New NOS1 dedicated PC. Less terrifying thanks to you. All help still welcome!

[rakeshpoorun]

So.

I have been thinking a little more about my choice of motherboard recently. It is not as with other things in the life of someone who cares about audio where if you are not completely satisfied with a discrete element of the playback chain you may just about get away with swapping it (and assuming that the fault does not lie elsewhere) and see what happens.

Someone wrote earlier in this thread that different motherboards have their own "individual sound." Whilst I doubt that one can generally audibly identify such differences, unless one has extensive experience building and listening to such audio pcs (like Peter here probably has for instance), I would think that in an audio pc which is acting as a "memory player," such differences as they might be are of little import. On the other hand, what is extremely important are other factors which can be determined by the choice of the motherboard. Now that I have spent time reading, researching and thinking about this project for a few weeks, and reflecting on what are the ideal characteristics that I would be looking for in a motherboard, the ranking in order of importance of these different characteristics, and what motherboard out there fits closest to this "ideal" mix of characteristics/features, there are still a few issues that require a final determination.

As a side note, I would add that, in my opinion, if anyone here is in my shoes and is thinking of a new audio pc, and has something that will competently do the job for now, then they would do well to wait until next year if they can, when Intel will have launched the new Skylake desktop processors and the Xeon DE processors (like the recently released D1540), which are low power  and multicore, and with the advantages associated to the new chipset such as USB3.1, even Thunderbolt 3 and even integrated TGbase Lan. For those planning to build their audio pcs next year, they are likely to be able to tick a long list of what I see as these ideal characteristics and then some.

For those who can wait, the future promises some exciting and promising solutions. The present does not have to be so gloomy though if one can balance wisely between occasionally competing and frequently conflicting demands.

I imagine that one might think that in order to inflict the least harm on the audio file while it is being processed by XXHighend, one might want to bring these processes within the confines of memory whilst also minimising demands on the PSU, something that I understand XXHighend does extremely competently.

I am quite happy with the location of Windows OS, XXHighend, Galleries and the music files in my proposed audio pc. Should I feel the need (urge?) to experiment between slightly different locations at a later date, changes such I envisage could be made quickly, cheaply and unobtrusively.

There is however one area where the decision must be made from the beginning and where there will be no going back. I do not/will not/cannot afford to change motherboards for years to come and so the question is as follows. Either I go for one which has no on-board graphics so no matter how much I can reduce power requirements elsewhere, I will never be able to get rid of the 8-12W that I imagine the Nvidia 210 (passively cooled) will draw. On the other hand, there are several eligible motherboards which have on-board graphics (via the Aspeed 2400 video chip on the motherboard typically). Is it the case then that such a motherboard will enable one to reduce the power consumption of the system as a whole even further? I ask this question more in hope than expectation as clearly it seems no one (who contributes here at any rate) has much experience with server motherboards like the one I am planning to use. Still theoretically (says someone whose understanding of the relevant theory is wafer thin), one would think that XXHighend will be able to reduce the demands made from the on-board graphics more than it could do with the PCIe slot GPU, whose power draw will never go below its "idle" state power draw. Any ideas?

And does it matter? Peter stated that his system's power draw is 45W at full load (with Windows OS and XXHighend working). Now a dual CPU motherboard, even one as power efficient as the Supermicros can be will draw in the region of 100W (I assume that the CPUs will not draw much at that point). So 10W is an additional 10 percent, surely not worth too much of a fuss.

Even so, does it matter? Well, this Nvidia 210 graphics card requires a PCIe slot and some of these motherboards have only three such slots. I need a minimum of three PCIe slots for SSD1 (Windows OS/XXHighend), SSD2 (Galleries) and the isolated Silverstone EC04-P USB card. Obviously with the Nvidia 210, I will need a fourth.

Any other considerations? Well, there might be. I like watching live or recorded concerts on Sky Art HD, music dvds and blu-rays. Is there a way XXhighend can process the digital two-channel signal from my SKy box or the OPPO player digital output whilst the picture is streamed to the display/projector via HDMI? I imagine that given that XXHighend is a memory player that there will be an insoluble problem with lipsynching and that this is simply not doable within a dedicated audio pc? Even if that is the case, can the NOS1 be used via a route that does not involve the "audio pc?"

I did say I like to ask only one question in any one post but I am also aware that it is a rule that I "(honor more) in the breach than the observance." My excuse is of course that the queries are all intrincally related to and derived from the original issue which is whether or not one would have a preference for a motherboard with on-board graphics or not when one has a choice, everything else being equal. Even if a preference can be found and the grounds are clear, how much importance should one attach to this issue generally when in reality one has to consider other factors as well? As I said earlier, I ask these questions more in hope than expectation. At least, they serve as pointers to some of the considerations that I face as I continue on this journey that will hopefully lead to some result worthy of the effort and grey matter I have invested in the process. It may also help someone else embarking on this journey as they may be able to avoid the many pitfalls that will become evident one day down the line when I have a working audio pc/NOS1 system and look back to comment on this journey with the benefit of hindsight.

Best regards
Rakesh

[PeterSt]

Hi Rakesh,

I don't think that you will find anyone even the slightest acquainted with any of the paragraphs in your last post. But of course anyone who is after all - is happiliy invited to respond.

If you try to set new standards, that is fine. Try all out and let us know the result. However, also please realize that you have no reference and so you can also have no result, really. If your sound is "the best" then, well, mine already was as well. And for all of the others, no difference.

It is not as with other things in the life of someone who cares about audio where if you are not completely satisfied with a discrete element of the playback chain you may just about get away with swapping it (and assuming that the fault does not lie elsewhere) and see what happens.

Already there you seem to be off by miles. It may count for you though ... in your mind and at this moment and nobody will have a problem with that. Of course not, why would he or she or me. The serious warning is and remains this :

You won't be able to judge.

