Just curious which HQPlayer settings you are using on DAC200 vs the ES9038Q2M. Since the two at least have different optimal settings.
For DSD, the newer ES9039Q2M is better than ES9038 though and there DSD512 is the only recommended output rate option. With this newer DAC chip model it is particularly important to choose optimal output rate settings as the objective performance largely depends on the operational parameters.
DAC 200: x512 ASDM7EC-super/fast 512+fs OR x1024 AHM7EC8B (to me sounds a little more resolving and open)
IanCanada Dual Mono ES9038Q2M: x512 ASDM7EC-super 512+fs
I try different filters but for comparisions I return to the default poly-sinc-gauss-long and poly-sinc-gauss-hires-lp.
Technically best to stick with ASDM7EC-fast at DSD256. If you feel that it is ātoo smoothā and you like more ESS-style sound, you could try switching to ASDM7EC-ul which (to me) has similar character as ESS modulator.
I use DAC200 also and owned Holo Spring 3 before that. I have similar experience with these two, Spring 3 being smoother sounding while DAC200 is more alive and dynamic. I havenāt heard or owned any Ian Canada gear. I recently compared Mola Mola Tambaqui with the DAC200 and MMT is probably the most transparent and resolving sound DAC Iāve ever heard. Itās very addictive first but can be a little bit too much in the long run. Especially if the rest of the setup is highly resolving. MMT was quite agnostic about HQPlayer. It didnāt make a clear difference if I fed it bitperfect signal or HQP upsampled signal, so I mostly used its own LAN connection for the sake of convenience. I had Mola Mola Makua preamp in demo also and compared different combinations I had at hand back then. In the end, I kept DAC200 but changed my preamp from Pass Labs XP-12 to Makua.
So my point was that DAC200 is not the absolute best in resolution. Yet itās highly resol ing but always stays easy to listen to. Itās versatility with HQP gives you tons of options to finetune the sound. If you want most resolution and clarity out of it, Iād try DSD256, ASDM7EC-fast and poly-sinc-ext2 filters. I like the halfband-filters these days but the refreshed ext2 filter family is super resolving IMO. More so than the default poly-sinc-gauss filters. DSD256 is vital, 512 is clearly smoother and more distant sounding. DSD128 is simply worse. DSD1024 Iāve never really gotten into, even though Iāve tried it multiple times with the AHM7EC8B modulator.
Thanks, I will try ASDM7EC-fast at DSD256. However I do not feel that DAC 200 sounds too smooth. Actually I find Ian Canada DAC to also sound smooth and cultural. It does not have a typical ESS style one does get with SMSL/Topping devices.
Same old discussion about pcm vs ess vs something =)
It has been proven for like ages that the final result of the entire device depends not only on the chip, or even not so much on chip, but on the design around it. There are lots of examples of good and bad DACs with same ICs, or for example RME ADI-2ā¦ they claim that the sound will be the same, regardless of AKM or ESS used, with one exceptionā¦ when embedded proprietary filters are being used. Personally, i think that a clean and stable power source has biggest impact to the sound quality, and IanCanada is obsessed with that one, supercaps everywhere. His devices are basically powered by Lithium batteries and supercaps as buffer/filter, the AC is used only to recharge those.
I always thought that hifi power cables are a myth, but when i changed those stock PC like power cables with 15$ hifi cables from aliexpress i was shocked how much the sound improvedā¦ and i started to understand why there are big buck XLR, BNC, COAX cables priced at thousands EUR.
I had to try over 20 different DACs before i understood that i was digging into wrong directionā¦ Even a $100 cheap Fiio R2R, sounds decent with an ultra low noise liner power supply.
Any updates regarding the device ? Ordered one, should arrive in few days. Iām curios to do some tests with it, together with SU-2 interface and a rubidium reference clock generator.
Also bought that Cosmos ADCiso with Scaler thing, I know itās not a proper measurement device but I think itās good enough to see the difference between different setup changes.
Thanks @patouskii, I think it was you who replied in a similar tone on another forum 
Over there, people werenāt really open to exploring possible answers; most just assumed something must be wrong with the rest of my chain. So, Iām trying again here, hoping for a more constructive exchange.
Regarding the comparison between the Spring 3 and the DAC 200, one could say that the DAC 200 sounds more resolving simply because it doesnāt roll off the highs as much as the Spring 3 does, along with its stronger dynamics and livelier presentation, which can make it appear more resolving.
