How much difference does it make?

[PeterSt]

As usual my controversial answer:

 

Room maybe 10% (in the end irrelevant if it is not too bad to begin with - avoid swimming pools and the like).

Power amp 5-10% as long as it measures well (rare to begin with).

Pre-amp, 100% detrimental. So ditch it. Same counts for any aftermath volume controls.

DAC ~25%.

PC / hardware source 30% or more.

Software 25%, but in combination with PC / hardware source and even DAC. This should not count to get to 100%.

Digital cables 15%.

Analogue cables 10%.

Speakers left out. *)

 

Add up: 10 + 5 + 25 + 30 + 15 + 10 = 95% (see power amp 5-10%, counted as 5%).

 

Do notice that the software works in conjunction with the DAC. Thus, what is omitted in the DAC (filtering) now happens in software. This is explicit.

 

Also notice that the software also works in conjunction with the PC and that the operating system in the end is an integral part of the software. At least this is how it has been set up explicitly. Change the OS means change a kazillion of software settings.

 

Only a year ago or so, the spread of 25% and 30% from DAC and PC respectively, would have been 50% and 5%. But this changed, getting more and more control over the PC and what happens in there for real. And mind you, very large threads are running about this (not our audio PC but in general) so it can't be ignored.

 

*) I leave out the speakers because it is not fair to incorporate them. Also they are the most hard to judge. This already counts for myself, because I really don't drag in new speakers for comparison. They will matter - they will be the most crucial, but still they become irrelevant looking at the other elements. I am serious here. Two quotes from customers which subscribe to a general thinking of mine:

- Peter, I like to cancel the order for the DAC. Your new software version sounds so damn good that I don't need a new DAC any more.

- Peter, my wife and I were 100% sure to buy new speakers and we had the (crazy) budget for it. It now appears that we don't need new speakers at all. You DAC does everything.

 

Mind you, these messages can be found in the Phasure forum of various kinds and from various angles. They are also not uncommon and you surely won't go look and find them. Point is that each of the elements wildly influence, and how they do relatively I laid out above. Please keep in mind my message about that list being subject to change (say each other year).

 

In my own forum you will also be able to find how people nag that we should produce USB cables while I refuse because they won't make a difference and the $1 cable provided with the DAC is OK to begin with. I did-not-believe in any of this impossible sh*t. Until people nagged so long that I sat back and started with it after all. All is spelled out historically on the forum. 1000s have been sold, one was returned (plus one who announced it in advance). It can't be ignored.

 

I'll stop here. But please be informed that we produce 100% of the whole audio chain (apart from power cords), so I should be able to know. It could be not so common to see a head to tail SNR of 135dB (thus from PC-software running on a PC to output of power amp). Some will say that this is totally unnecessary. Perhaps that is correct. But maybe it allows to hear through matters. Perhaps this is also why more and more we end up in "tweak" areas which seemed impossible to influence in yesteryear.

Maybe we just don't want to know.

 

Peter

 

 

[PeterSt]

50 minutes ago, STC said:

Room acoustics is everything. It can make lousy system to sound great. Just like when you sing a bathroom where the acoustic changes how you sound and more preferable. 

 

This  is very funny. I agree with that.

 

50 minutes ago, STC said:

Just like when you sing a bathroom where the acoustic changes how you sound and more preferable. 

 

Also funny. I agree again.

 

51 minutes ago, STC said:

The acoustics in the bathroom are basically perfect to start a little private concert.

 

Apart from that I just said to avoid swimming pools (and I did not read your post !).

 

So try to believe in the other way around now (I suppose nothing new for you because I said it before):

 

The better the system, the less room treatment is necessary. It even is so that it removes any audible standing waves (low frequency as we know them, and high frequency as most do not know it - buzzing).

My room consists of over 50% glass in the walls. The remainder is hard material. Only the floor carpet (few mm of height) is soft.

 

So the "funny" thing is that you just said the same thing. But most 99% probably you won't agree with my stances ...

haha

[PeterSt]

4 minutes ago, Rexp said:

Yes the source is key, why is this so hard for some to understand

 

Because for a close to life time we were spoon fed with the speakers being the most important factor ?

Maybe back in the days they were (say that I am from back in the days 😀) but back in those days the source just could-not matter as much as today. I am not talking about the wax cylinder era (that would make me quite dead) but from the LP-only era, yes. But look how easily we drift apart ... half of people swears that LP is the by far very best compared to digital. Well, they should dedicate the speaker the most important in the chain. I think.