This, perhaps sadly, means that you first must obtain something "normal", to next go the unexplored route of ... well, anything you can make up according the current state of technique etc. Only then you can compare.
If it were so that for example me would tell you "sure, the more of that all the better SQ will be !!" ... but I explicitly do not. The contrary and no need to repeat it I'd say.

The only thing what remains for me at this moment is telling you that no normal SSD connects to PCIe although it will be possible if really needed. Just SATA(III).
Also, nobody buys a MoBo for the years to come. At least that is what I think. Sure, if one first makes himself so dependent on all around it that hardly anything else is possible later plus the $ investment is too high otherwise ...
But things were then put upside down to begin with.

Anyway ... and I really have to emphasize it or otherwise it is misunderstood :
Nobody can really say / state in advance that this all is not useful. So by all means go ahead with it and set the standards (ehm, if it works out). But how can you tell that it is "the" standard without reference ?

Best regards,
Peter

[rakeshpoorun]

Hi Rakesh,
...
If you try to set new standards, that is fine. Try all out and let us know the result. However, also please realize that you have no reference and so you can also have no result, really. If your sound is "the best" then, well, mine already was as well. And for all of the others, no difference.
...

The serious warning is and remains this :

You won't be able to judge.

This, perhaps sadly, means that you first must obtain something "normal", to next go the unexplored route of ... well, anything you can make up according the current state of technique etc. Only then you can compare.
...

So by all means go ahead with it and set the standards (ehm, if it works out). But how can you tell that it is "the" standard without reference ?

Peter,

It is kind of you to suggest that I may be trying to set some standards, given that it sounded more correct to interpret your meaning as being that I have no standards when it comes to a music pc.

The more unkind interpretation is in fact the correct one here. I have no standards or benchmark against which to judge any results that my "audio pc" may lead to.

But then my objective is not to set any new standards. In a reply to Alain who made a similar point to yours earlier in this thread, I wrote:

...

In all honesty, I do not know how this build has come to be regarded as a "super powered pc."

If it is indeed a "super powered pc," this was not my intention and the whole thing is a darned mistake. Given how little I know about computers generally, it is not at all the situation I was aiming for. And a mistake of the worse kind, because I have already acquired quite a lot of the parts required for this project.

...

I imagine it is never really too late and I could always just leave out the Supermicro board, use only some of the RAM and go for something simpler.

But Alain, I thought the idea of having two cpus, which are both low powered and have 12 cores each, made a lot of sense (the CPUs are rated at 75W each, half of what many here are using). I never intended using Windows Server OS but rather Windows 8 64bit that I thought most people over here use. I did not know that the RAM could be too much, I had read in a thread somewhere that Peter worked out for high resolution and upsampled music one would be looking at 24GB and Juan also seemed to have added to his Ram to get to 24GB. I thought 32Gb was fine but because I had two CPUs, I thought let me make sure each CPU can access 32Gb...Maybe the reasoning above is "rubbish," using Peter's helpful terminology , but that's where my thinking got me.

I clearly have a choice. Rethink the whole thing and go for a single processor motherboard or since I have advanced so far, bite the bullet and see what comes out of it.

...
Rakesh

I am inclined to heed your advice. As I said earlier, I might have liked to use a slightly more modern counterpart of the X79 Extreme 4M (i.e. the Asrock X99M announced in April but not for sale yet) by Asrock, so I can use the LGA2001 R3 processors that I have. But the company is intent on releasing products that are then nowhere to be found for sale.

The only thing what remains for me at this moment is telling you that no normal SSD connects to PCIe although it will be possible if really needed. Just SATA(III).
...

You are right. No "normal" SSD can be fitted/connected to the PCIe slot.

You are also wrong. A cheap AHCI SM951 128GB SSD by Samsung will do it without breaking a sweat. See the links below.

http://www.anandtech.com/show/8979/samsung-sm951-512-gb-review

http://www.flexxmemory.co.uk/solid-state-drives-ssd/samsung-sm951-128gb-m-2-ngff-pcie-solid-state-drive-ssd-2280-mzhpv128hdgm-00000/?gclid=CjwKEAjwkcWrBRDg5u6SuPS11C0SJAChLLAHUJvym0KuEUrpLgOdLZRKkCUu0Dln-cWYQIZDrE1PphoCd43w_wcB

The 128GB AHCI version costs less than 100 euros and I imagine that two of these might be perfectly suitable for fast access to Windows OS/XXHighend and Galleries (yes they are indeed bootable and the NVME versions, which are even faster, are also bootable provided the motherboard has the updated UEFI, and the Asrock X99 motherboards have all been so updated recently).

Also, nobody buys a MoBo for the years to come. At least that is what I think.

Best regards,
Peter

I agree. But for the reasons you mention, I still hope that I will get a good three years of service from whatever I buy now. We shall see. I wonder how many Asrock X79 mobos will be around then; time will tell...

But how can you tell that it is "the" standard without reference ?

Best regards,
Peter

I sincerely hope that someone not far from me will be interested in carrying out some comparisons. I could get together with that person or exchange audio pcs and see...

In any case, I am not too sure I any longer agree, if I ever did, that there is a "benchmark" when it comes to what different people here are using audio pc wise. There are too many differences in terms of execution, implementation...

In many ways, in my mind, my proposed audio pc is nothing really out of the ordinary, merely the logical consequence of a few principles that have their origin in the execution of your own "XXHighend PC" and others before who have followed or advocated similar solutions as on the website that Juan kindly referred us to:

http://www.thewelltemperedcomputer.com/

It was certainly well worth a close read.

Thanks for the advice, Peter. There is clear wisdom in it but it is not quite as black and white and logically the inexorable conclusion that you make it appear to be. A solution could have been found half-way where some more modern components could have been used (for example replace the motherboard and suddenly I can use my processor and memory -  but I am not asking, as I did so already and I know that answer)...


