Become a Site Supporter and Never see Ads again!

Author Topic: master md's>analog out>24 bit???  (Read 10004 times)

0 Members and 1 Guest are viewing this topic.

Roving Sign

  • Guest
  • Trade Count: (0)
Re: master md's>analog out>24 bit???
« Reply #30 on: August 22, 2007, 08:51:15 AM »


 I go Standalone (optical) > JB3...realtime - but at least its a "digital" transfer - (I wouldn't be surpised if there is an A/D step in there...rather than an software based ATRAC>WAV conversion.)

Why would there be an AD step here? Your passing a digital signal to a digital recording AFAIK. I could be wrong, but I think this is the best way to go if you are starting with a "normal" non-HI MD recorded in the highest old school md quality ???

It was rank speculation!

Well - yes it is a digital signal - but do we know for sure that process that produced that signal is ATRAC > WAV? Just speculating...I dont have an link - only a recollection of someone saying this once...it might take more computing power to do a software based conversion? MIght have been easier just implement a ADAD process...

Offline prof_peabody

  • Trade Count: (4)
  • Needs to get out more...
  • *****
  • Posts: 4339
  • Team Houston
Re: master md's>analog out>24 bit???
« Reply #31 on: August 22, 2007, 09:35:40 AM »


 I go Standalone (optical) > JB3...realtime - but at least its a "digital" transfer - (I wouldn't be surpised if there is an A/D step in there...rather than an software based ATRAC>WAV conversion.)

Why would there be an AD step here? Your passing a digital signal to a digital recording AFAIK. I could be wrong, but I think this is the best way to go if you are starting with a "normal" non-HI MD recorded in the highest old school md quality ???

It was rank speculation!

Well - yes it is a digital signal - but do we know for sure that process that produced that signal is ATRAC > WAV? Just speculating...I dont have an link - only a recollection of someone saying this once...it might take more computing power to do a software based conversion? MIght have been easier just implement a ADAD process...

Going optical out from a standalone to jb3 involves converting the ATRAC to WAV essentially a digital interpreter, some might consider this a D2D I guess.  The difference between doing this and transferring the ATRAC via USB to a computer is that in the later method you are using a different ATRAC to WAV algorithm.  It seems most people prefer the D2D sound of the standalone unit over the software route.  I don't think anyone has done a detailed comparison of the standalone algorithm verses the software algorithm, so it's really up to your ears what you prefer.  Chances are the two algorithms are not written the same and will interpret an ATRAC file differently. 

Offline sunjan

  • Trade Count: (0)
  • Needs to get out more...
  • *****
  • Posts: 2003
  • Gender: Male
  • Taping since 1988, 28 years of fine recordings...
    • Just a handful of stuff I put on etree
Re: master md's>analog out>24 bit???
« Reply #32 on: August 23, 2007, 09:36:16 AM »
This was just covered in another thread.  It was also researched on minidisc.org and discovered to not be true.  A fellow there xferred his files and checked the file(s) in Adobe Audition and saw no degradation as has been described.  Once again, one test is worth a thousand opinions.  Use the RH1 to U/L all your legacy MD's with no fear of having them degraded.

As for saving them as 24 bit; well, if you care to manipulate them on the PC I understand it is a better way to do it, in 24 bit.  It will not improve the sound but will add fewer artifacts during manipulation.  You will have to resample and dither back to 16 bit.

Cheers
I seriously wish I could believe this because it would save us all a lot of time. Did you read my posts carefully? http://taperssection.com/index.php/topic,76899.msg1027426.html#msg1027426
I never said SP/LP2 files are being converted to ATRAC3+ as a standard or as an option that I accidentally might have selected (I surely didn't), SS does never inform you about this step, when you convert to WAV your output files are .WAV but they have been converted to ATRAC3+ (no idea on the bitrate) prior to being converted to .WAV, apparently that is the current workaround for the ATRAC-for-the-PC codec as was stated by someone on the German minidisc forum.
This discussion has nothing to do with HiMD recordings, transfers of those are all lossless and no one has claimed otherwise anywhere, just to make this clear once and for all (not saying anyone in this topic did say so, just trying to keep the information as straight as possible in here).