If that makes sense ...

[PeterSt]

6 minutes ago, Rexp said:

Actually back in the day when Linn were at the forefront of high-end

 

This could be no coincidence for a TT manufacturer, right ? haha

But still it would be true, yes. However, the difference to be made there is "bland". So whether you have the groove etc. noise at -30dB or -35dB (just making up some figures which will be relatively close) - that matters 5dB. Today, with digital, we talk well over -100dB. Well, maybe. Maybe it is -80dB only (look at poweramp specs). But there's the thing - this is a difference of 20dB already. This, while a few days ago I showed better than 130dB (SNR). Oops. And this has to be in the source as well because that was about the complete chain.

 

So for SNR only ... -35dB (at best ?) vs -130dB.

or what the subject was about to some extent:

-25dB to -35dB vs -80dB to -130dB (think source only). Thus 5dB of diifference vs 50dB of difference in the source(s).

 

Btw, a speaker doesn't even make noise. It can show a lot of THD though. A lot (like well over 1% in the lower regions or 5-10% at best when you thought to add a sub woofer).

Wait ... so a speaker can matter just the same ?

or wait more ... the source thus does not matter because the speaker is so poor at it (THD) ?

 

Maybe this is why I counted out the speaker. It is unfair. Sure, mine does better than 0.3% THD at 19Hz +/- 0.5dB or even 0.3% at 17Hz +/- 3dB (89dBSPL). Brag brag brag brag. But this is far from normal and should be counted out. Of exceptional importance, but still not to be counted. Not realistic. Just as the 118dB sensitivity is not realistic (though still real over here).

 

So what sort of remains is that SNR of 30dB of back in the days (I forget the cassette tape decks to stay positive) which won't have improved once it reaches the speaker. That speaker, in my view, in that context has prevalence for many quality aspects. I mean, the difference between 10% THD and 0.3% THD would still be there, also in the old days. A tweeter would do 16KHz or 30KHz (today only full bandwidth drivers may have a problem there because they are not-so-full bandwidth at all). Today ? no, that too existed back then.

I am drifting off; When the source is lousy, the end of the chain receives only lousy and probably even more lousy. Never better. Today's sources vary wildly and are of way better specs than the speakers anyway.

 

Now choose. Technically I am not really able to (as in : how should a THD+N of 0.00090 sound better than a THD+N of 0.04 (standard (NOS) unfiltered Redbook), while the speaker is at 0.3% at the very best for LF (and well over 1% for almost everybody) ?

It's just a different "THD" I'd say (digitally/technically wrong, vs mechanically impeded <-- more natural ?).

[PeterSt]

3 hours ago, STC said:

Linn discovered that by making better TT, it improved the sound. He discovered it by placing his TT in another room which improved the sound. His reasoning was and correctly too for mechanical device like a TT that the sound from the loudspeakers causes degradation in the playback. 

 

The same would be so for the DAC (these days / still). It is about as mechanical, but the relative forces and what is influenced are different (think infinitely more low jitter opposed to the TT's wow).

[PeterSt]

53 minutes ago, SJK said:

I guess I should hang out here more often, I’ve never heard of mechanical vibration affecting solid state chips.  Pressure, sure - vibration?  That’s a new one.

 

But of course it is not new.

Blame the clock (oscillator) in there as the first to be hammered upon. Next will be chips if not overwhelmed by the former in the first place.

 

So footers for your DAC don't matter ? May I ask, what is your DAC ?

[PeterSt]

Here's vibration management:

Runs even on batteries.

It has been well experimented with either.

[PeterSt]

3 minutes ago, jabbr said:

a whole new market  for the Lush!

 

Shortest is 40cm. Nobody wants it shorter. I have head no requests for pink yet.

 

But we better quit this subject quickly.

[PeterSt]

This *is* a good moment to announce that usually I hear best through the music when the hoover in the kitchen is on (me being in the kitchen just the same, the kitchen being open, adjacent to the listening room.

 

Of course this is (very) noisy, but at stages necessary and I won't shut up the music because of that.

It seems to work like adding noise to a photo (sharpening).

[PeterSt]

OK, you are not going to believe this ...

 

9 hours ago, STC said:

 

Put this on your DAC, speakers, amplifier.

