Denon PCM Encoding in 1970s. Is it different than Sony CD PCM?

Discussion in 'Audio Hardware' started by Mr Bass, Aug 7, 2016.

  1. crooner

    crooner Well-Known Member

    Location:
    San Diego, CA
    Sounds super clean, which is what was expected of digital at the time. Almost antiseptic with quick fades at the end of the songs to hide microphone and studio noise. It is overall well recorded although a little weak in the bottom end. The faults I find are of the very early Mitsubishi digital recorder and perhaps the CD transfer using the dreaded sample rate converter. It is also mastered a little low in volume, which makes it sound thin unless cranked. If it were made today it would sound much more spacious...
     
    Last edited: Aug 14, 2016
    MonkeyMan likes this.
  2. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
    This is why I am concentrating on pre Redbook digital source LPs. The introduction of Redbook complicates the comparisons and introduces more bad digital than good. Even the Redbook sourced LPs suffered from it in the 80s.
     
    crooner likes this.
  3. I do not have this on LP but I have it on CD from the early 80's. Sounds good even with SRC. Nice tone on the solo flute, although a pretty dry recording with a bit of natural decay and reverb.
     
    crooner likes this.
  4. testikoff

    testikoff Forum Resident

    This 1972 PCM Jean-Pierre Rampal recording of Telemann's 12 Flute Sonatas has been also re-issued on Denon Blu-Spec CD in 2010 (with no CD pre-emphasis).
     
    Mr Bass likes this.
  5. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
    I found a reference to another Denon digital recorder the DN-035R.It was introduced in 1979. The earlier 034R appeared in 1977 and the 023R in 1972 and 1974.

    (Specifications)
    Modulation pulse code modulation
    Coding :16-bit 2's complement code
    Transmission clock frequency: 3. 0713 MHz
    Audio sampling frequency: 47. 25 kHz (44—54kHz selectable)
    Transmission waveform: standard TV signal
    Number of audio channels: 4/2 selectable
    Usable tape transport:U-matic VTR4-head-type VTR
     
  6. crooner

    crooner Well-Known Member

    Location:
    San Diego, CA
    Cool! I knew Denon had kept up with advances in digital technology and had a 16 bit library of recordings available when CD was launched...

    On the other hand, I am sure anything recorded by Denon after 1979 is 16 bit including LP releases.
     
  7. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
    What is interesting about the DN 035R is that it has a selectable sampling rate range from 44-54kHz. So unlike the Soundstream or 3M they could have made 44.1/16 recordings.
     
  8. crooner

    crooner Well-Known Member

    Location:
    San Diego, CA
    Which probably explains why so many of these Denon CDs sound so good!
    The Mitsubishi recorders were 48 kHz, even well into the 1980's.
    The EIAJ tried to standardize bit depth, but did not do much for sampling rates. The 14 bit EIAJ standard was pretty much history by 1981...
     
  9. sunspot42

    sunspot42 Forum Resident

    Location:
    San Francisco
    Well, you can thank Sony for killing the 14-bit standard.
     
  10. EddieVanHalen

    EddieVanHalen Forum Resident

    I've always wanted to know the models and specs of the Mitsubishi digital recorders Michael Jackson's Bad was recorded with.
     
  11. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
    We can also thank them for the blasted Redbook and their lousy encoders.
     
  12. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
    I'm not sure about Bad but the pre Redbook Mitsubishi was the X 80, a largely forgotten model. This is all I could find on it from several sources. I think the X-800 followed it.


    The Mitsubishi X-80 2-track 1/4 inch digital recorder from 1980 predated the ProDigi format and has many similarities, although it used an unusual 50.4 kHz sample rate, and is not directly compatible. However, Mitsubishi did build the capability to play back tapes created on an X-80 into the X-86 series machines. Only 200 X-80's were manufactured. Then, Sony created a 24track digital recorder where the Mitsubishi Corporation created a different 32 track digital recorder. The Mitsubishi recorded their data differently and it could be edited, the old-fashioned analog way, with a razor blade and splicing tape(X-80 left which was the last tape editable digital recorder). The Sony used 1/2 inch tape whereas the Mitsubishi used 1 inch wide tape.

