valid market research is a signed purchase order.
I just want to add that all this controversy was because some accused the digital copy (masterisation does not mean necessary modification of the original source) to be responsible of some losses of quality, while i pretend their is more luck those losses come from the deterioration of the original analog master tapes, when it is about old records.
This include the lack of dynamic, losses of trebles, drops out etc.
jan.didden, you are out of reality. Catalogs have a great value and it is about their 'conservation'. Sergent Pepper is not in the charts and radio top ten anymore ;-)
Re-masterisations are not a 'commercial' work. Conservation of the heritage is the purpose. It is done systematically by all the editors and record companies, even if they don't publish the new master on the market. And, for the few that i'm aware of, they really try their best to make transparent copies, or, at least, as close as the original was at the origin as possible.
This include the lack of dynamic, losses of trebles, drops out etc.
jan.didden, you are out of reality. Catalogs have a great value and it is about their 'conservation'. Sergent Pepper is not in the charts and radio top ten anymore ;-)
Re-masterisations are not a 'commercial' work. Conservation of the heritage is the purpose. It is done systematically by all the editors and record companies, even if they don't publish the new master on the market. And, for the few that i'm aware of, they really try their best to make transparent copies, or, at least, as close as the original was at the origin as possible.
Last edited:
jn,
You understand step quantization. Depending on the converter design, the issue becomes where the error is at a maximum. For many designs this occurs around half scale which can in some cases be used as AC zero level.
Dick,
Nice to know you understand why an 80 input console looks great in the studio but really isn't useful anymore. As to the 118 dB capture range, we could argue a bit, as I think today the same research would come up with a higher number.
Actually the Fairchild that built these items was the Fairchild Semiconductor folks. They were showing off what their uA741 op-amp could do!
SY,
Still waiting..
I am familiar with dither averaging of an input that is being converted used to attain information below the LSB size, but not so with dither on the output of a DAC as your Q seems to infer. My question had more to do with a 2 volt sine being played using existing cd, and a 20 millivolt sine being played on the same system. For example, using the same digital info, but one track uses full scale, the second shifts the info 3 bits leaving MSB and next two as zero.(ignoring half scale equals zero of course).
I make the assumption that the converter is monotonic. So, my expectation is that there is never a step greater than 2 LSB.
jn
Last edited:
There is the issue. I do not expect a 24 bit converter to be that good. I expect 18-20 bits accuracy over the full range. 20 being the minimum for a first rate recording, if there is enough bandwidth.
I used to help out at one of the older studios in town. The owner bought all of his gear used after the newness fad wore off. He had just about every classic piece: Telefunken microphones, Pultecs, Fairchild and AKG reverbs, Fairchild compressors, Ampex 350's, Scullys, Mavecs and a bunch more. Always found it interesting what they sold for when he retired. (Still have a 350 myself.)
Just a short update on wireless standards for audio --- that DTS and Qualcomm (AllPlay) are new on the scene and both will "wirelessly stream high-definition audio" (DTS) - up to 24/192k and is able to" tosupport standard CD-quality and high resolution audio file formats, including WAV, FLAC and AIFF ........" While Qualcomm will do same as far as quality of signal is concerned. They differ in other ways..... just read about it in this issue of AudioExpress. So.... maybe No expansion to get dynamic range but more support for HiRes down loads and wireless streaming. So, the HiRes or Hidef is becoming part of main stream. Something for everyone, this time.
Some of this acceptance comes from consumer research which found that average Joe consumer liked the HiRes sound better when watching their movies.
What ever you think about CD vs HiRes as sources, the industry has moved on once again and included CD now as a legacy catagory. It just keeps getting better and I, for one, like that. But we have work to do for best use of those files in a Hi-End home system. I suggest going to high output - use pro standards - for the levels.... to get the dynamic range and full use of the 24 bits. For, one thing.
THx-RNMarsh
Some of this acceptance comes from consumer research which found that average Joe consumer liked the HiRes sound better when watching their movies.
What ever you think about CD vs HiRes as sources, the industry has moved on once again and included CD now as a legacy catagory. It just keeps getting better and I, for one, like that. But we have work to do for best use of those files in a Hi-End home system. I suggest going to high output - use pro standards - for the levels.... to get the dynamic range and full use of the 24 bits. For, one thing.
THx-RNMarsh
Last edited:
What are you showing? Is this a depiction of a 2mv signal bare out of a brick?
hmm. perhaps you need to buy better converters? I would expect a 24 bit unit to at least make 23 and change. Long term dc stability, sure it'll drift.
But even your worst case at 18 bits, you really believe there will be audible artifacts at 18 bits? Remember, even if you got 18 bit performance, it's not an 18 bit "hole" at the zero crossing of a 24 bit unit.
jn
I am welcome to suggestions for a true 24 bit A/D that will do 100 ksps.
Recording is more important than playback as when you have the file you scale it so you don't clip. Harder to do on a live recording.
Having looked at the commercial products that do 24 bits the best I saw was only 20 real bits, many 18.
One of the first digital audio delays was 12 bits accurate with 4 more thrown in for the fun. Problems showed up requiring the improved version to go to true 16 bits. Even that didn't really work in some churches.
A Sabine designed church had 1.5 seconds reverb time and a noise floor of NC25-30 depending on the day of the week. Sunday was of course the quietest. The pipe organ had a 32' pipe (or even larger don't recall precisely.) and could easily lift one's spirit physically. The main nave seats about 1200. So 18 bits probably would be borderline.
- Status
- Not open for further replies.