The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Could this be the Best way to mix ITBs? (Plural) Consoles
Old 2nd September 2011
  #31
Quote:
Why don't you route all your tracks to a stereo bus, create a new stereo track with the input on that bus and then hit record and export the recorded file after that.
If I understand what your suggesting, it would mean mixing using the a/d then d/a part of the converter. Although my converters sound pretty good with the black lion upgrade, I would rather not have them go through those 2 conversion processes which do color the sound and add errors of some sort. That is the reason I went through the process I described at the beginning of the thread which avoids using the a/d or d/a conversions.

Added: It seems others have experienced degradation on bounces as mentioned in forums but I am thinking it is very specific to certain mix elements that come together. It may be incompatible software or routing issues which are compounded with large CPU busting projects. I still have to go back to the problem projects when time permits later today. It's good to hear that most are finding bounces to work as advertised.
Old 2nd September 2011
  #32
Quote:
Originally Posted by AudioSoundzz View Post
I could have discovered a new and maybe important improvement in the way to mix ITBs. If this is old news please forgive my enthusiasm and making myself look like a fool. If this is new and news, I would love to get other people thoughts and possible experiments with the technique. First I would like to share a brief overview of my reasons for this discovery.

I am a rather old school engineer/producer. Started recording on a 2 track Revox and then moved up to a 4 track, then 8 track, then 16 track Tascams. Finally graduated to a 24 track Otari with a Trident board and too much outboard gear. I have recorded 100's of albums over the somewhat long years of my career. I have fought hard to keep the analog sound alive but at the same time invested in some pretty good Black Lion Converter upgrades, Wave plugins, TC Powercore 6000 and most recently the Wave MPX and Slates VCC. I am happy to report that the VCC and MPX have taken away the digital edge and added that stereo imaging and mid range fullness I was used to hearing off the board.

Finally the ITB mixes are sounding as good (to me) as my analog mixes and sometime even better.

But I found I still had a problem. My mixes are often large and also have many virtual keyboards so I am pushing my CPU and usually have to run the buffer at 1024. The sound of the mix coming out of the speakers is fantastic but when I bounce to disk the stereo imaging goes out the window and somehow some of the harsh digital sounds seems to come back which the VCC and MPX so elegantly improved.

Then I tried the obvious which was to play the DAW into the converter and record the song as it played in real time. Comparing this mix to the bounced version showed that the real-time playback through the converter a/d then d/a sounded a little better in the stereo imaging department but had added digital edginess. Not a slam dunk. Then I got to thinking about when I transfer DATS to the DAW, I use the AES/EBU out of the DAT into the converter which other than a few errors along the way is a pretty exact copy of the DAT and avoids going through converters. Could this be done for mixing off a DAW?

1. How I created a real-time digital mix (without bouncing and not going through any a/d or d/a conversion process) is thus:

2. Be warned: This takes 2 computers and 2 converters. Computer 1 is connected to converter 1 and computer 2 is connected to converter 2.

3. The 2 converters are best timed using the same Clock source (any good timer will do) for both DAWs. If you do not have a timer, then select converter 1 as the clock source in both DAWs (or visa versa - not really sure on this one).

4. Use a digital grade cable and connect the AES/EBU output of converter 1 into the AES/EBU input of converter 2.

5. Computer 1 is for playback and Computer 2 is for recording. Make sure your DAWs on the 2 computers can see the AES/EBU connections.

6. Set all the main DAW mix track outputs in computer 1 to be AES/EBU. Set the input of the computer 2 to be AES/EBU. Both DAWs need to be set at the same sample and bit rate that the project is in.

7. Set the playback computer (1) at the highest buffer setting available. From my experience, the higher the setting, the better the playback quality.

8. Set the recording computer (2) and the lowest buffer setting. From my experience, this gives the best recording quality.

9. Finally start recording on computer 2 and then start playback off of computer 1.

10. This new recorded mix has avoided any a/d or d/a conversions, the pitfalls of bounces, and from what I hear sounds exactly like the original mixing coming off the DAW.

If I am off my rocker on this please let me know but I am liking what I am hearing.

invincible media - less hassles - more music
if it's in the groove look no further
So... just what do you imagine is being accomplished here -- aside from making the system more complex and introducing more jitter from the added transfer and slaving of one clock to another?

Your DAW is still summing the tracks to a mix. That mix is still being written to disk. The only thing is the added Rube Goldberg elements -- and they are more likely to degrade the sound.
Old 2nd September 2011
  #33
Quote:
So... just what do you think is being accomplished here? Aside from making the system more complex and introducing more jitter from the added transfer and slaving of one clock to ahtoher?
I would think a bounce still needs to use the clock (not sure on this though), so therefore no extra jitters would be added as I use 1 clock for both converters. If some extra are added it still seems better than going through the a/d and d/a.

