Become a Site Supporter and Never see Ads again!

Author Topic: Sharing a Show in 32/48?!?  (Read 619 times)

0 Members and 1 Guest are viewing this topic.

Offline cmstewart

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 33
Sharing a Show in 32/48?!?
« on: September 02, 2018, 03:20:06 AM »
As the title says, I had a strange idea to share a show in 32-bit float. I know this begs the question, why? It's obvious overkill and a waste of space on an HD. I may be nit-picking, but there is a potential reason in this case if someone would prefer the least-colored version of the show.

I discussed it in some detail elsewhere, but I was having some weird issues when exporting my 24-bit recording in audacity down to 24-bit and 16-bit. I say "down to 24-bit" because whenever you import a 24/48 file in audacity it automatically sets the track to 32-bit float for editing. And as I was advised, I leave it that way if I want to do a great deal of editing. When I exported with a dither, I discovered that for an unknown reason, peaks that were not previously clipping, now were. It's a multitrack recording so it's hard to pinpoint what the cause was when summing the tracks during export, but clearly some rounding along with the dither, etc. caused something to happen. Ultimately I just edited the peaks more to keep it clean.

So from this I concluded when I export, either I use a dither which (though very slightly) does alter the recording, or I don't use a dither which leaves it vulnerable to SBEs. But if I export at 32-bit float I can avoid both, and subsequently I can compress it down with WavPack. Which surprisingly compressed it to a tolerable size. 16/44.1 flac = 700Mb; 24/48 flac = 1.27Gb. 32/48 wv = 1.48Gb. Much smaller than the 24/96 shows I've owned. I know I'm nit-picking and it's an odd idea, and it's true maybe no one would care either way, but these are the files I've been listening to for my own pleasure and sonically I have noticed some differences. I just wondered what others would think of it. If I shared it that way would anyone appreciate that, or would it get the opposite response? Are collector's priorities more on the size/format of the recording, or the purity of it? Is this issue too small for people to accept something so unconventional (since I've never seen a 32-bit recording shared before)?

Offline vanark

  • TDS
  • Trade Count: (29)
  • Needs to get out more...
  • *****
  • Posts: 6847
  • If you ain't right, you better get right!
    • The Mudboy Grotto - North Mississippi Allstar fan site
Sharing a Show in 32/48?!?
« Reply #1 on: September 02, 2018, 07:50:55 AM »
I would expect an odd response. Lots of folks use Audacity and I've never heard of anyone wanting to do what you are describing or having the challenges you describe. I don't think I've ever seen a show shared that way. It isn't its native format so I think your description of purity doesn't make a lot of sense to me.
If you have a problem relating to the Live Music Archive (http://www.archive.org/details/etree) please send an e-mail to us admins at etree(AT)archive(DOT)org or post in the LMA thread here and we'll get on it.

Link to LMA Recordings

Link to Team Dirty South Recordings on the LMA

Mics: AKG C480B/CK61, CK63 | Church Audio CA-11 (cards) (with CA UBB)
Pres: <empty>
Recorders: Tascam DR-60D | Sony PCM-M10

Offline noahbickart

  • phishrabbi
  • Site Supporter
  • Trade Count: (31)
  • Taperssection All-Star
  • *
  • Posts: 1888
  • Gender: Male
  • So now I wander over grounds of light...
Re: Sharing a Show in 32/48?!?
« Reply #2 on: September 02, 2018, 08:03:00 AM »
How could dithering create SBE problems?

This makes little sense to me. Put a brick wall limiter at .01db and share as normal. It can’t be more pure at 32bit because it wasn’t recorded that way.

And, if you’re serious about post production, perhaps use a more sophisticated DAW like Reaper or Logic.
Recording:
Capsules: Schoeps mk41v (x2), mk4v (x2), mk22 (x2), mk3 (x2), mk21 & mk8
Cables: 2x nbob KCY, 1 pair nbob actives, Darktrain 2 and 4 channel KCY extensions:
Preamps:    Schoeps VMS 02iub, Naiant Littlebox, Naiant IPA, Sound Devices Mixpre6
Recorders: Sound Devices Mixpre6, Sony PCM m10

Home Playback: Mytek DSD 192> McIntosh MC162> Eminent Tech LFT-16; Musical Fidelity xCan v2> Hifiman HE-4XX / Beyerdynamic DT880

Office Playback: Grace m903> AKG k701 / Hifiman HE-400

Offline Ronmac

  • Trade Count: (0)
  • Taperssection Regular
  • **
  • Posts: 170
  • Gender: Male
Re: Sharing a Show in 32/48?!?
« Reply #3 on: September 02, 2018, 08:09:20 AM »
What was the headroom available on the original recording? What processing did you do in Audacity at 32 bit?


Offline Gordon

  • Trade Count: (20)
  • Needs to get out more...
  • *****
  • Posts: 10616
  • Gender: Male
    • my list
Re: Sharing a Show in 32/48?!?
« Reply #4 on: September 02, 2018, 09:26:42 AM »
You mention it being multitrack. When mixing 2 or more sources it will raise the overall gain.  In wavelab I make my aud and and both -3db. More often than not when mixed they are -1-0db.
Neumann ak40 > Nick mod lc3 > Naiant PFA or km140 > Sound Devices MixPre-6

https://archive.org/details/fav-gordonlw

Offline cmstewart

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 33
Re: Sharing a Show in 32/48?!?
« Reply #5 on: September 03, 2018, 02:32:50 AM »
How could dithering create SBE problems?

This makes little sense to me. Put a brick wall limiter at .01db and share as normal. It can’t be more pure at 32bit because it wasn’t recorded that way.

