VideoHelp Forum
+ Reply to Thread
Results 1 to 9 of 9
Thread
  1. Member
    Join Date
    Nov 2006
    Location
    United States
    Search Comp PM
    Hello.

    Years ago I borrowed a buddie's D8 tapes, rented a Sony Walkman D8 player & copied the tapes via firewire & burnt to data dvd's (20 minutes each disc).

    Today I re-borrowed the same tapes and this time used a D8 camera via firewire to save the files to a larger storage drive (no breaking up the files in 20 min sections).

    When viewing the audio portion of the AVI files in MediaInfo, it shows the earlier files as PCM 16b 48k. When viewing the new copy of the files it shows PCM 16b 32k.

    Since I'm using firewire, isn't that just coping the files from the tape to a storage device ? I'm not capturing anything. And since the tapes are digital, shouldn't the quantization always end up the same as the source?
    Quote Quote  
  2. aBigMeanie aedipuss's Avatar
    Join Date
    Oct 2005
    Location
    666th portal
    Search Comp PM
    32 and 48 are both allowed. but 32kHtz is 12 bit not 16. see if you can change it to 48kHtz in settings.
    --
    "a lot of people are better dead" - prisoner KSC2-303
    Quote Quote  
  3. Member
    Join Date
    Nov 2006
    Location
    United States
    Search Comp PM
    You mean change the settings in the D8 camera? That would be for new recordings. I spent hours coping to the PC, then moving those files to an external device so I do not want to re-copy again just for 48k from 32k.

    But being how I copied the same tapes via firewire, even though they were with different players, shouldn't the quantization be the same no matter what any settings are? The audio quantization is baked in the tapes correct?
    Quote Quote  
  4. aBigMeanie aedipuss's Avatar
    Join Date
    Oct 2005
    Location
    666th portal
    Search Comp PM
    the recorded audio setting should be on the tape. not sure how it got messed with but maybe the deck was set to only export 16 bit 48k. 12 bit 32k was an option so that 4 audio tracks could be recorded per video. usually the normal stereo audio and a second stereo commentary track.
    --
    "a lot of people are better dead" - prisoner KSC2-303
    Quote Quote  
  5. Member Cornucopia's Avatar
    Join Date
    Oct 2001
    Location
    Deep in the Heart of Texas
    Search PM
    very good possibility that 16bit is just how the capping app wants to save the capture - to pad to 8bit boundaries.

    the true quantization is baked into the tape, for that particular clip. DV format of all stripes support 2ch 16bit at 48kHz (preferred) or 44.1kHz, or 4ch 12bit at 32kHz. Capping apps, however, often are only able to capture 2 ch at a time, so is quite possible to see a file w 2ch 12bit 32kHz.
    if padding to 16bit is occurring, it should be quite easy to tell by checking the bottom 4 LSBs of those files' audio - they would be all Zeros.

    also a good possibility that the way you capped it carried over throughout clip boundaries, and then there could be the possibility that those clips weren't consistent in their settings, so the app might not have known how to properly deal with them. if that is the case, best to cap again but only by individual clip segments, so that your files are all monotonic (as most containers only support monotonicity).


    Scott
    Quote Quote  
  6. Member
    Join Date
    Nov 2006
    Location
    United States
    Search Comp PM
    If it helps at all, the original 2008 copy was done with Pinnacle Studio (w/the Walkman) and the 2021 copy is with Vegas Pro and a D8 camera as the player. Keep in mind these are the exact same tapes used in both transfers (or copy; whatever the proper terminology for moving the information from the digital tape(s) to a PC).

    Attached is a tree from the new copy. I couldn't find the four LSBs you mentioned.
    Image Attached Images  
    Quote Quote  
  7. Member Cornucopia's Avatar
    Join Date
    Oct 2001
    Location
    Deep in the Heart of Texas
    Search PM
    You would look at LSBs (least significant bits, aka bits 12-15 of a 16bit sample) by exporting (demuxing) your audio to a 16bit Pcm WAV file, and looking at the bits in a hex editor to see if there are repeating patterns of Zeros. Or open in an audio editor and compare granularity (when zoomed all the way in) against a true 16bit file. Or better, if you have Audition you can look in the (Wav) file info stats to see the true bitdepth.

    There is quite a difference in quality between 16bit48k and 12bit32k, so it might actually be worth your while to test multiple clips with different systems to see what you truly have or ought to have.

    Do you have tapes where you have examples in both the old & new capture? play them back-to-back and you should be able to tell which setup got it right and which got it wrong (for that clip at least).


    Scott
    Quote Quote  
  8. Member
    Join Date
    Nov 2006
    Location
    United States
    Search Comp PM
    I have Sound Forge Pro 13, so I can zoom in quit a bit there.

    Yes, I do have files of a few of the same tapes from the old and new captures.

    Since MediaInfo shows Sampling rate of 32k, that means the Bit depth is 12? (it shows 16 here)
    Quote Quote  
  9. aBigMeanie aedipuss's Avatar
    Join Date
    Oct 2005
    Location
    666th portal
    Search Comp PM
    digital sampling at 32k limits the high frequency to a top end of 16khtz vs 24 khtz at 48k. 12 bit only has @ 4k discrete sample steps vs 16 bits 65k samples.

    it's possible to have 32k16bit, just not found often.
    --
    "a lot of people are better dead" - prisoner KSC2-303
    Quote Quote  



Similar Threads

Visit our sponsor! Try DVDFab and backup Blu-rays!