Phasure NOS1 vs. Pacific Microsonics Model Two

[gkoones]

So, we now have the "game changing" dac, how about some of the more down to the earth details. What's the price of the unit? How long is the warranty? What’s the financial stability of the manufacturer to back up that warranty? Is it sold directly or through retailers or distributors? What are the dimensions of unit? Finally, exactly how was the determination made that it outperforms the storied and legendary Model 2?

 

[manisandher]

"Finally, exactly how was the determination made that it outperforms the storied and legendary Model 2?"

 

Purely subjectively.

 

Mani.

 

[barrows]

"Purely subjectively."

 

As this is the only relevant criteria to judge an audio component! And this is one person's opinion, of course I am sure if you go to the Phasure forum you can get the opinion's of others. As for your detailed questions, I would suggest that you send PeterST a PM for detailed answers.

Also remember, by nature this DAC avoids SPDIF and operates asynchronously by nature, this fact alone should result in improvements in performance of the PM Model 2.

 

 

 

[PeterSt]

Wait ... please ...

 

If this means that people may be offended, or this is appreciated as inapproriate otherwise, I am happy to answer questions - but not if it really shouldn't be done here.

 

Yes, maybe I like to share because I'm proud a little too, but it really would be ok for me to answer questions on my own forum.

 

On the other hand, apologizes if I read this as a message while it is not one at all. English you know ...

 

Regards,

Peter

 

[SteveChicago]

To echo Barrows, I wish this Phasure DAC were compatible with MAC computers. I looks like a great new concept / product. In addition, Mani's endorsement carries weight with me.

 

Q

 

[TimDH]

Peter, I am really enjoying this, and I think others are too. Thanks for your responses to my questions- mostly answered now.

 

To recap, the upscaling is key to getting the non-ringing performance, correct? The data are upscaled and then put through your software to get the analogue-like behavior. I shouldn't have used "massive" upsampling, but it's what it takes to get from the source resolution to 384 khz.

 

Also, can you comment more on the PC aspect. Many in this forum try to avoid CPU load, and you say that the processing is in "parallel". Does this mean that the CPU is working hard while XXHighEnd is playing, but it doesn't affect the sonics, or is the CPU load light? I'm also interested in why you say that the energy savings features of laptops make them worse than a standard PC.

 

Thanks again,

TDH

 

[Elprior]

Peter,

 

no please, keep posting here !

The thread is all about the NOS1 being a great dac.

 

And it's always fun (and informative) to read (and try to understand) your explanations.

 

Elp

 

[SteveChicago]

Peter, you should be very proud of your hard and successful work. Do not apologize to anyone!

 

Your English is fine (remember, it's Mani who has the accent) :-)

 

Steve Q

Chicago

 

[ted_b]

(however, the self-congrat comment is too over the top and not warranted IMo). I think readers of this thread ought to know what the investment is like; they've been told what the return is likely to be. BTW, the latest price I have seen is about $3900 US. The dimensions are not on the website bu the picture I posted above shows the unique footprint, not exactly "rack mountable".

 

Barrows, I always look forward to your posts and your logical approach, and agree about 99.8% of the time. However, if I understand your last statement correctly (and I probably don't) I will argue that not all asynchronous signal paths are inherently therefore better than a PM Model 2. Heck, we have asynch USB DACs at the $1k level; they are decent but not state of the art just because of asynch operation.

 

[ziggyzack]

I'd like to hear Peter explain those

 

'Height - this is straightforward. Depth - now it becomes a little more confusating. It's like a depth, except it's not, because in fact the depth has some depth of it's own, so it's hard to give atraditional depth figure. As for the width, well, that's the subject of some debate, and even I am not really sure. We don't yet have a device capable of measuring it...'

 

More please Peter ST!

 

[PeterSt]

Ok, thank you guys. That makes me smile again.

(but if Chris interrupts this ...)

 

Barrows,

 

Actually this is the most interesting question.

But before I will give some sort of an answer, I was thinking ...

 

You wouldn't mean DC Offsets which may be eliminated by digital filtering, or do you ?

 

If not, for fun and honesty ... what would you think will be acceptable for a(n electrical) DC offset ?

 

Btw, by accident Mani already knows the answer, so I'm afraid I can't lie (later).

 

Peter

 

 

[manisandher]

I'm sitting at the back of the classroom with my arm all the way up - a first for me.

 

Steve, don't get me started on accents...

 

Mani.

 

[PeterSt]

Yes, easy answers obtain priority :

 

17.32" x 17.32" x 7.5cm.

 

The height excludes any feet (3 or 4) of 2" more or less because they are not provided.

 

Oh, for people not common to inches and stuff, I calculated 2.5cm per inch here.

 

And by these larger dimensions (feet, inches or otherwise) all the connections are on the inside. So, sticking out nowhere (see earlier picture).

