Jump to content

Chord Electronics Owners & Discussion Thread


Recommended Posts

5 hours ago, legend said:

In an exchange on their web site soon after they added sox in 2016 Jriver would not discuss its implementation saying that they were not an 'audiophile' site.

I'd guess from that the question wasn't asked in an objective way.     Something like, can you please tell me what parameters you pass to SoX? ...  wouldn't have got such a response   ;) 

52 minutes ago, Music2496 said:

Yes, again... the analogue section designs are different, right?

AFAIUI we don't have exact information, they are both of very similar design.    An op-amp, connected to a discrete transistor amplifier in a nested feedback loop.    This may have oversimplified it some ..... but CH is really talking about a different issue  (ie. what amplifier circuit to connect a typical DAC IC to..... datasheet, or something else, and some DA ICs are very very fussy/complex).

Quote

With discrete circuitry, the only limit is your imagination. You are free to adjust the topology of the circuit, the brands of the parts, the active devices, the bias current in each stage - anything you can think of.

 

1 hour ago, Music2496 said:

Qutest and Dave vary here too...

Aside from Dave having mains conversion inside .... do you have a link handy as to how?

1 hour ago, Music2496 said:

The Charles Hansen post is well worth a read and re-read every now and then....

Chord avoid/address many/most of the issues he raises....  The post is really targeted towards "designing for a typical DA IC", where the power supplies required often need to be more complex/separate and stiffer than people have been warned, and the amplifier recommended in the datasheet isn't the greatest.

1 hour ago, Music2496 said:

as he says in the post, designed his own digital filters.

Quite a different outlook to Chord :) .

2 hours ago, legend said:

am not sure how much difference it will make to SQ

My understanding is the real important differences are in the DSP, and there is less 'pulse array' elements.

2 hours ago, legend said:

and whether with their FGPA technology they really do upsample again after the M-Scaler's best output and for what benefit?

I tried to cover this in my previous post, but perhaps it was too cryptic.    Yes, they really do upsample (further).    Why?  Because they must - it is how the DAC functions.

 

The DAC is an array of 5 bits DACs runnin at 104Mhz ... and so obviously the data needs to be interpolated (resampled, upsampled) get to that speed.

 

Incoming signals are resampled to 16fs (eg. 705.6 or 768khz).   If you input a 705.6/768khz signal, then nothing happens here.

 

There are two additional stages of 'WTA' interpolation until the data is 2048fs .... then 5bit 104mhz modulation happens, and the analogue signal is reborn.

Link to comment
Share on other sites



9 hours ago, davewantsmoore said:

 

I tried to cover this in my previous post, but perhaps it was too cryptic.    Yes, they really do upsample (further).    Why?  Because they must - it is how the DAC functions.

 

The DAC is an array of 5 bits DACs runnin at 104Mhz ... and so obviously the data needs to be interpolated (resampled, upsampled) get to that speed.

 

Incoming signals are resampled to 16fs (eg. 705.6 or 768khz).   If you input a 705.6/768khz signal, then nothing happens here.

 

There are two additional stages of 'WTA' interpolation until the data is 2048fs .... then 5bit 104mhz modulation happens, and the analogue signal is reborn.

Thanks Dave I am now truly enlightened - as a loudspeaker designer you can't assume too much digital knowledge with me? Presumably the interpolation in the two additional stages is at such a high rate/fine detail that the shape of the 'impulse'  which the M-Scaler has taken so much trouble to perfect is little affected by any errors in these later stages?

Edited by legend
Link to comment
Share on other sites

10 hours ago, davewantsmoore said:

I'd guess from that the question wasn't asked in an objective way.     Something like, can you please tell me what parameters you pass to SoX? ...  wouldn't have got such a response  ;)

 

Actually the guy did ask Jriver in an objective way - he wanted to know if/how he could vary the phase (linear, minimum etc) in their sox upsampling.  But Jriver basically said they weren't interested as they were not an audiophile company - which is one reason I have not upgraded my Jriver subscription from MC 23.