I've done frequency analysis and spectral analysis on files transferred via standalone deck (TOSlink > TOSlink connection) and compared them to the RH1>USB transfers and there were obvious differences, referred to as cut-offs in prior posts, most noticeably on SP recordings, not so bad with LP2 recordings (which all seem to have drop-offs at around 16kHz, as is obvious since the bitrate is really low, 128kbps, I think). You might want to give this a try yourself if you own a deck - I also read some of the minidisc.org discussion and couldn't open the attachment, other than that I just saw someone who "believed" something. Again, try this for yourself and post the results, I'd love to see a new version of SS/the ATRAC codec handle this differently. I'm on 4.2 still afair (am at work right now so can't check).
Shortly after this was being discussed on ts.com the DSEE feature was introduced to SonicStage (coincidence?), it adds something similiar to DSP to the playback of files, thereby making them sound/appear fuller than they were in the first place. So you might get even different results when playing back your .oma in SS and recording the digital output of your soundcard (internal or external) than when doing a straight transfer via USB or via a deck.

Hey nardo,

+T for researching this topic!
I'd be really grateful if you could take screen shots of the side-by-side analysis you did (optical realtime tranfer vs. RH1>USB>SS). Why don't you post your findings here: http://forums.minidisc.org/index.php?s=&showtopic=19709&view=findpost&p=125800

It would be really interesting to hear what the guys at Minidisc.org has to say about your results, if you descibe your method in detail.
Their theory is that you didn't highlighting the entire waveform in the edit window, clicking "scan" in the frequency analysis window. This is what caused the erroneous analysis.

Thanks,

Jan
Mics: A-51s LE, CK 930, Line Audo CM3, AT853Rx (hc,c,sc),  ECM 121, ECM 909A
Pres: Tinybox, CA-9100, UA5 wmod
Recorders: M10, H116 (CF mod), H340, NJB3
Gearbag: High Sierra Corkscrew
Photo rig: Nikon D70, 18-70mm/3.5-4.5, SB-800
MD transfers: MZ-RH1

Offline nardo

  • Trade Count: (0)
  • Taperssection Regular
  • **
  • Posts: 232
    • The Comatorium - The Mars Volta News
Re: master md's>analog out>24 bit???
« Reply #33 on: August 23, 2007, 11:15:31 AM »
This was just covered in another thread.  It was also researched on minidisc.org and discovered to not be true.  A fellow there xferred his files and checked the file(s) in Adobe Audition and saw no degradation as has been described.  Once again, one test is worth a thousand opinions.  Use the RH1 to U/L all your legacy MD's with no fear of having them degraded.

As for saving them as 24 bit; well, if you care to manipulate them on the PC I understand it is a better way to do it, in 24 bit.  It will not improve the sound but will add fewer artifacts during manipulation.  You will have to resample and dither back to 16 bit.

Cheers
I seriously wish I could believe this because it would save us all a lot of time. Did you read my posts carefully? http://taperssection.com/index.php/topic,76899.msg1027426.html#msg1027426
I never said SP/LP2 files are being converted to ATRAC3+ as a standard or as an option that I accidentally might have selected (I surely didn't), SS does never inform you about this step, when you convert to WAV your output files are .WAV but they have been converted to ATRAC3+ (no idea on the bitrate) prior to being converted to .WAV, apparently that is the current workaround for the ATRAC-for-the-PC codec as was stated by someone on the German minidisc forum.
This discussion has nothing to do with HiMD recordings, transfers of those are all lossless and no one has claimed otherwise anywhere, just to make this clear once and for all (not saying anyone in this topic did say so, just trying to keep the information as straight as possible in here).

