View Single Post
Old 21st January 2017
  #18
Airwindows
 
chrisj's Avatar
 

Quote:
Originally Posted by A Fak View Post
Ok so I did the same test but instead used BitShiftGain to boost the signal to hear the difference. If I understand BitShiftGain correctly 1bit = 6dB of gain. I boosted HOFA 4U +24bits (144dB?) and DMG Track Control +18bits (108dB?) to get them around the same level.

First thing that jumps out to me is the extra 6bits between the two. HOFA 4U nulls 36dB (to be fair probably 32dB's but had to go up/down 6dB with BitShiftGain) further down than DMG TrackControl.

Level aside would you say the difference I'm hearing between these two and PurestGain is the errors\resolution between them? With the assumption that PurestGain being the most accurate?
I think it's safe to assume BitShiftGain is the most accurate thing around. It's literally a carefully controlled multiply at exactly the right amplitude to do a bitshift (and no other processing: no noise shaping or whatever). I've had people test it with bit-display meters and it moves a block of active bits up and down, just as you expect, losslessly. Gain trims are normally multiplies anyway: this is just set up to not even expand the word length.

The catch is, you have only 6dB gain increments but within that context, it's the reference for accuracy.

Last edited by chrisj; 30th January 2017 at 01:51 AM.. Reason: huh? how'd I substitute PurestGain but still be talking about 6 db increments?