 

How could I be so "smart" with that toy eh ? well, because I for 100% real, once, for Santa, gave such a thing. This was witnessed by our 10 year old son (and 10 years ago by now). I had an outrageously good story accompanying: "dear, this is so you can test the influence of vibrations of the DACs you build; just hold it against it at various places and check whether you can see it in the output". Btw, it was as obfuscated as the one I showed. "It does 10K Vpm" I added.

"Yeah, sure !" our son said. "Tell that Santa !".

 

f*ck

 

So ST, you don't know what you just woke up in me.

[PeterSt]

Anyway, thou shall check ur vibration management on a most regular basis.

 

[PeterSt]

4 minutes ago, rando said:

a 16" woofer

 

I always wanted to have those. Will you be the first making them ?

[PeterSt]

14 minutes ago, STC said:

Like looking for the micro SD card which just disappeared right in front of my eyes when I was opening the package. It just happened.

 

Start to practice soldering 0201 SMD.

I'd already need sharper vision to see the things in the fist place. 

[PeterSt]

10 minutes ago, shtf said:

Yeah, I’d say 60% is a pretty safe conservative estimate.

 

Btw, I wouldn't throw in credentials (as you just did). Seems to weaken.

Thanks for a good post.

[PeterSt]

I just saw that Eberhard Sengpiel passed away 3 months after I discussed with him whether math could exist to calculate (fluid) material properties for damping all sorts of influences at the same time. This sprung from this little project:

 

 

where floating on normal water would do a lot (against floor vibrations but also from music impact itself on to the case) - but that still not doing anything for the inside electronics "movement" itself (with the easy example of capacitors).

 

In my view there was nothing this man could not calculate, as long as it was audio related. I think many will know about his website, which kind of ceased to exist but is still there : http://www.sengpielaudio.com/Calculations03.htm

 

Anyway, he admitted to me that he could do a lot with math, but what I wanted there was beyond his capabilities. This gave me the idea that it really could not be done.

It is only today that I learn that he probably was very Ill already when I asked.

 

PS: The water level under the DAC there needed to be maintained at a quite "same level" or else the sound degraded - or at least changed already. Water level drop was 5cm per day or so. I used this for maybe 6 months before I was fed up with watering my DAC.

 

[PeterSt]

What you are working on is great but so special and complicated
that I must say that I don't have the knowledge to help you
in doing calculations for you.

Don't forget that I'm old as a dinosaur.

 

R.I.P.

 

[PeterSt]

Maybe there's enough talk about preamps already. So maybe a next time. 😊

 

24 minutes ago, JoshM said:

 

Having gone through a bunch of power amps, all of which measured well

 

I'd say that that "problem" must be solved first. And that problem is ... are you sure ?

I don't measure power amps very regularly, but of those I did measure, only one was up to my own ideas about it (say same for noise floor as I showed from my own although that was the complete chain in one go). All of the others I saw passing by (Internet) don't show good at all. One thing: with NCore's as the example, don't forget to look beyond 20KHz as well. Thus, showing a noise line nicely way down (and flat) up to 20KHz is one thing, but not showing it beyond there - because it will creep up sky high - is something I don't like at all. And my tweeters told me they don't like that either. So no need to show the Hypex (based) stuff without showing a fair band beyond 20KHz.

 

I suppose (I never tried) it will be relatively easy to find measurements which fulfil the "art" of a decent straight (very) low level noise line (search Stereophile or something) but now look whether those are the power amps you went through. We could compare a few and decide what actually means "measure well". And just saying: I don't rule here 😆 but while DIYAudio was mentioned a couple of times the past few days as the example of where the good minds hang out (I am actually saying that myself), I never really see good measurements passing by in there (of any "device") while everybody raves about what's presented.

 

Possibly we can say in advance that such comparisons or show cases are relatively useless, because when "we" like, we claim that we won't be able to span more than 70dB of dynamic range anyway, and thus why would noise be any good or for the better when it is 71dB down. Only 69dB would be a kind of detrimental. And thus is -71dB a testimony of good measurement ?

Thus the point is that we seem to decide for ourselves what exactly "good" is. That the one tells that being on the moon is OK while the other prefers to be in an other universe, makes it a bit difficult.

 

In my view, over-spec'ing is a key element to good sound. I do it with everything where I deem myself capable (or I am idiot enough to set myself some kind of stupid target).

