Filter brewing for the Soekris R2R

[vergebnisse]

I loaded the package you recently posted onto a dam1121 running f/w 1.21 and it locked on right away. When I then loaded a different package (from the older sets,) I was back to a flashing LED (looks like a pattern of six flashes followed by a different-length flash) and no joy. Reloading the OEM has ot running again.

The assistance is appreciated. I’d assumed/hoped it was a common enough issue as to be answerable with, “you’re out if luck,” or, “yeah, happens all of the time, here’s the fix that you didn’t find.” Troubleshooting audio gear is part of my job anyhow, do im certainly game to work this put in steps if you/the community are up for it.

Thank you for the time/patience,

Eric

 

[vergebnisse]

As a last-ditch try before heading into AmpTown for the say, I tried loading the DIYA_1021filt_NoDC… pack and it, somehow, actually runs. Is it that the other filters I tried only function on a pre-1121 firmware version?

I imagine this is a significant data point.

Thank you again,

Eric

 

[janho12345]

Eric,

The dam1121 is way younger than the dam1021 rev. 1 for which the filters you want to use were intended for.

Did you not like my filter set?

If you had the coefficients of the old filters you could write them into a new filter file and make an skr out of it.
But,
As many of the old filters did only come with a skr file and not txt file, you will not be able zu use them.

 

[TNT]

DIYA_1021filt_NoDC

The F4 of the set is great ;-)

//

 

[ghitus]

Hi guys, is there a good filter I can use with a dam1021 rev 6 with 1.23 firmware?

Thanks in advance,
Gaetano.

 

[gumisb]

I noticed in rev7 that second stage filters don't switch according to selected filter. One S2 short filter is serving all the first stage filters.

Can it be fixed?

# filters

05 2 FIR1, 44.1 Ksps, F5
14 S2 1.5, 384 Ksps -> 1.5 Msps, (f6) c

# exit

L044
F7


dam1021 uManager Rev 1.24 20210530 FPGA Rev 1.24 Press ? for help.

# filters

07 4 FIR1, 44.1 Ksps, F7
14 S2 1.5, 384 Ksps -> 1.5 Msps, (f6) c

# exit

L044
F5


dam1021 uManager Rev 1.24 20210530 FPGA Rev 1.24 Press ? for help.

# filters

05 2 FIR1, 44.1 Ksps, F5
14 S2 1.5, 384 Ksps -> 1.5 Msps, (f6) c

12 S2 1.5, 352.8 Ksps -> 1.5 Msps, (f4)
12 S2 1.5, 384 Ksps -> 1.5 Msps, (f4)
13 S2 1.5, 352.8 Ksps -> 1.5 Msps, (f5) c
13 S2 1.5, 384 Ksps -> 1.5 Msps, (f5) c
14 S2 1.5, 384 Ksps -> 1.5 Msps, (f6) c
15 S2 1.5, 352.8 Ksps -> 1.5 Msps, (f7)
15 S2 1.5, 384 Ksps -> 1.5 Msps, (f7)

 

[TNT]

It's a pity that this whole filter thing couldn't be properly described. Why have a customer benefit and half hide its function is beyond me.

Do the right thing and make a "Application Information" paper of the user interface for custom designed filters. Its sold - so deliver it.

//

 

[janho12345]

If you look at the txt file for the rev. 7 board, there is just no other filter defined than the 14 S2 for the second stage. So there is no fault in the switching.

If you want other second stage filters, you have to make your own set.

 

[gumisb]

According to the pattern from the header of the original filter file for rev7, I made my own set and the switching does not work properly in it.

This is orginal

--
-- DSP Filter for the Soekris dam1xxx, rev 1.24 dated 20210505
--
-- Format:
-- signature, samplerate, interpolationrate, type, numbercoefficients, multiplier
-- type 1 = FIR1, 2 = FIR2, 16..30 = IIR, 30 = Deemphasis
-- type 4..7 = FIR1 new format, add 1000 for shared filters
-- type 8..11 = FIR2 new format, add 1000 for shared filters
-- type 12..15 = FIR2, new 1.5M final rate, add 1000 for shared filters.
--
-- IIR Coefficients is b0, b1, b2, a1, a2. a0 is assumed to be 1.0000
--
-- FIR1 filters for 44.1K/48K Samplerate

dam1021,352800,4,1014,239,4
14 Minimum Phase FIR2, 352.8 Ksps -> 1.5 Msps, 120 Khz -1 db, 60 db/oct

dam1021,384000,4,14,239,4
14 Minimum Phase FIR2, 384 Ksps -> 1.5 Msps, 130 Khz -1 db, 60 db/oct

 

[janho12345]

And would you share the headers of your implementation? So we can see and learn and maybe help

 

[gumisb]

Yes.

dam1021,352800,4,1012,239,4
12 S2 1.5, 352.8 Ksps -> 1.5 Msps, (f4)

dam1021,384000,4,12,239,4
12 S2 1.5, 384 Ksps -> 1.5 Msps, (f4)


dam1021,352800,4,1013,239,4
13 S2 1.5, 352.8 Ksps -> 1.5 Msps, (f5) c

dam1021,384000,4,13,239,4
13 S2 1.5, 384 Ksps -> 1.5 Msps, (f5) c


dam1021,352800,4,1014,239,4
14 S2 1.5, 352.8 Ksps -> 1.5 Msps, (f6) c

dam1021,384000,4,14,239,4
14 S2 1.5, 384 Ksps -> 1.5 Msps, (f6) c



dam1021,352800,4,1015,239,4
15 S2 1.5, 352.8 Ksps -> 1.5 Msps, (f7)

dam1021,384000,4,15,239,4
15 S2 1.5, 384 Ksps -> 1.5 Msps, (f7)

 

[janho12345]

That looks fine to me. You could try and see what happens if you upload a filter set where filter 14 is missing.

 

[gumisb]

Good idea for troubleshooting but leads me nowhere, and I think that only Soren can fix it.

 

[soekris]

I decided that a basis butterworth lowpass filter was fine for the last FIR filter, and just implemented that in the 1.5M filters, and even hardcoded it in the uC code, meaning it only look for "filter 14".... It's really the first FIR filter that's important anyway....

 

[TNT]

In my last filters which I haven't published here (yet?), I did alterations to FIR2 and that was a real big improvement to me.

//

 

[ghitus]

Hi @TNT, some time ago, I upgraded my rev6 dam1021's firmware to 1.23 version. From what I recall, when the firmware is upgraded, the other filters which have been loaded previously are erased. Could the last filters you were speaking about be uploaded on my rev6 dam1021? If yes, did you publish them somewhere?

Thank you,
Gaetano.

 

[TNT]

No.

//

 

[ghitus]

Thanks.

Gaetano.

 

[psilonaught]

hey guys,

I was hoping one of you guys might be able to help me test and potentially re-flash a dam1121-01 board. I think the firmware is corrupted and I need someone who is familiar with these boards to take look ideally

Someone in the EU ideally, so postage is more easy. I'd be more than happy to offer a fee for the service.

Thanks

 

[gumisb]

Improved time domain optimized filter. Only one (F4) due to rev7 limitations and support for only one 2nd stage filter. Works best with volume set to -2. Supports 44.1kHz, 48kHz, 88.2kHz and 96kHz.