CHORD M-Scaler Review (Upsampler)

[Deleted member 19122]

Hey at least he's paying for Robb's vacations, according to his own recent posts on head fi,he's off to Disney World Florida with the family and then can't make London CanJam due to a family cruise around Norway.The volume of posts and time he spends there to me is a clue of exactly what Chord is about and why the suckers he massages over there come here to try and defend this nonsense product without even a hint of proof.It's snake oil being hustled to gullible people with that calculated friendly salesman pitch,that knowing touch/attention from the designer of your overpriced gizmo goes a long way with making people feel "connected" with their purchases.It's marketing psychology 101 or what the biz refers to as "emotional motivators" and seeing how the founder of Chord(John Franks) qualifications seem mostly to be in marketing it's no surprise.

 

[pma]

19 pages of replies and no one noticed that for $5500 they put the wrong BNC connectors on the back? They are 50ohm connectors. The S/PDIF spec calls for 75ohm termination. I know it doesn't make a practical difference contributing at most a couple of ps to jitter, but for $5500 I expect at least the right connector to be used.

This is quite pointless with the relatively slow signal with not very fast edges as S/PDIF is. The difference in BNC impedance would be an issue if you were dealing with GHz signals. The reason to use 50 ohm BNC connectors is the mechanical construction, compared to 75 ohm versions they have thicker insulation around the middle pin and thus higher lifetime. If you wanted to complain on connectors, then rather about 90% of other manufacturers, who happily use RCA-CINCH connectors (with unspecified characteristic impedance) for S/PDIF and I have not seen much complaints here.

 

[bidn]

I've been wondering for a while why upscaling with lots of taps is something you can't just do in software, either in real time, or output as a higher bitrate file. It's an operation entirely in the binary domain! You can play the 768KHz/32-bit output to an excellent DAC over USB for under $/£300, so if there's any point at all to the taps, why on earth is the only implementation in hardware, and at £4000??

My suspicion was you can do it in software, but that wouldn't make any money, and there'd be no point because it sounds the same.

So these file links got my hopes up at verifying this (or having some remarkable revelation) in a blind test:


But it turns out the output was recorded in 44.1KHz, so these recordings seem completely pointless I'm actually a bit unhappy they're presented as evidence, but maybe I've missed some advertised claim that the upscaler can improve the sound so much that it's still improved when you downscale it back to the original (/s).

I don't know if there is equipment readily available to perfectly capture a 768KHz digital signal (which would be ideal for isolating the M-Scaler from other factors), but I'm pretty sure you can at least record the resulting analogue signal at that rate.

Quite right.

On Head-Fi, years ago (July 2018), Chord was exciting people about the coming launch of the M-Scaler on a pre-launch thread. Then I posted a message asking Rob Watts, : What is the point of the M-Scaler given that I can have my computer do the upsampling before sending the audio stream through USB to the DAC:

What difference, and why, in SQ would there be between the following audio chains:

- PC → USB → digital upscaling with M-Scaler → digital signal → DAC

- digital upscaling with PC → digital signal → DAC ?

Because the M- Scaler upscales and outputs a digital signal, I don't understand how the output upscaled digital signal would differ from the same produced by software running on a fast PC.

https://www.head-fi.org/threads/chord-electronics-hugo-m-scaler.884801/post-14381071

Rob Watts posted this answer :

It's about how accurate the system does it; the PC is not capable of reconstruction to anything like the same accuracy that the M scaler is capable of.

That was so vague!! Definitely a non-answer for me.

Then I asked again:

Thank you,
but I would be interested if you would elaborate on this difference and on the reasons allowing the M-Scaler to produce a higher accuracy than a powerful PC.

But then I never received any answer...

And Indeed I have been upsampling for free with the software that came with my RME ADI-2 PRO DAC.

 

[Jomungur]

This is not an excuse for not doing listening test.

I don't understand the vitriol. If you prefer the sound of the M Scaler, go for it. You don't even have to blind test. Amir explained why he believes the listening tests/samples would not add value to his review. Remember that the pause from switching from 4x to 16x is noticeably longer than the near instantaneous pause from switching from bypass to 2x. I don't recall the pause from 16x to bypass, as I never really did that. (I think the longer pause has to do with the signal going from single BNC to dual BNC, someone else above mentioned this too).

