Filter brewing for the Soekris R2R

A few comments about the posts above:

- the pops do not sound "like everything breaking up" at all. They appear only at the very peak of the signal and only if it is recorded close to 0db. Solo piano can be excellent for testing, it even reminds of slightly worn out vinyl.

- completely agree with Spikestabber about the 4k filters vs Paul's. No comparison at all

- not sure i like anything processed by HQPlayer. Or even not processed. Have recently compared JRiver running locally vs HQPlayer running through NAA with all processing switched off. Against all prejudice JRiver imho provides a much better sound. Some of HQP filters are interesting and impressive after a short listen, but i cannot find one i consistently prefer as a substitute to Paul's filters.
This test has been done by upsampling in HQP to 384 in order to bypass the DAM filters. It is perhaps interesting to try a lower upsampling frequency together with the NOS filter.

When is Paul releasing his dac? Please sign me up :)
 
A few comments about the posts above:

- the pops do not sound "like everything breaking up" at all. They appear only at the very peak of the signal and only if it is recorded close to 0db. Solo piano can be excellent for testing, it even reminds of slightly worn out vinyl.

- completely agree with Spikestabber about the 4k filters vs Paul's. No comparison at all

- not sure i like anything processed by HQPlayer. Or even not processed. Have recently compared JRiver running locally vs HQPlayer running through NAA with all processing switched off. Against all prejudice JRiver imho provides a much better sound. Some of HQP filters are interesting and impressive after a short listen, but i cannot find one i consistently prefer as a substitute to Paul's filters.
This test has been done by upsampling in HQP to 384 in order to bypass the DAM filters. It is perhaps interesting to try a lower upsampling frequency together with the NOS filter.

When is Paul releasing his dac? Please sign me up :)

Hmm interesting i find that hqplayer filters do a better job than C960_140dp with my dac1541 with removed dc blocking filter.
 
Yup I am back on c128dp linear phase and am loving the heck out of it once more, it's even better since the extra mods. Wow. I just don't get how the stock filters can be considered good, it's great knowing the old filters do work in the newer firmwares. I am just playing straight 44.1 material no hqplayer no nothing, the dac just comes to life like an second unleashing. There is a reason why I used this one before for a good 2 years. If magic was the single word to explain it, whatever Paul put into this one it sure is something special. Even things like brass in film scores feel just more in depth 3d, its surreal.
 
Last edited:

TNT

Member
Joined 2003
Paid Member
Okey - here is my filter for 44,1 without DC blocker... (in the linear / F4 position).

For me, this sound much more real than the c128dp which I find sound like it all take parts in a "box" and that box follows every recording - live or studio. It's not bad but still a very obvious signature - makes it disqualified for me.

//
 

Attachments

  • TNT2.skr.zip
    36.6 KB · Views: 54
Phew thought i had bricked my dac as the flashing was always terminating midway, but it seems the serial ttl usb adapter was not getting enough power from the keyboard, i plugged in the other usb cable end for extra power on the rear of my pc and seems everything is fine so far, i did need to reflash the full firmware file though.


Bought a FTDI serial cable too thinking that this one was crapping out so i wasted 30$ on that too.


Okey - here is my filter for 44,1 without DC blocker... (in the linear / F4 position).

For me, this sound much more real than the c128dp which I find sound like it all take parts in a "box" and that box follows every recording - live or studio. It's not bad but still a very obvious signature - makes it disqualified for me.

//


Will your filter work for 48khz too ?
 
Last edited:
Okey - here is my filter for 44,1 without DC blocker... (in the linear / F4 position).

For me, this sound much more real than the c128dp which I find sound like it all take parts in a "box" and that box follows every recording - live or studio. It's not bad but still a very obvious signature - makes it disqualified for me.

//


Hmm comparing your filter to HQplayer poly-sinc-xtr 384khz which i usually upsample too, your filter sounds more distant, possibly larger soundstage like i'm sitting several rows further back in a live performance.


Still unsure if i find this presentation better or not, but i'm interested in the details of the filter.
Is it short, long, leaning towards time domain, or frequency domain etc.
 
Last edited:

TNT

Member
Joined 2003
Paid Member
I only added new stuff for 44,1. It's just very picky on mirror/aliasing products so trading BW for suppression just on the low side of fs/2 - not the other side as many filter seem to do. And totally committed to the sampling theorem - linear basic - text book. As it should theoretically do best if the HW does it job properly. With this filter I recognise a concert hall to a greater extent than what I have had with other.

It can a time be aggressive in the top - I have wondered if this is really the filter or actually the recordings - how could they all be good? Maybe I just hear more /it all now...

I agree that the presentation is "released" compared to other filters - I play a lot of classic music so important that a real live event is faithfully created. But I play a wide veriity of music really.

I'll put up the text file in a while...

//
 
Last edited:
I only added new stuff for 44,1. It's just very picky on mirror/aliasing products so trading BW for suppression just on the low side of fs/2 - not the other side as many filter seem to do. And totally committed to the sampling theorem - linear basic - text book. As it should theoretically do best if the HW does it job properly. With this filter I recognise a concert hall to a greater extent than what I have had with other.

It can a time be aggressive in the top - I have wondered if this is really the filter or actually the recordings - how could they all be good? Maybe I just hear more /it all now...

I agree that the presentation is "released" compared to other filters - I play a lot of classic music so important that a real live event is faithfully created. But I play a wide veriity of music really.

I'll put up the text file in a while...

//


Hmm further listening with your filter i find that bass rumble suffers a little with the larger presentation, and some background details in very complex passages get's a little lost due to sounding more distant but i can still hear them if i concentrate on it.
 
Last edited:
I only added new stuff for 44,1. It's just very picky on mirror/aliasing products so trading BW for suppression just on the low side of fs/2 - not the other side as many filter seem to do. And totally committed to the sampling theorem - linear basic - text book. As it should theoretically do best if the HW does it job properly. With this filter I recognise a concert hall to a greater extent than what I have had with other.

It can a time be aggressive in the top - I have wondered if this is really the filter or actually the recordings - how could they all be good? Maybe I just hear more /it all now...

I agree that the presentation is "released" compared to other filters - I play a lot of classic music so important that a real live event is faithfully created. But I play a wide veriity of music really.

I'll put up the text file in a while...

//
Hmm do you have a version of your filter without the DSD filters removed, i kinda want to have additonal functionality to playback the few dsd64/dsd128 files i have.