I've done frequency analysis and spectral analysis on files transferred via standalone deck (TOSlink > TOSlink connection) and compared them to the RH1>USB transfers and there were obvious differences, referred to as cut-offs in prior posts, most noticeably on SP recordings, not so bad with LP2 recordings (which all seem to have drop-offs at around 16kHz, as is obvious since the bitrate is really low, 128kbps, I think). You might want to give this a try yourself if you own a deck - I also read some of the minidisc.org discussion and couldn't open the attachment, other than that I just saw someone who "believed" something. Again, try this for yourself and post the results, I'd love to see a new version of SS/the ATRAC codec handle this differently. I'm on 4.2 still afair (am at work right now so can't check).
Shortly after this was being discussed on ts.com the DSEE feature was introduced to SonicStage (coincidence?), it adds something similiar to DSP to the playback of files, thereby making them sound/appear fuller than they were in the first place. So you might get even different results when playing back your .oma in SS and recording the digital output of your soundcard (internal or external) than when doing a straight transfer via USB or via a deck.

Hey nardo,

+T for researching this topic!
I'd be really grateful if you could take screen shots of the side-by-side analysis you did (optical realtime tranfer vs. RH1>USB>SS). Why don't you post your findings here: http://forums.minidisc.org/index.php?s=&showtopic=19709&view=findpost&p=125800

It would be really interesting to hear what the guys at Minidisc.org has to say about your results, if you descibe your method in detail.
Their theory is that you didn't highlighting the entire waveform in the edit window, clicking "scan" in the frequency analysis window. This is what caused the erroneous analysis.

Thanks,

Jan
Highlight a waveform? I used AnalFreq for that, it's a realtime analysation software, you can "watch" the frequency spectrum. I did spectral view using CEP, EAC and Wavelab but was mostly interested in the frequency analysis. I am at work at the moment and probably won't be able to take screenshots until tomorrow. I'll try and get one off an SP and an LP2 show and will at least post them here if not over at minidisc.org, too.

A little personal tidbit, I was extremely excited over being able to transfer legacy MDs faster than real-time using the RH1. Then one of my friends posted one of his LP2 masters to a torrent site - and the torrent was promptly banned for being "mp3 sourced". That's why I even started looking into this. The show (a Living Colour recording) did look perfectly fine when transferred via my MDS-JB980 deck.

Offline boojum

  • Trade Count: (1)
  • Needs to get out more...
  • *****
  • Posts: 3633
  • Gender: Male
Re: master md's>analog out>24 bit???
« Reply #34 on: August 23, 2007, 12:50:28 PM »
Have you tried posting your transfer of Living Colour???
Nov schmoz kapop.

Offline sunjan

  • Trade Count: (0)
  • Needs to get out more...
  • *****
  • Posts: 2003
  • Gender: Male
  • Taping since 1988, 28 years of fine recordings...
    • Just a handful of stuff I put on etree
Re: master md's>analog out>24 bit???
« Reply #35 on: August 24, 2007, 05:12:06 AM »
Highlight a waveform? I used AnalFreq for that, it's a realtime analysation software, you can "watch" the frequency spectrum. I did spectral view using CEP, EAC and Wavelab but was mostly interested in the frequency analysis. I am at work at the moment and probably won't be able to take screenshots until tomorrow. I'll try and get one off an SP and an LP2 show and will at least post them here if not over at minidisc.org, too.

A little personal tidbit, I was extremely excited over being able to transfer legacy MDs faster than real-time using the RH1. Then one of my friends posted one of his LP2 masters to a torrent site - and the torrent was promptly banned for being "mp3 sourced". That's why I even started looking into this. The show (a Living Colour recording) did look perfectly fine when transferred via my MDS-JB980 deck.

Hi again nardo,

I get the impression the guys at minidisc.org are using either Adobe Audition/CEP or Audacity for their frequency analysis screenshots. Hence the reference to "highlighting the waveform".
I'm extremely excited about the possibility too. I've got a deck with optical out, but I found too tedious with realtime transfer of 200+ hours of MD masters. Under the current circumstances, I'm hesitant to shell out $350 for a RH1 only to use for legacy transfers, if I can't be sure that it's superior.