But since you're talking about listening tests, I want to share my personal experience. I've had the M Scaler in my system for about 2 years. DAVE is the DAC. Hifiman Shangrila headphones, Airtight 300B amp, loudspeakers have varied but right now are the BW 802D4s that came out last year. I've also tried the M Scaler on other quality speaker systems that I no longer have. I also have some pricey components in the middle which I'm now having a lot of doubts about (in terms of value add). Still, the system sounds great, I love it. My family loves it, friends come over to listen.

I'm sharing this not to brag but so you know that I'm using the M Scaler with what most people would consider a high end system, where presumably it is supposed to make the biggest difference.

Even since I took the M Scaler out of the equation when I sent the unit over, I have noticed *zero* difference. I asked my older son yesterday, who doesn't even know what an M Scaler is and that I had one, how the speakers have been sounding because I plugged into the amp into a different outlet. (That was a lie, but I had to come with a reason to ask). He told me they sound better lately.

I have more reason than most people here to want the M Scaler to work, to do *something* that makes a difference. But I can't recommend anyone to buy it. If it does anything it's too subtle to care about. Perhaps if it was 10 times cheaper and if you are someone one who likes to take chances on being an early adopter of potential innovations.

 

[bidn]

When I went to a presentation, specifically on the M-Scaler, a Dave was used. Their rep did the before and after and had the remote controls. I definitely heard a difference, the M-Scaler in use supposedly presenting a deeper sound stage over the bypassed setting done via one of th ebuttons on the front. The thing is, I'm certain the mean volume level was different, having vivid memories of trying to compare two things audibly and getting totally confused. In this case, the 'quieter' one sounded better in an audiophile sense. The audience, all 50+ in age and most I'd say 60+ too (so with age degraded hearing I'd suggest), nodded sagely at Rob W's presentation and seemed to enjoy the dem. If we were told the upscaling amount, I can't remember the setting.

I can relate to your experience, your point about the differing volume levels is very important, and I am proposing an explanation as to why many people say that they feel that the M-Scaler increases depth and soundstage.

At my dealer, I made comparative auditions of the Chord Dave DAC alone versus the chain M-Scaler → Dave.
As is typical of such subjective tests (at dealers or shows), nothing is provided for volume matching.
With the M-Scaler in the chain, the produced sound volume is lower, i.e. things sound as if they were further away. Now when experiencing such a hyped and very expensive, innovating device, one has the unconscious bias that things are improved. So one would interpret the distant (softer) sounds as a ultimate improvement in depth and soundstage, even if the device actually does not bring in any factual improvement.

Now that we have facts about the M-Scaler, I am lucky I never bought none of these super expensive devices, but I will still be interested to see the results of the Dave, if one day Amir gets one shipped by an open-minded, nice owner like presently @Jomungurfor did for the M- Scaler.

 

[Laphr]

At my dealer, I made comparative auditions of the Chord Dave DAC alone versus the chain M-Scaler → Dave.
As is typical of such subjective tests (at dealers or shows), nothing is provided for volume matching.
With the M-Scaler in the chain, the produced sound volume is lower, i.e. things sound as if they were further away. Now when experiencing such a hyped and very expensive, innovating device, one has the unconscious bias that things are improved. So one would interpret the distant (softer) sounds as a ultimate improvement in depth and soundstage, even if the device actually does not bring in any factual improvement.

Which one is it? Louder equals better sound, or is it quieter that is better.

 

[mansr]

Dual Coax output at 768 kHz is proprietary so nothing is capable of capturing it.

Isn't it just two parallel S/PDIF streams?

 

[mansr]

You can upsample audio files in real time via this Foobar plugin https://www.remastero.com/foo-pggb-rt-guide.html, or offline via this program https://audiowise-canada.myshopify.com/products/pggb-it2 (free for 2 million taps, which is crazy already).

Using software like Matlab, it is trivial to resample whole files at once using all the taps. You simply load the file, do an FFT on the whole thing, pad to the desired length, and do an inverse FFT. The result is equivalent to using a filter as long as the file. A longer filter would be pointless anyway since the parts extending beyond the input wouldn't be contributing anything to the output.

