I’ve been trying to find out which filters the gain cell dac uses. But nothing came up on the forum search.
So, what filters are used, how many choices and what is the sound difference between them?
Also, how are they implemented? As in, what does the filtering and where is it in the signal path?
Ps: I have a marantz hdamp1 with 2 filters. And I love the one and not the other.
Thanks in advance.
Danny
(1) low slope linear phase, (2) high slope minimum phase, (3) high slope linear phase
Higher the slope = more ringing.
Linear has equivalent per-ring and post ring but has no phase issues.
Min phase has lower pre ring higher post ring and phase distortion.
There are digital filters in the ESS during the up-sampling stage to filter out aliasing.
I would like to add that while additional sound quality can be had from DSD upconversion in Software, the built in filters are very musical and very well done.
Okay, thanks for the responses!
I have another set of questions though. Why are there filters in the first place? Why is ringing there? What is it in the dac that makes these issues with phase and pre ringing?
Did dacs always have these problems, or are they modern dac problems?
Also, I’ve seen “impulses” in explanations, but in music what does pre ringing actually do?
Thanks again guys.
Danny
If you don’t filter you won’t suppress aliasing so you will have tons of high frequency noise above Nyquist.
Linear phase have more latency thus produce pre-ringing which is a pre-echo. This can have a bad blurring effect if the filters are too steep. As long as the slope is low order the ringing is kept minimal. I believe linear phase produces more latency due to the more intensive math calculations to maintain phase coherence.
Min phase filters sacrifice phase so they are lower latency.
To my ears, min phase are not musical and sound analytical. I do not like the phase issues, instruments tend to have improper depth to my ears.
The only way you avoid these issues is by getting an R2R DAC which have no upsampling. In those DACs however you can’t play hi res or DSD formats, they are limited and quite $ in comparison. They also have other issues.
I recently acquired a Gain Cell DAC, which I use with Audirvana to play ripped CDs, high resolution downloads and SACD ISOs. Audirvana provides iZotope sample rate conversion up to DSD 128 and PCM 384/24. When running PCM files through iZotope SRC, the signal is filtered for steepness, length, anti-aliasing, phase (min to linear) prior to upsampling. If I output to DSD 128 from Audirvana, the Gain Cell DAC cannot filter that DSD stream, but if I output to hi-res PCM, the DAC’s filters will be active. Does anyone have an opinion about which of the three preset filters in the Gain Cell DAC would have the least impact on a signal that is already filtered and upsampled in Audirvana? This scenario should apply to other music library softwares, such as JRiver.
If you output to DSD, none of the 3 selectable filters will be active in the Gain Cell, they are only for PCM. I would keep the PCM upsample at linear and use filter 1 on the Gain Cell to maintain a linear slow roll off filter.
Also, try Sox upsampling, run it 97% bandwidth, 100% linear with anti aliasing on at max. I prefer Sox now, Izotope sounds a bit muffled in comparisons. In A+ use filter C for DSD.
Thanks for the suggestions, cxp. I’m trying SoX with the settings you recommend, in combination with filter 1 on the DAC.
Welcome, Undertone 29!
Thanks, Elk!
Hi there
This is just to update my post.
I got a Gain Cell DAC and a S300! Also, got a used pair of Dali Minuets for my desktop system.
The filters are interesting to use in person, but, I’ve found that after adding in my S300 (broken in, 300 plus hours), my system sounds a little thin, maybe kinda hollow, I don’t really know how to explain it. It’s the same as the other two class D integrateds I have bought in the past in this respect, but much less so. In this way, I actually prefer my cheapest integrated (Marantz PM5003) as a power amp, as it doesn’t feel as hollow or paper thin. This is through all music and genres, and particularly with stringed instruments, like guitar, piano and harp…
Buuuuut, otherwise I love the Stellar “stack”. Sound staging, imaging, the decay of reverb, the ability to easily hear the tonal difference and timbre of instruments, and my fave is the melodies! When it comes to melodies, I can follow even the really simple ones so much better. I also don’t really know how to explain this, but this is all the GCD, as I hear this on my headphones too. MMnnnnn, yummy!
I’m thinking it’s the class D sound I am hearing, lots of negative feedback… Class Low volume class AB sounds great, (remember, this is a desktop system, ie low volume) and here in Japan most of us live in Apartments, and always think of our neighbours… So I can’t ever listen over maybe 80db if that.
Am I crazy? As I hear nothing but perfect praise of the S300… Maybe , like me, others have been apprehensive to criticize due to the overwhelming praise…
Maybe I need a Schiit Loki in my life? I dunno. My wife could tell after the first week that I wasn’t 100% happy, as I was when using the GCD -> PM5003 combo.
