I was strongly recommended the NOS filter and that is what I mostly used. I read the results depend a lot on the audio chain also so there are no guarantees. I'm tempted to try it again in my headphone setup that i didnt have at the time.
Singxer SDA-6 DAC (AK4499)
roderickvd
100+ Head-Fier
I took the plunge and ordered a SDA-6 PRO. When all is well, I will be selling my Ares II.
R2R to SDM convert? I might just be! I mean, I have been eyeing an upgrade particularly in the clock department. Gustard R26 R2R DAC with a clock synthesizer? No thanks. I think the features on this Singxer are out of this world, particularly at this price point, and should still provide me with a velvet sound and NOS possibilities. Maybe even more NOS than Denafrips when I get to deliver HQPlayer DSD to the chip output stage, rather than having Denafrips still oversample but just not filter.
I also own a SA-1 with DC jumper mod. Pity that the form factor isn’t the same as I think the stack would have looked prettier.
More importantly it seems to me that the SDA-6 has a DC coupled output stage. If so, I should remove the jumpers and make the SA-1 AC-coupled again. Can anyone confirm?
R2R to SDM convert? I might just be! I mean, I have been eyeing an upgrade particularly in the clock department. Gustard R26 R2R DAC with a clock synthesizer? No thanks. I think the features on this Singxer are out of this world, particularly at this price point, and should still provide me with a velvet sound and NOS possibilities. Maybe even more NOS than Denafrips when I get to deliver HQPlayer DSD to the chip output stage, rather than having Denafrips still oversample but just not filter.
I also own a SA-1 with DC jumper mod. Pity that the form factor isn’t the same as I think the stack would have looked prettier.
More importantly it seems to me that the SDA-6 has a DC coupled output stage. If so, I should remove the jumpers and make the SA-1 AC-coupled again. Can anyone confirm?
labeelen
100+ Head-Fier
Hi Roderick,
I use the same combo sda-6 and sa-1 (no jumper mod) in combination with modified he6se v2.
Upsample to dsd256 using HQPlayer (poly-sinc-gauss-long & poly-sinc-gauss-hires-lp filters; ASDM7ECv2) and small EQ in its matrix menu.
Never looked back.
roderickvd
100+ Head-Fier
Awesome how great it is to have forums like these. Thanks for chiming in.
I am not new to signal processing, but I am new to HQPlayer as a software. What hardware do you run it on? And with these settings how taxing is it or how much load does it give?
I have a fanless Celeron J3455 lying around that I was hoping to repurpose to this task and do USB in, USB out.
I am not new to signal processing, but I am new to HQPlayer as a software. What hardware do you run it on? And with these settings how taxing is it or how much load does it give?
I have a fanless Celeron J3455 lying around that I was hoping to repurpose to this task and do USB in, USB out.
labeelen
100+ Head-Fier
In HQPlayer there are a lot of filters some of which can be a challenge for even the most powerful systems.
First I ran it on a RPI4 but if you want to upsample to DSD you need more power.
I bought a used 8th generation Intel Nuc i5 that works fine uptil dsd256 (with the filter settings I mentioned in my previous post)
But due to the load the Nuc fans will generate enough noise to make you want to put it another location and add an additional (passive cooled) RPI4 running HQPlayers Network Audio Demon (naa) that connects to your sda-6 via usb.
I stream mainly Qobuz via Roon.
roderickvd
100+ Head-Fier
Pretty sure my fanless Celeron won't pull DSD256 then. I'll fiddle around and see what it can do at lower rates or computationally less expensive filters.
roderickvd
100+ Head-Fier
Just turned it on and it's awesome from the get-go. Can only imagine how it will be after burning in the caps some more.
Observation: the output indeed is DC-coupled. When fed into a SA-1 with the jumper mod (or any other DC-coupled amp) this causes loud pops when switching to unlocked inputs. I removed the jumpers from the SA-1 and all is well again. I did not bother to measure the DC offset on the SDA-6 outputs.
Question: does the oven remain on temperature when the SDA-6 is in standby?
Observation: the output indeed is DC-coupled. When fed into a SA-1 with the jumper mod (or any other DC-coupled amp) this causes loud pops when switching to unlocked inputs. I removed the jumpers from the SA-1 and all is well again. I did not bother to measure the DC offset on the SDA-6 outputs.
Question: does the oven remain on temperature when the SDA-6 is in standby?
roderickvd
100+ Head-Fier
MAN this thing is good. Last night I went to bed way too late... Just one... more... track....
Chain was SDA6 PRO -> SA-1 (unmodded) -> Beyerdynamic Amiron Home with hybrid pads, all balanced. Amazing synergy. Warm yet detailed and spacious. Coming from R2R I would never have thought to say: this Sigma-Delta DAC sounds really analog.
Chain was SDA6 PRO -> SA-1 (unmodded) -> Beyerdynamic Amiron Home with hybrid pads, all balanced. Amazing synergy. Warm yet detailed and spacious. Coming from R2R I would never have thought to say: this Sigma-Delta DAC sounds really analog.
Try upsampling everything to DSD and you might be even more surprised.
I upsample everything in roon to DSD512 before outputting it to my DAC. (Smsl D300).
I have the Singxer SA1 as well (no mods).
