• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

Introducing the Buffalo III-SE-Pro 9028/9038

Jump to Latest
T

twluke

Member
Joined 2012
Russ White said:
That's correct - in this mode you would only need 90/98 clocks if you want PCM at ~700Khz or DSD1024 - but lower sample rates would still work just fine as long as they are in sync with the master clock.
Click to expand...

Working with the lower rate is probably reflecting the clock divider function given to 9038/9028 chips, though only based on the reading somewhere.

I think Twluke's initial experiment was hampered by a sagging supply at the very high sample rates - otherwise it should have worked as expected.
Click to expand...
You are quite right. After re-building the case for the 9038/Mercury combo, the 9028/IVY-III combo was moved into the case originally for the old 9018 and I was too lazy to take care of PSU for them, only supplying the power from an LCDPS sharing with BBB through a 9V+9V trafo!

Though this has been quite okay for the DAC in async mode, I noticed that the BBB/Hermes/Cronus tends to stop abruptly with this SyncMode firmware, particularly with the DPLL to the lowest for DSD512, which is wonderful in SQ.

Well, now it appears the time to renew the PSU, together with building another Mercury, remaining still unsoldered. :D Thank you for your proper pointing.
 
T

twluke

Member
Joined 2012
Russ White said:
I believe all that "sync mode (128-fs)" does is remove the DPLL from the picture. :) I will be doing some more experimenting along those lines when I have some time.

The nice thing is - if that theory proves true - we have 4 switches available to be mapped to different purpose - because there is no use setting any DPLL bandwidth value. :)
Click to expand...

Oh, thanks for this correction of my misunderstanding.

Regards,
 
T

twluke

Member
Joined 2012
Russ White said:
I am really glad to hear of your experimenting and success so far. Myself - I am very pleased with the sounds I am hearing from the DAC in true "sync mode".
Click to expand...

Hi Russ,

This weekend, I hooked up this SyncMode firmware to another B3SE pro (9038) which was connected from BBB-Herms/Cronus and to Mercury. Playing upsampled DSD512 from HQ player/Roon via BBB went quite smooth and I found the 90/98 MHz clocks to work well without noise, though the playing rate was twice faster than normal as seen in the 9028 pro experiment.

The combination of B3SEpro and Mercury with the firmware set to sync mode gave me surprisingly high quality sound of DSD512, probably in a different league, I must admit, from the one in async mode.

It's quite amazing to hear the DAC sound of no band width.

BTW, while collecting the info on 128fs in Sabre 9038 pro, I came across the description in a manual of a 9038 pro product like below:

the MCLK should be equal to 128fs when the input sample rate is 384kHz or below. When the input sample rate is above the 384khz, MCLK should be equal to 64FS.
Click to expand...
As I don't have the data sheet, I'm just wondering this can be an original feature of the 9038 DAC chip or not.

Again thank you for your great work.
 
T

twluke

Member
Joined 2012
Russ White said:
The main point is that in 128fs sync mode with 90mhz clock 44.1khz music plays perfectly normally.
Click to expand...

Thanks. Here is a good news. I finally could manage the 90/98 clocks to work normally for DSD512 and other lower formats with sync mode!

I'm using a debian linux with Botic driver on the BBB for playing music by MPD or using as an audio conduit as NAA for HQ player. When the Hermes and Cronus are applied to this system, it's my understanding that the Cronus takes over the master clock rate using jumper pins: 1/2 for 45/49 and 1/4 for 90/98. So I thought there would be no need for command line setting of MCLK in /boot/uEnv.txt.

As there was no idea how to solve my double speed problem with 90/98 clocks for DSD512, I just modified the command line in /boot/uEnv.txt like below:

Code: 
cmdline=coherent_pool=1M quiet snd_soc_botic.clk_44k1=45158400 snd_soc_botic.clk_48k=49152000
To my surprise, my problem has totally gone and finally I could enjoy the music of DSD512 in sync mode with 90/98 clocks. Though not sure, this empirical result may suggest that the MCLK defined by 128fs for DSD512 play must be explicitly indicated in the boot option as long as the 90/98 clocks under the Botic system is concerned.