 

Yes, I promised pictures, and promised myself each day. Will happen tomorrow. Maybe someone should remind me and keep me to it.

 

 

 

 

 

[1audio]

When used with a Lynx, RME or Mykerinos card using external word clock from the Model 2 the system is effectively asynchronous, since the master clock is inside the Model 2 and the upstream chain is locked to the Model 2. This is clumsy since every sample rate change requires manual intervention at the Model 2 but it does a good job. None of this addresses the effects of RF leakage and conducted emi in the cables between the boxes. Keith Johnson spends considerable time checking for problems when he sets up the Model 2's and he uses isolation transformers on the power to reduce these effects.

 

 

 

[souptin]

Mr Rimmer wants to know when he can have his badge back...

 

 

Joking aside, my best wishes for this endeavour

 

[PeterSt]

Hi,

 

Earlier in the thread you asked a one line question. I think I thoroughly answered that. For your rememberance, this was your question :

 

Peter, but what about doing the upconversion offline?

 

Now, what about thinking about the time the answer took me ?

And how easy it would have been to say something like :

 

"No Ted, please think further. I don't think that's a real option."

 

Maybe I got carried away, and too extensive answers are not appreciated.

I will try to keep some senses.

 

Peter

 

PS: We posted along eachother (and never saw your last one); otherwise I would have said this first.

 

 

[ted_b]

upconversion? And I was quite happy with your answer; I am no longer thinking offline upconversion makes sense for those reasons. What is this last post about?? I have no issues with you....I simply said I think asking about size, weight, cost and warranty are good questions, and his calling your posts self-congratulatory are not needed here. What's wrong with that? (Wow, this is getting a bit testy.)

 

[manisandher]

I think this was a genuine misunderstanding.

 

It happens to the best of us... just ask my wife.

 

Mani.

 

[manisandher]

"When used with a Lynx, RME or Mykerinos card using external word clock from the Model 2 the system is effectively asynchronous, since the master clock is inside the Model 2 and the upstream chain is locked to the Model 2. This is clumsy since every sample rate change requires manual intervention at the Model 2 but it does a good job. None of this addresses the effects of RF leakage and conducted emi in the cables between the boxes. Keith Johnson spends considerable time checking for problems when he sets up the Model 2's and he uses isolation transformers on the power to reduce these effects."

 

Although I would never consider that I could match KOJ's setup skills, I do slave the interface to the Model Two in exactly the way described. I also use ground isolators between the interface and the Model Two and power both units from a PurePower2000 AC regenerator running in battery mode.

 

Mani.

 

 

 

[barrows]

Ted:

 

"...However, if I understand your last statement correctly (and I probably don't) I will argue that not all asynchronous signal paths are inherently therefore better than a PM Model 2. Heck, we have asynch USB DACs at the $1k level; they are decent but not state of the art just because of asynch operation."

 

No, of course I did not mean to suggest that asynchronous DACs will always sound better regardless of implementation. My comment contained the presumption that both the PM Model 2 and Peter's Phasure NOS are well designed and implemented DACs. And to address Demian's point of a clock linked PM-2 being asynchronous at the same time, yes I agree, to a point. The PM Model still uses SPDIF, and this requires the attendent SPDIF receiver, which I believe it is correct to say, will still add (some) jitter even in a clock linked scenario. (Not to mention the accumulated jitter caused by the clock linking itself, cable losses/reflections, signal transmitter losses, etc.)-inherently, a system using a single fixed frequency clock, adjacent to the DAC chip(s), without having to send that clock anywhere else is ultimately going to result in the lowest possible jitter-assuming equal implementation.

 

Peter: re DC offsets. I was referring to the analog stage, and offsets at the analog outputs of the DAC. This was a round about way to try and get an idea of the circuit design-to deal with DC offsets there are basically three ways: design a circuit which can be carefully trimmed adjusted to have no/minimal offset, this usually requires a somewhat complex circuit. Use a servo feedback loop to eliminate offsets, or AC couple the circuit using caps. My feeling is that each of these approaches has advantages and disadvantages. My opinion is that the lower the offset the better, below 50 mV is good, but I would rather see DC offset in the single digits. I was just wondering about your approach.

 

 

[PeterSt]

(the self-congrat comment is too over the top and not warranted IMo).

 

Man, I am sorry. Please understand that this only takes some more seconds to read this correctly (but blame my englisch not being able to do it within the second). I will take the blame fully on me, but this is how I read (again, too fast), and allow me, it's just for your fun (hopefully) :

 

a. Gkoone's questions should be more private ? eh, huh, who said that ?

 

b. No, nobody did. Hmm. Ah, wait, DISagree. Ok. Still nobody said that.

 

c. The quote above (confused because of the before) ... self-congrat. Man, see, it *was* a message, and here it is repeated. Not warranted. Ok, so not. Well, I can agree.

