The case to eliminate the .01dB increment

I recently made a poll about frequency increments. This post is specifically to make the case for why the official increment of any dB knob, mouse move, or any other method of moving 1 value upward of volume…you stop using 100ths of a db (.01db) and stick to a minimum value of .1db. I will make a great case for this and you’re going to understand why very quickly, I think!

Reason #1: Most humans hear loudness in 1db increments. It would be hard, if not impossible, to hear a .2db increment, let alone a .1db increment. Yet I’m still willing to make this the minimum, which I will explain why in a minute.

Reason #2: Helps all developers establish a universal standard in how many steps/values a dB knob takes per dB. This is helpful for reasons I’m sure you can imagine.

Reason #3: MIDI controllers would have an easier time syncing the exact dB value, when you’re scripting and displaying text values on your knobs based on degrees that might not quite line up with the DAW. In other words, the value on the screen would match the value on your controller much more closely.

Reason #4: We are using 301 degrees on the dial for most dB and frequency dials. Not 300. It’s 301. 150 negative degrees from 0, 150 positive degrees from 0, and then 1 degree for zero itself.
When you have a -/+15db dB knob like an SSL channel strip, and you go in .1dB increments, you have 301 possible values. That’s 301 degrees - 1 value per degree.

We need to start thinking about dB values as degrees and not digits. How many degrees from zero. -37 degrees from 0 would be 3.7dB; you get the idea. You would never have to wonder what the dB value is at that position. It’s also easier to think about 0 as a degree, rather than a digit, when we’re talking about dB steps, since 0 can always be the off position. 1 degree is always .1. 2 degrees is always .2 in whichever direction the knob is being turned.

Reason #5 (final reason): It gives us code that we can share easily. If you know that you’ll always have 301 maximum steps for a dB knob, the code becomes much more standardized when everybody is dividing by 301 to calculate step amounts from the range of max values (that type of thing).

To those wondering how you would get 301 steps on your MIDI knob when your MIDI goes from 0-127, we are talking about that in another thread (NRPN vs other 14-bit potential solutions).

If we are willing to accept what an increment means, and we are willing to accept that 100ths of a dB is a waste of everyone’s time, we can start standardizing 14-bit more, and we can start sharing more with each other so that everybody can have high resolution easily, instead of the select few who try and figure it out for themselves.

To me this seems like a bit of a waste of development resources.

I have never felt constricted by the way things currently are so I don’t need any change. If I don’t hear 0.1dB it still doesn’t prevent me from doing anything right now.

As for control isn’t MIDI 2.0 32-bit resolution? So how does that standard fit into this?

What made me realize how ridiculous it all is was that we have terrible hearing to hear small increments, we have terrible MIDI gear with low resolution that makes coarse increments, yet we try and manage that with this massive resolution on the screen that we don’t need, and makes it harder to map 1:1 on our controllers without spending a fortune, or hoping and praying that a hardware company comes along and makes just the right piece.

We don’t need to wait for MIDI 2.0 anymore because this awesome developer made this MIDI plugin for the Stream Deck+ that gives you total control of things that we never had control of before (no hardware company ever made anything like it, so we were all stuck living with mediocre controllers).

It’s true that midi faders (1.0) uses 0-127 limiting the number of resolution steps - can’t do anything about that.

But the audio return attenuators belong to Cubase. The fader knob just assumes a best fit approximation of where that would be based on a log db scale. They use 01.db as it maximum resolution most likely because three digits plus a decimal point easily fit acrost a narrow mixer channel, and nobody could ever accuse them of not having enough resolution like midi.

Cheers
:slight_smile:

1 Like

I’m working on doing something about the 0-127 issue, as I have a guy in another thread helping me through some things with NRPN and 14-bit values, to get as many steps as we need.
In the meantime, I need everybody to give up .01db and .01khz increments so that I can release a script that people can load and have just work for their encoders.
I’m releasing a Stream Deck product soon for the community and I hope that it opens some eyes, and some doors.