Kind regards
Rakesh

[rakeshpoorun]

HARD DRIVES: IN OR OUT?

It is really strange that after two months of reading about the subject, here and elsewhere, that I should feel the need to revisit this topic.

The whole idea of externalising hard drives makes so much sense that in some ways there can be no controversy. As Alain said above, the objective is to reduce variables and the demands on the power supply in the "audio pc."

I agree with the idea in principle but maybe we should stop for a moment and consider the consequences of our desire to reduce demands on the main "audio pc" to as close to zero as possible.

By getting rid of hard drives from the main system, are we not creating a monster further along because the PC or NAS where the audio files will be held will itself have a power supply which will be itself spurting all its electronic noise and RFI into the lines where the rest of the whole electronic playback system draws power? Peter advocated this solution as well but perhaps he has his music files in single enclosures which do not draw much power?

I am surprised that the accepted consensus seems to be to have two computers linked over a LAN. Unless one has the luxury of being able to plug the second pc over a separate mains circuit (I ran a separate circuit from the external power source for the house which was dedicated to my audio system years ago but that is not a common solution I think), are we not just creating a multi-headed monster after spending so much time and effort subduing the first one? And this right where we have the rest of our playback?

On the other hand, if we have our music hard drives in the same case as our "audio pc," don't we find that these drives spin down into inactivity once the audio files have been transferred (I am assuming to Playback partition in Ramdisk) and the PSU, faced with the very low demands made on it as a result of the operation of "XXHighend," is taken, blissfully for us, out of the picture as far as is possible?

I wonder.

Best regards
Rakesh

[AlainGr]

HARD DRIVES: IN OR OUT?

It is really strange that after two months of reading about the subject, here and elsewhere, that I should feel the need to revisit this topic.

Hi Rakesh,

The whole idea of externalising hard drives makes so much sense that in some ways there can be no controversy. As Alain said above, the objective is to reduce variables and the demands on the power supply in the "audio pc."

I mentioned in one of my previous posts that at a certain time I found it practical to have the OS drive in an enclosure, but that it is not required anymore. I was swapping between two drives at the time, so having the OS drive outside of the box was making my life easier.

I also power my OS drive with a linear power supply, but even this can be adapted, as I will have a cable inserted into the PC to still power the SSD (from the LPSU). So it will stay independant from the PC PSU.

Reinserting the SSD inside the PC will also reduce the length of the sata signal cable. Having an external sata (Esata) cable requires a certain length and though the Esata cable is shielded properly, it is still a lot longer than what would be required if the drive was inside the PC.

As you see, putting back the OS drive inside the PC chassis and still have it powered with linear power will only bring good things, unless it can be shown that the SSD will add noise (electromagnetic or else) inside the chassis. I doubt I would be able to hear the difference, but who knows...

Another point is that a mechanical drive will bring some vibrations in the PC chassis. I am entering in an unknown territory here, but Peter would be better than I to explain what this could be about. Of course, there are ways to prevent these vibrations, but in a PC chassis it will probably be more complicated. A mechanical drive in an external enclosure, the enclosure being suspended with some rubber bands with a "U" shaped piece of wood (or else) - that is quite easy. (My OS drive being a SSD, this is not a problem).

As for the music drive, I really appreciate that I can move it in different places or even swap drives (the enclosure I have allows for this very easily, as it is a "trayless" enclosure.
I use this: http://www.newegg.ca/Product/Product.aspx?Item=N82E16817707329&cm_re=startech_drive_enclosure-_-17-707-329-_-Product
By the way, the external music drive is also powered with a linear power supply and suspended.

I agree with the idea in principle but maybe we should stop for a moment and consider the consequences of our desire to reduce demands on the main "audio pc" to as close to zero as possible.

By getting rid of hard drives from the main system, are we not creating a monster further along because the PC or NAS where the audio files will be held will itself have a power supply which will be itself spurting all its electronic noise and RFI into the lines where the rest of the whole electronic playback system draws power? Peter advocated this solution as well but perhaps he has his music files in single enclosures which do not draw much power?

Linear power supply

I am surprised that the accepted consensus seems to be to have two computers linked over a LAN. Unless one has the luxury of being able to plug the second pc over a separate mains circuit (I ran a separate circuit from the external power source for the house which was dedicated to my audio system years ago but that is not a common solution I think), are we not just creating a multi-headed monster after spending so much time and effort subduing the first one? And this right where we have the rest of our playback?

On the other hand, if we have our music hard drives in the same case as our "audio pc," don't we find that these drives spin down into inactivity once the audio files have been transferred (I am assuming to Playback partition in Ramdisk) and the PSU, faced with the very low demands made on it as a result of the operation of "XXHighend," is taken, blissfully for us, out of the picture as far as is possible?

This is a matter of trust. I do trust Peter, even when I do not follow his path in all details. It was easy for me to implement the two PCs configuration and it also will allow to use a tablet to control the server through which I can access the audio PC with RDC... In some way, there is a "split" of work and I believe it can be for the better.

But you will be able to test these different configurations only once you have you PC built. I suppose you already have another PC for the frequent internet access or for any other task. It will be easy for you to determine if the two PC configuration is worth it or not, depending on your experimentations. You may (or may not) find this useful for you.

Like they say, "Your mileage may vary" (YMMV)

You have many good questions. Many of them will be answered by yourself Some of them will require that you spend some money, but as many of us have discovered through time, there is an evolution and this journey can be very interesting, as they will also be "ear openers"

Only you may fix the limits but from what I have seen up to now, you are preparing "the ground" quite thoroughly. But no matter how many questions you will ask, you will discover a lot of things by yourself.

