One final question. How would you subjectively compare HQP option three, to PGGB 88.2KHz upsampling, both played via your NOS DAC.
Interesting point again Ken, did not had that comparison on the radar.
so next I will listen to the 88 PGGb files and listen to the original 44.1 with HQP on 88 as well and with the Polynominal-1 filter setting
and report back
Good point Mark,
I have all non negative feedback tube amps in the chain...
6N30 - 27 - 300B SE - Speaker: Avalon Sentinel clone
I also like my LINN lp12 with Kandid and Garard301 and Ortofon Silber meister
may be a correlation good be found if others reply as well who have a NOS DAC and like it?
Still, listening to the hardware filters and "cheap" filter in some players, I always come back to no filter (may be till now, but I will see)
That could be. Another possibility is that they are all known for producing a specific kind of sound. One that seems to, for whatever reasons, flow in a subjectively natural, low-fatigue, sounding way.
Of course it is. There is a big no-no for nested feedback amplifiers so popular these days, no matter how fast slew rate they claim. Non-feedback class A is optimal for such kind of setup, a good tube amp even better. Also an internal conversion method is important. Passive give good results, much better than a poor active stage. Audio GD use internal ACSS link, it is a current loop design, similar to the Krell CAST. I/F conversion is moved to the receiving side of the link. Speaking of other R2R, Soekris says that there is NOS firmware on his DACs, but those who tried it say it doesn't sound as expected. NOS DAC must have a capable output stage.
My amplifier is a budget Denon PMA-520AE, still very good in my opinion. Interesting, I was doing echo test on speakers, not a (much more resolving) Sennheiser HD-600. I don't know. I heard more reverbation changes on speakers.
Last edited:
The soekris boards have the option for passive output. Soekris NOS filter (made by community, never endorsed by soekris ) is ZOH OS filter operating at high sample rate, similar but not quite.
Also using SS amps without feedback loop.
NOS DACs I've used tended to sound worse with OS (good quality software options), dam1121 (passive output) sounded much better with well made custom OS filters than the ZOH filter.
My current perspective, liable to change, is positive attributes of NOS sound is related to jitter and clock rate and S&H/ZOH itself is not good... more distorted, veiled, probably more effect downstream.
Comparing ZOH filter at different rates with a NOS dac gives some good insight.
Analogue reconstruction filters are of interest but no experience yet.
Last edited:
Hi Ken, just did some listening….
I used the PGGB 88 from the earlier test. Best distinction with the GOD song…
PGGB88 was more dynamic and spatial than the HQP88 with FIR
But compared to HQP88 polynomial-1 the latter had more of the typical NOS touch without giving up the smoother cleaner sound I described earlier
Switching back to the original 44 track I had this extra NOS dynamics which is so hard to explain, but you know it when you hear it. But again consistent with earlier tests, I specifically with the GOD track, had the tad more roughness in the voice….
Again, all differences are small, but recognizable. In the meantime I feel I am pulled towards the higher FS and polynomial filter as it sounds NOS and so clean at the same time. But but but… not convinced yet, I want this for every type of music and always… 100% pure NOS can keep you awake and sitting upright in your chair , kind of….
Like wine? Not always it need to be fruity and soft. Some heavy tannin can give a good drinking experience as well (not drunk writing this btw ;-)
Look forward to your 176 test tracks - so I can do the same thing
Isn't it better to make a theoretically correct DAC and if one don't like it - pursue what else in the chain that might not be correct? I mean, this strategy will lead ultimately to a better system rather than introducing a fault to probably compensate for an other. Because with a permanent fault in a system, one will always experience no or negative improvement when an other competent is actually improved. One halts the development of the system really...
But I suppose as long as one have a grip of the above situation I suppose its fun to experiment with distorsion... i.e. an extra burst of hf...
//
But I suppose as long as one have a grip of the above situation I suppose its fun to experiment with distorsion... i.e. an extra burst of hf...
//
"...fun to experiment with distortion i.e. an extra burst of hf..."
Sometimes people assume that's what its about, that people like a little distortion.
Not necessarily so IME.
IHMO some people will tolerate a bit more measured distortion in return for improvement in some other factor, which itself may not be normally be very well measured. For some people it may be stereo imaging that they want improved, maybe better depth; for others it may be an improved sense of dynamics (i.e. transient response).
Good to remember that steady state measurements don't show everything. The belief of many is that if steady state measurements are good enough, then everything else should be good too. Again, IME it ain't not necessarily so (to paraphrase Gershwin).
Sometimes people assume that's what its about, that people like a little distortion.
Not necessarily so IME.
IHMO some people will tolerate a bit more measured distortion in return for improvement in some other factor, which itself may not be normally be very well measured. For some people it may be stereo imaging that they want improved, maybe better depth; for others it may be an improved sense of dynamics (i.e. transient response).
Good to remember that steady state measurements don't show everything. The belief of many is that if steady state measurements are good enough, then everything else should be good too. Again, IME it ain't not necessarily so (to paraphrase Gershwin).
Last edited:
Hi, TNT,
I'd say that there currently are a fair number of commercial DAC which are essentially 'theoretically' perfect. From it's objective specifications, we've had "perfect sound forever" since 1983, when CD was introduced. Except, for many of us, our ears have informed us otherwise. While 'typical' OS digital playback has subjectively much improved since then, (which, makes me wonder how "perfection" was improved on
) many of us still hear some important sonic character differences between OS and NOS playback. Some of which, but not all of which, favor NOS.So, we DIY hobbyists are investigating an aspect of what else in the playback chain may not be correct or not performing at a sufficiently high level. As subjectively judged by listening. Low, and behold, we appear to identified that most OS FIR interpolation-filters do not perform at a sufficiently high level. What, subjectively, appears required is either, an FIR filter which performs much closer to an perfect SINC-function than is typical (which, of course, can never actually be perfect), or to dispense with OS interpolation-filtering altogether and go NOS.
What remains, is to test whether the highest performing 176.4KHz interpolation, of which I'm presently aware, can result in yet better sound. Our next experiment.
Last edited:
I agree with all. Improvements in some factor can be a matter of personal preferences. Some imrovements can be easily simulated in studios, by example a wide stereo imaging is no difficult to achieve at all. Some other like depth of stage require coherence of harmonics, no reverse-phase tricks can achieve it.
As for an "extra burst of hf...", it is not a random noise or a correlated distortion. A presence of Nyquist images is required to reproduce fast pulse response in a lab, with a limited bandwith there is inevitable ringing. These images increase sensitivity of our receptors (which are not analog, despite of a popular belief) and reconstruction in our brain result in more natural, better articulated sound.
The images are intermodulation products with multiples of the sample rate.
With no anti-alias filtering and zero-order hold reconstruction, you can only reproduce a square wave as a square wave when its frequency happens to be at the sample rate divided by an integer > 1, and even then the duty cycle may be wrong. It works very nicely when the duty cycle is 50 % and the integer even.
Pulses narrower than one sample period will either disappear or be extended to one sample period. No idea what any of this has to do with music reproduction, as no-one in their right mind would record without anti-alias filtering anyhow.
With no anti-alias filtering and zero-order hold reconstruction, you can only reproduce a square wave as a square wave when its frequency happens to be at the sample rate divided by an integer > 1, and even then the duty cycle may be wrong. It works very nicely when the duty cycle is 50 % and the integer even.
Pulses narrower than one sample period will either disappear or be extended to one sample period. No idea what any of this has to do with music reproduction, as no-one in their right mind would record without anti-alias filtering anyhow.
Last edited: