Guys, Looks like Allo got the Boss and Piano drivers in the official 4.9.16+ (to be 4.9.17) kernel today.
Also it's the Piano "plus" driver that can do the PLL bypass (use the MCLK if present) -- I presume the boss driver has always done this, didn't trace thru the code.
Plan to compile and see if it works.
BTW did do a full 4.4.50* compile a week ago... with cleanly applied boss/piano patches, and it never did work for me.
I've been stuck on volumio (it uses 4.4.9) with my manually started squeezelite to use the boss. 🙁
*that rev needed to apply allo's github patches
Also it's the Piano "plus" driver that can do the PLL bypass (use the MCLK if present) -- I presume the boss driver has always done this, didn't trace thru the code.
Plan to compile and see if it works.
BTW did do a full 4.4.50* compile a week ago... with cleanly applied boss/piano patches, and it never did work for me.
I've been stuck on volumio (it uses 4.4.9) with my manually started squeezelite to use the boss. 🙁
*that rev needed to apply allo's github patches
Last edited:
Yes congratulations Allo. This will help the community. We have both drivers fully integrated to 4.9.17 for the next pCP
Ok. The day before yesterday they weren't in on 4.9.17.
Just downloaded the latest 4.9.17. Now they seem to be in. Nice.
Been using the patches for a while.
However. To me it seems that the patches are not properly implemented.
1. To me it looks like an old version of the patch.
The misspelling of "allo" reminded me! It's been fixed a while ago.
(I've been working with them a little in the background on the subject)
2. They also missed out to properly setup/patch the kernel _defconfig (=m)
for the drivers. They are "off" by default.
I'll check with Allo.
Just downloaded the latest 4.9.17. Now they seem to be in. Nice.
Been using the patches for a while.
However. To me it seems that the patches are not properly implemented.
1. To me it looks like an old version of the patch.
The misspelling of "allo" reminded me! It's been fixed a while ago.
(I've been working with them a little in the background on the subject)
2. They also missed out to properly setup/patch the kernel _defconfig (=m)
for the drivers. They are "off" by default.
I'll check with Allo.
Last edited:
Haha, yep -- was just trying it now.
Yesterday had to manually add this to the default config:
The piano plus worked, but boss would not compile. Looks like they committed changes to that just a few hours ago. Will see if the default config works also...
Yesterday had to manually add this to the default config:
Code:
vi ./arch/arm/configs/bcm2709_defconfig
CONFIG_SND_BCM2708_SOC_ALLO_PIANO_DAC_PLUS=m
CONFIG_SND_BCM2708_SOC_ALLO_BOSS_DAC=m
The piano plus worked, but boss would not compile. Looks like they committed changes to that just a few hours ago. Will see if the default config works also...
Attachments
Last edited:
What does this mean for those of us who aren't very Linux literate?
From now on every linux RPI distro - with most current 4.9.19 and onwards kernels - is
able to offer Boss and Piano 2.1 support without kernel customization.
Usually just the GUI needs to be adapted to offer available drivers to the user.
One thing some of you might look at (I do!) is 384k support.
The driver is prepared for it. However. Some other files in the
kernel need to be patched on top of the Allo provided stuff.
Not a big deal, but somebody has to take responsibility for it!
Therefore 384k for I2S and Boss is unfortunately not available
on the default RPI kernels.
I just saw, that the patches were only approved (committed?) to the 4.9.x kernel tree
and not to the still commonly available 4.4.x tree.
It would still require a customized 4.4.x kernel to offer the driver. I have no idea if
Allo also committed a patch to the 4.4.x tree.
Last edited:
Boss in Official 4.9.19
Clean compile and no tinkering after the latest commit (4.9.19+) -->
Tested master and slave mode, works good. Sounds good with squeezelite.
The Piano *Plus* is still not in defconfig:
./arch/arm/configs/bcm2709_defconfig
CONFIG_SND_BCM2708_SOC_ALLO_PIANO_DAC_PLUS=m
Clean compile and no tinkering after the latest commit (4.9.19+) -->
Tested master and slave mode, works good. Sounds good with squeezelite.
The Piano *Plus* is still not in defconfig:
./arch/arm/configs/bcm2709_defconfig
CONFIG_SND_BCM2708_SOC_ALLO_PIANO_DAC_PLUS=m
Attachments
The Piano *Plus* is still not in defconfig
Yep. I reported that already. Not nice, rather cosmetics. In bcmrpi_defconfig still both are missing!
However.
What matters is that it works now!
Jeez, all of this, and I'm pretty sure extremely few people will be able to hear a difference when compared with a 35 EUR plug and play IQAudIODAC. Even less people would be able to say "It definitely sounds better", and would be able to recognise it in an A-B test.
Jeez, all of this...
Look. What we do the moment, has no meaning for the normal user.
That should be clear.
However.
Perhaps you might get an idea what it takes to get a system and devices
properly going. U might get an idea how many people spent long hours
- many of them free of charge - even for users like u, who're just consuming
and complaining. Luckily that's the negligible amount of users. The vast majority appreciates what's being done.
Enjoy your IQaudio.
Trust me I do have a very good idea, I'm an electrical engineer, and avid DIY'er. I also contribute to the Linux Kernel since 20 years with thousands of hours and thousands of lines of code (thanks accepted). But still, I hate seeing effort and intelligence wasted on a product that, when it works, at higher prices,it will bring really no tangible advantages when compared with existing solutions. Perhaps the focus could be set on more meaningful projects... but that's just me ranting.
avid DIY'er.
DIY-Audio join date = "Dec 2016" 🙄
Perhaps the focus could be set on more meaningful projects... but that's just me ranting.
Look. Just start your own thread, with your own project. It's that easy.
This project focuses on the Allo Boss DAC. Nothing more and nothing less.
If u have anything meaningful to contribute, contribute it. Otherwise don't waste my/our time.
Just tested 352k8 and 384 with the latest kernel + patches.
I think it's worth it.
To me it just sounds cleaner without the on-dac filters in the chain ( or whatever's causing the difference).
I think it's worth it.
To me it just sounds cleaner without the on-dac filters in the chain ( or whatever's causing the difference).
Question for any owner of the Boss Dac: Does your Boss dac has a resistor on the place where I put a circle on the photo?
Attachments
Last edited:
Thanks.
I did, I followed the instructions but without success.
I desoldered the resistor. Then I connected J4. It didnt work. I tried another 5V powersupply. I didn't work. I made another J4 connection cable. It didn't work.
So I compared the photo of the PCB on the Allo website with the Boss which I have here. And there was a difference. As the photo (post 95) shows in the red circle resistor R828 is missing. And on the photo of the Boss on the Allo Website the R828 is present. Hence my question. Is R828 missing on my Boss Dac? Should I solder a 0 resistor on this R828 place? Maybe then it works?
Last edited:
Just saw that Pin 1&2 of J19 is 5V.
I'm wondering if external 5V also can be feed through these
pins once R826 (0R bridge) is removed.
I'm wondering if external 5V also can be feed through these
pins once R826 (0R bridge) is removed.
- Status
- Not open for further replies.
- Home
- Source & Line
- Digital Line Level
- My new Boss - another PI-HAT