  • Like 1
Link to comment
Share on other sites

4 hours ago, legend said:

as a loudspeaker designer you can't assume too much digital knowledge with me?

I am in the same boat.    I know enough to understand what others are doing/why and follow along.

4 hours ago, legend said:

 Presumably the interpolation in the two additional stages is at such a high rate/fine detail that the shape of the 'impulse'  which the M-Scaler has taken so much trouble to perfect is little affected by any errors in these later stages?

Yes....   to close the loop, let's say you used a completely different DAC.   It is possible that in order to improve the analogue output, that you might need to alter the digital data you input in a different way.

 

3 hours ago, legend said:

Actually the guy did ask Jriver in an objective way - he wanted to know if/how he could vary the phase (linear, minimum etc) in their sox upsampling.  But Jriver basically said they weren't interested as they were not an audiophile company - which is one reason I have not upgraded my Jriver subscription from MC 23.

If we treat the DA converter as a "distorting black box" .... then it depends on the DA converters imperfections, as to how we might want to massage the data going in, to get a more perfect output.

I feel for these guys, as I think they often don't communicate very clearly.    What they mean is that they want to spend their limited time budget on "media centre features".

 

Their attitude is that if you want to resample your audio .... Then do that.   Pipe the output of Jriver into your resampler and do whatever you want.

3 hours ago, legend said:

But Jriver basically said they weren't interested as they were not an audiophile company - which is one reason I have not upgraded my Jriver subscription from MC 23.

 

It was only a couple of posts into the SoX page on JRMC where they confirmed there would be no configuration options... and that they use -v -s  (very high quality, steep, linear phase).

 

You could say these are the default/recommended options from SoX themselves .... although in reality SoX don't recommend these settings because "they are overkill".

 

 

Link to comment
Share on other sites

3 hours ago, legend said:

Actually the guy did ask Jriver in an objective way - he wanted to know if/how he could vary the phase (linear, minimum etc) in their sox upsampling.

Ahhh, right I see.

 

JRMC:  Here is SoX, there'll be no configuration options

User:   Can I have configuration options for SoX

JRMC:  No :thumb:

 

 

There might be a way through this, using https://sox.fandom.com/wiki/OPTIONS  ;) .... or as mentioned, just go outside the box (tell JRMC to play to a virtual audio device, and massge the output elsewhere)

 

 

Link to comment
Share on other sites



18 minutes ago, davewantsmoore said:

It was only a couple of posts into the SoX page on JRMC where they confirmed there would be no configuration options... and that they use -v -s  (very high quality, steep, linear phase).

 

You could say these are the default/recommended options from SoX themselves .... although in reality SoX don't recommend these settings because "they are overkill".

 

 

I think that the M-Scaler and @Ittaku's work on extreme upscaling have subsequently shown that "they are overkill" to be incorrect - though may have been true when SoX was written.

 

But I guess JRiver have the right to decide what niche market they are in rather than try to satisfy everybody!

  • Like 1
Link to comment
Share on other sites

5 minutes ago, legend said:

I think that the M-Scaler and @Ittaku's work on extreme upscaling have subsequently shown that "they are overkill" to be incorrect - though may have been true when SoX was written.

Agreed. I think that although I'm not a fan of the sound of Chord's DACs per se, with their upscaling, I'm quite sure that he was on the right track to try and approximate a perfect sinc function. All other filtering approaches to date sound "different", where sometimes linear sounds good and sometimes intermediate and sometimes minimum sounds good; with a super long sinc function, linear sounds better with everything I've tried on.

Link to comment
Share on other sites

 

16 minutes ago, legend said:

though may have been true when SoX was written.

I don't really see what has changed....  but certainly, these options have been tuned for filter lengths which are much shorter than typical.