And then, we will ask you questions too

Regards,

Alain

[PeterSt]

A very general remark could be that we are a long way from conclusions around "the PC". And, people may trust me - that even could be justified BUT it is to be noted that I usually only travel paths that others came up with first. What am I saying really ?

Well, that it is only a year ago that I started to work out the means of set up I am still using today, but that prior to that I approached it a sort of other way around and was satisfied with that just the same. You first must know the difference and prior to that must have the idea in the first place (and the ideas usually come from others which is just statistics).

If we go back to the first glimpse of this all making a difference, then I recall Paul (AKA Scroobius) who detached a DVD drive and was sure it sounded better. Just detach it - it wasn't in use anyway.
But mind you, this was maybe 5 years ago (I really did not count) and only two years ago I started thinking about it and worked on it for what ? maybe 6 months. You can't see it, but by sheer accident Alain was involved. It was the goal to have exactly NOTHING in the PC. And this can work, but it takes so much d*mn time that I had to give up. But what came from it was all the support in the software for such a thing and now, for example, the "disk I/O" speed over the LAN is net faster than when the music would be on SATA connected hdd(/ssd etc.).

I am not sure what my subject is, but it should be about 6 months of work to only find out after that whether it is helpful (to SQ) or not. So this is all really crazy stuff. First though, there must be sufficient indication that the result can be beneficial and with bits and pieces this comes forward. A Paul detaches a DVD drive, a Mani detaches a Monitor, a Paul removes a video card (with a full weekend of trialling because the stupid PC doesn't like to boot without it - but worked out), and me ? Oh, I only notice that a phone remote sounds better than a tablet. Why beats me so far, but I think it is so.

So Rakesh, now you have some back ground. And moral : in a year all will be different again.

Peter

[rakeshpoorun]

Hi Alain,

Thanks for the detailed response. I have carefully noted the points you make and they have helped me to think more clearly about what I am trying to achieve.

The fact that you are very happy with the sound quality of your source is a hugely weighty factor even though at times, I do not follow the reasoning behind certain decisions. As you say, some of these decisions are based on trust.

But one should also note that many of these decisions seem to be based on a 3-5 years old approach, at a time when the rules of engagement were markedly different. For instance, at the time, Intel processors were not as efficient as they are nowadays. Most of these configurations assumed that Windows OS/XXHighend and Galleries would be on HDDs or the first generation SSDs. The relative newness of the SSD technology and the Sata bottleneck meant the starting point was very different to what it would be today.

The use of Ramdisks for the operating system, XXHighend and fast low power SSDs mean that we can approach the conception of an "audio pc" on a very different footing. I am nearly sure that if you put Peter, yourself, and others more knowledgeable than myself in a room, with time and resources, that you would come up with something very different, and undoubtedly superior to anything I can come up with from my particularly  limited vantage point. I do not have your wealth of experience, practical knowledge and your understanding of theory. In short, this is the sort of venture where the odds are so heavily stacked against me that I wonder everyday why I even bother on persisting with this project.

But I am me. And I am persistent. So something will happen, good or bad and more likely bad but at least I will have learnt.

...
As you see, putting back the OS drive inside the PC chassis and still have it powered with linear power will only bring good things, unless it can be shown that the SSD will add noise (electromagnetic or else) inside the chassis. I doubt I would be able to hear the difference, but who knows...

...

This is a matter of trust. I do trust Peter, even when I do not follow his path in all details. It was easy for me to implement the two PCs configuration and it also will allow to use a tablet to control the server through which I can access the audio PC with RDC... In some way, there is a "split" of work and I believe it can be for the better.

But you will be able to test these different configurations only once you have you PC built.

...
Regards,

Alain

So Alain, I take it you use a Linear PSU or maybe more.  It is not very clear from your signature what is the power supply arrangement of your system. The Intel processor you use, the i7-3930K is 130W, which even in a minimalist system like yours, alongside the X79 motherboard must mean a peak load of say 170W. Obviously the normal power draw for a system dedicated to a single purpose like yours will be considerably less -  still your PSU must be able to cope with a load of 150W when your computer is booting up. I suspect the Linear power supply you refer to somewhat enigmatically is the HDPlex 100W, but this looks to me as if it is not powerful enough for the task unless you are combining it with the HDPlex 160 or 250 DC-ATX PSU? I would be fascinated to know the power supply configuration of your system even though it will not work in the case of my proposed dual cpu "audio pc" which will likely have a power draw of 100W at idle, 170W at boot-up and 250-300W at peak load (where it will never be admittedly given it will be dedicated for use with the NOS1).

In the future this will be hopefully different, and better. The next generation of Skylake processors will maybe offer 12 cores for a total power draw of 45W someday (including motherboard) and the HDPlex 100W LPS will then be amply sufficient (or maybe even unnecessary given that so little demand will be made on the power supply.

Again thanks for the detailed and helpful response. I appreciated that. It helps me feel a little less left to my own devices in a field where I have no expertise of any kind to draw on, so the chances of me answering questions instead of asking them, as you kindly surmise will happen someday, are both very slim and very remote.

Best regards
Rakesh

[rakeshpoorun]

Spinning around

This has nothing and everything to do with my Phasure NOS1/ Audio PC project. I thought, as there is some interest in horns over here, that you would not mind me posting what is arguably the most irrelevant post/attachment ever.

I placed this order ages ago, in late February, but the whole project had been discussed, researched and fermented over a number of years. I would like to recoup my costs but since I learnt everything I know from forums like this one and my personal research, I am happy to answer questions in a frank, open and transparent manner in the hope that others may learn from my experience and my mistakes.

It is however important that I do not tempt the devil and jinx the whole project which is already behind schedule. So I will wait for the horns to be finished (I am obviously not the one making them - I bought a lathe but then sold it when I realised I did not have the requisite skills) before I publish too much about it - I do not want to tempt fate. I have been let down by so many amateur would-be horn builders and myriad events and circumstances which conspired to delay and distract this project.

What you can see is a small (and the first) section of a fairly large 115Hz tractrix horn spinning.

Enjoy!

Best regards
Rakesh

[acg]

Spinning around

This has nothing and everything to do with my Phasure NOS1/ Audio PC project. I thought, as there is some interest in horns over here, that you would not mind me posting what is arguably the most irrelevant post/attachment ever.

I placed this order ages ago, in late February, but the whole project had been discussed, researched and fermented over a number of years. I would like to recoup my costs but since I learnt everything I know from forums like this one and my personal research, I am happy to answer questions in a frank, open and transparent manner in the hope that others may learn from my experience and my mistakes.

It is however important that I do not tempt the devil and jinx the whole project which is already behind schedule. So I will wait for the horns to be finished (I am obviously not the one making them - I bought a lathe but then sold it when I realised I did not have the requisite skills) before I publish too much about it - I do not want to tempt fate. I have been let down by so many amateur would-be horn builders and myriad events and circumstances which conspired to delay and distract this project.

What you can see is a small (and the first) section of a fairly large 115Hz tractrix horn spinning.

Enjoy!

Best regards
Rakesh

Ha, I'm building exactly the same thing myself!  Those big horns are a LOT of work to turn, so in a way I am not surprised that it has taken several months so far for you to get them especially from an amateur builder (such as myself).  There is a mob in the US that make these for their horn systems and they allow three weeks construction for two workers...so one guy in a back shed is going to be much longer.

[rakeshpoorun]

Old faithful in audio and stupendous customer service

I came home a few days ago and opened the doors to the unmistakable smell of burnt electronics. It turned out that the capacitors of my old faithful Grace M902 headphone amplifier/DAC, manufactured in 2005 and bought by me in 2006, had burnt to a crisp. Although I only use the unit intermittently, I realised that I missed it straight away. Fair enough, it may have its foibles but the combination of this neat little unit with Sennheiser HD650 is something quite special indeed. Then I discovered that the company which supplied it in the UK and had serviced it and upgraded it in the past was not a Grace Design Distributor any longer. Having already called Grace Audio years ago, I remembered a very helpful, courteous down-to-earth conversation that addressed briefly and succinctly all the many queries that I had about the unit (I have always liked asking questions!).

So where normally I might have contacted the companies/individuals I turn to when I need such help, I called Grace Design and spoke to a real gentleman that blew away my usual skepticism when it comes to audio companies, that in the main I regard as instruments of the devil himself. Yes he was happy to help. No they could NOT UPGRADE to the latest M920 "flagship" unit but I had a choice of returning to their European distributor (unknown to me) or ship directly to them.

What shocked me and what is the point of this post is that he felt confident enough to diagnose the problem there and then send me an invoice for a derisory amount including parts, labour and shipping! I was left speechless. This is not a feeling that comes often in this hobby. I remember my dealings with Naim here in the UK in the past that left a bitter taste and turned me away from their whole product line forever, and quite a few others, who in spite of their reputation, were so obscure or obscene (figuratively speaking of course) that I became in the main disillusioned with audio companies. On the other hand, companies like ATC offer a stupendous warranty that they back to the hilt (I took my ATC active SCM20-2a 6 years after having bought these from a dealer with whom I fell out and they replaced drivers no question asked, and even changed the voltage on one rare unit I had imported from the US).

So what is my point? Sometime we audio people forget that in buying products from dealers/distributors/manufacturers, we forget that if you tear down all the hullabaloo about a change in settings that means your mother-in-law can shimmy away to "dance music" (what the hell is that anyways? - I thought a set of Bose speakers and two huge subs would do the job) in the kitchen, presumably this being the metaphysical g-spot sweet spot where the system is at its most resolving...you are left with real products which deliver day in day out, and if you run into difficulty, there is someone you can easily turn to.

So my Grace Design M902 is off to be repaired and I await its return with impatience.

So that's the point. But is there a moral to the whole thing? I think there is. I realise today that some of these companies that have most impressed me (I mentioned two here, Grace Design and ATC - there are others but not many) have something in common. They are not part of the mainstream industry and do not go out of their way to serve the high-end audiophile market, which like FIFA, corrupts everything that it touches. Of course, when they do realise that the audiophile community likes their products, they are likely to start behaving very differently. Even Grace Design and ATC are not immune to this contamination (see the pricing of the M920 and the burl 20th Anniversary special edition ATC SCM-50).

One truth which is rarely voiced and admitted, is that ironically and ultimately, the problem lies more with us the electorate of that corrupt and tainted industry, because we in many different ways behave in a manner which encourages these crazy notions such that speakers consisting of 5 drivers in a nice looking enclosure will easily cost upward of $50,000, and pre-amps can cost north of $10000. I feel so bad about this industry that I will make sure that of my many interests and hobbies, this is the one that I will not pass on to my children. They are welcome to my ATCs (the horns speakers are too much hassle for them) and I wish to god that's all the interest they have in audio. If they continue with the piano as they have been doing for the past two years, and develop a habit of going to regular live concerts of classical music, and if they need to get some audio equipment to replace the Grace Design or ATC I send them off with, I know who I will recommend they go to.

This is a unique and quite tragic hobby I find, where although I do enjoy it, I would not wish it on my children and for that matter, on anyone who has a choice.