 

[Shadders]

Hi,
@amirm
I am on another forum, and they are having kittens about the review. Did you use the Chord Hugo 2 or the Chord Hugo TT2, as the dual-BNC graph indicates the TT2.

Thanks and regards,
Shadders.

 

[Jimbob54]

Hi,
@amirm
I am on another forum, and they are having kittens about the review. Did you use the Chord Hugo 2 or the Chord Hugo TT2, as the dual-BNC graph indicates the TT2.

Thanks and regards,
Shadders.

Perhaps @Jomungur can clarify as they were his but I am pretty sure it was the TT2 (as you say, the portable Hugo 2 doesnt have dual bnc and wouldnt be the ideal DAC for Mscaler use)

 

[Jomungur]

I sent a Hugo 2. The Hugo 2 does take work with the M Scaler. You just need a reverse split or Y cable that takes 2 BNC outputs and combines into the input jack of the Hugo 2. I included this too. The Hugo 2 input sphere changes color depending on input sample rate so you know it's recognizing the signal rate.

This is straight from the M Scaler Manual:

The Hugo M Scaler uses the world’s most advanced filter technology to upscale standard 44.1kHz digital audio up to 705kHz (16x CD’s 44.1kHz native resolution), ready to be passed to a suitable DAC; Hugo M Scaler extends its upscaling performance to 768kHz (from 96kHz input data) for dual-BNCinput Chord Electronics DACs: (at the time of writing) DAVE, Qutest, Hugo 2 and the new Hugo TT 2.

And from the Hugo 2 specs:

Coax via (same) 3.5mm jack (Dual data mode: Yellow): 44.1kHz to 768kHz – 16bit to 32bit

Ironically, I thought the Hugo 2 was a better test because the M Scaler should help the Hugo 2 more than it would the DAVE. Because the DAVE filter uses significantly more taps than the Hugo 2 (164,000 vs. 49,152), and so has less to gain. But some people seem to be taking issue with the use of the Hugo 2?

 

[JSmith]

I will still be interested to see the results of the Dave

Well in lieu of an AP measurement by Amir, we have this here by @BlinkDL for now;

Chord DAVE measured by Cosmos ADC

Hi guys. I measured my Chord DAVE using Cosmos ADC (grade A). Cosmos is set to 4.5Vrms in both XLR and RCA measurements. Note DAVE has a HF filter which affects the FR and probably cause the rise of IMD after 18kHz. I will remeasure it later without HF filter. DAVE has higher noise and...

www.audiosciencereview.com

... and the Mojo 2 scored well here recently as well. Alas it seems because this M-Scaler doesn't do anything... somehow Amir has it in for Chord.


JSmith

 

[DSJR]

Which one is it? Louder equals better sound, or is it quieter that is better.

My experience first hand is that a A-B dem (by instant switching) will almost always favour the (very very slightly) louder one. The M-Scaler dem wasn't like this, it was half a track or so followed by a pause as the rep pushed buttons and then the same half track repeated. In a high end audiophile type system, the slightly quieter one will appear to give a broader and deeper soundfield I feel. The volume difference 'seemed' very slight I have to say, but it was enough for me to be a bit uncomfortable when everyone there was so convinced (Rob W had already got me going with his mix of sound science and audiophool bullsh*t, very artfully blended together - I wish I was properly qualified to call him out on so much of the BS side and give proper evidence, especially as the attendees were lapping it up so fully, but I can't).



P.S. Another can-o-worms I can't prove as it varies from person to person and system to system, but some stereo's are more sensitive to 'absolute phase' than others (where the + and - is reversed on both channels from amp to speakers). If one is sensitive to it along with the speakers and th emusic recording is carefully chosen, the perceived image can also appear to push slightly back or forwards depending on the absolute polarity. Again, a subjective vibe here and it's not something you can instantly change. Interestingly, my Bryston BP25P preamp had a phase switch on it which didn't seem to do anything at all, but doing the same thing at the speaker terminals seemed to make a bigger difference (the amp can see the crossover differently - apparently...).

