Open-source USB interface: Audio Widget

Jump to Latest
#1,921
In the audio-widget branch, the latest update adding revised VID/PID into WidgetControl was two months ago. The latest update adding the productid4=0x3e9 into widget-control.c was seven months ago.
It looks like widget-control.c may not be in sync, but I'd leave final analysis to Alex or Borge.
 
#1,923
Hi,
I'm amazed by a change in the Widget Control.

Has someone tried iir filter instead of fir one?
With the AB 1.2 and AK4430 dac it seems to me it improves bass deepness and articulation and trebles become more polite and focused.
Distortion seem to be a lot less.

For me it's a lot better than with fir filter.

Frank
 
#1,925
I've used Fir filter since the beginning.
Then I did some mods:
- two LifePO4 supplies for the Xos
- three crcrc linear supplies and three Salas shunt regulators for 3,3v and 5v
- Mkp bypassed with teflon cap in the LPF filter
- Dale resistors rn55 in the LPF filter
- Dale resistors rn55 in the I2s lines

Sound has improved a lot with IIR filter.

This is my setup:
Headphone____Laptop - Jplay SW - Violectric V200 - Rudistor Chroma MD1 and AKG701.
Speaker____Laptop -Jplay SW - B4 Preamp - Ncc200 Amp dual mono - Audio Physic Brylon - Cabasse Santorin25

Test has be done with classical and jazz.

Frank
 
#1,926
Interesting stuff with the regulators! Do you have schematics to share?

The fir/iir setting shouldn't be doing anything on the AB-1.2. The AK4430 doesn't have a filter selectr input. The functionality was put in for the PCM5102 of the AB-1.12 prototype.

I doubt the fir/iir setting does anything else in firmware, but I'll have to check to verify.

Børge
 
#1,927
For example sound doesn't improve when I change filter and use AB 1.1 with the ES9022 ( I have also to change usb receiver with later one).

I've asked in the forum to verify if " I'm crazy or not" 🙂)
I know this setting should do anything to the sound but I can hear clearly an improvement with iir and AK4430 dac.

Can you try with your setup?
Frank
 
#1,929
Catal said:
Hi,
I'm amazed by a change in the Widget Control.

Has someone tried iir filter instead of fir one?
With the AB 1.2 and AK4430 dac it seems to me it improves bass deepness and articulation and trebles become more polite and focused.
Distortion seem to be a lot less.

For me it's a lot better than with fir filter.

Frank
Click to expand...

Where is the filter you are talking about? The AK4430 has no filter options.
 
#1,930
It's the filter setting in the widget control software.

As I said to Borges, I know that this selection is "only" possible for the PCM5102 dac, ... cause AK4430 datasheet says clearly that only fir filter can be used, ... that I should hear no difference between these two options.

Frank
 
#1,931
Hi guys,

I found a bug in the feedback code I wrote back in October. Even with the bug in, stability seems to be good. Still, what I believe to be improved code is checked into branch audio-widget-experimental and compiled into awx_20121206.elf

Please help out with testing the UAC2 playback on Linux, Mac and Windows. My own tests on Win7-64 are good.

Also, there have been some reports that UAC1 was noisy on Mac OS X. If you can help out with UAC1 tests of the newest updates, that would be very good. I hope recent changes for Macs haven't broken UAC1 support elsewhere. I don't have a Mac to test with here. (See below for details.)


The UAC2 code uses something like a 4-point P-regulator. (The October version had a bug which actually turned it into a 3-point P-regulator with asymmetrical thresholds.) I found the bug and added even more stability to the regulator using a simulator in Octave. Let me know if anyone's interested in the source code for that 🙂

Before October the UAC2 code used a 2-point thermostat regulator. The UAC1 code in this update also uses a 2-point thermostat regulator. If there are observations of rarely-occuring mid-music noise in UAC1 I can port the 4-point P-regulator to UAC1. But I don't think I'll try to fix that before it turns out to be a problem.


Cheers,
Børge
 
#1,932
Catal said:
It's the filter setting in the widget control software.

As I said to Borges, I know that this selection is "only" possible for the PCM5102 dac, ... cause AK4430 datasheet says clearly that only fir filter can be used, ... that I should hear no difference between these two options.

Frank
Click to expand...

I can't find a reference to filtering in the AK4430 datasheet. You get what it has, only.
I don't think that gpio goes anywhere on the AB1.2. Maybe Borge can enlighten us?
 
#1,936
I just upgraded mine to the latest fw using cli on win7. worked fine. (whew!)

I see this on linux, now:

Code: 
aplay -l
**** List of PLAYBACK Hardware Devices ****
card 1: AB12 [QNKTC USB DAC AB-1.2], device 0: USB Audio [USB Audio]
  Subdevices: 0/1
  Subdevice #0: subdevice #0

this changed the naming in my alsa state file:

Code: 
state.AB12 {
        control.1 {
                iface MIXER
                name 'Clock 2 Playback Switch'
                value.0 true
                value.1 true
                comment {
                        access 'read write'
                        type BOOLEAN
                        count 2
                }
        }

and it used to be

Code: 
state.DG8SAQI2C {

this caused me to have to re-edit my alsa state file (I do weird things with it, so I have to, lol) but was there a reason the name had to change?
 
You must log in or register to reply here.
Top Bottom