 

Unless you were going to run longer filters (which JRiver isn't) then their performance seems OK.

 

16 minutes ago, legend said:

But I guess JRiver have the right to decide what niche market they are in rather than try to satisfy everybody!

This seems to infer that "audiophile options" ... and "media centre software features" are both niche markets.  ;) 

 

You only have to look at their support forums to see how much noise "resampling" (and things like it) causes them.

 

The simple fact is that this (being a flexible resampler) doesn't need to be integrated inside their software.... so why would they?  (cost/benefit)  :) 

Link to comment
Share on other sites

16 minutes ago, davewantsmoore said:

 

This seems to infer that "audiophile options" ... and "media centre software features" are both niche markets.  ;) 

 

You only have to look at their support forums to see how much noise "resampling" (and things like it) causes them.

 

The simple fact is that this (being a flexible resampler) doesn't need to be integrated inside their software.... so why would they?  (cost/benefit)  :) 

We probably agree that one of the hardest things to do in strategic marketing is, given a company's non-infinite resources, to consciously decide who NOT to try to satisfy as much as who to try to satisfy.

 

I think JRiver have decided to try to satisfy the 'media centre' market/niche while HQplayer and Amarra etc try to satisfy the 'audiophile' market/niche - and I have no problems with that given I can make a choice.

 

Likewise I think Chord with their M-Scaler have largely tried to currently satisfy their Chord DAC market with the dual BNC connections etc (of which I am one) as well as audiophiles with deep pockets (of which I am not one) - though this may change if/when this market saturates and/or other competitors enter the market.

  • Like 1
Link to comment
Share on other sites

6 hours ago, legend said:

Thanks Dave I am now truly enlightened - as a loudspeaker designer

It's a pity we never got together before you crossed over to the big island....  as this is what I'd much prefer to ruminate on, and I'm sure the enlightening would be 2-way.   :) 

Link to comment
Share on other sites



I have to wait another 12 weeks to get an MScaler, but in the meantime Ive been experimenting with SOX and I like what I hear upsampling to 16s through USB into my Qutest!  Cant wait to hear the MScaler.

 

For anyone using LMS Server, there is a plugin called C-3PO which has a GUI and presets to SOX which makes experimenting with it very easy. The plugin is free and should be listed in your 3rd Party Plugins list in the Server Settings.

Link to comment
Share on other sites

Interested to know what sort of percentage in cost a device like the m-scaler is of the total system investment around theses parts.

 

From what I can gather it's not going to make strawberry jam out of rocking horse shite so surely it should cost less than the individual cost of the source, the dac, the pre-amp, the amp and the speakers to be seen as good value to the owner?

 

Or does it create some sort of magic where you put a bunch of above average components together, you add an m-scaler and all of a sudden your system is complete?

  • Like 1
Link to comment
Share on other sites

54 minutes ago, Hi-Fi Whipped said:

Interested to know what sort of percentage in cost a device like the m-scaler is of the total system investment around theses parts.

 

From what I can gather it's not going to make strawberry jam out of rocking horse shite so surely it should cost less than the individual cost of the source, the dac, the pre-amp, the amp and the speakers to be seen as good value to the owner?

 

Or does it create some sort of magic where you put a bunch of above average components together, you add an m-scaler and all of a sudden your system is complete?

The DAC still remains the most important part of giving your digits a sonic signature. I'd seriously only consider the m-scaler if my DAC was a Chord DAC with the proprietary dual BNC input, otherwise you won't even be using it to its full capacity, and also only if most of my source material was 16/44.

Edited by Ittaku
  • Like 2
Link to comment
Share on other sites

9 hours ago, Hi-Fi Whipped said:

Interested to know what sort of percentage in cost a device like the m-scaler is of the total system investment around theses parts.

 

From what I can gather it's not going to make strawberry jam out of rocking horse shite so surely it should cost less than the individual cost of the source, the dac, the pre-amp, the amp and the speakers to be seen as good value to the owner?

