• 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.

DSD512 on Buffalo IIIse

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
nautibuoy

nautibuoy

Member
Joined 2010
Russ, over on the Botic thread we (twluke, miero and I) have been making good progress with getting a BBB running as a boticized Network Audio Appliance (NAA) for use with HQPlayer (HQP). In so doing, we've been experimenting with HQP's upsampling and have found the BBB/Botic is capable of PCM at up to 384KHz. It is also capable of up to Octa-rate DSD (DSD512) - almost!

twluke and I are both using Buffalo IIIse DACs.

With DSD512 we are experiencing short bursts of a static-like noise, which then fades away. This is maybe a good point at which to pick up the story:

http://www.diyaudio.com/forums/twisted-pear/258254-support-botic-linux-driver-182.html#post4776058

Do you have any thoughts about what is going on?

TIA

Ray
 
nautibuoy

nautibuoy

Member
Joined 2010
Thanks Brian.

I'm not using Amanero at all, just BBB.

Just to ensure clarity, twluke is using hermes/cronus with his BBB whereas I'm using brand X isolator/reclocker. We're both using B3SE and we're both getting the same symptoms.

I'll see if I can monitor the trident output voltages.
 
Last edited:
T

twluke

Member
Joined 2012
BrianDonegan said:
You are experiencing the noise with both the BBB and the Amanero?

Try checking local supply (Trident) voltage when you are hearing the noise. Higher bandwidth music demands more current, which *might* be causing a voltage sag.
Click to expand...

I checked the supply voltage of trident when playing DSD512 via each of (a) BBB/Hermes/Cronus with noisy condition and (b) a prototype of BBB/Botic system with no audible noise. The CCS on Placid HD supplying trident was 720mA (= 720mV). The results are as follows:

VDD/VDD_XO/DVCC
(a) 3.2/3.2/1.1
(b) 3.2/3.2/1.1

I tried the CCS up to 850mA but the situation of noise from Hermes/Cronus remained unchanged. So I made the CCS back to the origina vlaue.

In summary it may be possible to say that the current noise issue is not relevant to the trident voltage sag.

Regards,
 
nautibuoy

nautibuoy

Member
Joined 2010
Thanks Russ. I've turned off spdif autodetect and, fingers-crossed, so far after about 30mins of listening at 22MHz DSD I haven't experienced the static noise.

Checking the B3SE integration guide for the spdif setting I was reminded of the IIF bandwidth options and have been experimenting with them; I think I marginally prefer the 70KHz setting to the previous 'Normal' setting but will experiment further over the weekend.

I need to work out how to update the firmware next.

Ray

NB. Should have said that all of the wiring is via short ufl cables.
 
Last edited:
T

twluke

Member
Joined 2012
Russ White said:
Make sure the SPDIF autodetect feature of the ES9018 is turned off - it can cause all sorts of problems with DSD.

Also at those rates wiring will be extremely critical.
Click to expand...

Thank you Russ for your kind advice. As I'm using a switch between I2S/DSD and SPDIF on B3SE, I have no problem with playing DSD512 as long as I use the BBB/Botic of original wiring by miero. The system became more stabilized after an increase of PSU current to 900mA (Placid HD, 5V) supplying the BBB.

However, when playing DSD512 with Hermes/Cronus for BBB/Botic with 45/49 Crystek clocks, something similar to white noise periodically appears, that is quite annoying.

What appears more weird is the Hermes/Cronus fails to send signal to B3SE when trying to play DSD512 with 22.5/24.9 Crystek clocks while the BBB/Botic is working as NAA. With the same clocks, it can send DSD256 signal to B3SE without problems. The issue is reproducible among all of three Hermes/Cronus systems I built.

I don't know why such things can happen but for now I gave up using Hermes/Cronus for playing DSD512 mainly because of being unable to eliminate the noise. BTW the B3SE is set async. So, the situation may change with sync mode.

Regards,
 
T

twluke

Member
Joined 2012
Russ White said:
22/24Mhz are too slow for DSD512 - as it most be at least 2 X bitclock

And with 45/49 you absolutely must disable SPDIF autodetect on the ES9018 via the firmware. I am not talking about the multiplexer switch. :) I am talking about the setting set by DIP switch.
Click to expand...

Hi Russ, thank you for enlightening me. There appears to be further things to do with Hermes/Cronus for high sampling rate play. I'll look for clocks 4 times higher.

Unfortunately disabling S/PDIF autodetect by DIP switch has resulted in no sound from B3SE. It appears that I should update firmware.

Regards,
 
nautibuoy

nautibuoy

Member
Joined 2010
I've been listening more and have now settled for running the BBB/Botic/NAA/B3SE at DSD256 max. After plugging in my objective head I compared DSD256 & 512 and, even with most of the noise issues resolved, I decided that DSD256 is preferable as I am perceiving some edginess at DSD512; 256 seems to be more relaxed. I'm not really surprised given that 512 is beyond the ES9018 specs.

I still consider the experiment to be a success though as I now have a very capable and nice sounding network DAC to use with HQPlayer. To achieve DSD256 rates is positive as I thought even that might be pushing the envelope, but to have gained high confidence that I can explore DSD512 further with a suitably spec'ed DAC is a bonus.

Ray
 
francolargo

francolargo

Member
Joined 2007
Paid Member
Russ,
Thanks for checking in! Could you comment on the workings of 'automute' in the es9018? I wonder if automute might be able to intercept and block the start or stop noise that Ray is getting with DSD512. [Ray, do you get the same noise at DSD256?] I see that the default settings are 104dB and, for time, "4" [=.68sec@384kHz?]. Thus, the time could be increased by a factor of 4 from default. ...not sure what params you have in the firmware...

Do you think that changing the thresholds could nip that start/stop noise?
TIA!
 
Last edited:
Russ White

Russ White

Member
Joined 2005
twistedpearaudio.com
DSD512 is actually not at all beyond the specs. :) As long as you obey the timing rules in regard to the master clock ratio. That said I can't think of any good reason to use it - DSD256 is already pushing the limits of "bits to bang" signal quality ratio. :)

As for automute - generally I don't use automute at all - I completely disable it. You can see the source code on github.

@twluke - even with the older firmware you should get lock on PCM/DSD with SPDIF detection off...(switch 2 - position 6 - off) so something is afoot. Maybe send me a pic of the switch settings you have right now.
 
T

twluke

Member
Joined 2012
twluke said:
It's actually the position 5 of SW2 according to lenonvb's Integration Guide. Position 6 is assigned to IR bandwidth...
Click to expand...

My bad! I realized that my DIP setting was not for B3SE but for B3. I'm so sorry, Russ, for my misleading post. My sincere apologies to you.

Now I've set the Position 6 on SW2 off and have been listening to DSD512 via Hermes/Cronus with 45/49 ***** for an hour. The noise is still present but seems much less pronounced and less frequent than before. The situation may be also dependent on the seection of filters provided by the music player. I'll continue listening with this setting for more hours to check if there is any further change in noise.

BTW, can I use a PIC microchip instead of an Atmel chip for firmware?

Regards,
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom