I am using FLAC, but I don't have option to set flac tp PCM. I only have options "NATIVE" or "DISABLED"
Guess what!?!?
Allo sent me the dual-mono firmware files.
They kept them in the pocket all the time. These....
The firmware package is a 121 replacement.
Basically all firmware files of all frequencies and samplerates will be swapped.
My guess. To avoid any "No Sound" drama out here and to avoid any driver changes, they just changed all files.
Ok.Ok. Meanwhile we know. That Dual-Mono thing is a quick and dirty solution (for now?). No complaints.
We are all hackers out here. And the whole thing is somewhat working - if you know how...
....now we know.
The "2.1" mode has to be chosen for dual-mono in Alsamixer, since there is no "1.1" mode to choose. The designer is well aware of this "workaround" btw. In "2.1" mode the right firmware files will be loaded into the DAC.
The outer RCAs will carry the signal. The others are dead. So - it's a "dual mono - single channel"
setup.
Benefits: At least crosstalk should be improved. And a little less load on one channel per DAC might
also do some good. Beside that ?????
On the CON side we have a DSP in the loop and we can't run 384kHz ?!?!?
I meanwhile think that the volume controls "digital" and "subwoofer" with alsamixer refer to the "DAC1" "DAC2" standard 51xx HW volume controls btw.
These controls should IMO be at 100% anyhow! I never liked the PCM51xx HW volume control.
BUT:
My right channel generates strange click noises/distortions every couple of seconds!
It plays music though.
I already switched cables and speakers. Replaced/disabled my realtimekernel/tunings asf.
No change.
Hmmh. What now?!? Are there any known HW issues with the Piano2.1 ?
Almost done!
PS: Let see if Allo decides to make the dual-mono firmware publicly available.
Allo sent me the dual-mono firmware files.
They kept them in the pocket all the time. These....
The firmware package is a 121 replacement.
Basically all firmware files of all frequencies and samplerates will be swapped.
My guess. To avoid any "No Sound" drama out here and to avoid any driver changes, they just changed all files.
Ok.Ok. Meanwhile we know. That Dual-Mono thing is a quick and dirty solution (for now?). No complaints.
We are all hackers out here. And the whole thing is somewhat working - if you know how...
....now we know.
The "2.1" mode has to be chosen for dual-mono in Alsamixer, since there is no "1.1" mode to choose. The designer is well aware of this "workaround" btw. In "2.1" mode the right firmware files will be loaded into the DAC.
The outer RCAs will carry the signal. The others are dead. So - it's a "dual mono - single channel"
setup.
Benefits: At least crosstalk should be improved. And a little less load on one channel per DAC might
also do some good. Beside that ?????
On the CON side we have a DSP in the loop and we can't run 384kHz ?!?!?
I meanwhile think that the volume controls "digital" and "subwoofer" with alsamixer refer to the "DAC1" "DAC2" standard 51xx HW volume controls btw.
These controls should IMO be at 100% anyhow! I never liked the PCM51xx HW volume control.
BUT:
My right channel generates strange click noises/distortions every couple of seconds!
It plays music though.
I already switched cables and speakers. Replaced/disabled my realtimekernel/tunings asf.
No change.
Hmmh. What now?!? Are there any known HW issues with the Piano2.1 ?
Almost done!
PS: Let see if Allo decides to make the dual-mono firmware publicly available.
Last edited:
hello!!
you have all the options because i have used the 7.9 too...
you have this options that are like this on default install:
FLAC AIF FLAC
FLAC FLAC Native
FLAC MP3 Disabled
FLAC PCM FLAC
you put like this way:
FLAC AIF disabled
FLAC FLAC disabled
FLAC MP3 Disabled
FLAC PCM FLAC
now all your flac will be decoded to pcm on server side.
you can see this if you click when file is playing on orange squeeze and see details...
it will show the flac is converted to pcm.
you have all the options because i have used the 7.9 too...
you have this options that are like this on default install:
FLAC AIF FLAC
FLAC FLAC Native
FLAC MP3 Disabled
FLAC PCM FLAC
you put like this way:
FLAC AIF disabled
FLAC FLAC disabled
FLAC MP3 Disabled
FLAC PCM FLAC
now all your flac will be decoded to pcm on server side.
you can see this if you click when file is playing on orange squeeze and see details...
it will show the flac is converted to pcm.
