[sdiy] OTA performance (was SSM chip reissue)

[Neil Johnson]

Tom Wiltshire wrote:
>
> On 24 Apr 2017, at 22:38, Neil Johnson <neil.johnson71 at gmail.com> wrote:
>
>>  0.24dB at Av = -40dB (after which they're too
>> embarrassed to tell you how bad it is).
>
> What, too embarrassed to say that the matching is a bit worse than a quarter decibel out at -40dB?!
> What's the smallest decibel difference people can detect? About 3dB, isn't it? Quite a bit more than 0.25dB.

Gain matching is important in stereo recording otherwise the stereo
image drifts away from where you want it.

> To be honest, if it were 1dB out by -60dB, who's really going to notice? 2dB out at -80dB? And that's a lot worse than the spec'd figures suggest is likely.

You're confusing single-channel gain errors with channel-to-channel gain errors.

http://www.soundonsound.com/sound-advice/q-are-matched-pairs-really-important

"Preamps actually stand a very good chance of being matched to far
tighter tolerances than any mic ever could be. A preamp only needs to
match its gain and frequency response. Modern electronics mean that
the frequency response can be flat from a few Hz to many hundreds of
Hz within a fraction of a decibel every time, and with a variable gain
control, gain matching between channels can be as good as your nulling
technique — and easily to within a fraction of a decibel, again."

Now imagine trying to make a computer-controlled mic pre-amp (uC
drives DAC to generate CVs to two VCAs) - and you need to match the L
and R gains within a fraction of a decibel over the controlled gain
range.  Sure, you could calibrate it out, but that adds cost and time
to the production line.

Neil
-- 
http://www.njohnson.co.uk