So some might ask, if you have used such a broad brush to paint such a grim picture of our hobby and the industry that's intimately bound with it, what do you make of our host? Well, he is still a mystery to me. There is clearly a sense of honesty and integrity that permeates the forum generally and this is also why I was prepared to take a blind punt (I knew nothing about the Phasure Dac until a friend told me about it and I read some of Peter's postings and thoughts here). For me, our hobby cannot be dissociated from the people behind the products that find their place in our audio equipment rack. I will feel happy to have Peter's dac in my room, even though ultimately I might not like his view and vision and interpretation as seen through the implementation and execution of the tools at his disposal. Peter is clearly going through a turning point in the evolution of his equipment portfolio. What the future holds, I do not know and I daresay nobody does. But in the main, with most audio equipment out there, especially the absurdly expensive ones, I feel nauseous and sick and guilty...I hope I am not the only one to feel so.

Best regards
Rakesh



P.S.

Ha, I'm building exactly the same thing myself!  Those big horns are a LOT of work to turn, so in a way I am not surprised that it has taken several months so far for you to get them especially from an amateur builder (such as myself).  There is a mob in the US that make these for their horn systems and they allow three weeks construction for two workers...so one guy in a back shed is going to be much longer.

Hi ACG,

I would not call my current builder an amateur although this is indeed the first time he engages in a project of this magnitude. He has access to CNC machinery and my own attention to detail, so in the end, I think it will be all fine. Good luck with your horns.

Rakesh

[Nick]

Rakesh ,

Blarr blarr blarr....

...So what is my point? Sometime we audio people forget that in buying products from dealers/distributors/manufacturers, we forget that if you tear down all the hullabaloo about a change in settings that means your mother-in-law can shimmy away to "dance music" (what the hell is that anyways? - I thought a set of Bose speakers and two huge subs would do the job) in the kitchen, presumably this being the metaphysical g-spot sweet spot where the system is at its most resolving...you are left with real products which deliver day in day out, and if you run in difficulty, there is someone you can easily turn to.

I don't oftern look in on this thread, too much navel gazing, but.

Clearly my repsonse to your pontificating in the "Nicks Chronos" thread has got under your skin. So you decide to have a pop masked in a point about "real product".

If I were you I would be seriously concerned if the caps had malfunctioned in that way. Might the cheap cost of repare be an embarrassed manufacturer ? Risk of fire, sock etc ? Interesting that the quote was over the phone, makes me wonder how many of these failures your "real product" manufacturer might have has seen before.....

If the certainty of getting a low cost repare in decades to come for treasured and possibly by then out of date equipment is important, stay with your "real product values. I am for experiment, inovation, progress and service.

Next time you take a pop have some evidance to back it up. You have absolutely no idea what you are passing comment on so better not to.

Nick.


Ps try asking your favorite "real product" vendors for their designs and for them to post them publicly on line, in the interest of transparency of course. Those that are offering well worn designs might help you but if there is inovation you will be told politely to go bother someone else.

[rakeshpoorun]

LEAVING HOSTAGES TO FORTUNE

Hi Nick,

You do yourself a disservice by posting in this thread. In the earlier "Nick's Chronos" thread you refer to, you had explained your position as follows:

I want to keep this quite simple.

The Chronos is a product, so beyond a high level description of what it is and does I do not propose to enter detailed descriptions here of how it is designed and how it works.

...

This is a clear position I hope. If there is genuine interest in the Chronos I ask please that people to contact me directly.
...

Nick.

You are right and I agree. It is a very clear position. And I for for one was prepared to accept it and close that chapter, which is why I did not reply to your post there. I have more important things to do with my time. Your tone has now veered to the aggressive and you have lost your manners in launching a fairly personal tirade against me.

Irrespective of the right or wrong of your attitude on a site where the ethos seems to be to share, assist others and be as transparent and honest as possible in one's dealings with others, which people are free to make of what they will, your response means that I will feel free to take the liberty this time of making a couple of observations.

You seem to have changed your attitude markedly from the time when you were writing on Juan's long thread about how he built his audio pc where he went on at great length and in some detail into the decisions and choices that he made in finalising his design,  and shared with us all in a post that must have occupied him for weeks and months,  as he updated the contents of his audio pc as it evolved over time. This is what you had to say early on in that thread:

Hi all,



Here is the spec for my PC build. Thanks to Juan and Mani for their recommendations which helped to spec the build.
...

Best,
Nick.

Later on, you refer to your debt of gratitude for the work done by Peter, whose "towering presence" you eulogise in glowing and grateful terms.

What I wanted to say is that IMHO there is one pioneer who posts regularly on this site. This person is well known by us all for not just relying on subjective experiment to discover improvements but who always seeks to address fundamentally the theories in his own developments our others proposals. Were it not for Peter's approach I believe that computer music reproduction would not have moved on like it has in the last 5 years, at times it has been is dizzying listening to what has been achieved.

So Juan thanks for the kind words but I am just a slightly obsessive and compulsive type enjoying trying things out and sharing the results. Pioneer is a little strong in such towering company 

Kind regards,
Nick.

Still further on, we have you referring to the cases used and "discovered" by Juan:

 

Bob Juan hi,

I have two of the fractal design cases in the link from Juan above. One is used for my main audio pc. The are very well built and tick the boxes for an audio pc that Juan mentions. Very nice kit.

Regards,

Nick.

This is just one thread where there are several instances of your willingness to borrow from others and make use of the work that they have done.

I clearly remember reading a thread which was about how you approached Peter with your miraculous discovery of "Dexa" clocks, of which you were using two and waxing lyrical about until Peter used his equipment to show you how poorly they were performing. Using equipment that you do not have.

I am sure, Nick, that this is the tip of the iceberg. If one were to dig deeper, one would find, I have no doubt, innumerable such instances where you borrowed and asked others for help.

Forgive me, Nick, but when you say that you have a unique solution, I am not convinced. Your solution is almost certainly a combination of several techniques, tips and ideas that you have gleaned, like I have, from this very forum. Pretending otherwise would be a lie.

So now I have said it, the thing that bothered me about your attitude. There is something inherently hypocritical and revolting about the way you seem to have taken full advantage of the open and sharing ethos of this forum, and then violated it in the same breath.