Hmmh.
There are problems with the firmware - I guess.
It seems that buffers are not properly cleared when changing tracks and/or configurations, or something like that.
This might lead to the crackles. I e.g. did some config changes in alsamixer and rebooted.
The first track played OK. The next track was messed up again.
I do also face phase issues. My phase test-track does not produce correct results.
Then . Sounds that usually are on right show up on left. If I run a left-right test-track it sounds OK again. Hmmh.
Not nice. Seems difficult to nail down. Running out of ideas. This goes beyond my capabilities now.
Cheers.
There are problems with the firmware - I guess.
It seems that buffers are not properly cleared when changing tracks and/or configurations, or something like that.
This might lead to the crackles. I e.g. did some config changes in alsamixer and rebooted.
The first track played OK. The next track was messed up again.
I do also face phase issues. My phase test-track does not produce correct results.
Then . Sounds that usually are on right show up on left. If I run a left-right test-track it sounds OK again. Hmmh.
Not nice. Seems difficult to nail down. Running out of ideas. This goes beyond my capabilities now.
Cheers.
Interesting. Just FYI did you notice my post #112 (this thread). You can run with the generic i2s driver and have both chips active... and be able to use 384K. But yeah L/R active both sides, probably no PLL bypass that the "plus" driver enables.
This is my fav config (for the piano 2.1)-->
(post #508)
http://www.diyaudio.com/forums/digi...y-dac-pro-hw-mods-anybody-51.html#post4945614 (Meant to follow up but too much time went by on that, sorry -- I did see you did not have time to try it out.)
Also as a quick aside, I like the Boss in certain ways, but this sounds much better to me... something about it's bass grates on me after awhile (meat grinder to my ears). I'd like to see a FR chart on it, it's so different!
This is my fav config (for the piano 2.1)-->
(post #508)
http://www.diyaudio.com/forums/digi...y-dac-pro-hw-mods-anybody-51.html#post4945614 (Meant to follow up but too much time went by on that, sorry -- I did see you did not have time to try it out.)
Also as a quick aside, I like the Boss in certain ways, but this sounds much better to me... something about it's bass grates on me after awhile (meat grinder to my ears). I'd like to see a FR chart on it, it's so different!
Last edited:
scramer - I hear what you're saying. StilI I'd prefer a properly implemented "1.1" mode.
I looked at the code again. I think a "1.1-single channel" mode would be easy to program.
Here's my input to Allo:
1. instead of disabling the "subwoofer" dac as done in "2.0" mode , it stays enabled in "1.1" ...
2. ...enabled without loading the DSP stuff and...
3. just one channel muted in the subwoofer volume control section. That's possible with the digital volume control function as far as I've seen it.
4. on the other DAC one channel gets muted too, Both muted channels are "hardcoded" in that "1.1" mode.
That's about it. ( The "1.1 - double channel" mode then would probably require the DSP in the loop again - but for now nobody knows is that mode is possible to implement at all)
Advantages:
1. No firmware required ( much less hassle for you guys!)
2. No DSP in the loop
3. 384kHz might be enabled (under certain conditions)
4. Kali masterclock mode can be used
5. No more dirty hacks in "2.1" mode
6. easy to document
384kHz
*********************
The driver currently doesn't allow 384kHz at all. The Piano is not running in I2S master mode like the Boss with its onboard clocks - I think. Not sure yet how scramer is able to run
384KHz with the standard I2S driver.
It would be nice if this limitation would be limited to the "DSP" modes "2.1" and "2.2".