The projects that I needed this for would not bounce correctly so I had to create a work around. I am looking into the reasons for the bad bounces. I initially thought is was just do to the fact they are very huge and CPU intensive but now I am thinking it's a combination of things as I mention in my last post. The good news is I got great and on time mixes with 2 problem songs using the technique I outlined and the client is happy.
Old 2nd September 2011
  #34
007
Lives for gear
 
007's Avatar
 

Quote:
Originally Posted by 007 View Post
OP...

Are you adding a Dither plugin on your master fader before you bounce?
yes? no?
Old 2nd September 2011
  #35
Quote:
Are you adding a Dither plugin on your master fader before you bounce?
yes? no?
No dither was added: On the master fader I was using the TC Powercore MD3 Multi Dynamics, Brick Wall Limiter and Waves SSL Master Buss Compressor. The Brick Wall Limiter has a dither function but that was turned off. The tracks were being bounce at the same sample and bit rate that they were recorded at (48k 24 bit). I usually dither only when changing the bit rate. Hope that helps.
Old 2nd September 2011
  #36
007
Lives for gear
 
007's Avatar
 

Quote:
Originally Posted by AudioSoundzz View Post
No dither was added: On the master fader I was using the TC Powercore MD3 Multi Dynamics, Brick Wall Limiter and Waves SSL Master Buss Compressor. The Brick Wall Limiter has a dither function but that was turned off. The tracks were being bounce at the same sample and bit rate that they were recorded at (48k 24 bit). Hope that helps.
Even if you are bouncing at the same bit depth, try a run with Dither turned on.
Old 2nd September 2011
  #37
Gear Maniac
 

I am a Logic user and have always contended that Bouncing to Disk is hit and miss...I either use a summer or assign everything to a bus arm a track with that bus as the input and print right into the session.

To the OP...I wouldn;t get involved in the argument that your are incorrectly Percieving differences...it most certainly happens with complex mixes that are taxing the system and can easily be proved by throwing the mix into your session lining it up to the sample reversing polarity and listening to the resulting sound...and there will be a lot of the time..

My buffer is always set to 1024...I track using hardware monitoring or a console...any thing over 30 samples of latency while tracking is unnacceptable.

Buffer settings can most definitely affect your mix if your system is being taxed any number of errors can occur not just clicks and pops...but seriously aren;t clicks and pops bad enough?

Computers make errors all the time...disks spinning 72000 times per minte 44100 samples every second times 60 tracks times 5 minutes 60 plug ins or more...accessing 100 regions fades automation etc...summing....c'mon people...they make mistakes...all the time actually

I don;t fight that part I just print it again if it doesn;t sound right

Some folks bounce 2 times in a row because it can be so innacurrate...in fact there is a setting in Logic for this very purpose

If you aren't actually mixing and are just playing 12 tracks and bounce that, then yes...it'll be the same...but mixing can get pretty intensive ITB.


I'm gonna try this actually...sounds interesting as the "mix deck" doesn;t have to do really any work at all...however I'd be careful with the clocking and make sure that the "mix deck" handles external clock well....they both HAVE to use the same clock as well.
Old 2nd September 2011
  #38
Gear Maniac
 

Quote:
Originally Posted by AudioSoundzz View Post
If I understand what your suggesting, it would mean mixing using the a/d then d/a part of the converter. Although my converters sound pretty good with the black lion upgrade, I would rather not have them go through those 2 conversion processes which do color the sound and add errors of some sort. That is the reason I went through the process I described at the beginning of the thread which avoids using the a/d or d/a conversions.
No, it's just internal routing. No AD/DA conversion
Old 2nd September 2011
  #39
Quote:
I am a Logic user and have always contended that Bouncing to Disk is hit and miss...I either use a summer or assign everything to a bus arm a track with that bus as the input and print right into the session.

To the OP...I wouldn;t get involved in the argument that your are incorrectly Percieving differences...it most certainly happens with complex mixes that are taxing the system and can easily be proved by throwing the mix into your session lining it up to the sample reversing polarity and listening to the resulting sound...and there will be a lot of the time..

My buffer is always set to 1024...I track using hardware monitoring or a console...any thing over 30 samples of latency while tracking is unnacceptable.

Buffer settings can most definitely affect your mix if your system is being taxed any number of errors can occur not just clicks and pops...but seriously aren;t clicks and pops bad enough?

Computers make errors all the time...disks spinning 72000 times per minte 44100 samples every second times 60 tracks times 5 minutes 60 plug ins or more...accessing 100 regions fades automation etc...summing....c'mon people...they make mistakes...all the time actually

I don;t fight that part I just print it again if it doesn;t sound right

Some folks bounce 2 times in a row because it can be so innacurrate...in fact there is a setting in Logic for this very purpose

If you aren't actually mixing and are just playing 12 tracks and bounce that, then yes...it'll be the same...but mixing can get pretty intensive ITB.


I'm gonna try this actually...sounds interesting as the "mix deck" doesn;t have to do really any work at all...however I'd be careful with the clocking and make sure that the "mix deck" handles external clock well....they both HAVE to use the same clock as well.
It's interesting your hearing bounce mix issues as well on large mixes. I have a lot of midi tracks playing the virtual keys and the timing of midi to digital is always off. Most DAWS have an offset adjustment to try and fix the problems. And yes your correct both converters have to use the same clock. I have a black lion clock I use for the 3 converters I use.