 

Or does it create some sort of magic where you put a bunch of above average components together, you add an m-scaler and all of a sudden your system is complete?

For me the approximate costs in AU$ as SRPs would be (accumulated over a long time and actually less for me being in the industry):

 

music source (computer or CD player)  $1000 4%

Chord M-Scaler  $7500   30%

Chord Qutest DAC   $2500  10%

Sony preamp $1000   4%

DEQX Ncore biamp  $4000   16%

Kantu 10Be speakers $8000    32%

cables  $1000   4%

TOTAL  25000  100%

 

So you can see the M-Scaler is a large % of the total and much more than I would normally want to spend on a source component - and why I have spent time over the past weeks researching upsampling alternatives.  Because when Sime brought his M-Scaler briefly into my system it did 'magically' improve the SQ of the system to a point where I would be happy just to listen to music with this system - which, although not chump change overall, is about as good as I have heard a passive loudspeaker system.  But I would strongly agree with @Ittaku that one only gets the full potential of the M-Scaler with a Chord DAC.

 

Edited by legend
typo
  • Like 1
Link to comment
Share on other sites



Another rave review (not sure if it has been listed here before):

https://audiophilestyle.com/ca/reviews/my-quest-for-a-new-dac-part-5-chord-electronics-hugo-m-scaler-hugo-tt-2-r781/

 

"What would this curious device, this culmination of Watts’ and Chord Electronics’ decades-long quest for “a million taps,” deliver? To find out, I placed it upstream of the TT2, connected by dual BNC cables, driven by my chain using USB, and sat down to listen. 
 

And couldn’t stop listening. I’d start a track and listen to the entire album."

 

"I’ve really had to think hard about how to describe the effect of the HMS. While it dramatically improved conventional areas of sound quality: tonality, extension, resolution of micro- and macro-details, and imaging, it also did more - almost along another axis of improvement I hadn’t registered before.
 
The best way I can describe it is temporal coherence. There is just something “right” about the sound in terms of its timing. Take dense orchestral music like the Mahler. Normally, with music like that when different instruments come to the foreground in terms of volume, one’s ear is drawn to them, and the others recede from attention. It’s akin to surfing - riding on the crest of each wave. Forgive the mangled metaphor - I don’t surf! What I found with the HMS is that as an instrument receded from its crest, it remained perfectly easy to follow. As a result, I found myself able to keep my attention on what instruments were doing off their peaks. On dense music, this is a revelation! It draws you so much deeper into the piece, both intellectually and emotionally. "

  • Like 1
Link to comment
Share on other sites

9 hours ago, Ittaku said:

and also only if most of my source material was 16/44

This is an interesting comment so are you saying it would be better to invest in Hi Res material such as https://www.onkyomusic.com/GB/mqa-rock these are all pretty reasonable prices.

I suppose it just common sense, how do you upsample something that is already at 24/352.8

 

Link to comment
Share on other sites

10 minutes ago, powerav said:

This is an interesting comment so are you saying it would be better to invest in Hi Res material such as https://www.onkyomusic.com/GB/mqa-rock these are all pretty reasonable prices.

I suppose it just common sense, how do you upsample something that is already at 24/352.8

I wouldn't be investing in MQA per se (I still have objections to MQA since it's also tying you into their filter which is also an imperfect compromise of sorts), but I always buy hires when I can. Upscaling something that's already 24/96 or 24/88 has only the smallest of gains with the mscaler. It's my opinion, confirmed by listening tests, that the most important step is getting away from 44 or 48 where the existing filters have effects in the audible range.

  • Like 2
Link to comment
Share on other sites

2 hours ago, powerav said:

This is an interesting comment so are you saying it would be better to invest in Hi Res material such as https://www.onkyomusic.com/GB/mqa-rock these are all pretty reasonable prices.