If singxer releases a DAC with the new AKM chips I will definitely getting one to replace my Cambridge cxn V2 setup
Last edited:
roderickvd
100+ Head-Fier
Can't try right now as the device I was eyeing for HQPlayer has to out for RMA.
Reading the AKM4499 datasheet you may be even better with DSD256, as this has 18 dB (!) better THD than DSD512. S/N is up too by 3 dB.
You see the same with PCM where the THD+N performance drops at higher sampling rates.
Right now I am using PCM with the low dispersion filter. Kind of like Archimago's goldilocks filter. RME also says so:
"In theory, a filter should have as little phase deviation as possible over the frequency range, have as short a settling time as possible, an acceptable decay time, and provide the maximum possible frequency range without deviation. The stopband attenuation should be high to prevent aliasing. A latency as low as possible would extend the application possibilities to more than just listening to music.
The filter with the bulky name Short Delay Low Dispersion approaches this ideal quite well. It does not have the early treble attenuation of the slow filters (see chapter 34.14), has a shorter settling time than Slow, an average decay time like Sharp, a phase maximum of only 9° at late 18 kHz (basically phase-linear in the audible range), and a latency of only 10 samples, so that it is also well suited for professional real-time monitoring. This makes it much more than just a successful compromise."
Reading the AKM4499 datasheet you may be even better with DSD256, as this has 18 dB (!) better THD than DSD512. S/N is up too by 3 dB.
You see the same with PCM where the THD+N performance drops at higher sampling rates.
Right now I am using PCM with the low dispersion filter. Kind of like Archimago's goldilocks filter. RME also says so:
"In theory, a filter should have as little phase deviation as possible over the frequency range, have as short a settling time as possible, an acceptable decay time, and provide the maximum possible frequency range without deviation. The stopband attenuation should be high to prevent aliasing. A latency as low as possible would extend the application possibilities to more than just listening to music.
The filter with the bulky name Short Delay Low Dispersion approaches this ideal quite well. It does not have the early treble attenuation of the slow filters (see chapter 34.14), has a shorter settling time than Slow, an average decay time like Sharp, a phase maximum of only 9° at late 18 kHz (basically phase-linear in the audible range), and a latency of only 10 samples, so that it is also well suited for professional real-time monitoring. This makes it much more than just a successful compromise."
Last edited:
roderickvd
100+ Head-Fier
To answer my own question, I think it does. Returning home after a weekend away with the SDA-6 in standby, the chassis remains warm to the touch.
roderickvd
100+ Head-Fier
If there is one nit I could mention: the SDA-6 has hardware volume control over USB, but the volume buttons on the remote don't do anything.
Otherwise this is one amazing DAC that seems very much underrated or at least overlooked. With all the rave for much pricier DACs this one will be hard to beat...
Otherwise this is one amazing DAC that seems very much underrated or at least overlooked. With all the rave for much pricier DACs this one will be hard to beat...
simon740
500+ Head-Fier
Hello,
what about soundstage? Depth? 3D? I had Ares II. And thinking about this SDA-6.
regards,
Simon
roderickvd
100+ Head-Fier
Coincidentally I also came from an Ares II which I mostly listened to in OS Slow.
In an uncontrolled comparison, I would say the SDA-6 has a wider soundstage with tighter bass and more micro-detail, without being analytical. ASR readers may scoff at AKM's "Velvet Sound" marketing, but I think it is very aptly chosen.
The Ares II has more bloom. I loved it but in hindsight think it was a tad too warm in my chain with an SA-1 to an Amiron Home or T1.2.
The depth is about equal but that is to be expected when listening on headphones.
Note that my SDA-6 PRO's Mundorf capacitors aren't fully broken in yet. From earlier experience with Mundorf SUPREME capacitors in a loudspeaker crossover, I do expect the soundstage to expand even further.
In an uncontrolled comparison, I would say the SDA-6 has a wider soundstage with tighter bass and more micro-detail, without being analytical. ASR readers may scoff at AKM's "Velvet Sound" marketing, but I think it is very aptly chosen.
The Ares II has more bloom. I loved it but in hindsight think it was a tad too warm in my chain with an SA-1 to an Amiron Home or T1.2.
The depth is about equal but that is to be expected when listening on headphones.
Note that my SDA-6 PRO's Mundorf capacitors aren't fully broken in yet. From earlier experience with Mundorf SUPREME capacitors in a loudspeaker crossover, I do expect the soundstage to expand even further.
roderickvd
100+ Head-Fier
Today I also set up HQPlayer on my 2017 iMac: DSD256 with crossfeed and loudness enabled. ASDM7 modulators are a tad too heavy, so I'm running with DSD7 256+fs and it's sounding great. I also chose poly-sinc-gauss-long as filter. Sounding very good so far. I don't find it earth-shattering, but I'll take it!
From the AKM4499 datasheet, it has a 3rd order SDM built-in so bypassing this entirely and doing 7th order externally is a nice upgrade. It is a multibit chip so 7th order modulators are fitting according to the HQPlayer manual.
Edit: differences are easier to hear with quiet music like acoustical and classical. More weight and spaciousness. With pop, rock or metal, I'm not sure if I would pass an ABX test.
Edit 2: now running ASDM7 with poly-sinc-short-mp, which should suit progressive rock better. What fun it is!
Last edited:
Users who are viewing this thread