Quote:
Originally Posted by AudioSoundzz
If I understand what your suggesting, it would mean mixing using the a/d then d/a part of the converter. Although my converters sound pretty good with the black lion upgrade, I would rather not have them go through those 2 conversion processes which do color the sound and add errors of some sort. That is the reason I went through the process I described at the beginning of the thread which avoids using the a/d or d/a conversions.
Quote:
No, it's just internal routing. No AD/DA conversion
I have no idea how this could be done as buss tracks and the master tracks cannot record.
Old 2nd September 2011
  #40
Gear Maniac
 

Quote:
Originally Posted by AudioSoundzz View Post
I have no idea how this could be done as buss tracks and the master tracks cannot record.
Instead of routing your tracks to your main stereo analog out, change the output of every track (including your master) to bus 1-2 for example (if it's not used), create a stereo audio track, set the input to bus 1-2 and record. Easy. Then after you just have to export the file.
Hope this helps
Old 2nd September 2011
  #41
Lives for gear
 
Mark D.'s Avatar
 

Please post both versions (internal, and PC to PC) so we can hear it.
Old 3rd September 2011
  #42
Quote:
nstead of routing your tracks to your main stereo analog out, change the output of every track (including your master) to bus 1-2 for example (if it's not used), create a stereo audio track, set the input to bus 1-2 and record. Easy. Then after you just have to export the file.
Hope this helps
I tried that out and it works great! Guess my elaborate way of doing the real time mix with 2 computers was not needed at all. So thank you Ambroise for your insights on doing a real time mix!

I have figured out one problem. I am using MOTU's BPM software and I am finding if it's in song mode and I bounce it to disk timings of individual instruments go off. Not just a little but way off. When I play if from midi files it seems to bounce fine. It's just a problem in song mode. The real time recording technique will save these projects until Motu can fix the problem.

Quote:
Please post both versions (internal, and PC to PC) so we can hear it.
I could post the BPM problem as that is a project of my own but that is very obvious. The other problem song of recent is for a client and I am not privileged to make it public. I will keep my eyes open a song that I can post with issues of the sound not being as good as the mix playback.
Old 3rd September 2011
  #43
Quote:
Originally Posted by AudioSoundzz View Post
I tried that out and it works great! Guess my elaborate way of doing the real time mix with 2 computers was not needed at all. So thank you Ambroise for your insights on doing a real time mix!

I have figured out one problem. I am using MOTU's BPM software and I am finding if it's in song mode and I bounce it to disk timings of individual instruments go off. Not just a little but way off. When I play if from midi files it seems to bounce fine. It's just a problem in song mode. The real time recording technique will save these projects until Motu can fix the problem.

I could post the BPM problem as that is a project of my own but that is very obvious. The other problem song of recent is for a client and I am not privileged to make it public. I will keep my eyes open a song that I can post with issues of the sound not being as good as the mix playback.
I think I posted the exact same thing on the 3rd or so reply...this is commonly known as recording to Fisk, instead of bouncing...using this method, as you have discovered, there is no way that you can't record what you're hearing, no ad/da is involved, neither is the interface itself. I can happily belive that offline bouncing a busy session could induce errors in instruments, but then I'd argue you should be printing your instruments as audio anyway, for future recall if nothing else.
Old 3rd September 2011
  #44
Quote:
I think I posted the exact same thing on the 3rd or so reply...this is commonly known as recording to Fisk, instead of bouncing...using this method, as you have discovered, there is no way that you can't record what you're hearing, no ad/da is involved, neither is the interface itself. I can happily belive that offline bouncing a busy session could induce errors in instruments, but then I'd argue you should be printing your instruments as audio anyway, for future recall if nothing else.
I was not aware of the terminology recording to Fisk but I understand. I completely agree about printing midi instruments although when they are still being edited printing is not practical. I print when the project is close to completion. If clients ask for a mix early on I cross my fingers that timing will be good when I bounce if lots of midi is going on. Usually it is good but now and then issues come up which will be the time to "Fisk". Seems like a good enough solution.
Old 3rd September 2011
  #45
Gear Guru
 

Quote:
Originally Posted by AudioSoundzz View Post
I was not aware of the terminology recording to Fisk..
Recording to Fisk is a technique developed by Nashville engineers where they would bus their audio to hard drives located on the campus of nearby Fisk University - a small, predominantly African-American institution with a strong liberal arts and science emphasis.
Old 3rd September 2011
  #46
Thats some very interesting information about Fisk.. Seems a bit funny that that event would have created a new word but I like it.
Old 4th September 2011
  #47
Lives for gear
 
Mark D.'s Avatar
 

Quote:
Originally Posted by AudioSoundzz View Post
I will keep my eyes open a song that I can post
Thanks, should you have a spare moment and an old mix you did that is of
a personal project, or one that can be shared would be great. If what you
say is true it should be audible on any project. I look forward to hearing it.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump