Filter brewing for the Soekris R2R

Member
Joined 2000
Paid Member
I'm probably going to get flamed for this and probably deserve it, but here goes anyway.

Has anyone created a filter that resembles the Schiit Audio Yggsidril's (Yiggy) closed filter? Similar at least by sound comparison? Obviously it's closed, so who knows what Schiit is doing. I have heard it several times and it is impressive. I know at least 4 guys that own it and they go on about how there's nothing like it.

I think that's a lofty statement considering how many bright people are out there, including here, that could engineer something like it. Maybe it's not the final statement in DAC filter technology or maybe it is. I don't know.

Has anyone compared original or diy Soekris filters to the Yiggy? I tried to compare my Soekris rev3 with soft filter, but when I got to the guy's house, he wouldn't compare the Yiggy to the Soekris. Only compared to his Gustard X20, which I felt wasn't better than the Soekris, just different. He said there wouldn't be any chance the Soekris was better or the same ballpark. I felt challenged, so I slowly started to improve on what I could on the Soekris 1021.

Today my Soekris is a different animal with better transformers, upgraded I2S to USB firmware, independent and discrete power supplies, etc. But there's still the question of the filters.

Ultimately I want my Soekris DAC to be better than the Yiggy, not sound like a Yiggy. I think it can or might already in someone else's setup.
Let me know what you think.

I like the idea of a $500 - $600 DAC beating out a $2300 DAC. Maybe I'm being a jerk. :rolleyes:
 
That question has been asked before.

I would think it was possible if someone knew the parameters of the filter you want to duplicate. I suspect SCHITT isn't about to let it be known.

I have read as many people who think the SCHITT is ho hum as much as I have heard it is the GREATEST THING ON EARTH. Typical with any component, especially one you own as we all know (and all of us are guilty of falling into the trap).

Also, one o'clock would be the man at this point to really know.

AND from him, I hope I am not getting this wrong, but inverting phase is not a big deal, though time consuming I suspect, since he made me two phase inverting filters so I could switch phase on my SOEKRIS with just selecting which filter to use.

There are four choices and I have each polarity with two filters, instead of four with all of the same polarity. With mine you cannot change polarity on the fly but you can change filters so this was an elegant solution that was suggested by someone here, maybe, it, too, came from the fertile mind of one o'clock. I cannot remember.
 
Member
Joined 2000
Paid Member
I'm not looking to unravel what Schitt has done. I'm interested in that level of palpability. There's a sensation\tension I feel when listening to it like it's presenting the whole room in the recording. I hear the room when a note reverbs with my DAC, but with the Yiggy the room is already there and the reverb fills into it. I don't know how to explain it well.

Very interesting. I have Paul's pack. I lost DSD with it, but would like to try it again. I want to try oneoclock's pack. The TXT had DSD settings.

Any other input is welcome. I understand this will take some playing on my part, but if someone has a similar experience, please share it and I will try it out.

Thanks,

Vince
 
If you are hearing lots of extra reverberation I bet what you are hearing, having never heard the SCHITT but having plenty of admiration for its designers, is likely a quieter power supply.

I know no one wants to hear it but I suspect the Vrefs are noisy and are masking the qualities you hear from the SCHITT. When I went with the nige2000 approach of bypassing the Vrefs and using those damn batteries I keep going on about there was a great reduction in the kind of noise you don't necessarily HEAR as something you can identify until it is gone.

The Vref is our output signal - it is modulated by the resistor stacks.

I am sure that what is there is good and for the cost of the board I doubt one could do better. But I do think we can do better with the battery.

Even at a lower voltage I do not lose output level - I do not run it wide open so maybe if one did they might hear a slight reduction in level.

Using a good regulator BEFORE the battery keeps the batteries at a steady voltage, in my case there was no difference as measured by my above average but not SOTA meter.

I like the filter package one o'clock put together. I alternate between the NOS and the filter one o'clock says is his favorite. Since we ALL get tired of the way our systems sound eventually it is a neat benefit being able to make the change for free!

Take care,
 
I'm probably going to get flamed for this and probably deserve it, but here goes anyway.

Has anyone created a filter that resembles the Schiit Audio Yggsidril's (Yiggy) closed filter? Similar at least by sound comparison? Obviously it's closed, so who knows what Schiit is doing. I have heard it several times and it is impressive. I know at least 4 guys that own it and they go on about how there's nothing like it.

I think that's a lofty statement considering how many bright people are out there, including here, that could engineer something like it. Maybe it's not the final statement in DAC filter technology or maybe it is. I don't know.

Has anyone compared original or diy Soekris filters to the Yiggy? I tried to compare my Soekris rev3 with soft filter, but when I got to the guy's house, he wouldn't compare the Yiggy to the Soekris. Only compared to his Gustard X20, which I felt wasn't better than the Soekris, just different. He said there wouldn't be any chance the Soekris was better or the same ballpark. I felt challenged, so I slowly started to improve on what I could on the Soekris 1021.

Today my Soekris is a different animal with better transformers, upgraded I2S to USB firmware, independent and discrete power supplies, etc. But there's still the question of the filters.

Ultimately I want my Soekris DAC to be better than the Yiggy, not sound like a Yiggy. I think it can or might already in someone else's setup.
Let me know what you think.

I like the idea of a $500 - $600 DAC beating out a $2300 DAC. Maybe I'm being a jerk. :rolleyes:

I agree with you on the sounding different part. I have the Soekris V3 and when I first compare it with my less than $200 Chinese es9018 dac with some tweak and Crystek clock upgrade, I can only say both dac is different rather than one is better than the other.

BTW, if you use the HQplayer software, you can select the closed formed filter which was claim to be similar to the one use by Schitt dac. I tried it and can't really say it's better than the other ones. In fact, I prefer the latest extreme filter over the closed form.
However, I am not sure for Soekris how effective those filters are as they are done in the PC and there is additional processing inside Soekris. And my Soekris doesn't like upsampled files and feeding it 192k/384k will sound bit digital (I am still on firmware 1, I believe) as compare to standard 44k
 
I would think it was possible if someone knew the parameters of the filter you want to duplicate. I suspect SCHITT isn't about to let it be known.

It can be reverse engineered to a fairly close approximation. That's because the impulse response of an FIR is in effect a read-out of all the coefficients in analog form. Stereophile has published the impulse response but the resolution of the graph probably isn't sufficient to get the coefficients accurately. But were someone to make a high resolution recording of the IR then the values could be gotten out, give or take a few LSBs.
 
I have read that the Schitt dac takes a long time to warm up b4 sounding good, a bit like tube. So how was the testing condition done?

My friends went to the Schitt DAC's house, so it's most probable that it was warmed-up. But I can't be 100% sure.

However, my friends are very competent regarding such matters. I'm fairly certain that the conditions of the test were fair.

Then again, I wasn't there, so..
 
Member
Joined 2000
Paid Member
I have read that the Schitt dac takes a long time to warm up b4 sounding good, a bit like tube. So how was the testing condition done?

When I was allowed to hear the Yiggy, (at different houses) it was on 24/7. Tthe impression was the same with two completely different systems. One system is a dedicated designed audio room with monster Evolution speakers and multiple isobaric subs in the ceiling. Something like 5 pairs. Spectral pre and amps. Synology NAS and Roon software.
The second house is a dedicated room, but not engineered for sound specifically. Used iMac and Audivana software with Vandersteens and McCormack amp and preamp that are probably 25 years old or older. Again, the impression was the same. This guy had the Gungnir, then upgraded it to the multi-bit, then got the Yiggy. I heard all three DACs and the Yiggy is a big leap even over the Gungnir multibit.

It could be improved power supplies, but I think its the filter design. Just my opinion.

I got to hear the Aqua Formula recently and I still think the Yiggy is better.
Heard one of the Lampizator DACs. (I changed the output caps for the guy.) That is a pretty great DAC too at 3x the price.

But again, I don't want a Yiggy. I want to make the Soekris better. I prefer the openness of the platform and the DIY aspect.

Thanks for your posts.

Vince
 
Last edited:
When I was allowed to hear the Yiggy, (at different houses) it was on 24/7. Tthe impression was the same with two completely different systems. One system is a dedicated designed audio room with monster Evolution speakers and multiple isobaric subs in the ceiling. Something like 5 pairs. Spectral pre and amps. Synology NAS and Roon software.
The second house is a dedicated room, but not engineered for sound specifically. Used iMac and Audivana software with Vandersteens and McCormack amp and preamp that are probably 25 years old or older. Again, the impression was the same. This guy had the Gungnir, then upgraded it to the multi-bit, then got the Yiggy. I heard all three DACs and the Yiggy is a big leap even over the Gungnir multibit.

It could be improved power supplies, but I think its the filter design. Just my opinion.

I got to hear the Aqua Formula recently and I still think the Yiggy is better.
Heard one of the Lampizator DACs. (I changed the output caps for the guy.) That is a pretty great DAC too at 3x the price.

But again, I don't want a Yiggy. I want to make the Soekris better. I prefer the openness of the platform and the DIY aspect.

Thanks for your posts.

Vince

It really sucks that Paul has given up on filter development, he and one-o-clock made a lot of progress in this very quickly but sadly the unnamed people who live just to crap on others hard work drove him away.
 
I heard all three DACs and the Yiggy is a big leap even over the Gungnir multibit.

It could be improved power supplies, but I think its the filter design. Just my opinion.
Vince

The Schiit website is actually reasonably helpful on the difference between Gumby and Yiggy. They claim the filter and the analog output stages are the same. The difference is the DAC chip. The Yiggy uses AD5791 while the Gumby uses AD 5781. The claimed result is 21 bit versus 19 bit resolution. Also, the Yiggy uses a better power supply with 12 locally regulated supplies and better regulators compared to Gumby with 8 local supplies.

In other words, the same approach to improving the DAM makes up some of the differences between the Yiggy and Gumby.