Best regards
Rakesh

[Nick]

Rakesh

Again it is clear I touched a nerve, look at the length of your post and the effort you are going to provoke a response ! You crossed a line and picked the wrong person a tit for tat post with you is a waste of my time..

Read back a further in the forum and you will find that much of what we do here with pcs i have been very involved with. Hanging out here progress and problems are all taken in an air of good will. I  love the problems, if you ask "why" apparent failiure becomes progress. You need to read more of the posts your quotes are from.

My advice is take it down a notch, you will get much further and discover more.

Recycled stuff ? You would like to think so, but truth is there is real innovation.  Other people developing serious server products are expressing interest.

Nick.

[rakeshpoorun]

Rakesh

Again it is clear I touched a nerve, look at the length of your post and the effort you are going to provoke a response ! You crossed a line and picked the wrong person a tit for tat post with you is a waste of my time..

Read back a further in the forum and you will find that much of what we do here with pcs i have been very involved with. Hanging out here progress and problems are all taken in an air of good will. I  love the problems, if you ask "why" apparent failiure becomes progress. You need to read more of the posts your quotes are from  .

My advice is take it down a notch, you will get much further and discover more.

Recycled stuff ? You would like to think so  but truth is there is real innovation.  Other people developing serious server products are expressing interest.

Nick.

Hi Nick,

I have actually read the whole of Juan's thread, twice now, thanks to you!  So in a strange way, yes you have helped me on my journey to understand and learn more about the many design decisions that go into making an "audio pc." I will gladly take this whole affair down a notch as you suggest, although please do not mistake my willingness to be amicable as weakness, or an admission that I agree with you or your approach in the slightest.

I would say this though. In my opinion, fundamentally, your approach is wrong, because instead of proceeding in a holistic fashion based on a deep understanding of how this very complex and audio unfriendly environment that is the personal computer works, you approach the "audio pc" in a random, piecemeal manner, that will ultimately paper over the cracks and not solve the deep-seated issues. Unfortunately, I think that everyone here has no choice but adopt this haphasard method. It is clearly not unscientific because as with many fields of knowledge significant progress can be made in this way. But it is not ideal.

And I find your constant reference to "innovation" in hushed, reverent and mysterious tones, like it was some secret to be handed down as you join a cult at an initiation ceremony, really grating and at odds with the whole atmosphere of transparency and mutual help that runs through this site. 

I would speculate this. What you are talking about is one of two things. Less likely it has something to do with your earlier experimentation with trying to isolate the usb card. Alternatively, and more likely, is the addition of a master clock which slaves both the dac and the "audio pc." Speculatively, this could be done for instance by adding a sound card like the RME (I think) which is capable of accepting an external clock. I am very doubtful about this method working, coming from a non-engineer and amateur like yourself, without meaning to be disrespectful but merely stating the obvious. In the right hands, say if Peter were to look into this, I imagine this could bear some promise. Is it worth the cost, effort and hassle? Personally, I do not think so.

All in all then Nick, far from touching a nerve, you merely made me think and for that thanks.

About the capacitors, you said:

....
If I were you I would be seriously concerned if the caps had malfunctioned in that way. Might the cheap cost of repare be an embarrassed manufacturer ? Risk of fire, sock etc ? Interesting that the quote was over the phone, makes me wonder how many of these failures your "real product" manufacturer might have has seen before.....

Seriously? This is a a piece of equipment that has on occasions not been used for very lengthy durations and capacitors, even good quality ones, can fail over time. I am not seriously concerned and there is no suggestion online that Grace Design products are prone to failure. Their five year warranty actually attests that the opposite is more likely to be true (do you know many audio manufacturers who offer a 5-year warranty on ALL their products? I don't).

....

Ps try asking your favorite "real product" vendors for their designs and for them to post them publicly on line, in the interest of transparency of course. Those that are offering well worn designs might help you but if there is inovation you will be told politely to go bother someone else.

As it happens, in 2004, Michael Grace was asked questions online precisely in the same vein that you were asked, and this even before the Grace M902 was launched. Here is his response (you may benefit from comparing it to yours when I asked about your "product"):

Hi All,
I thought I would shed a little light on the m902. The m902 replaces the 901 in our lineup (although we still have a few 901s left!). It has the same headphone output amplifier as the 901 which is based on a high current, high speed transimpedance amplifier (AD815). This is where the similarities end though.
The first important difference is in the level attenuator. The m902 uses a digitally controlled analog attenuator which provides 95dB of attenuation with precision .5dB steps. I think there are several advantages to this type of control. First off there are no contacts to become contaminated, there is no switch capacitance that varies from setting to setting, and there is no wire or connectors to color the sound. Also, the digitally controlled attenuator has a very wide volume range which makes the m902 handle a variety of headphones from super sensitive types like the Shure E5 to the not-so-sensitive types like the HD600 or K1000. Anywhere in the entire range you can find the exact listening level with .5dB steps. (final firmware might switch to 1dB steps at attenuation levels of –85dB and below). The attenuator IC that had the best transparency and musicality also happened to have four channels so we hooked up the extra two channels to a line output so that you could use the m902 to control speakers or use it as a DAC.

Speaking of DACs, the m902 contains our “second generation” DACs which have some notable improvements over the DACs in the 901. While the 901 used a very high performance chip (CS43122) we chose the Burr Brown PCM1730 for the m902. This DAC is one of the few that has current outputs and allows us to use our own current to voltage converter. The current to voltage converter uses very high speed transimpedance amplifiers which assures that reconstruction of the analog signal is done without the non-linear slew rate limiting of regular op amps. While you won’t see the benefits of this type of topology in the measured performance I think it makes a significant improvement in the clarity and resolving power of the DAC. The m902 also employs a second PLL which we call s-Lock. The s-Lock PLL re-clocks the recovered clock from the digital inputs (including the USB) using super stable crystal oscillators and delivers a sample clock to the DAC with very low intrinsic (self generated) jitter and well over 100 times attenuation of incoming jitter from external sources. Call me old fashioned but I’m just not a believer in sample rate converters. While the 901 DAC is capable of fine performance with a low jitter source the clock recovery circuitry is not capable of attenuating incoming jitter so its performance degrades with long cables and jittery sources.