There might be no need to have that limitation in "2.0" and "1.1" mode without DSP in the loop....
""IF"" a Kali is used (with overlay configuration "glb_mclk").
Without Kali 192kHz would still be the limit.
Ah. And I've also learned that the DAC digital volume control comes before the DSP section. Another reason for keeping it at 100% all the time. Looks like a TI design flaw to me.
##################################################
But all that won't matter - if - I don't manage to get the earlier mentioned flaws under control.
I'm wondering is those guys who're raving about "double mono" operation are all running the Kali!?!? I can't use/try mine, since it is running in my main system. And I do not intend to rip it apart.
I looked at the code again. I think a "1.1-single channel" mode would be easy to program.
Here's my input to Allo:
1. instead of disabling the "subwoofer" dac as done in "2.0" mode , it stays enabled in "1.1" ...
2. ...enabled without loading the DSP stuff and...
3. just one channel muted in the subwoofer volume control section. That's possible with the digital volume control function as far as I've seen it.
4. on the other DAC one channel gets muted too, Both muted channels are "hardcoded" in that "1.1" mode.
That's about it. ( The "1.1 - double channel" mode then would probably require the DSP in the loop again - but for now nobody knows is that mode is possible to implement at all)
Advantages:
1. No firmware required ( much less hassle for you guys!)
2. No DSP in the loop
3. 384kHz might be enabled (under certain conditions)
4. Kali masterclock mode can be used
5. No more dirty hacks in "2.1" mode
6. easy to document
384kHz
*********************
The driver currently doesn't allow 384kHz at all. The Piano is not running in I2S master mode like the Boss with its onboard clocks - I think. Not sure yet how scramer is able to run
384KHz with the standard I2S driver.
It would be nice if this limitation would be limited to the "DSP" modes "2.1" and "2.2".
There might be no need to have that limitation in "2.0" and "1.1" mode without DSP in the loop....
""IF"" a Kali is used (with overlay configuration "glb_mclk").
Without Kali 192kHz would still be the limit.
Ah. And I've also learned that the DAC digital volume control comes before the DSP section. Another reason for keeping it at 100% all the time. Looks like a TI design flaw to me.
##################################################
But all that won't matter - if - I don't manage to get the earlier mentioned flaws under control.
I'm wondering is those guys who're raving about "double mono" operation are all running the Kali!?!? I can't use/try mine, since it is running in my main system. And I do not intend to rip it apart.
I'm using the Kali but not specifically raving about 'double mono' as some people are - I hear no difference between that and 'double stereo' mode with RCAs still in outer positions. Both sound better to me than standard stereo though.
Just to let you know. I tricked the Piano!!
Just tried a quick and dirty hack:
I disabled the MUTING of the subwoofer DAC in "2.0" mode in the driver:
Then I removed all firmware to force the DAC into 2.0 mode - without DSP .
That'll happen without any firmware files under /lib/firmware/allo/piano.
Guess what happened.
"2.0" mode and two DACs playing music. ( something for you to try scramer )
And. Most important!
The DACs play clean. No more distortions, phase issues or other hickups.
Next challenge would be to mute one channel per DAC.
Enjoy.
Just tried a quick and dirty hack:
I disabled the MUTING of the subwoofer DAC in "2.0" mode in the driver:
Code:
--- allo-piano-dac-plus.c.orig 2017-05-19 14:35:08.301592489 +0200
+++ allo-piano-dac-plus.c 2017-05-19 14:09:55.300236678 +0200
@@ -112,7 +112,7 @@
if (mode == 0) { /* 2.0 */
snd_soc_write(rtd->codec_dais[1]->codec,
- PCM512x_MUTE, 0x11);
+ PCM512x_MUTE, 0x00);
glb_ptr->set_rate = rate;
glb_ptr->set_mode = mode;
glb_ptr->set_lowpass = lowpass;Then I removed all firmware to force the DAC into 2.0 mode - without DSP .
