Ah yes - sorry I was looking originally at the Xmox board for usb-i2s with a 1211 and one of them provides pin outs for saying what it is currently interpreting the byte stream as and the rate.
Yes - its strange actually...
Playing 44,1....
Filters all:
....
08 Linear Phase Soekris FIR2, 352.8 Ksps, 102 Khz -1 db, 177 Khz -130 db
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
....
Filters ->
04 Linear Phase TNT5, 44.1 Ksps, 17 Khz - 0.3 db, 22.04 Khz -183db
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
29 DC Blocking IIR, 352.8 Ksps, 2 Hz HP 1st order
30 Deemphasis IIR, 352.8 Khz Samplerate, hiself f=5600 Hz, Q=0.485, gain=-10.1
So the DAM seem to have choosen filter 29 and 30 because of 44,1 beeing played, but for some reason:
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
instead of
08 Linear Phase Soekris FIR2, 352.8 Ksps, 102 Khz -1 db, 177 Khz -130 db
???
Why is that Sören?
//
Then I have an other filter handling observation:
Did just play a red-book song - playback stoped:
# filters
04 Linear Phase TNT5, 44.1 Ksps, 17 Khz - 0.3 db, 22.04 Khz -183db
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
16 230 2 1
17 1,5k 2 3
18 6k 2 -1,5
19 70 3 -1,5
20 HS4k 0,5 1*
21 900 2 -1
22 12k 1,5 2,5*
24 650 1 -1
25 4k 1 -1
26 3100 0,5 -1,5
27 1000 2 -1*
28 L90 2 -3
29 DC Blocking IIR, 352.8 Ksps, 2 Hz HP 1st order
30 Deemphasis IIR, 352.8 Khz Samplerate, hiself f=5600 Hz, Q=0.485, gain=-10.1
# exit
L044
Playing a radio web feed (48k)
L048
+++
dam1121 uManager Rev 1.08 20160523 FPGA Rev 1.08 Press ? for help.
# filters
04 Linear Phase TNT5, 48 Ksps, 17 Khz - 0.3 db, 22.04 Khz -183db
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
29 DC Blocking IIR, 384 Ksps, 2 Hz HP 1st order
30 Deemphasis IIR, 384 Khz Samplerate, hiself f=5600 Hz, Q=0.485, gain=-10.1 dB
#
Where did my PEQs go? OK, they are Fs dependent...
# exit
L048
+++
Change to play a red-book song
#filters
04 Linear Phase TNT5, 48 Ksps, 17 Khz - 0.3 db, 22.04 Khz -183db
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
29 DC Blocking IIR, 384 Ksps, 2 Hz HP 1st order
30 Deemphasis IIR, 384 Khz Samplerate, hiself f=5600 Hz, Q=0.485, gain=-10.1 dB
Hmm, where are my PEQs - I hear music... but it sound strange and soundscape is tilted to the right... I do:
# exit
L044
WOW - sound is OK and in centre as I entered "exit" - same instance. +++
# filters
04 Linear Phase TNT5, 44.1 Ksps, 17 Khz - 0.3 db, 22.04 Khz -183db
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
16 230 2 1
17 1,5k 2 3
18 6k 2 -1,5
19 70 3 -1,5
20 HS4k 0,5 1*
21 900 2 -1
22 12k 1,5 2,5*
24 650 1 -1
25 4k 1 -1
26 3100 0,5 -1,5
27 1000 2 -1*
28 L90 2 -3
29 DC Blocking IIR, 352.8 Ksps, 2 Hz HP 1st order
30 Deemphasis IIR, 352.8 Khz Samplerate, hiself f=5600 Hz, Q=0.485, gain=-10.1
- - - - - - - -
So, why is filters not engaged at media Fs change until one does "exit" if one happened to be in umanager (+++) ??
Problem: If you are in umanager after +++, i.e. at promt, filters will not be changed at Fs change until you do "exit".
//
Did just play a red-book song - playback stoped:
# filters
04 Linear Phase TNT5, 44.1 Ksps, 17 Khz - 0.3 db, 22.04 Khz -183db
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
16 230 2 1
17 1,5k 2 3
18 6k 2 -1,5
19 70 3 -1,5
20 HS4k 0,5 1*
21 900 2 -1
22 12k 1,5 2,5*
24 650 1 -1
25 4k 1 -1
26 3100 0,5 -1,5
27 1000 2 -1*
28 L90 2 -3
29 DC Blocking IIR, 352.8 Ksps, 2 Hz HP 1st order
30 Deemphasis IIR, 352.8 Khz Samplerate, hiself f=5600 Hz, Q=0.485, gain=-10.1
# exit
L044
Playing a radio web feed (48k)
L048
+++
dam1121 uManager Rev 1.08 20160523 FPGA Rev 1.08 Press ? for help.
# filters
04 Linear Phase TNT5, 48 Ksps, 17 Khz - 0.3 db, 22.04 Khz -183db
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
29 DC Blocking IIR, 384 Ksps, 2 Hz HP 1st order
30 Deemphasis IIR, 384 Khz Samplerate, hiself f=5600 Hz, Q=0.485, gain=-10.1 dB
#
Where did my PEQs go? OK, they are Fs dependent...
