Are you saying a DAC should be tuned to compensate for limitations elsewhere in the system?Bill, I’m thinking things must be fine tuned to your room @ lp..... the system as a whole.
Even here anecdote is not data!
And at least you agree with me that audio measurements are 40 years out of date. But they are still the only common ground we have. THD at a single frequency is not a useful measure once you are well under 1%.
BTW I have no idea who JohnW is and have no great desire to know. This is DIYaudio not guru designers fireside chat . But that's just me.
And at least you agree with me that audio measurements are 40 years out of date. But they are still the only common ground we have. THD at a single frequency is not a useful measure once you are well under 1%.
BTW I have no idea who JohnW is and have no great desire to know. This is DIYaudio not guru designers fireside chat . But that's just me.
DAC design as well as preamp, amp & cables.Loudspeaker designers have to make many assumptions about the average listening environment so compromises are inevitable. I don't see much similarity with DAC design?
We are talking about an area of audio that people have been arguing about forever. It looks to me like a lot of work would have to be done to settle most of the disagreements scientifically.
In that regard, Jakob once opined to the effect that if as much money was spent on audio reproduction audibility research as the food industry spends on its research, probably most of the disagreements would be settled already.
Personally, it seemed a particularly salient insight.
In that regard, Jakob once opined to the effect that if as much money was spent on audio reproduction audibility research as the food industry spends on its research, probably most of the disagreements would be settled already.
Personally, it seemed a particularly salient insight.
Would you mind sharing the details of that evaluation setup? Doing so may encourage others to visit him (& you) for listening comparisons.Jam has a very different system that is for evaluating one change in an entire system for overall better or worse, but there are limitations to it.
But then they (the usual suspects) will make it seem like their hearing and or electronic system is superior and therefore it should be viewed as a proof.if more people said 'prefer' then we would have less arguments on here,
I would say: It's not entirely impossible that we do not yet have enough measurement tools or methods, to completely dismiss such opinions on perceived sound.
Better to allow for the possibility than to completely dismiss it flat out, could well be there's some correlation between the perceived "sound" and some other measurement under certain conditions. Most measurements are made in very limited scenarios, IE sine with fixed tone, multi-tone or sweep. If it was possible to single out distortion figures and other measurements, using something like pink-noise mixed with low frequency square waves, could be that we'd get very different, much more meaningful numbers.
The measurement gear, the methods, the equipment you're measuring, it's all more or less optimalized for measuring with sine, especially 1khz. So in most cases you'll get good performance when you measure just that..
Well said.
dave
Just like Gulliver and the little endians vs big endians.
There is no need to make up stories about what might be then believe the stories as though they are reality. If your mind is already made up that there is nothing more to hear than you already do, then fine. Just say it, say your mind is made up and that's that. Period.
Beats the **** out of talking about decay tails?
A few things about missing decay tails.
1) They're "missing", i.e. if this is a proper description this is a black/white issue.
2) They're decaying so the levels should be low with little to mask them.
3) They're "tails" so you know where to look for them.
If no one can come up with a measurement akin to the one I showed they aren't trying. Last I looked DAC's output two voltages, any claims of hidden information are extraordinary. BTW the ESS "hump" plot was trivial to produce and after literally years of controversy the problem was obvious.
Last edited:
A few things about missing decay tails.
All good and interesting points. Thank you.
However, not what I am spending my time on right now. Have a list of projects to work on before coming around to those things again (if ever, if it requires ESS dac chips). It means its too late by now for me to have interest in the hump although I will remember the suggested fixed-tone time-domain residual technique.
Regarding reverb tails, I may have to come back to that one if I don't figure out what I want to know by other means first. Its still something of interest.
By the way, its no accident that I am not saying everything I experiment with and observe blog-style this time around. It can wait.
Are you saying a DAC should be tuned to compensate for limitations elsewhere in the system?
I think more dacs should have full dsp options to fit the sound to your personal space/system and taste. Used in place of a preamp one could dial it in to whatever nirvana (or nervosa!) you desired.
Like EH says, it’s all about the room.....so let’s have the ability to dial it out (or in, depending on how you look at it)
By the way, its no accident that I am not saying everything I experiment with and observe blog-style this time around. It can wait.
Trying to be all mystic really doesn't do anything .
@Bob: What you have described is a miniDSP with dirac. Something Mark has dismissed as not of suitable quality!
perhaps...
You should take a look at the measurements of a DAC using the AK 4499 before inserting foot so far in mouth.
The AK4499EQ utilizes the cheap modern high distortion delta-sigma modulation method. The abundant quantization error noise and high frequency harmonic transients generated cannot be brushed under the rug and will impact the audible range.
The electrical characteristics of CMOS devices are horrible for the task of processing audio signals.
Op-amps are ill-suited for I/V conversion and filtering. Why use crappy op-amps in nonindustrial artisanal approaches at all?
Any attempt at analysing such a strongly nonlinear abstruse system is simply ridiculous. I love the hypocrisy, spiritualism and wishful thinking in audio.
You should take a look at the measurements of a DAC using the AK 4499 before inserting foot so far in mouth.
Yes more of that Bill....I’m actually on the fence about getting one. (SHD)
And yes the factor of quality is one of my issues.
And yes the factor of quality is one of my issues.
Last edited:
Trying to be all mystic really doesn't do anything .
It was just a statement, no need make up attributions of intent when you don't know.
@Bob, have fun, man. That's why we call it a hobby. I wouldn't recommend some of what you suggest for the purpose of ending up with a good system by the shortest route, but its all good for learning what stuff sounds like (and I know you will listen carefully as you always do 🙂 ).
Last edited:
And yes the factor of quality is one of my issues.
Kevinkr started a thread on his miniDSP SHD aventures. The SHD studio is digital in and out. If you want to process at 192kHz then you need to up the ante or roll your own in linux (which once it's working is generally rock solid).
I want active speakers (rightly or wrongly) so DSP is a pragmatic solution to that problem. My personal choice to chase the easy gains rather than stuff 120dB down. Shame about that damned piano...
The good Roon based i7 stuff is all $2k+ ....I think maybe holding out for the good stuff might be beneficial in the long run. It can even do multi channel !
I've got 2 channel Roon endpoints in a couple of spots around the house (Allo and RPI based) and I use an SHD (the full version) as a black box processor in my system. Yeah, it's only 24/96, but I can't hear 40kHz and really barely 1/4th that at my age. I'm thrilled with it, a huge improvement over the pair of BSS Varicurves I used for speaker/room EQ - less noise, less distortion, substantially better resolution - a clear upgrade. Quite transparent, and preserves the character of the rest of my signal chain. I've become quite pragmatic in recent years as Bill can attest.
The good Roon based i7 stuff is all $2k+ ....I think maybe holding out for the good stuff might be beneficial in the long run. It can even do multi channel !
I'm running ROCK on an Intel Nuc i5 with an EVO 960 250GB NVM drive for the OS and Core, an EVO 860 1TB drive for the music library which is currently just short of 1000 albums. I don't need to do a lot of transcoding as the DAC I use in the primary end point can handle anything currently available without conversion, the other end point has a 32 bit 384K PCM only dac and I use Roon DSP for volume control (horrors) and transcoding dsd to pcm - funny thing is this all sounds so much better than the Sony HAP it replaced and everything ran about what I paid for the Sony including a 1 year Roon license.
- Home
- Member Areas
- The Lounge
- If it's purely an engineering challenge why bother designing yet another DAC?