Phasure NOS2

[PeterSt]

Just in case people are still not sure what the April's Fool joke was ...

[PeterSt]

Still, this is one of the bigger secrets. Or mile stones, actually.

With a bit of prayer that it will work ...

[vrao]

Peter,
Very interesting.

On the names for the new DAC/amp

Something out of the box, A different path!

Poincaré conjecture edition (NOS-PC edition)

One of the most interesting problem only recently solved by an enigmatic mathematician. In short explanation of a hyper sphere (sphere in a higher dimension).

http://en.m.wikipedia.org/wiki/Poincaré_conjecture

Tesseract is another option (NOS The Tesseract)

http://en.m.wikipedia.org/wiki/Tesseract

In many ways we can consider bits to be another dimension with computers becoming more and more sophisticated.

It's an amp, DAC, DSP, modular. Something that has been not done before ...

VJ

[rakeshpoorun]

Still, this is one of the bigger secrets. Or mile stones, actually.

With a bit of prayer that it will work ...

The picture above makes one thinks of the stained glass windows one sees inside a church. One could take inspiration from this and come up with the following names:

Phasure Trinity

[/b]

or

Phasure Omega

[/b]


Well, these names can be linked to what Peter is trying to achieve. Trinity suggests strongly the multi-functionality of the device as well as its supremacy. Omega draws a sly arrow towards the Alpha dac but suggests, that whilst the Alpha is the beginning, Peter's device is the end of the journey.

Best regards
Rakesh

[acg]

There is already a Trinity dac and it is not unlike the original NOS1 in some ways.  It costs $50k or something silly like that.

[PeterSt]

Man, I woke up at 3:30 am and couldn't sleep because of all these names floatng through my head.

I saw VJ's Tesseract and liked it very much. But when back in bed I already couldn't tell that name any more. And even now I had to look 4 times before having it typed correctly.

Then, being in bed, something stupid happened;
I first thought of (Thunderbird's) Zero-X which I always have found a super name. But I couldn't find the resemblence much. Then from there I came to Zero-One and while this was all just in my mind, a "pfff" came outlout from my lips because of so daft, but also so true.
Although true for very much digital, a bit strange and not all that much saying; If you imagine a 0-1 instead then things get more strange, and also more true. DSD blahblah and such.

Then, only a few hours ago I thought to fall back to an internal code name I have been using a year ago, and which I even presented to you in the forum (an in between the liner).
An attempt in the next post ...

[rakeshpoorun]

Peter,
Very interesting.

On the names for the new DAC/amp

Something out of the box, A different path!

Poincaré conjecture edition (NOS-PC edition)

One of the most interesting problem only recently solved by an enigmatic mathematician. In short explanation of a hyper sphere (sphere in a higher dimension).

http://en.m.wikipedia.org/wiki/Poincaré_conjecture

Tesseract is another option (NOS The Tesseract)

http://en.m.wikipedia.org/wiki/Tesseract

In many ways we can consider bits to be another dimension with computers becoming more and more sophisticated.

It's an amp, DAC, DSP, modular. Something that has been not done before ...

VJ

If names like Tesseract or Thunderbird appeal, other names that can be derived from popular culture are:

Phasure Skyfall

[/b]

after the James Bond film which deals with both Q, who can more do more damage on his laptop over a cup of Earl Grey, or the villain who can cause nations to rise and fall by hacking, and in fact hacks M16's network

or

Phasure Matrix

As above, construction of virtual reality world

or

Phasure Serenity

From the name of the spaceship in Firefox, cult sci-fi tv series

I quite like Phasure Serenity...

Best regards
Rakesh

[acg]

Ahhhhh.....the Serenity.

Quote from a famous Aussie movie.

I like it.  Good name.

[PeterSt]

Well, what's in a name !


Here's an attempt to make clear what exchangable modules are in there :

• Main Board
  
  May seem a strange "module", but all the logic is on there, and logic may change ...
  This of this as a PC motherboard. It concists of a half related to the PC interface and a half which is fully isolated from that (same approach as the NOS1a regarding this).
  
  
• Interface
  
  Computer Interfaces may change, so this "obviously" is a separate module.
  Notice that there's also S/PDIF and AES inputs.
  
  
• Clock Module
  
  A smaller module holding the oscillators, obviously subject to change.
  In this case this goes way beyond oscillators only, because there's a whole "clock tree" with buffering and all which is required to control the further away ladders.
  Also the whole clock line is setup differentially and in 50 Ohms (this is also PCB stuff) while at this moment the oscillators itself are not differential ones (so the transport of the clock signal still is). But this is only because the differential oscillators existing are not the better ones. But this may change ...
  
  
• DSP
  
  The DSP module is prone to change because of the development in speed and capabilities. Also notice that what we are talking about here is a complete graphical interface (provided to make the filters) and that at this moment it is not 100% clear yet whether what we want is sufficient for bandwidth (and this is just the at this moment lacking of the programming experience for the DSP (chip) concerned).
  Also notice that this is only one example of the reason of the modularity, which is not only future update possibilities; it is just for ourselves to see whether what was decided in the design turns out as expected; when not, only a small part needs to be re-done. This counts for almost all of the DAC.
  
  
• Ladder Module
  
  The Ladders too are modules and they are the most subject to change, which is because the most "vague" logic is on them. It is here where the R/2R stuff really happens and what switches on and off the bits. I say "vague" because it is also here where a completely new means is applied for this logic.
  
  A Ladder Module always comprises of two R/2R ladders and is differentially setup throughout.
  It it designed to hold 28 bits and resolve to 26 bits.
  