Regards,
 
Russ White

Russ White

Member
Joined 2005
twistedpearaudio.com
twluke said:
Hi Russ,


The combination of B3SEpro and Mercury with the firmware set to sync mode gave me surprisingly high quality sound of DSD512, probably in a different league, I must admit, from the one in async mode.

It's quite amazing to hear the DAC sound of no band width.

Again thank you for your great work.
Click to expand...

I was a bit busy last night when I originally read this - but I just wanted to say thanks!

I really appreciate that feedback. I Have been listening in "pure sync" mode all week - and I can't disagree with you. It's really fantastic.

I am making some changes to the firmware I will push up to github soon - since any setting regarding DPLL are totally irrelevant in that firmware :)

Cheers!
Russ
 
F

fred_com

Member
Joined 2009
Hi Russ,

thanks for this "true sync" firmware - I really enjoyed listening to music with it. Music seems more alive, so to speak :)

Automute seem to be working, but I still hear a pop when starting a new track, and lock led is lit all the time, regardless of playback.

Waiting for the updated version. :)

Thanks,
Fedor.
 
Last edited:
L

Luca72c

Member
Joined 2013
Hello Russ,
would it be possible to have a replacement chip with the new true sync firmware onboard?
I'm not equipped (and able, without clear instructions) to flash the new firmware to my Buffalo IIIpro IC; more, i would like to keep spdif input capabilities "by hand"...
Obviously i would pay for it.
 
Russ White

Russ White

Member
Joined 2005
twistedpearaudio.com
fred_com said:
Hi Russ,

thanks for this "true sync" firmware - I really enjoyed listening to music with it. Music seems more alive, so to speak :)

Automute seem to be working, but I still hear a pop when starting a new track, and lock led is lit all the time, regardless of playback.

Waiting for the updated version. :)

Thanks,
Fedor.
Click to expand...

Did you set the automute switch? Switch 2 posistion 1 needs to be off (binary 1) for automute to be enabled. :)

With automute enabled I have no objectionable noise at all on any transition when using Amanero/Hermes/Cronus in pure sync mode. With it disabled I do.

Cheers!
Russ
 
Last edited:
Russ White

Russ White

Member
Joined 2005
twistedpearaudio.com
Luca72c said:
Hello Russ,
would it be possible to have a replacement chip with the new true sync firmware onboard?
I'm not equipped (and able, without clear instructions) to flash the new firmware to my Buffalo IIIpro IC; more, i would like to keep spdif input capabilities "by hand"...
Obviously i would pay for it.
Click to expand...

The firmware still needs some polish - but when I am happy with it I will talk with Brian about multiple firmware options.
 
F

fred_com

Member
Joined 2009
Russ White said:
Did you set the automute switch? Switch 2 posistion 1 needs to be off (binary 1) for automute to be enabled. :)

With automute enabled I have no objectionable noise at all on any transition when using Amanero/Hermes/Cronus in pure sync mode. With it disabled I do.

Cheers!
Russ
Click to expand...

Yes, I've set it to off, and the mute led do light up, when I pause the playback, so I guess it's working. I can reproduce this pop when unpausing the playback or starting a new track. It does not happen when pausing or when track changes by itself in the playlist.

But I'm using the Hermes/Cronus/BBB, so this may be because of the software (MPD most likely).

But it's not a big deal, really, so please don't waste much time on it :)
 
P

Paul M

Member
Joined 2005
I finally got the modified firmware on a new ATtiny85 (somehow I managed to brick or fry the original since couldn't even reset the fuses with HVSP).

I can confirm that modifying the firmware as in post #511 results in having both PCM/DSD in "pure sync" mode and SPDIF in async mode. It works well using the original method of switching between the two inputs.

In pure sync mode I can echo what Twluke and Fedor have already said: the sound is more open and lively while remaining "full bodied". Sounds very good! The only issue I have is that with DSD512 I get a high pitched constant noise superimposed on the music, and only in the right channel. Changing back to DSD256 the noise completely disappears. Not sure if this could be a firmware issue, or any of the other parts that come before the buffalo?
As for the pop on pausing the music, I cannot reproduce it, but maybe it's because I have the Buffalo hooked up to the NTD1 which has coupling caps on the output?

Paul
 

Attachments

  • IMG_3229.jpg
    IMG_3229.jpg
    1 MB · Views: 494
You must log in or register to reply here.
Top Bottom