Metrum Octave Dac - What are the Chips used

[ThorstenL]

Hi,

Originally Posted by SoNic_real_one

In my polar world I saw engineers bad of good. Doctors bad and good.

In my polar world, your comments on a range of topics here seriously disqualify from being any kind of judge of engineers. In most cases the comments are merely repetitions of marketing claims and folklore, devoid of supporting evidence, combined with fundamental misunderstandings of specifications and their meaning, as well as possible impact they may (or may not) have on sound quality.

In fact, I would suggest that in future you try to avoid technical arguments regarding topics you do not understand at least as thoroughly as the person your are trying to debate with and stick to "I like the way it sounds and that is all I need", which I will happily accept as both honest and true, even if I do not your share your preferences.

I cannot speak of your qualifications to judge doctors. In the interest of your continued health and existence I most sincerely wish for you that they may be significantly better than the first one...

Ciao T

PS, I have not seen any further comments on the 10dB more HF noise than signal measurements I showed for SACD. Considering how adamant you where on the point that there was no such noise, one may at least expect a public correction of your position, in light of the evidence presented...

[ThorstenL]

Hi,

Originally Posted by Ken Newton

Yes, managing an NCO, for example, to match the exact long term average frequency of the source oscillator, but at the DAC, is probably the best, but also sophisticated, solution.

I helped AMR to implement precisely this for their latest product. It is both harder than it sounds at first and much easier than it may seem after some investigation.

Originally Posted by Ken Newton

Such an implementation is, unfortunately, probably not an option for the more advanced diy audiophile.

This may be, but if we cannot fix the problem completely (be it by using asynchronous USB or suitable clock management for SPDIF), one could at least use the receiver with the best source jitter attenuation available...

Instead I routinely see people selecting the ones that for all intents and purposes have zero audio band source jitter rejection...

Ciao T

[ThorstenL]

Hi,

Originally Posted by Ken Newton

he's unlikely be an expert at all facets of product design, development, and production. Too much to expect of one person, I should think.

Why?

The device under discussion is quite primitive in nature.

One single person most certainly can hold all required competencies. It is perhaps not all that common, but by far in the realm of what I see every day.

Ciao T

[Ken Newton]

Originally Posted by ThorstenL

Hi,



Why?

The device under discussion is quite primitive in nature.

One single person most certainly can hold all required competencies. It is perhaps not all that common, but by far in the realm of what I see every day.

Ciao T

I've not seen the Octave PCB layout, and so have no opinion regarding it's board nor the PCB layout expertise of the person who designed it. I was merely suggesting a reasonable explanation for why the PCB might evidence some suspect ground scheme, if in fact it does, which abraxilito objects to.

While universal competency is a much different thing than universal expertise, in my view, kudos go to anyone who gets a product developed and to market succesfully, essentially on their own. Which apparently includes the founder of Metrum.

[jitter]

The person in question seems to hold more than just one competency. From the website of Acelec Engineering comes the following quote that reads a bit like a resumé:

Acoustics, under water acoustics, robotics, marine electronics, hifi electronics, industrial electronics, consultancy.

There are some really specialist fields of expertise there. No doubt contacts in the industry helped him to get a DAC like the Octave manufactured professionally.
When it comes to the design of the Octave, design decisions must have been made with the intended selling price in mind, hence the choice for a $2 transformer and not one of double digits. That does leave the opportunity of improvements to be made for us diyers, though...

Somewhere I read Cees Ruijtenberg is designing a new model. I wouldn't be surprised if that took the good bits of the Octave and added some features like more digital inputs and sampling rates up to 192 kHz. Time will tell.

The Hifi Critic review says that the occasional Octave will work at 192 kHz, but the review model didn't. Mine also doesn't. What is the limiting factor for this, the SPDIF receiver or the transformer?

[pieter t]

Originally Posted by jitter

The Hifi Critic review says that the occasional Octave will work at 192 kHz, but the review model didn't. Mine also doesn't. What is the limiting factor for this, the SPDIF receiver or the transformer?

The transformer will not be the limiting factor; it should have ample bandwidth reserve.
I had a discussion with Cees when I experimented with the input circuitry; he confirmed that the pulse transformer was there to prevent ground loops.
At that time I had a critical look at the digital interface between my Rotel CDP and the Octave, because several components were redundant: the interface held two pulse transformers, and more DC blocking capacitors than necessary.
I cleaned up by removing the crappy Rotel transformer, removing the redundant blocking capacitors, and replacing the Murata transformer of the Octave by my own pulse transformer (wound on a tiny cobalt amorphous alloy toroidal core).
SPDIF signals looked clean now with much better rise time, and sound quality of this combo became on par with the better quality Kenwood CD drive/Octave I also had on hand.

[gaetan8888]

Originally Posted by SoNic_real_one

Not if they are properly filtered. That proper filtering requires OS too...
I can't deal with NOS without serious filters. Because once you get used to an bad/unatural sound, even the real instruments will sound "bad".

Hello

How much DB of low pass are serious filters for you ?

Bye

Gaetan

[abraxalito]

Originally Posted by jitter

The Hifi Critic review says that the occasional Octave will work at 192 kHz, but the review model didn't. Mine also doesn't. What is the limiting factor for this, the SPDIF receiver or the transformer?

DIR9001 is only guaranteed up to 108kHz so it will depend on the particular chip's characteristics.

[jitter]

In that case I don't understand why the Octave is guaranteed to work up to 176.4 kHz. Is there an upgraded version of the DIR9001 or a successor that will do 176.4 kHz (guaranteed)?

[abraxalito]

Could be that they're production tested at 176k. I'm not aware of anything in that same SO28 package which locks at 176k guaranteed. TI has parts which go up to 216k but they're in 48pin flat packs.