Allo Boss DAC review

#42

Same excellent impression from the Boss DAC here. I now use it in my main rig. Tried to power the Pi with an Uptone lps1 and did not notice a difference. Will do some further tests but I doubt my first impressions will change.

(John B) #43

This looks very interesting, can I clarify tow things. On the website it says

what does this actually mean?
Can I can add this to my allo sparky as well as a my Pi 2 and 3?

Does it accept DSD and up to what rates.

Sorry if these are silly questions I really am a bit of a N00b wrt these things.

sjb

(johan ) #44

Hat size basically it has the right dimensions for a HAT , but not compliant since it has no camera grove etc.

DSD only as DOP as its a i2s DAC

(johan ) #45

Boss is a master DAC will not work with Sparky, only with RPI (rpi accepts i2s as master or slave , sparky has only master…cannot have 2 masters on i2s)

(Jesper Billing) #46

Hi Johan@Allo,
I just read this review of your Boss DAC and got really interested in getting one, BUT if I get one I’m only interested in the upgraded version, like the one Janis tested above. I live in Sweden. How to make SURE to get this new one?
Kind regards,
Jesper

1 Like
(johan ) #47

All mass production units come with right low pass filter (upgraded version like you called them). So no worries… we will never ship faulty units.

(Michel Picard) #48

Hi @allo.com
I read in an other thread that you’re going to make a case for RPI + Boss.

  • Will it be an open box (like for the vana player) or a closed box ?
  • Will it be made of acrylic or metal ? (please make a closed box in Aluminium :slight_smile: )
  • And last question : is there an ETA for this case ?
    Thanks a lot !
(johan ) #49

First acrylic box is coming soon (3 weeks max)… it will be closed but with vents and not very expensive

Second will be made in ALU closed…coming in 5 weeks more expensive

(Henry Larsen) #50

@allo.com I have just installed dietpi on the Boss, that just arrived yesterday. It works fine, but not with the 384, when I install that from Dietpi, the dac is not recognised in Roon. Any solutions for that?
It would be nice to use it beyond 192

(Michel Picard) #51

Thanks @allo.com for the quick answer, this is awesome news !

(johan ) #52

I can ask our team to look at it on Monday, but I suggest you ask the magician aka @Dan_Knight :slight_smile:

1 Like
(Jesper Billing) #53

Thanks Johan!
I can safely get it then.

I have another question regarding the Allo Boss. The PCM5122 DAC from TI used in Boss, TI states the following:

“Members of the PCM512x family integrate preset audio processing functions with programmable coefficients, allowing developers to change the characteristics of the interpolation filter, speaker EQ, dynamic range controls, and average volume control in their products.”

The “speaker EQ” I’m interested in, as it can give major improvements correcting the speakers EQ to the room. Could the Boss be programmed using a room correction program to measure the room acoustics and then input the EQ settings into the PCM5122?

I guess it´s not set up for that right now, but what can be done now, if any?

Kind regards,
Jesper

(Dan Knight) #54

Hi Henry,

I’ve replied to your post on our forums, please see below:
http://dietpi.com/phpbb/viewtopic.php?f=11&t=1480#p6163

(Henry Larsen) #55

Thank you Dan. I hope that a solution can be found at some time.

(johan ) #56

Frankly speaking…its very difficult to program the DSP functions from TI ic… you need to download some software (and its hard to get TI people to reply) and then learn how to use it.

(Jesper Billing) #57

Thanks for the answer Johan. Maybe not the answer I wished for though. TI ought to learn from companies like Apple to make “developers kit” so that their hardware products can get even more useful. Tis is the big reason for their success!

Room EQ abilities that can be done but are ignored seems like waste of resources to me, right?

I guess, if no one asks nothing happens, but if you show interest into it, it could make a difference. I understand that companies like TI of course, like many companies, look to help their big customers first.

You have very competitive products today, fairly new to the RPi sector. What I really like is the way you listen to your customers - you have too much to win on doing that! Just see the feedback you got from Janis to correct your product, going from good to excellent. Then think if your products suddenly had room correction abilities - that would for sure beat the competition!

(Henry Larsen) #58

@allo.com - I agree with he comment from Jesper about your willingness to listen. But to understand the problems with the questions I asked you and Dan from DietPi: With the driver for the Boss, 352 and 384 kHz is red, DSD 64 is yellow and the rest is red in Roon. Although the Boss hardware-wise is ready for 384, and possibly also DSD 128 and 256, it is not possible because of software problems with TI. Is that the case? Or is it a problem with Linux in the current version?

(johan ) #59

The main problem is that DSP cycles in DAC ic are very limited…and to do room correction you need lots of processing power. So its not possible.

Besides why would you do that in DPS? You can use a software like BruteFir and do it directly on the CPU of RPI.

(johan ) #60

No problems with the TI, but 384Khz is no part of Rasbian. There was a guy (brilliant) CliveM that wrote the code , it was never accepted in rpi mainline (never submitted)…so most audio distribution took his patch and used it. However , every new version of Raspbian , makes it more and more difficult to use that patch. I think that 384Khz will completely disappear from RPI soon enough if Clive is not coming back.

(Jesper Billing) #61

Thanks again Johan for enlightening into some of the valuable implications into doing both EQ and 384 kHz. Now I read up on some things in the mean time.

When it comes to sampling rates (384 kHz etc) there is the interesting fact of how small time differences we can hear. Like some displays are said to be “retina” display, meaning small enough pixels to not be able to see the pixels by the eye, there are the same fact of sample rates. When the sample rate is 192 kHz each time between the samples are 5,2 microseconds. Our hearing of time differences are very good, about 6-7 microseconds, which makes a sample rate of 192 kHz small enough to not distinguish any smaller difference. But of course it´s easier to make a good filter on a higher sampling frequency, so the DAC in Allo Boss (TI PCM5122) is working internally on 384 kHz. This means the signal is up sampled to 384 kHz. So internally the filter then can be more optimal having 384 kHz than “only” 192 kHz, but taking samples above 192 kHz is probably not going to make things much better.

What does matter is having a very good clock signal, like Allo Kali have. Actually, Allo Boss I didn’t see data on the clock jitter. Maybe you have data on how the Boss’s clock compare to All Kali’s?

The other factor is having a stable and clean voltage feed to the analog circuits and clock generator, as this becomes direct distortion added to the output signal. The Allo Boss is designed, as far as I understand, with this philosophy in mind. Correct me if I got it wrong.

So the gain in audio quality is much bigger handling jitter and a stable voltage feed than 192 to 384 kHz to my reality.