Re: [LAD] announcing envy24control, mudita (*) edition.

From: Niels Mayer <nielsmayer@email-addr-hidden>
Date: Mon Jul 26 2010 - 02:55:10 EEST

On Sun, Jul 25, 2010 at 3:23 PM, <fons@email-addr-hidden> wrote:
> - The meters just indicate level in a linear way. Half scale
> is -6dB etc.

I originally talked about doing this in a previous thread, along with
various other things that ended up being impossible, like integrating
jkmeter :-).

I figured there's enough changes in this first round that need
testing&feedback and decided for "release early and often" rather than
implementing yet another feature...
This sounds like an incremental feature for 1.1 once the other changes
I've made are vetted/tested and working.

Note that the original meters essentially used the same value-to-pixel
mapping for the peak meters -- they just didn't tell you that
half-scale was -6dB.

> - The analog gain sliders behave strangly. They seem to 'detent'
> on the scale marks, it's quite impossible to set a value *near*
> a scale mark while at the same time the resulotion halfway
> between scale marks seems to be OK.

This is a side effect of routines I added to draw markings in the
scale widgets. GTK automatically auto-detents at the markings -- it's
certainly not what I wanted. (See volume.c: e.g.
draw_24bit_attenuator_scale_markings(), draw_dac_scale_markings() &
draw_dac_scale_markings() called out of
envy24control.c:create_analog_volume()).

If anybody knows how to get gtk_scale_add_mark() to not setup the
"detent" then I could make this an option. (or at least weaken the
level of hysteresis)... Otherwise, it's either detent, or no markings.

As workaround, and perhaps a better choice for the finer control
offered -- use the up-arrow and down-arrow, as well as page-up and
page-down keys to move the selected slider. I've found that for the 24
bit attenuators feeding the digital mixer, the 1.5dB step-size pretty
much necessitates using up/down arrow to get the right level of
control. Alas, this aspect is unchanged from the 0.6.0 version of
envy24control. Due to the 1.5db step-size at the top end, It would
make a logarithmic control of these inputs quite "jumpy"...

> - *Please* make the first analog gain fader size the same as
> the others. The presence of the 4dBu button should not affect
> this.

Hmm... what 4dBu button?? Sounds like you have a delta1010? Could you
send me a screenshot of what that ends up looking like?

It is supposed to look like this
http://nielsmayer.com/envy24control/Screenshot-Envy24Control-AnalogVolume.png

Yep, I saw code-paths I'd never invoke with my hardware, which is why
I wanted to see testing on devices like the M-Audio Delta 1010 and
1010lt.

> - (personal preference) Don't group the faders into stereo
> pairs.

Do you mean on the digital mixer inputs in "Monitor PCM Outs" and
"Monitor Inputs" -- those were already stereo paired to begin with. I
don't like it that much, but it would have been a big change.
http://sourceforge.net/projects/kenvy24/ uses a "Pan" knob on each
channel, which makes more sense, and wastes less space. (Alas kenvy24
doesn't work correctly, and appears long neglected).

Or do you mean the L/R alternating labels on the sliders? For the
"Monitor PCM Outs" and "Monitor Inputs" panels, I did it that way to
reduce redundancy. Lots of repeated scale markings wasting screen
real-estate. For the "Analog Volumes" I did it out of habit -- one of
these days I'll expand my horizons to ambisonics. I guess this should
be an option. To be really unixy, I should just have it be something
really geeky like "--meter_channel_modulus" and if you set it to 2 you
get the existing behavior. :-) Seriously, I guess the stereo pairing
should be a binary command-line option in a future version.

Thank you for your feedback.

Oh, by the way, are you also seeing the high CPU usage issue that
others have reported, or is it doing as intended -- which is acutally
much lower CPU usage than the stock ALSA /usr/bin/envy24control .

-- Niels
http://nielsmayer.com
_______________________________________________
Linux-audio-dev mailing list
Linux-audio-dev@email-addr-hidden
http://lists.linuxaudio.org/listinfo/linux-audio-dev
Received on Mon Jul 26 04:15:04 2010

This archive was generated by hypermail 2.1.8 : Mon Jul 26 2010 - 04:15:04 EEST