But if we define āresolvingā as the ability to reveal information across the entire frequency range, from bass through mids up to treble, I honestly wouldnāt call the DAC 200 highly resolving. My system is transparent and neutral. The DAC 200 is fed by a Raspberry Pi running the Signalyst NAA image, connected through an Intona isolator and Intona USB cables, so it really has a good setup to show what itās capable of.
Still, what I hear feels as if thereās a kind of veil over the sound. Maybe thatās not the perfect word, but itās how it feels to me - like looking at something beautiful through a thin layer of fabric: you can recognize the shape and beauty, but only when the veil is removed do all the fine details appear clearly.
Thatās why it feels odd to listen to music I know very well and realize how much of it seems hidden by the DAC 200. Thatās where my question comes from. Since the DAC 200 is often regarded as one of the most resolving native DSD DACs, I wonder whether this is an inherent characteristic of 1-bit DSD designs, that they offer a certain analogue-like smoothness, but not ultimate resolution?
Or perhaps there are native DSD DACs that are built or voiced differently, without that limitation?
Regarding what @Leonid_Bachevich said, yes, I fully agree that the quality of power supplies makes a big difference. My IanCanada DAC, powered from good linear supplies, sounds OK, but itās only when each rail runs entirely off its own supercapacitors that it becomes truly impressive. With DSD in True Sync mode and DPLL disabled (not officially supported by ESS Sabre), an output stage with no op-amps, and DSD upsampling from HQPlayer, the result is extremely resolving yet smooth, natural, and with a heart-warming timbre. With PCM, it loses that charm and starts to resemble a typical high-quality ESS DAC.
Building such a DIY setup takes time, effort, and skill, and costs almost as much as a DAC 200. But itās allowed me to hear whatās possible when resolution, tonal balance, and timbre all come together without harshness, in a way that simply invites you to keep listening.
Thatās why Iām genuinely puzzled. Based on what Iāve read, native DSD DACs are supposed to deliver that natural, analogue-like sound. With capable engineering and the DAC 200ās price level, I would expect it to easily outperform a DIY build of similar cost. As Jussi said, everything matters, especially short paths, handling EMI, grounding, and how clocks are implemented, among other things. The DIY setup I have is an amateur construction. Is it just the excellent power supplies and/or clocks (better than the ones that are used in commercial DACs) that make such a significant difference?
So, Iād really like to hear what others think. Are my observations consistent with those of anyone else? Is this simply how 1-bit DACs behave, or have some of you found designs that combine that analogue character with actual high resolution?
Marcin
Just out of interest, do you use DAC Corrections with HQP?
Do you use DAC correction with these? Because thereās no roll-off or anything like that when you use DAC correction. Multiple aspects of the output signal get corrected.
Also my recommendation is to keep the 60 kHz output filter enabled in DAC 200.
At least to me, the āultimate resolutionā referred by some for example in ESS is distorted harshness in the highs, causing listening fatigue. So the highs become āshimmeringā and sound brighter, but also less clean. Like sun shining through scratchy windscreen of a car. So the highs get diffused over time and spectrum. Extra brightness highs, or louder output donāt equal to more detail, although it can be mistaken for such.
Feeding ESS with DSD reduces this effect (given correct rate for the particular model and clock design), compared to PCM inputs. But to my ears it doesnāt disappear completely. It only disappears with a bit-perfect DAC.
Generally, for a DAC the most important and critical pieces are:
Apart from VREF, the typical PSRR is pretty high, so the power supply is less critical.
Interesting thoughts, thanks for sharing.
Spring 3, being a NOS only DAC, truly has rolled off highs but only with lower quality source material like 16/44,1. When you upsample to high DSD rates with HQP, itās not the rolled off highs causing the smoothness but something else in the sound signature. It also lacked punch compared to DAC200.
DAC200 is highly resolving DAC for sure but as I wrote above, there are even better ones when it comes to resolution and pure detail (like Tambaqui). But to me it was too much (in my setup). I could now own the Tambaqui but decided to keep the DAC200. Btw. I forgot to mention the DAC correction earlier, but that one for sure makes DAC200 more resolving and articulate sounding. Iām not using it currently though, since thereās a limit of how much resolution and detail you want before it makes the sound fatiguing in the long run. Itās always a balancing act between extreme resolution and listening pleasure. You can have both with very good gear though, like DAC200, but ultimately itās always a compromise between these two qualities of sound. You can possibly beat the DAC200 resolution even with some cheap Topping DAC but is that DAC a pleasure to listen to in a long run? Possibly not.