# exit
L048
+++
Change to play a red-book song
#filters
04 Linear Phase TNT5, 48 Ksps, 17 Khz - 0.3 db, 22.04 Khz -183db
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
29 DC Blocking IIR, 384 Ksps, 2 Hz HP 1st order
30 Deemphasis IIR, 384 Khz Samplerate, hiself f=5600 Hz, Q=0.485, gain=-10.1 dB
Hmm, where are my PEQs - I hear music... but it sound strange and soundscape is tilted to the right... I do:
# exit
L044
WOW - sound is OK and in centre as I entered "exit" - same instance. +++
# filters
04 Linear Phase TNT5, 44.1 Ksps, 17 Khz - 0.3 db, 22.04 Khz -183db
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
16 230 2 1
17 1,5k 2 3
18 6k 2 -1,5
19 70 3 -1,5
20 HS4k 0,5 1*
21 900 2 -1
22 12k 1,5 2,5*
24 650 1 -1
25 4k 1 -1
26 3100 0,5 -1,5
27 1000 2 -1*
28 L90 2 -3
29 DC Blocking IIR, 352.8 Ksps, 2 Hz HP 1st order
30 Deemphasis IIR, 352.8 Khz Samplerate, hiself f=5600 Hz, Q=0.485, gain=-10.1
- - - - - - - -
So, why is filters not engaged at media Fs change until one does "exit" if one happened to be in umanager (+++) ??
Problem: If you are in umanager after +++, i.e. at promt, filters will not be changed at Fs change until you do "exit".
//
Last edited:
Is the output switching any different between the two rates - something that could identify what rate it actually is with an osciliscope on the LVC clock pulses?
Well, there seem to be 1 filter for each hierarchy - right? Did you not supply these?
08 Linear Phase Soekris FIR2, 352.8 Ksps, 102 Khz -1 db, 177 Khz -130 db
08 Linear Phase Soekris FIR2, 384 Ksps, 111 Khz -1 db, 192 Khz -130 db
//
Well. I downloaded filters from # 1833. And they play me a lot better than the ones from # 1700. F4 sounds more open, more resolving. The instruments are more tangible - it is more pleasant to listen to. F7 plays very quietly.
TNTseems like you did a good jobThank you.
Glad the worked for you. It probably means that you have a quite balanced system and appreciate a reproduction that mimics reality. That would be my bet anyway.
//
My DAC is very well developed with a buffer on d3a tubes. Soekris tile after many modifications. Speakers JBL L100 century and Crossover Sword and Dennis Murphy. Class D amplifiers - hypex NCore nc500. Yes. I think this system is balanced.
I strive for the most natural sound possible. Your filters help me a lot. Thank you again.
Regards
TNT filter is easily the best, out of the rest I like c128dp but they all have a similar issue in sound that makes them less enjoyable for long listening session.
Since that filter has more attenuation more I usually listen to TNT filter at 0dB and others with a few dB attenuation from digital volume to get desired level, there is the combination of the filters having greater headroom and absence of any extra digital attenuation with TNT filter, one or both could be part of the reason so I want to try editing the other filters.
I think by adjusting the multiplier it should be pretty easy to adjust filter levels.
If so, is it ok to play around with values to get a similar level or does it need to follow some rules?
say so you dont end up with -3.19287482dB instead of -3db
Also it uses stock soekris FIR2 or modified version?
Since that filter has more attenuation more I usually listen to TNT filter at 0dB and others with a few dB attenuation from digital volume to get desired level, there is the combination of the filters having greater headroom and absence of any extra digital attenuation with TNT filter, one or both could be part of the reason so I want to try editing the other filters.
I think by adjusting the multiplier it should be pretty easy to adjust filter levels.
If so, is it ok to play around with values to get a similar level or does it need to follow some rules?
say so you dont end up with -3.19287482dB instead of -3db
Also it uses stock soekris FIR2 or modified version?
I'm not so sure it matters whether you implement the attenuation in the volume control, filter or multiplier. They are all arithmetic and floating.
The only multiplier that is truly transparent is 2X 4X (0,5 0,25) etc i.e. 6 dB steps and implemented with a right or left bit shift. That aint gonna happen in a DAM.
TNT filter uses stock FIR2.
//
and floating.The only multiplier that is truly transparent is 2X 4X (0,5 0,25) etc i.e. 6 dB steps and implemented with a right or left bit shift. That aint gonna happen in a DAM.
TNT filter uses stock FIR2.
//
Last edited:
To my best understanding, it can. I don't own that DAC. Maybe someone else here has done it?
There should be no harm loading it - you can always revert to the origignal filter file. No FW change involved. But maybe you should be careful as you try it for the first time? Headphone on very low volume just to check - watch your ears!!!
Just to be really safe - I think it will work.
//
There should be no harm loading it - you can always revert to the origignal filter file. No FW change involved. But maybe you should be careful as you try it for the first time? Headphone on very low volume just to check - watch your ears!!!
Just to be really safe - I think it will work.
//
Hi,
It worked, using on uManager download and then filters all...
The F4 filter for 44.1 is really very enjoyable for long listening session with headphone. It is very resolving, well balanced and really the best match for HD800s
The only issue i have is that since i downloaded the filters i have clipping for Line output volume (0) and no more at -1 !
Is it possible to add a little attenuation more ?
Or at least get txt and a bit of how to ?
Many Thanks for your work....