R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: [1]   Go Down

Author Topic: converter calibration  (Read 5310 times)

j.hall

  • Moderator
  • Jr. Member
  • *****
  • Offline Offline
  • Posts: 60
converter calibration
« on: June 13, 2011, 02:05:04 pm »

ok, i would consider myself moderately technical with my studio knowledge and practices.  my room is calibrated to a general operating level.

so knowing that, here is a quick story.

about a year ago i picked up an older mytek 8 channel (max 96k) ADC.  it was a good deal, and i just wanted something "better" to use for input then my lynx aurora (which i do like, but i have some complaints).

the mytek right away was better.  lowend was tighter and deeper, stereo image felt wider and in general, the mixes printed through the mytek just felt better.  so i moved on with life.

a year later i'm starting to have some complaints about some "2buss" things i've been noticing in my chain.  after many random, and more esoteric, conversations with Brad Blackwood (who does ALL my mastering and happens to be a good friend) we narrow the field to the make-up gain amp on my SSL 384 buss compressor, and the mytek ADC.

i realized, through some questions from Brad, that i had never calibrated the Mytek.  a little embarrassing, but it was time for some testing.

each pair of the mytek was set differently.  but 1 and 2 are what i print mixes through 100% of the time.

they were really close to -15dbfs.  so i tweaked them a touch and printed a mix not touching the 384.

then i set them to -20dbfs and printed the same mix with all the same settings.

i then repeated this process but this time setting the make-up gain amp on the 384 to unity.

the results blew me away.  i figured it would be a slight change and i would just pick which calibration i liked better.

the -20 calibration was dramatically different then -15.  i couldn't believe how much better it sounded.  the depth of field alone was amazing.  transients felt better, stereo image was better, top to bottom spectrum was richer.

the differences between the 384 make up gain amp settings were noticeable but not as extreme as i originally suspected.

moral of the story, spend some time playing with your converter calibration.  find a place they sound best to you.  well worth the 2 hours i spent testing.
Logged

Hollis

  • Newbie
  • *
  • Offline Offline
  • Posts: 4
  • Real Full Name: Andrew Hollis
Re: Converter calibration
« Reply #1 on: October 12, 2011, 03:28:05 pm »

What is your console? I presume your -15 dBFS cal = +4 dBu, thus 19 dBu max level. At -20 dBFS you'd have +24 dBu max. Did you compensate for the change in level when making your comparison?

Did you do this cal on your outputs as well? From a playback perspective, I would expect the -15 dBFS cal to be better sounding, because you won't be crushing analog mixer headroom off the bat with the way most people record with every track peaking to 0 dBFS. You're also 5 dB below your converter's maximum capability, which gives you a cushion away from the THD "fuzz" zone where I think a lot of gear starts to sound bad approaching it's operating limits, despite how high their marketing numbers go...

If you only adjusted the ADC, then you've demonstrated these points when you preferred the -20 dBFS cal--you gained 5 dB headroom and avoided that fuzz zone by running the same mix level. Now, the trap I think people fall into is they just go push their mix levels back up--I guess if one can have the discipline to leave that area alone they will have better sounding mixes as you observe.

This thread I think makes a pretty good case for using a lower maximum operating level +18 over the +24 SMPTE standard in the wild-west world of music recording on an analog console. The back and forth with jonnyclueless, minister, and The Alamo is where to look:

http://www.gearslutz.com/board/high-end/48600-digi-192-calibration-levels.html
Logged

Viitalahde

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 99
  • Real Full Name: Jaakko Viitalähde
Re: converter calibration
« Reply #2 on: October 21, 2011, 05:10:10 pm »

Low analog levels are a great thing. My chain is calibrated for -8dBFS = 0.775v RMS. I dropped the internal level of my chain to -10dBFS = 0.775v RMS some months back, and the results were so good I dropped it even more to -8. I'm running HEDD192 myself.

Highly recommended, really.
Logged
Jaakko Viitalähde
Virtalähde Mastering, Kuhmoinen/Finland
http://www.virtalahde.com
http://www.facebook.com/VirtalahdeMastering

Mo Facta

  • Newbie
  • *
  • Offline Offline
  • Posts: 23
Re: converter calibration
« Reply #3 on: September 21, 2012, 07:04:28 am »

Well, the whole point is to match the gain structure of your hardware and converters, isn't it?

For instance, my Chilton console's direct outputs clip at +20dBu and so does my Lynx Aurora 8.  This means that my gain structure is consistent from digital through to the analogue units on input.

However, I also run a Dangerous DBox and its summing inputs clip at +27dBu.  This means that the top 7dB of headroom will never be used, which makes me feel like it's a little redundant.  Thus, one could say that the headroom of the summing bus is offset by 7dB.

Of course, there is no way to calibrate the stock Aurora.  Only the VT has a calibration facility.

My point is that I wish I had a way to calibrate the inputs and outputs separately to compensate and to maximize the operating ranges of my equipment.

Cheers :)   
Logged
Quote from: Slipperman
That'll learn those arrogant LA *bleeps*.

http://shiftedaudio.co.za
http://gregbester.xp3.biz

Watyariz

  • Newbie
  • *
  • Offline Offline
  • Posts: 1
  • Real Full Name: Watyariz
Re: converter calibration
« Reply #4 on: January 07, 2015, 12:00:37 am »

I can be available or not.

endarn

  • Newbie
  • *
  • Offline Offline
  • Posts: 2
  • Real Full Name: yoyo brrt
Re: converter calibration
« Reply #5 on: January 23, 2015, 12:05:34 am »

I think that happens very difficult.

Jim Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 545
Re: converter calibration
« Reply #6 on: January 23, 2015, 11:20:28 am »

Lower your analog levels and the noise floor will creep up. Modern opamps show decreasing THD with levels up to the clip point. If a design doesn't do that, something was missed. A sweep through an audio analyzer using a THD vs amplitude test will show if your ADC inputs are behaving differently at different signal levels. Then re-check with a THD vs frequency sweep to determine if it falls apart at higher frequencies.

I run my BurrBrown converters very hard at +14~16 dbu, about 4 to 6 db below the zero line. They sound the same if I lower the levels, but the noise increases.
Logged
Pages: [1]   Go Up