I suppose it just common sense, how do you upsample something that is already at 24/352.8

 

The Mahler 2 recording of the audiophilestyle review I quoted above is available in 96/24 (I have it) and the reviewer seemed to think the M-Scaler made a very significant difference to its reproduction - though it may be because he was also using a Chord DAC.

Link to comment
Share on other sites

I have loads and loads of Hi Rez. The Scaler with 44.1 trumps anything a HiRez file can achieve. It’s not always simply about the data rate. 

Link to comment
Share on other sites



1 hour ago, Sime V2 said:

I have loads and loads of Hi Rez. The Scaler with 44.1 trumps anything a HiRez file can achieve. It’s not always simply about the data rate. 

Well this suggests the upscaler is doing something that highres is not, even though the whole point of the upscaler is to turn regular resolution data into ultra high resolution data. Are you suggesting the mscaler somehow makes things sound better at equivalent resolution than music originally sampled at that resolution, or is it simply because you can push to 700+ kHz whilst highres files are lower?

Link to comment
Share on other sites

1 hour ago, Sime V2 said:

I have loads and loads of Hi Rez. The Scaler with 44.1 trumps anything a HiRez file can achieve. It’s not always simply about the data rate. 

And are you saying that a 44.1 with M-Scaler also trumps a hires file with M-Scaler - or just that a 44.1 with M-Scaler trumps a hires file without M-Scaler?

Link to comment
Share on other sites

7 hours ago, Ittaku said:

Well this suggests the upscaler is doing something that highres is not

I don't necessarily find that to be particularly controversial.

 

 

"highres" tells us nothing reliable about the audio.  The fact that you got the audio delivered to you as 24/192 or whatever, tells nothing reliable.   In an ideal world .... if we had identical audio formatted at 48khz and 192khz .... and we played them back, and then truncated everything above 20khz from the analogue output - they would still be identical.

 

We also have to be extremely careful we are not comparing the "high res" re-release, to the "original CD" as they are very often not the same content (CD can be superior).

 

 

If CD rates vs higher rates sound the same  (when encoding the same content) ..... then CD, which has been resampled (improved*) could sound better than a higher rate encode.

 

* Not all resampling should be expected to "improve".    It' very contingent on the specifics.

  • Like 1
Link to comment
Share on other sites

8 hours ago, davewantsmoore said:

I don't necessarily find that to be particularly controversial.

 

 

"highres" tells us nothing reliable about the audio.  The fact that you got the audio delivered to you as 24/192 or whatever, tells nothing reliable.   In an ideal world .... if we had identical audio formatted at 48khz and 192khz .... and we played them back, and then truncated everything above 20khz from the analogue output - they would still be identical.

 

We also have to be extremely careful we are not comparing the "high res" re-release, to the "original CD" as they are very often not the same content (CD can be superior).

 

 

If CD rates vs higher rates sound the same  (when encoding the same content) ..... then CD, which has been resampled (improved*) could sound better than a higher rate encode.

 

* Not all resampling should be expected to "improve".    It' very contingent on the specifics.

I agree that one has to be vary careful about the term 'hires'.  It is used by record companies to sell more copies but often just means a 44.1 version upsampled to 96/24 or higher with dubious upsampling pedigree.  Mark Waldrop of Aix records has a nice analogy for it - it is like just transferring a dinner to a bigger plate, sometimes badly.

 

However some 'hires' are genuine as was the Mahler 2 (recently recorded at 96/24) that the reviewer in the audiophilestyle article found was significantly improved by the M-Scaler.

 

I guess the only way to answer my question  "does 44.1 sound better than hires both using the M-Scaler"  would be to compare with a case of true hires downsampled to 44.1.  I have plenty of such cases on SACDs (which have hires and CD layers) - but not an M-Scaler ?

  • Like 1
Link to comment
Share on other sites



  • Recently Browsing   0 members

    • No registered users viewing this page.




×
×
  • Create New...
To Top