    [​IMG]
     
    EddieVanHalen likes this.
  13. MrRom92

    MrRom92 Forum Supermodel

    Location:
    Long Island, NY
  14. EddieVanHalen

    EddieVanHalen Forum Resident

    Thank you! If I'm not mistaken Bad was recorded a year (or two) before Mitsubishi's ProDigi hit the market but I think I read somewhere some time ago it used a sampling rate of around 50 Khz. That may be the main reason (higher sampling rate, less impact of filtering on the audio range) why Bad sounded different to the typical Sony PCM recorded album. I also think that being Michael Jackson at the top of his game he'd use what he, Jones or Swedien considered what was the best equipment available equipment at the time. Maybe I'm all wrong...
     
  15. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
    Mitsubishi Digital recorders

    According to the Asst Engineer on Bad they used the X850 not the x80. I believe the X800 series was ProDigi and was Redbook.

    Assistant Engineer Russ Ragsdale: “Everything gets recorded in super duper S T E R E O. At one point, during the making of ‘Bad’, I saw two Mitsubishi 32-track X-850 digital machines, one Studer 24-track, and a one-inch 16-track machine, all synchronized together to play in sync….what an experience.” (Source: In The Studio With Michael Jackson, p. 72).


    Here is a review of the earlier X80 which provides some technical info. Note that the Mitsubishi X80 like the Soundstream used Analogic converters.

    Engineer Review from 1999

    Over the past 18 years I have recorded no less than 60 albums on the Mitsubishi X-80 and I always love the way it sounds. Designed by audiophile Kunimaro Tanaka, the X-80 is totally discrete, Class A solid state. Passive filters with a corner frequency of 25 kHz and huge high-quality input and output transformers are part of the reason this machine sounds good.

    I recently mastered a project that was recorded directly to the X-80 in 1983Sonically, the X-80 holds up to any of the new 24-bit PCM systems. The actual A/D and D/A converters were made by Analogic and are built into large metal enclosures mounted directly onto the PC boards, next to the passive filter enclosures. The filters contain lots of inductors to get the kind of slope required for anti-aliasing.

    The X-80 is designed to look and feel like an analog recorder, with a heavy-duty, die cast transport, oversized reel and capstan motors and disc brakes. The electronics for the transport are mounted in a card cage below the deck, while the analog and PCM electronics are housed in a separate chassis.

    The transport and the electronics are interconnected by four locking multipin cables. Balanced XLR connectors interface the X-80 with the outside world while a pair of RCAs are provided for additional unbalanced pretransformer outputs.

    Two large VU meters let you look at the RMS level while peak reading LEDs show the absolute digital level (red is over). Both input and output levels are controlled by precision stepped attenuators (not pots). A four-position interlocking push-button switches input, EE (A/D, D/A), reproduce and analog (for scrubbing). Input can be compared with EE or Repro while recording, which is a nice feature that few products have today. A headphone jack with level control also lives on this same front panel.
     
    crooner likes this.
  16. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
    It's interesting but I can't find any other reference to the so called Mitsubishi Digital Mastering System other than this album. Of course Mitsu did have digital recorders by 1982, the X80 and derivatives. Also the X800 was introduced mid year 1982. It would have been nice if hey had used a less idiosyncratic term so we would know what they actually used.
     
    crooner and 56GoldTop like this.
  17. EddieVanHalen

    EddieVanHalen Forum Resident

    It seems like MJ's Bad was recordedat 44.1 or 48 Khz after all. Despite this Bad sounds very different to the typical Sony PCM recording of the era. The way it was engineered may be one of the reasons why, the overall quality (and superiority) of the Mitsubishi digital recorders may be another one.
    Next question I also wonder is related with bit depth. I think it was 1990 when I saw the first CD from Sony Music stating that the original recording was done with 20 bit equipment and mastered with Super Bit Mapping to get the most out of these 20 bit recordings on a 16 bit media as CD. I know Super Bit Mapping is just a way of dithering and noise shaping, one among many available. Has anybody any idea when was the first over-16 bit (18, 20 or whatever) recording made? The first recording I know being done to 24bit is the soundtrack for Independence Day (actually only mixed down as it was tracked to analog) but sure 24 bit digital recording was tested before 1996.
     