 

d. Heck, why does Ted say this. He asked a question himself. Ok, answer was not appreciated I guess. Well, why answer longer than the question ever was in the first place.

 

Ad c.

Yes, I left out the "however" (never read that), and I guess all it needs now, is to see that "comment is" should really be "comments are" to make my failure here not happen.

All it further needs is your comment about USD 3900 which sounds a lot more to me than EUR 2900 (but which sadly is so at this moment ... just wait :-), and some comment about things not fitting in any rack. Disaster is about to happen (my side). Of course I continue to read that this time you won't agree with Barrows (who shows interest in, say, "me"), and at this moment I am biased enough to think you have an investment in the Metric Halo or something. All it needs further is to briefly browse throuhg this thread and meet your post titled "Ignorance is showing here".

 

Again, this was for your fun, but this is what happened. Even waited an hour before clicking "Save" to the post, because I really couldn't imagine you "behaved" like this. Still I did against all odds.

 

I was wrong.

Peter

 

 

 

[PeterSt]

I'll try to be brief ^-)

 

The subject has been passed numerous times, and I think I never have interfered in any of those threads about it. But IMO :

 

A (connected) Word Clock as a master can't do much in a system (DAC) that is really bit clock driven (the Word Clock formally derived from that, or from the Master Clock operating on an even higher (or the same) frequency).

For SPDIF I am not 100% sure about this all, because SPDIF is (a clock signal) "derived" to begin with. Still, in-DAC a bit clock would be some master of it all first, and I don't think there's a good way to derive the bit clock (as a devider ?) from the Word Clock. Of course, the Word Clock determines jitter (letting the samples go and such), but still the bits must be readily available first.

So, for SPDIF maybe, because I just don't know it really, but for I2S, no, I can't see how (btw, the length of the wire with the clock signal creating jitter already).

 

Peter

 

[manisandher]

"My opinion is that the lower the offset the better, below 50 mV is good, but I would rather see DC offset in the single digits."

 

I'm measuring 5mV at my NOS1 analog outputs.

 

Mani.

 

[PeterSt]

Let's start with this :

 

My opinion is that the lower the offset the better, below 50 mV is good

 

Haha, you just must have asked Mani.

Ok, kidding, but this is exactly what I said to Mani of what would be within specs if I had to determine them.

 

You didn't ask, but a well behaving DAC would be under 20mV on all output pins (which are 4, thinking balanced). A very good one may be all under 12. My own does 15. Usually two will be under 5, and one will be 0. It depends on how things "balance out" if I'm allowed to say it like this.

 

to deal with DC offsets there are basically three ways: design a circuit which can be carefully trimmed adjusted to have no/minimal offset, this usually requires a somewhat complex circuit. Use a servo feedback loop to eliminate offsets, or AC couple the circuit using caps. My feeling is that each of these approaches has advantages and disadvantages.

 

Maybe you didn't want to say it exactly like this, but I don't see any advantage. Well, except for keeping the DC low of course.

 

Earlier I told you that the subject is the most interesting, because, well, it just is. It is when you don't like to apply any circuitry (up to one cap) like I do. And hey, isn't this why you asked the question in the first place. And so I don't. Well, that's what I told not even in between the lines.

 

Now, to your surprise perhaps, I don't like much to tell what happened to let this happen, but what I can tell at this moment is that it depends a bit on the environment so to say. People may receive 50mV or even more, and although totally harmless it clearly is not the best for the best sound. In theory that is. Right now I just have another little project behind me, which guarantees the lowest possible DC a DAC can inherently have. With "inherently" I mean the internal current flowing which is unavoidable. But, it makes it all independent of the outside world to the DAC, so people who now may have a DC offset of 50mV because of environmental influences, *will* end up with that under 20mV of which I know can be the inherent DC (max that I have seen so far).

I am not 100% finished with this yet, but when I am people will be able to "upgrade" themselves.

 

No real answer, but I hope it will do, that indeed (still) nothing is in the signal path except this one resistor. Anybody can correct me, but I think it is still much better to have a little DC (which is about stuff like absolute phase only -> 50mV against 1.5VRMS which is about 42 times more and that against some wave shape and phase "offset" (per pin)).

 

Lastly before I forget (and hoping nobody made it till here because I'm almost too embarrassed to say it afer being 18 months or so being plain against it) : gain is done by an opamp. It really turned out to be the only way, and when I at last gave in, it turned out to even improve SQ which I never imagined. People who followed all a bit will know how many times it failed before I suddenly found a way which just worked. Is that in the signal path then ? let's say it is. But it depends on how it's used, or better : what it is used for. That is why I count it out.

 

Peter

 

 

[PeterSt]

Trust me that I didn't see Mani's post before hitting Save.

But somehow I had rather seen the two posts the other way around.