  Take it that actually a complete D/A chip has been mimiced which all its features as far as could be determined - but now in discrete form. A quite undoable task in itself, were it about improving over the commonly known D/A chips for the cheer technical merits. But the advance : no active components to form the signal. Just resistors to create it.
  
  A tremendous amount of work has gone into the net applicable jitter during the D/A process. I say that this too is completely new and that actually no discrete design exists that takes into account what's all in order here (without laying out what that is). It would be a fact though, that all what normally happens in-D/A-chip regarding jitter, certainly has not been taken into account (no specs exist about it anywhere). The fun is though that when you rebuild (or re-engineer) such a chip, you can easily see what's all there to be detrimental to jitter.
  For this reason alone it should be the lowest jitter DAC ever.
  
  A Ladder Board goes into the Main Board like a PCI (etc.) card goes into a PC mother board. There's sockets for 8 of them.
  
  
• FPGA Firmware
  
  Not a real module of course, but changable by means of firmware files. And, also necessary, thinking about future updates of any kind.
  Each differential Ladder Board contains its own FPGA.
  
  
• Bit Module
  
  Yes, now it gets crazy. Even one bit is on its own small PCB;
  Think 10 or so components on such a PCB and it forms the R or /2R part.
  The real reason for the bits ending up on their own PCBs is a real strange one : The design of the Ladder Module (PCB) was out of control otherwise. So it's a bit like a computer program : when all is stuffed into one main section, the overview is lost and it is all a redundant mess. If something needs to change, it needs changing (in our case) 224 times. Not now, because only that one Bit PCB needs the change and it is just produced 224 times (again).
  
  So yes, you read right - 224 small PCBs for this alone for 2 channels.
  They go into sockets on the Ladder Module - parallel to it.
  
  A next dimension of modularity within the Bit Module is the resistors going on there. Remember, with the discrete D/A converter it's the resistor making the sound.
  All can be on there what we want, and the module is designed for both SMD and Through Hole resistors.
  Notice that it even will be so that the MSBs (Most Significant Bits) can be layed out with different resistors than the LSBs (Least Siginificant Bits), just because the MSBs are more crucial to the sound.
  Never forget that resistors too cost money and the better "sounding" they are, the more they cost.
  
  
• Output Module
  
  The Output Module bears the line output stages which comprises of buffer or gain or a combination, depending on the real needs.
  
  Think about the Output Module as a long small board going across the width of the DAC cabinet, near the back wall, sticking to the Main Board in parallel; it contains all the output line terminals hard-wired.
  Outputs are RCA and XLR and true 75 Ohm BNC. So max 24 in total for 8 channels.
  
  
• Power Amp Module
  
  Not really subject to change (once it works as expected), but modular in itself apart from the number of them (one per required channel). So, which channels bear power amplification can be decided (including none at all) and how much power each channel needs to provide can also be decided.
  
  Each one module is differentially set up from head to tail and the speaker outputs go by wire into the cabinet's back (one for plus, one for minus).
  Envision the (max) 24 line outputs (see above) to be on the bottom in the back of the cabinet and a row of (max) 16 speaker outputs above that.
  
  I rate the output power at ~52 Watts but this depends on the power supply and what's feasable there.
  So just notice : peak power can be 221 Watts and in continuous RMS it can be 78 Watts. But see below.
  (output voltage is +/- 15.8 (so total 31.6V))
  
  THD+N is theoretically rated at better than 0.00003%.
  
  
• Linear Power Supply
  
  Comes in a separate cabinet at max 1.5m power cable distance.
  Power Amplification for 8 channels assumed, 56A needs to be provided.
  Now we run into a few problems because of the transformers needed, how large they are allowed to be to avoid humming and how large that cabinet is to be and what we can still bear (weight !).
  So while 1KVA transformers are required to deliver this power, I'd say this is too much of it and we stick to 600VA, that devided over 2ch per one transformer of 150VA. And that will be heavy enough already.
  If we now think that nobody is going to use the maximum RMS = continuously and redo the math for 150VA transformers then 55.71 Watts remain. And I made that 52W so I never claim too much.
  
  Please keep in mind that all still depends on the possible heat dissipation, which is exactly why this needs to be tested in practice.
  
  

All done by heart and possibly I forgot a "module" here or there, but for now it's enough I think.

Peter

[rakeshpoorun]

Phasure PhD1

Excellent!

"Ph" for phasure, "d" for digital, and "1" to draw link to NOS1? And "PhD" to suggest the intellectual challenge/achievement presented by this new device

This is very very clever indeed.

Congratulations

Best regards
Rakesh

P.S. As it happens, my first serious phonostage (which I still own) is the battery-powered PhD by Sutherland in the US. And I like it very much indeed. Some coincidence... Look it up!

[PeterSt]

is the battery-powered PhD by Sutherland in the US.

Well well !

"Ph" for phasure, "d" for digital, and "1" to draw link to NOS1? And "PhD" to suggest the intellectual challenge/achievement presented by this new device

There's even a third dimension in that. But you can't know.

[Matt]

PhD1
Phasure/DSD/1 ?

Excellent

Matt

[Matt]

Peter,
is it possible to get a PhD1 without DSP module?

Matt

[PeterSt]

Hi Matt,

No. So as I said, it will always be in there, but will be passed by when not in use.

Why ?

Peter

[PeterSt]

Btw, I said No, but I'm not even sure why. The Main Board is not in production yet, so it is a matter of passing by in hardware. But I must look into that.

Still : why ...