That'll happen without any firmware files under /lib/firmware/allo/piano.
Guess what happened.
"2.0" mode and two DACs playing music.
( something for you to try scramer )And. Most important!
The DACs play clean. No more distortions, phase issues or other hickups.
Next challenge would be to mute one channel per DAC.
Enjoy.
Last edited:
While I was at it, I added <=384kHz support to the driver and tested it in my new "2.0" mode:
Doesn't look and sound too bad.
Code:
me@pi # cat /proc/asound/card0/pcm0p/sub0/hw_params
access: MMAP_INTERLEAVED
format: S32_LE
subformat: STD
channels: 2
rate: 384000 (384000/1)
period_size: 15360
buffer_size: 61440Doesn't look and sound too bad.
I am also not using the Kali because of the same reason you mentioned. Nevertheless, also without the Kali, the Piano 2.1 sounds very good in the double mono setting. BTW I never tried the double stereo setting.
Thanks for your posts about the Piano 2.1. They are very interesting, I read them all.
Scott, I just wrote this on another thread, but in case you haven't run the Boss 24/7 for 2-3 weeks, you might want to put it in a test setup and do that and listen again... here's what I said:
"Do note that like all capacitors that I've used, these Supercaps take some time to form and settle in. For the 1st 2-3 weeks of on-time (I leave my gear on 24/7 when I don't have any fear of thunderstorms), bass will be over-emphasized & there is an overall 'hashiness' to the backgrounds. In my experience, these both diminish in that third week and they seem totally settled after that. But depending on your setup and your own tolerance for this characteristics, you may have to grit your teeth at times during this period.
So far, IN MY SETUPS AND TO MY EARS, I have used them and liked them on the power leads to RPis and Allo Kali's & Boss and Piano DACs, and installed them on the main DAC (and where appropriate clock) power supplies on a Mamboberry & a couple of Sony HAP Z1-ESs.
After settling, IN MY SETUPS AND TO MY EARS, I still hear enhanced bass power, deliniation, and depth along with an overall enhancement in instrument & voice 'solidity'... and an increase in perceived dynamics."
About 1 week into the break-in time on my Boss, I started planning an extreme series of modifications to fix the issues I was hearing. But I held off on execution because I knew it was still breaking in.
After a bit more than 2 weeks constant run time, I started liking it.... and now at 3 weeks, I'm pretty darned happy with it (see my comments on it here:
http://www.diyaudio.com/forums/pc-b...e-cheap-dac-hat-raspberry-pi.html#post5083801 )
SO if yours did not have that amount of runtime, you may want to do that and give it another chance. OTOH, I can see that it won't be everyone's cup of tea due to system differences and personal preferences... as always, YMMV!
Greg in Mississippi
Last edited:
@Soundcheck,
I haven't yet, though I agree it is a good target of opportunity.
I'm also looking at the other power-supply bypass caps in both the Boss & Piano & comparing them. There may be other areas that could benefit.
BUT there is little room to do much cap substitution on the current Piano layout. We may not hear the benefits of applying the Boss techniques to the Piano setup until Allo does an upgraded Piano (and maybe Boss too?). I know they are watching & listening.
Right now I'm more focused on the IsolatorPi beta. Anyone got some good 192 & 384 source material I can get for full checkouts? Or links to same?
Greg in Mississippi
I haven't yet, though I agree it is a good target of opportunity.
I'm also looking at the other power-supply bypass caps in both the Boss & Piano & comparing them. There may be other areas that could benefit.
BUT there is little room to do much cap substitution on the current Piano layout. We may not hear the benefits of applying the Boss techniques to the Piano setup until Allo does an upgraded Piano (and maybe Boss too?). I know they are watching & listening.
Right now I'm more focused on the IsolatorPi beta. Anyone got some good 192 & 384 source material I can get for full checkouts? Or links to same?