And, if you’re serious about post production, perhaps use a more sophisticated DAW like Reaper or Logic.

I didn't say the Dither caused SBEs. I said if I DIDN'T use a dither then it's vulnerable to SBEs. I didn't need any features that Audacity doesn't have. Thank you though.



I would expect an odd response. Lots of folks use Audacity and I've never heard of anyone wanting to do what you are describing or having the challenges you describe. I don't think I've ever seen a show shared that way. It isn't its native format so I think your description of purity doesn't make a lot of sense to me.


My goal wasn't absolute purity or anything like that. No matter what bit rate I export in it's going to be converted to FLAC, SHN, etc. afterward. I'm just saying the 32 bit had less issues than the others during export. It was an odd situation and when I learned wavpack can compress 32 bit float wav, I thought -- why not. It's different, and I for sure am different... So it's fitting. But I never intended to share that copy. I just did it for me. I don't care either way, I just wanted to share it and hear other thoughts on the issue.


What was the headroom available on the original recording? What processing did you do in Audacity at 32 bit?


It was a very dynamic set. I had 2 stereo tracks and if I averaged it over the whole 110 minutes about 4-8 db of room (on each), but some songs were MUCH louder than others with peaks that (when mixed) would clip. Some songs that were acoustic had 8+ dbs of room. I mainly had to edit the levels to insure they weren't clipping when summed together. That was about 80-90% of the editing I did. I also did some eqing on certain parts that required it.



You mention it being multitrack. When mixing 2 or more sources it will raise the overall gain.  In wavelab I make my aud and and both -3db. More often than not when mixed they are -1-0db.

It was my first multitrack live recording so I didn't use a standard adjustment as you mentioned. I adjusted the entire show some, and then I went through it... especially on the louder songs... and edited down the peaks that still were clipping when I did a test export. I went song by song until when I mix & rendered the whole thing it was clear of any clipping. It wasn't till I exported in a new bit/sample rate that some new bits were clipping that weren't earlier. Can't explain why, but I ended up just re-adjusting the levels around those points to resolve it. It was pretty frustrating trying to find the cause.
« Last Edit: September 03, 2018, 02:43:11 AM by cmstewart »

Offline scb

  • Eli Manning should die of gonorrhea and rot in hell. Would you like a cookie, son?
  • Site Supporter
  • Trade Count: (10)
  • Needs to get out more...
  • *
  • Posts: 8586
  • Gender: Male
Re: Sharing a Show in 32/48?!?
« Reply #6 on: September 04, 2018, 12:45:52 PM »
I didn't say the Dither caused SBEs. I said if I DIDN'T use a dither then it's vulnerable to SBEs.

This makes no sense, for a variety of reasons.


Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 12597
  • Gender: Male
Re: Sharing a Show in 32/48?!?
« Reply #7 on: September 04, 2018, 05:05:00 PM »
Dither has no connection to file-split points and CDR sector boundaries.  Sector boundary errors are only an issue when splitting files intended for burning to redbook audio format CD.  The "split on sector boundaries" option incrementally adjusts the track cut points so that they will land on the physical CDR sector boundaries and not create an audible gap between tracks.  It is a file-cut-point option, and is not specifically bit-length or dither related, other than the file output format also needing to be 16 bit to be redbook compatible.

I adjusted the entire show some, and then I went through it... especially on the louder songs... and edited down the peaks that still were clipping when I did a test export. I went song by song until when I mix & rendered the whole thing it was clear of any clipping. It wasn't till I exported in a new bit/sample rate that some new bits were clipping that weren't earlier. Can't explain why, but I ended up just re-adjusting the levels around those points to resolve it. It was pretty frustrating trying to find the cause.

If you experience this problem in the future, I'd just go back and reduce the level of all sources being mixed together, each by the identical amount, until you get no clipping upon export. That will be far less work and will keep the output identical other than overall level. Select the whole track, each source in its entirety, to adjust the level of the entire thing at once.  If you lower all sources by the identical amount, their relative balance will be identical, only the overall level of the resulting mix will be lower.

If you have threshold-level-sensitive filters active on a source (such as compression or whatever which would be altered by lowering the level of the input file) save the individual sources as individual files first with that processing applied, so as to print that filtering to file without altering how the thresholds are tripped.  Then pull those back into a session with no filtering other than lowering the level of each source by the same amount, just for the summing operation so as to output without clipping.
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

Offline cmstewart

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 33
Re: Sharing a Show in 32/48?!?
« Reply #8 on: September 07, 2018, 04:35:25 AM »
Thanks for the advice, I definitely can see how that would be easier. Since this was my first entire-show recording I kind of became a perfectionist about it I guess and only wanted to affect what needed to be edited, and tried to leave alone what was already good. I'll just have to wait and see if it ever comes up again. And maybe try a new DAW next time (which will take some getting used to) and see how that helps resolve things. At least nothing was terribly wrong with it. The issues I had weren't critical. Just annoying at times.

I didn't say the Dither caused SBEs. I said if I DIDN'T use a dither then it's vulnerable to SBEs.

This makes no sense, for a variety of reasons.


.

Okay. If I know nothing about this teach me. I'm just reciting what I thought I was told here. SBEs were brought up in another thread. I read what they were, and I asked since I recorded in 24-bit, how to avoid SBEs if I want to reduce to 16-bit for CD quality. I was told to use a dither at export. Which told me logically that without a dither, SBE's at lease can be an issue. Sorry I got it wrong. I probably shouldnt be trying anyway.
« Last Edit: September 07, 2018, 04:45:49 AM by cmstewart »

 

RSS | Mobile
Page created in 0.22 seconds with 36 queries.
© 2002-2018 Taperssection.com
Powered by SMF