/Jan
Mics: A-51s LE, CK 930, Line Audo CM3, AT853Rx (hc,c,sc),  ECM 121, ECM 909A
Pres: Tinybox, CA-9100, UA5 wmod
Recorders: M10, H116 (CF mod), H340, NJB3
Gearbag: High Sierra Corkscrew
Photo rig: Nikon D70, 18-70mm/3.5-4.5, SB-800
MD transfers: MZ-RH1

Offline nardo

  • Trade Count: (0)
  • Taperssection Regular
  • **
  • Posts: 232
    • The Comatorium - The Mars Volta News
Re: master md's>analog out>24 bit???
« Reply #36 on: August 24, 2007, 07:58:32 AM »
Highlight a waveform? I used AnalFreq for that, it's a realtime analysation software, you can "watch" the frequency spectrum. I did spectral view using CEP, EAC and Wavelab but was mostly interested in the frequency analysis. I am at work at the moment and probably won't be able to take screenshots until tomorrow. I'll try and get one off an SP and an LP2 show and will at least post them here if not over at minidisc.org, too.

A little personal tidbit, I was extremely excited over being able to transfer legacy MDs faster than real-time using the RH1. Then one of my friends posted one of his LP2 masters to a torrent site - and the torrent was promptly banned for being "mp3 sourced". That's why I even started looking into this. The show (a Living Colour recording) did look perfectly fine when transferred via my MDS-JB980 deck.

Hi again nardo,

I get the impression the guys at minidisc.org are using either Adobe Audition/CEP or Audacity for their frequency analysis screenshots. Hence the reference to "highlighting the waveform".
I'm extremely excited about the possibility too. I've got a deck with optical out, but I found too tedious with realtime transfer of 200+ hours of MD masters. Under the current circumstances, I'm hesitant to shell out $350 for a RH1 only to use for legacy transfers, if I can't be sure that it's superior.

/Jan
I used CEP this time, highlighted the entire wav, opened the Frequency Analysis and hit "Scan". The difference seems to be less obvious compared to e.g. AnalFreq where you can see in real-time that there are a lot more peaks in the deck-transfer as well as more "happening" in the frequencies above 14kHz, there's not just a constant amount of noise/fuzz in case you were wondering.

Alright, first pic is a Keith Caputo (rock) show, Sanken COS-11s>Sharp MD recorded outdoors in SP mode, JB980QS TOSlink transfer:

Same recording but transferred via RH1, converted to WAV using HiMD Renderer:


Now I understand that in the SP ATRAC there won't be much info recorded above 14-15 kHz, yet there is a difference in the freq analysis (even though at very low db levels - maybe there is some kind of low-pass filter at those higher frequencies to "clean up" the sound a little and get rid of what could be considered mostly useless information? doesn't change the fact that it seems to alter the sound) and when playing back these files the RH1 does seem, to my ears, thinner, or sterile even. The bass seems less pronounced than when transferred via standalone-deck, which is actually weird since the low frequencies do look pretty much the same in both ways of transferring but my ears tell me otherwise, I don't notice much of a difference in the higher frequencies or even the mids when listening to these files.

And like I said before, much less of a difference with LP2 recordings, both when looking at the frequency analysis as well as when listening back. This is from a Living Colour recording, outdoors, recorded using AT831b>Sharp MD, first pic is the deck transfer:

Here's the same recording transferred via RH1:


Any thoughts where those slight differences could stem from? Since both are digital transfers I would expect them to be even closer than they are.

Have you tried posting your transfer of Living Colour???
I haven't, no, he only had me transfer one song for comparison reasons, at the moment I only have that saved here with me but no worries the taper does plan on torrenting it (again, as a deck transfer) at some point.