    Plan9 likes this.
  18. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
    All I can offer as possible explanation for the subjective impression of difference is the extensive and lengthy studio production that went into Bad. Most albums were not worked over to that extent. Since Bad is very synth based, there is no sonic penalty for constantly shaping and tweaking the sound. I mean that relative to a standard acoustic instrument where we know what it sounds like in a room. What I heard on Bad was a meticulously almost obsessively arranged series of songs. So the typical sound of say a digital synth played in the studio and processed through Sony style ADCs was modified and constantly tweaked and EQed. With enough of that the typical sonic signature might well be removed.
     
  19. EddieVanHalen

    EddieVanHalen Forum Resident

    Another album came to mind, it is Gloria Estefan 1989's Cut Both Ways which was recorded digitally (no brand or model are stated) at Criteria Recording and Mastered by Bob Ludwig at Masterdisk. I haven't listened to it in years but I remember it sounding very digital, ultra clean and dinamic but also very aseptic, but not dry and sterile as the typical Sony PCM recordings of the era, I wonder what kind of equipment was used to record it. If we compare it with another digital recording of the same year, Madonna's Like A Prayer, they sound like day and night. I've never been a fan of how LAP sounds, it doesn't sound harsh to my ears, but it does sound dry and "mechanic", and it's not an ultra-clean sounding album IMO. And I think in 1989 Madonna had more resources and a higher available budget for her albums than Gloria Estefan.
     
  20. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
    There is a reason I based this thread on my reaction to an acoustic (classical music) digital LP. If I had had the same reaction to a non acoustic pop album I wouldn't know where the source of the sonic advantage would lie. With respect to pop albums from the 80s to now the actual recording (and mic'ing) of anything on the album is probably way down the list of influences on the final sound. With acoustic instruments we have an independent basis for judging accuracy. If I hear something more realistic on a Denon digital LP of a string quartet compared to any redbook digital LPs I can more securely infer the recording process had a major effect rather than EQing and digital processing.
     
  21. EddieVanHalen

    EddieVanHalen Forum Resident

    You're right, with acoustic music performed with real instruments mic'ed one can be objective, with synth Pop music one can't be as there's no way to know what is due to the intention of the producer and what is distortion or other artifacts introduced by the recording equipment.
     
  22. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
    In reviewing the posts in this thread, some summary points come to mind:

    1. The pre Redbook recordings/LPs can have very good to excellent sonics. At least in some cases they display greater realism than many if not most Redbook sourced LPs
    2. The 14-15 bit restriction does not seem to be a barrier to creating an above average PCM recording.
    3. The build quality of the PCM converters from Denon, Soundstream, 3M and Mitsubishi far exceeded the Sony PCM Redbook and subsequent converters.


    In thinking about point 3, there are obvious considerations of audiophile design of the analog stages. However I wonder if the larger dimensions of these converters also was beneficial. I'm thinking in terms of the reduction of EMI by spacing circuits farther apart.
     
    McLover likes this.
  23. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
  24. sunspot42

    sunspot42 Forum Resident

    Location:
    San Francisco
    Interesting that the 3M deck is already listed there. I don't think it became commercially available until '78 or '79 though, did it?
     
  25. Mr Bass

    Mr Bass Forum Resident Thread Starter

    Location:
    Mid Atlantic
    Well you are talking about AES and digital standards. Any big player with a machine whether on the market or being assembled is going to send someone to the meetings. The info I've seen is that a prototype/test system was given to the Sound80 label in mid 1978. Given that this meeting took place in Dec 1977, it doesn't show much time between.
     

Share This Page