Any thoughts anyone?
Thanks
Love ya
Just an update on my getting more meaty sound out of the s300. What I ended up doing was using my toslink output from my computer to the gain cell DAC, and I’m using filter option 2 for 44.1 content. And using some Oyaide XLR balanced cables to my s300. Although that changed the sound some more, now to be a little less transparent and a little more opaque but more meetier.
Maybe my computer was way way too noizy for the USB? I don’t know, but things seem a little more fleshed-out and less hollow now.
I might also try the different XLR set of cables and try a different speaker cable to change the sound a bit more to what I want.
Isn’t it great that we can tailor the sound for our own tastes!?
I use Atlas Element incl. XLR balanced to my active speakers. The cable to my iMac is 5 m USB. I get a good sound stage, certainly considering the position of my speakers, being close to the wall.
In a different thread or forum I read the USB can profit from longer cable length. I don’t know as I don’t have a choice, but in my system it sounds very good. Technical explanation might be, the impedance increases with length, thus that dampens the noise on the power leads of the USB cable.
Anyway you could try a longer audiophile USB cable, my Atlas Element SC 5 m USB was EUR 129.
Interesting, sounds like something I should try.
Thanks. Will get one and update my impressions.
I am certain your XLR cables contributed to the improvements you made already.
I tried them when I first got the gain cell dac,on and off throughout but I didn’t like the results until recently. They still sound a little like they are obscuring the very very low-level micro details and the audio signal I don’t know. My Audioquest Red River RCA cable still, to me, sounds less obscured. Maybe that’s just the increased noise level that I’m hearing constantly. I don’t know but the noise from the speakers is the same regardless with no signal.
Regardless, the combination of the above [toslink to gcd filter 2, XLR to S300] has resulted in a much beefier sound. If that sentence even makes any sense. Ha.
Hi Danny, if you are experiencing an elevated noise floor, I recommend searching for SHCD on this forum. I noticed there are some experiencing elevated noise levels. While others have none of those issues.
I hate saying this, as it doesn’t help your case at all. But my SGCD is dead silent and I enjoy the richness of details and sound that does not fatigue. It is really strange that you get better sound with RCA’s than XLR’s.
I happily accepted the Atlas Element XLR offered by my dealer. They are fabricated utilizing cold weld connections preferable over soldered connections due to lower resistance, more consistent and mechanically stronger. They were EUR 295 per pair.
If I ‘d live in Japan or you in The Netherlands, it would be so nice to exchange cables and try and compare.
Maybe you have fellow Stellar users close by.
After all you’ve got the super fast and comfortable Shinkansen 
).
I loved the ride from Tokio to Kyoto, what an experience. The time planning and implementation of that train may be more stable and accurate than the clock of many DAC’s ;-).
The new Kyoto station is an experience all together.
I haven’t posted a full review of my setup yet. There’s a few issue’s I’m sorting out with PS Audio’s Customer Service (RMA in place). Nothing negative to report. It’s an issue they are aware of and it hasn’t dampened my opinion of their company.
Regarding the SGCD DAC performance :
I like it’s PCM reproduction via the S/PDIF RCA coaxial input but it’s louder( SGCD vol. set 20 on my Sony via XLR outputs to the SGCD’s XLR inputs/vol. set to 10 on the coaxial input).
I currently can’t utilize the I2S HDMI input so can’t comment there
USB input sounds thin but it’s more my PC’s fault than the SGCD. Maybe one day I’ll but the Sonore or Matrix USB to I2S HDMI converter.
I leave Filter 1 on all the time. I did RTFM and tried to other two filters but Filter 1 is my favorite. Any drastic changes I need to do are done on the Bose Series 2 Dog’s Breakfast EQ.
Bass via the Coaxial input is a tad more extended via the coax input fed from my Sony UBP-X1000ES. Mids and highs to me sound the same whether I use the Sony’s PCM DAC or the SGCD’s PCM DAC.
DSD playback for now is fed thru the SGCD’s analog input for 2 channel 5.6MHz & 11.2MHz files from the Sony UBP-X1000. For 5.1 DSD & PCM playback I use the Sony STR-DN1080 built in amps for the center and surrounds and my Stellar Stack for the front L & R. All volumes are level matched.
The rat’s nest behind my Entertainment Cabinet is wired as shown above. The volume control is bypassed coming out of the Sony AVR’s XLR’s (top). But all the audio modes still function.
I too am using Audioquest Red River throughout the audio chain including a pair of Golden Gate’s for the RCA outputs form my Sony Disc “Transport” to the RCA analog inputs of the SGCD.
