CHORD M-Scaler Review (Upsampler)

[deniall83]

Legitimate reason is that they confuse one measurement for another when they are either different or presented differently. Earlier this claim was made that stereophile measurements showed 120 dB of attenuation while mine showed 80 to 90 dB. I pointed out that the stereophile showed the same as their "zero" reference as -40. So the differential was 80 dB. This happens fair bit because people just go by the punchline text in the measurements and not look in detail.

The bigger reason is that they don't want to believe the results. So they resort to talking points and other nonsense. They need to be reminded that I have tested around 1,200 products. If measurements were wrong, you would see a line mile long from manufacturers saying the measurements are wrong. We don't see this because it simply is not the case. While mistakes can happen from time to time, they simply are not routine. And at any rate, could easily be shown with identical setup and explanation which almost never happens.

Bottom line, listen to manufacturer providing own measurement with different outcome. This is something I welcome in every review. If this doesn't happen, then that is it. Folks complaining because they have websites/youtube channels doesn't count.

Thanks for your response Amir. I had assumed that was the case.

 

[Deleted member 19122]

Just curious does Rob Watts the engineering guru of Chord actually hold an engineering degree in a relevant field?A cursory glance of the internet doesn't seem to list any qualification and a recent online forum that he was on he stated he studied for and EE when he was 18 in the UK but nowhere does he claim he graduated or holds a certification.Either way that might explain a lot

 

[amirm]

I do wish @amirm would remeasure the Topping 70 for the weird non-harmonic result. If it is the same, I personally would be curious about trying to find out why. Even if he is not, confirmation something wasn't amiss seems in order.

The results were so odd that I had already repeated them for the review. But here you go again.

44.1 kHz:

The spikes on the right were moving up and down so I tried grounding and made no difference. I switched to Toslink to break electrical connection to the analyzer but results are the same. But watch what happens when I change sampling rate to 48 kHz:

The pattern is odd but different. Something is really broken. Here is the D70s with 48 khz sampling over same Coax connection (but now fed from APx555):

Cleary there is a problem here. I need to move on to testing other gear. Chord really needs to investigate this.

 

[abdo123]

Let me ask a silly question.
As far as I understand that thing connects to dac through spidf,right?
What kind of spidf plays more than 24/192 so it can receive the X16 resampling?
All the dacs I know only play 24/192 through spidf.

It’s using 2x BNC connectors that are only available on Chord DACs.

 

[Blumlein 88]

The results were so odd that I had already repeated them for the review. But here you go again.

44.1 kHz:

View attachment 217035

The spikes on the right were moving up and down so I tried grounding and made no difference. I switched to Toslink to break electrical connection to the analyzer but results are the same. But watch what happens when I change sampling rate to 48 kHz:

View attachment 217036

The pattern is odd but different. Something is really broken. Here is the D70s with 48 khz sampling over same Coax connection (but now fed from APx555):

View attachment 217038

Cleary there is a problem here. I need to move on to testing other gear. Chord really needs to investigate this.

Thanks for humoring us with the double check. That is very curious. Could it be inadequate dithering that varies with time?

 

[bennetng]

The results were so odd that I had already repeated them for the review. But here you go again.

44.1 kHz:

View attachment 217035

The spikes on the right were moving up and down so I tried grounding and made no difference. I switched to Toslink to break electrical connection to the analyzer but results are the same. But watch what happens when I change sampling rate to 48 kHz:

View attachment 217036

The pattern is odd but different. Something is really broken. Here is the D70s with 48 khz sampling over same Coax connection (but now fed from APx555):

View attachment 217038

Cleary there is a problem here. I need to move on to testing other gear. Chord really needs to investigate this.

Thanks. Could you show the linearity plots with this signal chain like what you did in other DAC tests?

 

[amirm]

Thanks. Could you show the linearity plots with this signal chain like what you did in other DAC tests?

I just took apart the setup. I need to test other stuff folks. This is for CHORD to debug....

 

[solderdude]

Seemed to work fine on the intended Chord DAC(s).
Seems to be an M-scaler / Topping D70 combination issue. As we don't know how it 'pairs' with other DACs we won't know.
All we know is it seems to work fine on Chord DACs and there is 'weird behavior' with the D70.
Chances are Rob W won't care about this, but he should.