 

[DSJR]

Does anyone here remember the late 80's early 90's Cambridge CD2 with 16 times oversampling (and four dac boards with a TDA1541 chip on each)? Good internal photos are on the Lampizator site, but his 'improvement' involved ditching the 16x oversampling and basically paralleling the chips up (I gave up after that and don't trust his stuff anyway). I've no idea, but how different is this Stan Curtis implementation to the M-Scaler here, at least on red book CD's? The CD2 often needed a day or two 24/7 power to fully bed in (some did and some didn't, using the track 'Everywhere' by Fleetwood Mac as evidence with the jangly tinkly sequence at the beginning) but for £700 here, it was a good player and I believe pretty reliable until the mechs wore out (Philips CD M4/9 I gather). The CD3 which replaced it had a better mech but simpler dac tech I gather.

 

[Shadders]

I've no idea, but how different is this Stan Curtis implementation to the M-Scaler here

Hi,
Given the time of implementation of the Cambridge system, i would expect at best, linear interpolation between samples for the 3 other samples, if 4x oversampling.

With Chord/Rob Watts, doesn't he use the sinc function with a high number of values/taps for the interpolation filter ? (at a guess, or from memory reading a few years ago).

Regards,
Shadders.

 

[Glint]

I'd assume the XC7A200T is on the underside obverse of all those bypass looking caps - it's an expensive part, and programming it with this fancy 'million taps' malarkey likely would cost somewhat. Add in overall design, case costs, the costs of bringing it to market, blahblah etc, and then again limited market for this, made that much smaller by it's ridiculous cost... Possibly not a very profitable venture?

Regarding looks, personally I love the design concept of chords recent stuff, all the shiny big colourful buttons looks lovely imho!

Still, I still don't understand why it exists! And why it seems to be a bit not very good :/ And the cost is just mind boggling for what it doesn't do, poorly.

 

[dananski]

You can upsample audio files in real time via this Foobar plugin https://www.remastero.com/foo-pggb-rt-guide.html, or offline via this program https://audiowise-canada.myshopify.com/products/pggb-it2 (free for 2 million taps, which is crazy already).

Excellent, thanks for those links. I did take a look many months back but probably not thoroughly enough. And my search back then was more for open source code as I thought performance can be an issue for real-time and I wanted to see why. I'll be sure to give these a try.

Generally this is my preferred way to be able to counter a claim about a device - disprove that the underlying proposed mechanism it is supposed to rely on does anything helpful. In doing that, you can put a stop to a whole dead end of scientific effort and free people up to try something else.

 

[Peternz]

I don't understand the vitriol. If you prefer the sound of the M Scaler, go for it. You don't even have to blind test. Amir explained why he believes the listening tests/samples would not add value to his review. Remember that the pause from switching from 4x to 16x is noticeably longer than the near instantaneous pause from switching from bypass to 2x. I don't recall the pause from 16x to bypass, as I never really did that. (I think the longer pause has to do with the signal going from single BNC to dual BNC, someone else above mentioned this too).

But since you're talking about listening tests, I want to share my personal experience. I've had the M Scaler in my system for about 2 years. DAVE is the DAC. Hifiman Shangrila headphones, Airtight 300B amp, loudspeakers have varied but right now are the BW 802D4s that came out last year. I've also tried the M Scaler on other quality speaker systems that I no longer have. I also have some pricey components in the middle which I'm now having a lot of doubts about (in terms of value add). Still, the system sounds great, I love it. My family loves it, friends come over to listen.

I'm sharing this not to brag but so you know that I'm using the M Scaler with what most people would consider a high end system, where presumably it is supposed to make the biggest difference.

Even since I took the M Scaler out of the equation when I sent the unit over, I have noticed *zero* difference. I asked my older son yesterday, who doesn't even know what an M Scaler is and that I had one, how the speakers have been sounding because I plugged into the amp into a different outlet. (That was a lie, but I had to come with a reason to ask). He told me they sound better lately.

Thanks for sharing. I would be delighted with all this information if I was the proud owner of an Mscaler. I could now sell it and spend the money on something that my family and I could enjoy, like travel and fine dining. A huge win all around.

 

[Katji]

^^^ You got it. I do it all the time. Especially the travel.

[edit:] The alternative value for money comparisons, I mean.]

 

[amirm]

Rob Watts posted this answer :


That was so vague!! Definitely a non-answer for me.

For sure. He should have at least talked about the numerical resolution of his resampling algorithm. It is entirely possible that it has far less depth than a computer software version making his claim completely invalid and backward.