Greg in Mississippi
This is slightly off-topic, but relevant to your assessment methodology. I have just gone to an Allo Boss DAC after my HifiBerry Digi+ failed. In order to get a sense of differences, I have used SoX at the command line to get good tracks to 384k, and then played them from the NAS and the SDcard. If, as some suggest, the musical content is fully represented at 98k or 192k, this upsampling the 384k for DAC processing considerations should work.
I'm looking forward to the Allo isolation card to work with my Boss DAC hoping to power the Allo units with a Silent Switcher.
Skip
I'm looking forward to the Allo isolation card to work with my Boss DAC hoping to power the Allo units with a Silent Switcher.
Skip
IMO the only reason for going 384 ( and manipulating the base material) is the bypass of internal DAC filters.
In any case there'll be a tradeoff.
Higher processing load (and related distortions) and DSP related losses might neutralize the gain achieved by bypassing the DAC filters.
All this is one of the reasons why I stayed with Sabre (for now) on my main system.
In any case there'll be a tradeoff.
Higher processing load (and related distortions) and DSP related losses might neutralize the gain achieved by bypassing the DAC filters.
All this is one of the reasons why I stayed with Sabre (for now) on my main system.
@Soundcheck,
What are you thinking for upgrading the AVDD cap? See my picture of the area of the Piano 2.1 with one of the 22uf PPS caps on the board. I just cannot see how to fit those in that space without a board re-design. AND that's the cap I'd like to use there... actually a stack of 2-3 of them.
The cap that's currently there is likely 100uf... & I suspect that the Piano sounds as good as it does due to the size of that & some of the other caps on the AVDD, charge pump, and DVDD pins. I would like to preserve that... and to do that with the SMD film caps (+ some small conductive plastics to add bulk C) will be difficult to impossible to do without a board redesign.
Also I'll go back and look at your blog post on upsampling again. I'm going to use these files (192 & 384) for testing the IsolatorPi with master mode DACs.
@Skip Pack,
Thanks for your comments on the processor load with either realtime upsampling or just playing 384 files. Useful.
On you question on use of the Silent Switcher with RPi DACs & isolation, that is one thing I'll be testing in my setup with Ian's Isolator Pi. I have one of the Dial Audio RPi DACs (see here: http://www.diyaudio.com/forums/digital-line-level/306117-dac-raspberry-pi.html ) and I have 2 +-15V regulator setups for it... a pair of small shunt regulators & a Silent Switcher. First try on the SS suggests it should work well... though I don't have a definitive read on a comparison between the 2 options yet.
Greg in Mississippi
What are you thinking for upgrading the AVDD cap? See my picture of the area of the Piano 2.1 with one of the 22uf PPS caps on the board. I just cannot see how to fit those in that space without a board re-design. AND that's the cap I'd like to use there... actually a stack of 2-3 of them.
The cap that's currently there is likely 100uf... & I suspect that the Piano sounds as good as it does due to the size of that & some of the other caps on the AVDD, charge pump, and DVDD pins. I would like to preserve that... and to do that with the SMD film caps (+ some small conductive plastics to add bulk C) will be difficult to impossible to do without a board redesign.
Also I'll go back and look at your blog post on upsampling again. I'm going to use these files (192 & 384) for testing the IsolatorPi with master mode DACs.
@Skip Pack,
Thanks for your comments on the processor load with either realtime upsampling or just playing 384 files. Useful.
On you question on use of the Silent Switcher with RPi DACs & isolation, that is one thing I'll be testing in my setup with Ian's Isolator Pi. I have one of the Dial Audio RPi DACs (see here: http://www.diyaudio.com/forums/digital-line-level/306117-dac-raspberry-pi.html ) and I have 2 +-15V regulator setups for it... a pair of small shunt regulators & a Silent Switcher. First try on the SS suggests it should work well... though I don't have a definitive read on a comparison between the 2 options yet.
Greg in Mississippi