 

And oh, once you're there with the better measuring amp (say you finally found it), you will come to the conclusion that all was in vain because of the analog volume control you applied in aftermath. Now it is back to square one (an unsolvable one in the realm of noise and linearity).

So that was the story about the preamp. Haha.

[PeterSt]

37 minutes ago, sandyk said:

The attached link may be of interest.

 

Yes, nice article. That is, good for some inside on what to actually consider. But it omits two major elements which may make you judge quite different than what the gist of the article is (although it nicely lets you decide for yourself).

 

Later ...

 

PS: At least I should have left in the one but last sentence in my previous post like this:

you will come to the conclusion that all was in vain because of the analog volume control you applied in aftermath, although a digital volume may even be more complicated to execute well.

But it didn't end well in that post, so I took that out again.

[PeterSt]

25 minutes ago, sandyk said:

but to compare your DAC with the majority of affordable offerings is perhaps a wee bit unfair ? 

 

WHY in heaven's sake, do you AGAIN start about our DAC ?

No answer needed, but you really should stop this. I can imagine that you always think I am talking about our DAC, but if I tell you I really never do unless I say so ... does that help ?

So please ... ⛑️

 

[PeterSt]

Maybe a little different subject, especially for you, Alex.

 

Why on earth shouldn't we all talk about the very best solution for audio, which in the end is not even related to $ really - only about best choices ? (in a following post)

 

Why would we need to comply to - and it is just an example - people who also like to stuff their whole HTPC, FM radios and what not, in their audio playing PC ? I mean, I understand that people (including me btw) may have the desire, but it is about audio(philia) or is it ?

 

So if you would be so kind to allow me - I talk about "best SQ" and such. And not about best digital radio supersh*t.

This does not mean that it is not allowed for people to use that, but debunking a good (preamp and such) discussion because they coincidentally like FM radios and more, No. That is no argument.

 

On again an other hand, might it be so that now *I* am out of line because of my ideas about good sound and such and that we really should take a step back because all is lost to begin with ...

[PeterSt]

24 minutes ago, PeterSt said:

which in the end is not even related to $ really - only about best choices ? (in a following post)

 

First off, I'd have to say that maybe 50% of people would agree about omitting a preamp. Following my math, this means that 50% does not agree. This is for various reasons, among which SQ itself. So 50% chooses a preamp for the better SQ it depicts and 50% omits it for the same reason - better SQ.

Funny. Funny, but also the truth and everybody is right.

 

Might one be able to choose

- a (very) low noise power amp;

- a DAC with sufficient output drive capability (but also sufficiently low output-noise);

- an other solution for their HTPC needs;

- a well executed digital volume (see next post)

 

then hardly with extra $ one would be quite ultimately better in the Sound Quality realm.

Of course, when one doesn't possess these goodies currently, it will cost $ to buy it. Otoh, don't I see another 50% of people buy new gear all the time ? (with great stories towards their partners meanwhile)

 

Part of the choices is also about "and what is actually important". I mean, we could talk Class A and less distortion and the like, but again show me the measurement of that amp on the noise line alone.

For example a GainClone will do very well, costs a virtual zero when put together in DIY fashion, but may take a grand to let it behave as could and not as the datasheet shows (and skip the Pass designs with it). Low noise, straight noise line and a THD+N IIRC of 0.0040% or so (at full gain) on Redbook source material (I don't think I measured them with 24 bits input signal, with again the remark that I always measure a complete chain and not via a signal generator).

Maybe it is that people think that if the amp does not weigh 40Kg / 75lbs it can't be much.

 

A choice for a DAC with driving capabilities ... it costs zero extra to produce. But it requires design strategy or thoughtfulness to do it. Also the thought of the small list above.

 

HTPC can go via computers all the way. This subject is a bit larger, but I watch movies through the main system for audio and don't have a preamp.

 

Digital volume ...

 

 

 

[PeterSt]

3 hours ago, sandyk said:

The attached link may be of interest.

 

3 hours ago, sandyk said:

https://www.soundstageultra.com/index.php/features-menu/general-interest-interviews-menu/311-what-s-wrong-with-digital-volume-controls

 

So indeed. A good article IMO. One major "fault": the presented perception you'd imposed to have of the digital attenuation.