I donāt see why IanCanada DAC couldnāt be on par with DAC200, especially if it costs nearly as much. DAC200 has tons of features which also cost money, like a proper analog preamp, plenty of connections and a high quality casing and remote control. IanCanada gear is mostly about the sound quality with close to zero features, not even a case. You should get better sound quality from IanCanada with less money.
As far as I can see, IanCanada DACs have practically no analog reconstruction filter at all. So they will spit out quite a lot of ultrasonic noise with DSD inputs and massive amount of ultrasonic distortion with PCM inputs. Because the analog signal reconstruction is missing the other half of the job.
I would at least avoid connecting such to any class-D amplifier.
Iāve just ordered a Teac UD507 (with their discrete 1 bit dac) . Would HQplayer be a worthwhile investment with that dac?
Oh lord, and I do use it exactly with 6 class-D amplifiers going into active speakers.
Maybe the reconstruction filter is located on the IanCanada output board (like his OPA861 I/V stage). If what I am facing is a lot of ultrasonic noise, shouldnāt the sound be less resolving and propably harsh and tiring?
Yes,I did use DAC correction for Spring 3 and and use now for DAC 200. In both cases, it is a worthy improvement.
And yes, I have 60 kHz filter enabled in DAC 200.
What Iām describing doesnāt seem to be about output level. The ES9038Q2Mās I/V conversion is done through 12 Ohm resistors, followed by step-up transformers.
Iām familiar with how harsh or bright some ESS-based DACs can sound, but my DIY setup is nothing like that. If thereās any distortion, I donāt have proper measuring tools to verify it, but to my ears, the DIY DAC sounds cleaner than the DAC 200, while remaining smooth, delicate, and even a bit euphonic. However, I would like to stress that my impressions are not limited to heights but resolution across the full frequency range, from the definition in the lower registers. Not necessarily impact or extension, but the timbre and texture of sounds in bass notes. The same applies to the mids, where thereās a more profound sense of presence and nuances of sounds and instruments that make the recording feel more alive and tangible. If all that is caused by ultrasonic noise, that is what puzzles me.
Thank you all for your comments. I really appreciate all input. I trust Jussi when he says that 1-bit DSD are better when it comes to delivering less distorted sound. So, the response to my first question is: No, there is nothing in the 1-bit DSD DACs architecture that limits their ability to resolve.
Therefore, it is either that the voicing of the DAC 200 does not align with my preferences, or my hearing is starting to degrade due to age, and only excessively bright, distorted ESS Sabre sound appears sufficiently resolving to me.
I would really like to try some other native 1-bit DSD DACs that are maybe voiced a bit differently than DAC 200. Any suggestions?
Many seem to think that the endgame DSD dac, or endgame DAC in general, is Playback Designs MPD-8. Of course itās in much higher price class but would definitely be an interesting DAC to audition. Unfortunately the brand is not available in Finland.
It looks like just I/V stage, maybe with first order filter. The pictures are not so detailed and I didnāt see any specifications for the reconstruction filter listed on the site.
But for example comparison, my DSC1 design has 4th order filter. Four opamps on the path, one for I/V, two for the active analog filter and one for the output / cable driver.
Class-D amps will alias input frequencies exceeding half their switching frequency, similar to ADC.
With DSD from HQPlayer that is just random noise, which would just lift the amplifierās noise floor.
With PCM that would be instead more like intermodulation distortion, but since the frequencies are related in a different way, it sounds different too. It makes the sound unclear, harsh and fatiguing.
Audio transformers to some extent act as a band-pass filter. Since they cannot pass the lowest nor the highest frequencies. In addition the distortion is increased, in particular at bass frequencies, but also to some extent high frequencies. So it will certainly reduce level of the ultrasonic tones, depending on the particular transformer.
So audio transformer combined with a DAC gives you a bit tubey-sound.
That is probably due to the transformer taming some of it, but it will certainly add itās own flavor too.
Iāve had a Playback Designs Merlot for many years. But these are not really bit-perfect DACs.
Maybe worth trying a T+A 3100-series DAC? It has two different operation modes, one is good for lower DSD rates and another is good for higher DSD rates. Certainly not cheap though.
Hello @jussi_laako I have a question. What is your opinion on 6-bit DACs? I have seen the new Garlubidor Divinity https://www.garlubidor.com/blank-3-1 and I wonder if this has advantages or disadvantages with HQPlayer compared to my iFi iDSD Pro and other devices such as Holo Cyan, etc. I guess you havenāt heard that DAC, but Iād like to know what you think. Thank you.
That sounds a lot like Denafrips DAC, including same wording. So maybe it is one of such, available under different brand name. If that is the case, it is not really a bit-perfect DAC.