<EDIT> I used the [img]-tag for the images but apparently they don't show up in my post, hence attached them to this post.</EDIT>
« Last Edit: August 24, 2007, 08:05:34 AM by nardo »

Offline Sparky191

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 1
Re: master md's>analog out>24 bit???
« Reply #37 on: August 24, 2007, 09:27:08 AM »
...
and when playing back these files the RH1 does seem, to my ears, thinner, or sterile even. ...

Wouldn't playback through the RH1 be influenced by the onboard amp.

Offline nardo

  • Trade Count: (0)
  • Taperssection Regular
  • **
  • Posts: 232
    • The Comatorium - The Mars Volta News
Re: master md's>analog out>24 bit???
« Reply #38 on: August 24, 2007, 09:50:02 AM »
...
and when playing back these files the RH1 does seem, to my ears, thinner, or sterile even. ...

Wouldn't playback through the RH1 be influenced by the onboard amp.
Should read "and when playing back these files the ones transferred via the RH1 does seem, to my ears, thinner, or sterile even."
I'm always listening to the transferred WAV files through the same exact stereo setup, not headphones.

Online yug du nord

  • uncleyug>Snowman>
  • Trade Count: (53)
  • Needs to get out more...
  • *****
  • Posts: 4876
  • made with natural flavor
Re: master md's>analog out>24 bit???
« Reply #39 on: September 01, 2007, 06:45:05 AM »
As far as I understand...  Sony's DSP (Digital Signal Processor) Type-R, produces the highest sound quality possible from an SP recording during playback.  DSP Type-S (which also retains Type-R), produces the highest sound quality possible from an MDLP (2 or 4) recording during playback.  So given that, I would tend to believe that the analog out of an MD player with DSP Type-R/S (depending what mode you're transferring) would provide a higher quality sound than a digital transfer (optical out or USB).  So... you may lose a little by transferring analog output, but the actual sound will be better.  That is what the DSP chips are for!  I know we all want digital outs, but with legacy discs and a DSP Type-S player...  that's as good as it's gonna get!  Unless Sony comes out with something better!!!  Just don't transfer while the MD's EQ is active!  If you're gonna EQ......  do it on the computer!!!

As far as recording goes with ATRAC compression...  ATRAC is used for SP mode, ATRAC3 is used for MDLP (2/4) mode, and ATRAC3plus is used for Hi-SP & Hi-LP mode.

Never done a comparison, so I may be wrong...  but I don't think so!!! :nightfevah:   
.....got a blank space where my mind should be.....

sml42

  • Guest
  • Trade Count: (0)
Re: master md's>analog out>24 bit???
« Reply #40 on: September 01, 2007, 07:12:30 PM »
So given that, I would tend to believe that the analog out of an MD player with DSP Type-R/S (depending what mode you're transferring) would provide a higher quality sound than a digital transfer (optical out or USB).

Disagree.

Where does the analogue out  come from? It comes from some analogue-to-digital converter inside the player. Where does the digital signal come from? the output from the DSP. The optical output also comes from the DSP...

...unless of course the optical output was truncated to 16-bit and the ADC was fed a 24-bit signal.

I've just had a look at the schematics for my sony mds-je470 (standalone player, atrac type-R), and indeed the DSP does give separate signals to the optical out and the ADC. Unfortunately my scope is out on loan at the moment, so I can't do any measurement... I'd be very interested to find out if there is any difference though... hmm!

best regards,
stephen

Online yug du nord

  • uncleyug>Snowman>
  • Trade Count: (53)
  • Needs to get out more...
  • *****
  • Posts: 4876
  • made with natural flavor
Re: master md's>analog out>24 bit???
« Reply #41 on: September 01, 2007, 10:45:59 PM »
So given that, I would tend to believe that the analog out of an MD player with DSP Type-R/S (depending what mode you're transferring) would provide a higher quality sound than a digital transfer (optical out or USB).

Disagree.

Where does the analogue out  come from? It comes from some analogue-to-digital converter inside the player. Where does the digital signal come from? the output from the DSP. The optical output also comes from the DSP...