There is no-single-best-way a digital volume can be applied in a DAC of the (implied) subject in that article. Only when the DAC of concern (mind you, it is about 16 bit CD players in there) has a voltage referenced volume directly controlling the D/A's output, *or* when it would contain an FPGA arranging for attenuation, the article could have a case for the pro's of digital.

 

So point is, the imposed digital volume may not even be in there in the first place (think CD player). It could be a ladder voltage divider controlled electronically by a remote or otherwise digital looking means (like nice numbers on a display) - which thus is just an analogue volume again, OR it would be a volume control chip of which everybody at Maxim's think it is useful for high end audio.

Today we have the more sophisticated (8ch) ships for it (something with C8 - Cypress ? I forgot) but in all events they are accurate (= matched per channel !!) to 0.5dB at best.

Now try to make a balanced situation of that. The THD will be screaming.

 

Eye opener ?

 

Yes. Maybe the eye opener is that a digital volume can normally only exist in software (this includes an FPGA or DSP chip if you want). Thus, something which is still ahead of everything, in the pure still digital stream/domain. Think of: no matter what, if I need to divide two channels by 256, that will be accurate to the unmeasurable dB (maybe a very small rounding error at the bit depth). This is different compared with the 0.5dB I talked about, which will not even be linear per step (like 256 to 1024 steps to attenuate 60dB, depending on the chip).

 

Next up is the analogue volume as being detrimental just because of its means of attenuation. And mind you, this was already my commercial story before I even was thinking about making DACs. This is just how a digital volume can be (should be ?) done, compared to how an analogue works (has to work);

 

Remember my mentioned 48dB range of attenuation which would work for 16bits source material in a 24 bit environment ? this was about shifting. Maybe the article talked about such means after all, but all my eye fell on was chopping off (the bottom). Of course not ... Nothing is lost at the bottom as long as you *have* that bottom and as long as it does not disappear in the noise.

This is not special PeterSt DAC technology - this is just (PeterSt) software.

Granted, that PeterSt next makes a DAC which takes this feature into account ... can he help it if others do not ? The software was first and it was all spelled out (again, as a commercial story back at the time).

Not to forget, if the power amp next adds the noise, then all is a little bit in vain again. Or, if we use a preamp or analogue volume control which is nicely capable of attenuating higher frequencies -or bands of it- differently (read the article and focus on impedance - I am sure it was mentioned in there), then what to do with shifting down bits in an already mangled-with area.

 

See ? this is how the better SQ emerges. And if people grasp what I am saying in my Dutch, then they see that it only costs less to achieve it. But make the proper choices.

Earlier on (or in the other thread perhaps) I talked about the manufacturers of audio needing to do this and how quite impossible this is (about inter-connecting things the right way). But if you focus on these matters, you can do it yourself.

If you deem it needed, of course.

 

Peter

 

 

[PeterSt]

21 minutes ago, sandyk said:

 You appear to be giving most affordable DACs far more credit for their performance than they deserve, when you mercilessly dismiss a high quality Preamp

 

Alex, the bit of the sad point is that everybody throws high quality pre-amps at me, but I don't even dare to listen to them.

With apologies to the owner (if he just doesn't announce himself then no harm can be done and that the manufacturer makes this of it - I can't help that) ... this is an USD 8K costing pre-amp. Maybe it was a bit more.

 

Same type of measurements as always.

 

-0dB (its own resistor ladder volume wide open).

 

Attenuated with 6dB. Start to get the idea ?

 

 Attenuated with -48dB.

In between the situations, everything changes all directions (I think this has 60 steps).

And then to think that this is in the end again about inter-connection and proper grounding (I am pretty sure of that). But I couldn't get it better than this.

 

Now I also have such pictures of a 20K costing pre-amp. I was asked to build a DAC in there. It looked even way worse than what you see above. In the end I refused. 40Kg was returned by UPS.

Do please note that it was the manufacturer (who I highly regard) himself asking this.

 

These are extremes. But I also have the screenshots of hard two years work of some engineer who started his own company and via-via wanted to let judge his new creation by me. Three of us worked a full day to get things right the best we could (after I had measured similarly and pointed out he'd better go back to his former employer) and when we thought we were done, my wife came home and she right away asked what the h*ll we thought we had created (while we, including myself were fairly satisfied)).

 

And indeed, the owner of the pre-amp of the screenshots above, is quite satisfied with the sound. And he knows the results. No anti-placebo there.

 

So you could be right.