 

[Jimbob54]

Seemed to work fine on the intended Chord DAC(s).
Seems to be an M-scaler / Topping D70 combination issue. As we don't know how it 'pairs' with other DACs we won't know.
All we know is it seems to work fine on Chord DACs and there is 'weird behavior' with the D70.
Chances are Rob W won't care about this, but he should.

Obviously nothing remotely conclusive here but it would be ironic if the MScaler connected by anything other than the dual BNC to any other DAC made things worse across the board.

But then, I wonder now many non DAVE/ Hugo TT DAC owners have bought this $6k magic box? Not many I would guess, even less again that use it on the likes of the Toppings of the world.

The real irony though is that the DACS this is "designed"/ intended for are already touted as having the most accurate filters anyway.

 

[mediahound]

All I know is I like the M Scaler. It upscales all my 44.1 CDs to high res. Plus all my streaming. The DAC is getting 768k out of the M Scaler and the audio improvement is there, albeit it may be subtle depending on the recording but I dig it. The M Scaler is akin to the Darbee video upscaler if anyone is familiar with that. It may not actually add resolution, but it sure looks (sounds) like it does.

I don't use it with a Topping DAC though, I use it with the Chord TT2.

 

[solderdude]

The real irony though is that the DACS this is "designed"/ intended for are already touted as having the most accurate filters anyway.

Yep.. but when one buys Chord they want even that 'improved' and if the guru says it does just that they will buy AND hear improvements.

 

[amirm]

All I know is I like the M Scaler. It upscales all my 44.1 CDs to high res.

It doesn't do that at all. It is like you saying if you change $1 into four quarters, you get more money! You do not. What has been discarded at production of music at 44.1 KHz, is gone forever. Remember, there is no intelligence in M-scaler. It is a simple math machine. Don't be confused with bigger numbers having more information. They simply cannot.

he M Scaler is akin to the Darbee video upscaler if anyone is familiar with that. It may not actually add resolution, but it sure looks (sounds) like it does.

No it is not. Darbee actually changed the video signal. The M-scaler does not do that.

 

[bennetng]

Chances are Rob W won't care about this

...and his believers.

 

[Jimbob54]

All I know is I like the M Scaler. It upscales all my 44.1 CDs to high res. Plus all my streaming. The DAC is getting 768k out of the M Scaler and the audio improvement is there, albeit it may be subtle depending on the recording but I dig it. The M Scaler is akin to the Darbee video upscaler if anyone is familiar with that. It may not actually add resolution, but it sure looks (sounds) like it does.

I don't use it with a Topping DAC though, I use it with the Chord TT2.

Now close your eyes.....

 

[Jimbob54]

...and his believers.

He clearly doesnt care much about them either ;-)

Well- he cares about one particular attribute of them

 

[bennetng]

one particular attribute

$ £€...

 

[Sokel]

It’s using 2x BNC connectors that are only available on Chord DACs.

Yes,the experts here already search it in depth

 

[theREALdotnet]

Seemed to work fine on the intended Chord DAC(s).
Seems to be an M-scaler / Topping D70 combination issue. As we don't know how it 'pairs' with other DACs we won't know.
All we know is it seems to work fine on Chord DACs and there is 'weird behavior' with the D70.
Chances are Rob W won't care about this, but he should.

Could this be an impedance mismatch (50 ohm M Scaler output into 75 ohm D70)? Perhaps the Hugo inputs are actually 50 ohm, too?

I would have never expected this to cause measurable distortion, perhaps a glitchy connection. But I’m not a user of coax S/PDIF myself, so I don’t know…

 

[garbz]

4- I'm not going to conclude that Rob Watts is a fraud.

Fraud is the wrong word. Someone produces something that customers want to buy. Whether or not it makes a difference is a problem for the person who requested it. Upscalers are a legit thing, and while it may not perform well that doesn't make it fraud. Additionally cost is related to economies of scale so if you produce something so niche it comes at a cost.

Waste of money? Yes. Fraud? No.

 

[voodooless]

Fraud is the wrong word. Someone produces something that customers want to buy. Whether or not it makes a difference is a problem for the person who requested it.

What is your definition of fraud? When is it okay to sell things that don't work? When is it not?