Like the 901, there are no electrolytic capacitors in the signal path. The inputs to the volume control are capacitor coupled with metalized polymer capacitors while the output amplifier is controlled with a DC servo. I think that a coupling capacitor sounds better than a servo since, when using a servo, you are not only listening to the integrator capacitor but you are also listening to the servo amplifier as its output is summed into the audio signal. However, in the case of the output amplifier we would need a very large value output capacitor to protect the headphones from DC offset so a servo is necessary here. (to keep the low frequency roll off below 10Hz while using 32 Ohm phones we would need around 5000uF which would only be available in an electrolytic type.)

Other enhancements to the m902 include:

All audio signal path resistors are 0.5% tolerance thin film surface mount types. Thin film resistors use the same type of metalic material used in thruough hole metal film types but have no lead inductance.

Improved capacitors in the output amplifier compensation network.

Larger power transformer with separate winding for digital circuitry. (the 901 derived digital supplies from the +analog supply)

Improved headphone jacks

4 layer pcb (this allows a continuous, low impedance ground plane and low inductance power supply distribution)

Separate power supply regulators for analog amplifiers, digital reciever PLL, DAC digital supply, DAC analog supply. 7 regulators in the m902!

Passive cross feed circuit.

Optional IR remote control. I wanted to be able to sit on my couch and have the m902 in my stereo cabinet so I could have it hooked into my loudspeaker system as a DAC…


Yup, there is more stuff in the m902 than in the 901…and it costs more. No cut corners. Just nice rounded ones on the outside.

Cheers,
Michael


A little further on, he posts some additional measurements for his product:

Hi all,
Here are the results of the USB jitter and bit tests. Please let me
know if proper etiquette would be to start a new thread with this sort of
technical stuff.

The Mac Power Book (OSX) will play 44.1kHz and 48kHz files over the usb
interface to the m902 with perfect bit accuracy. Again, as I mentioned
in a previous post, you must open the MIDI/Audio setup panel and set the
output sample rate to match the file being played or else sample rate
conversion occurs. (We will have these details in the owners manual)

SACD:
As far as I know there are no SACD players that have digital output when playing SACD disks with the possible exception of units with the I-link port. I think this is a firewire-like interface but the data is encrypted and (correct me if I'm wrong) asynchronous which would provide for jitter prone transfers. In the professional world DSD equipment uses SDIF-3 which consists of separate clock and data signals on BNC connectors. I think EMM Labs makes a modification for a SACD player so that it will output DSD data to a DSD dac...

USB jitter:
Measured at the spdif output of the PCM2902 the recovered clock from the PCM2902 measures about 240 pico seconds of RMS jitter when playing a 44.1kHz file and 390 pico seconds when playing a 48kHz file. Note that the 44.1kHz performance is very near to the noise floor of the Terrasonde Digital Audio Toolbox which is about 200 Pico Seconds when measuring spdif data streams. To really know the jitter of the PCM2902 would require access to the sample clock inside the chip. Oh well. However, this is not shabby jitter performance. I was expecting worse. With a good secondary PLL (like the s-Lock circuit) this can be reduced to less than 40 pico seconds.

If there was any confusion earlier in this thread, the usb interface is limited to a *maximum* sample rate of 48kHz and 16bit word length. 44.1kHz works fine. While there are USB solutions that support higher sample rates and bit depths, they are marginaly reliable. For higher sample rates and bit depths the only viable solution is IEEE 1394 (firewire) as it is designed for audio and video streaming. USB 2.0 will probably not become a popular audio transport for its lack of clear standards for audio streaming.


On another note a friend of mine who is a Linux guru is working on determining the requirements for making the m902 work with Linux.
I'll report as soon as we know anything.

Michael

So Nick, there you are. This is what happens when you ask a professional details of the product that the potential customer is considering. Very different to your response, quite clearly. Nobody asked you for your schematics (not that I think you have any because you merely assemble a bunch of parts), just a simple explanation of what your product is, what it does, how much it costs, why you think it represents an improvement on what's available elsewhere.

On the other hand, you will have noticed that this is very much the approach adopted by Peter. He is very forthcoming with what he is trying to achieve, the technical hurdles that he faces, and what solutions he has adopted. This is very different to what you have offered here.

So I rest my case.

I am not interested in your "product," Nick, and am unlikely to ever be (you will be relieved to hear). To end, if I may be allowed to quote myself:

...
There is clearly a sense of honesty and integrity that permeates the forum generally and this is also why I was prepared to take a blind punt (I knew nothing about the Phasure Dac until a friend told me about it and I read some of Peter's postings and thoughts here). For me, our hobby cannot be dissociated from the people behind the products that find their place in our audio equipment rack.
...

Best regards
Rakesh

[manisandher]

A very quick post - just no time...

Rakesh, IMO you are VERY wrong about Nick and his product. Nick makes some fundamental changes to the mobo itself, which in my own experience make a big difference to the SQ.

But I understand your (and Peter's) frustration with the way things are being communicated on this forum. I can't speak for Nick here, but I think he has his reasons for wanting to stay vague about exactly what he does to the mobo. I mean, just ask Peter what happened with M2Tech a few years and you'll understand why Nick might be a bit cautious.

Paul Pang is the only other person I know who is approaching things in a similar (though different) way to Nick.

HTH.

Mani