...unless of course the optical output was truncated to 16-bit and the ADC was fed a 24-bit signal.

 :tomato:  I guess the analog output would come from the DA section of an AD/DA (analog to digital/digital to analog) converter that is inside the player.  And I guess that the digital signal would come from the minidisc itself.  Which if analog out...  would have to go through the DA converter, or if digital out...  no DA would be needed.  It would stay digital!  So the DSP may be active with both analog and digital outputs?  I don't have a deck with an optical out, so I guess I'm not too sure!  So if that's correct...  USB output would still be the least quality in sound?   

An MD recorder is only 16 bits, so the discs would have already been truncated or dithered.  So I don't think that's an issue for this situation. 

This is an interesting question...  something I've wondered for a long time, since I have over 100 legacy discs to deal with.  Some are in LP2 format, and since there is no deck that has an optical output and supports DSP Type-S (at least not that I know of), the analog out on my Hi-MD seems the best for my situation.  But for SP discs, maybe a DSP Type-R deck with an optical output would be better?  That is, if the DSP is active through the optical output!  Either way, it's sounding like using the DSP would be superior to transferring through USB!!!  What do ya think???  I like this minidisc talk...  not enough of if around!!! :realhappy:   And I might be completely wrong about any/all of this...  I know what I think I know, but what I think I know, I might not know!!!   ;D     
« Last Edit: September 01, 2007, 11:01:52 PM by uncleyug »
.....got a blank space where my mind should be.....

Offline boojum

  • Trade Count: (1)
  • Needs to get out more...
  • *****
  • Posts: 3633
  • Gender: Male
Re: master md's>analog out>24 bit???
« Reply #42 on: September 02, 2007, 01:13:14 AM »
How you transfer your legacy files is of no matter to me.  It is your trip.  However, I uploaded an SP file recorded on an MZ-N1 about three years ago to my PC.  I did it with my RH1 through SS.  The intermediary files was an OMA and the out put was a WAV.  The charts with the frequency analysis are in another thread.  Whether the chart was regular or logarithmic the frequency response was well up to ~18KHz which is keeping for what SONY says SP freq range is.  Had it run through this intermediary degradation which so many talk about it would have been degraded by ~4 or 5 KHz.  There is little low end in the freq scans as the recording is an interview so what you have is two tenor or baritone voices talking.

As for 24 bit conversion.  Well, you will have a 24 bit copy of a 16 bit file.  Scan at the highest resolution a fuzzy picture and you will still have a fuzzy picture.


Cheers
« Last Edit: September 02, 2007, 01:16:08 AM by boojum »
Nov schmoz kapop.

Online yug du nord

  • uncleyug>Snowman>
  • Trade Count: (53)
  • Needs to get out more...
  • *****
  • Posts: 4876
  • made with natural flavor
Re: master md's>analog out>24 bit???
« Reply #43 on: September 02, 2007, 01:38:56 AM »
How you transfer your legacy files is of no matter to me.  It is your trip. 

boo...  I'm with ya on that!  But...  I do want to squeeze as much juice out of them ol' apples as I possibly can!!!  I wouldn't think that there could be too much difference though!  If only Sony would cough up some facts about some of these questions.  May MD live long!!!

                                                       :drunk:
                                                       :iamwithstupid:   
.....got a blank space where my mind should be.....

Offline nardo

  • Trade Count: (0)
  • Taperssection Regular
  • **
  • Posts: 232
    • The Comatorium - The Mars Volta News
Re: master md's>analog out>24 bit???
« Reply #44 on: September 02, 2007, 07:05:18 AM »
Some are in LP2 format, and since there is no deck that has an optical output and supports DSP Type-S (at least not that I know of)
My deck, the MDS-JB980QS, does have DSP Type-S and optical-out (as well as coax-out and analog out).

 

RSS | Mobile
Page created in 0.275 seconds with 37 queries.
© 2002-2017 Taperssection.com
Powered by SMF
Website Design by Foxtrot Media, Inc., a Baltimore Website Company