[Leporello]

So, and I have said this many times - to me the M scaler is absolutely essential for good sound quality and for musicality. To say a Dave with an M scaler using stock cables and simply plugging it into the mains sounds worse compared to a sole Dave to me is a completely crazy suggestion. So please stop trying to imply the opposite.

And many would agree with you. But enquiring minds want to know: does the M-scaler really sound any different? How has this been demonstrated?

 

[OneEyedHito]

And many would agree with you. But enquiring minds want to know: does the M-scaler really sound any different? How has this been demonstrated?

Tossing my hat in the ring on this one.

I have used the M-Scaler with both the Dave and the TT2, it sounds different. How I demonstrated it was I plugged it into power, then connected the Dual BNC cables on the outputs to the inputs on the Dave/TT2, I then selected a few of my most often heard tracks, pressed play in Roon and heard a difference. I even compared it to using the Upscaling in Roon and still heard a difference.

As with all things in audio and food you must try it for yourself. There is not a test to tell how a bell pepper tastes to one person versus another. Our taste buds are translated by our brains, just like our hearing is translated by our brains.

I know we want to see the ¼ mile times for a Tesla Model S Plaid before we plunk down our sometimes hard earned $100K on it just to flash it up on IG but she doesn't care about how fast it is, just how good it sounds and in the case of the Plaid it sounds like a Slot Track car from our childhood.

 

[Leporello]

Tossing my hat in the ring on this one.

I have used the M-Scaler with both the Dave and the TT2, it sounds different. How I demonstrated it was I plugged it into power, then connected the Dual BNC cables on the outputs to the inputs on the Dave/TT2, I then selected a few of my most often heard tracks, pressed play in Roon and heard a difference. I even compared it to using the Upscaling in Roon and still heard a difference.

As with all things in audio and food you must try it for yourself. There is not a test to tell how a bell pepper tastes to one person versus another. Our taste buds are translated by our brains, just like our hearing is translated by our brains.

I know we want to see the ¼ mile times for a Tesla Model S Plaid before we plunk down our sometimes hard earned $100K on it just to flash it up on IG but she doesn't care about how fast it is, just how good it sounds and in the case of the Plaid it sounds like a Slot Track car from our childhood.

Not very convincing, I am sorry to say. While I am happy that you are happy with your purchase, your testimonial suffers from well-known problems which should be addressed before accepting it as evidence. "You must try it for yourself" may be a necessary but certainly not sufficient condition for it.

 

[GuiltyRocker]

Just because I don't post about comments that to me are completely wrong, does not imply any form of agreement. I take the view that people have different opinions when regarding sound quality or musicality, and I am sure that people often disagree with my posts. If I were to post every time I fundamentally disagree with a post, I would be here wasting my time all day and not doing important design work. Life is too short, and I have a lot to do.

So, and I have said this many times - to me the M scaler is absolutely essential for good sound quality and for musicality. To say a Dave with an M scaler using stock cables and simply plugging it into the mains sounds worse compared to a sole Dave to me is a completely crazy suggestion. So please stop trying to imply the opposite.



It is absolutely nothing like the PGGB files I heard early on. IMHO PGGB is very much the wrong approach, and when I heard the test files I said they sound like apodizing filters, which seriously damage transient time reconstruction. Problem is some like the soft bloated bass from apodizing, as they prefer the timing problems. And when the website went live they claimed it was an apodizing filter. It doesn't matter if it's 10 taps or 10 billion taps - apodizing is the wrong approach.



No absolutely not. It reconstructs the timing of transients to better than 16 bits, as the coefficients drift from sinc below 16 bit values. Indeed, using a 6kHz -301dB signal through the filter structure gives perfect rendition of -301 dB using the dual BNC outputs. So a 24 bit signal will maintain it's SNR perfectly when you use the dual BNC outputs. The single BNC does add some noise, as I use the better sounding Gaussian dither which is a little higher than the usual dither, but this will be inaudible.

Thank you for the answers Rob.

 

[Triode User]

And many would agree with you. But enquiring minds want to know: does the M-scaler really sound any different? How has this been demonstrated?

The pursuit of this line in the name of 'enquiring minds' has a somewhat scurrilous whiff about it.

The scurrilous smell is brought on by the request for peer reviewed evidence and design data for a consumer audio product. This is not a vaccine which is being claimed to work. It is a commercially produced audio product.

The normal situation with a consumer audio product such as this one is that the potential customer will first listen before buying. If no difference is heard they will not buy it. That is the end of the matter. If they buy mail order and cannot hear a difference they will return it for a full refund (certainly that is the position in the UK). Again, that is the end of the matter.

I am sorry but it is difficult to conclude that you have any other motive in mind apart from trying to stir up trouble.

And please do not fall into the trap of thinking that the recently posted measurements and so called 'review' elsewhere lends any credibility to the possibility that the Mscaler does not sound different. I am pretty sure that a peer review of that would find it sadly lacking.

With a bit of luck you guys will again push it beyond the acceptable line once again and can again have another clean up in the thread.

If your motive is merely that of an enquiring mind then you might be best advised to take the Mod's earlier advice and start a new thread in the Sound Science section.

 

[GuiltyRocker]

I questioned it by auditioning the device for over a month as i have stated previously. My dealer was kind enough to lend it for that long. I don't accept people's claims, when it comes to audio I try the equipment. I tried my Paradigm Personas at home, the Qutest, the M Scaler, the Parasound JC-5 and JC-2, I also tried many other pieces that I returned, because I didn't hear any improvement. I just don't take measurements from some guy in a website that claims science but bans anyone who questions, I don't subscribe to cults, i try the equipment myself.

 

[AxelCloris]

Another reminder that this thread is for the discussion of subjective impressions and discussion of our experiences with the Hugo M Scaler. General science discussions, like blind testing and peer reviewing, are always welcome in the Sound Science forums. Let's please keep the discussion in this thread on topic and within Head-Fi's guidelines. If folks have any questions regarding the thread or our guidelines, please feel free to shoot me a PM. Thanks everyone.

 

[flyte3333]

apodizing is the wrong approac

Don't apodizing filters fix actual real issues - with common ADCs and 44.1k releases ?

But your non-apodizing filters leave these issues un-fixed ?

Its not an issue with top notch classical and jazz productions of course and most hi-res - but a real issue for modern 44.1k

 

[Sampajanna]

The normal situation with a consumer audio product such as this one is that the potential customer will first listen before buying. If no difference is heard they will not buy it. That is the end of the matter. If they buy mail order and cannot hear a difference they will return it for a full refund (certainly that is the position in the UK). Again, that is the end of the matter.

This says it all. Very well put.… Go listen. If you like it, buy it. If not, leave it and find what does bring you joy. I love what the MScaler does. It sounds brilliant. SO I bought it. I wouldn’t want to prove why others should buy it. There is no end all be all product for everyone.

 

[Rob Watts]

Don't apodizing filters fix actual real issues - with common ADCs and 44.1k releases ?

But your non-apodizing filters leave these issues un-fixed ?

Its not an issue with top notch classical and jazz productions of course and most hi-res - but a real issue for modern 44.1k

There are two intents with apodizing or slow roll off filters - the first is to remove pre-ringing and the second intent is to attenuate aliasing from common half band ADC decimation filters within the 20 kHz to 22.05 kHz band. As I have talked about before, removing pre-ringing creates huge problems for transient reconstruction, in order to correctly reconstruct the timing of transients the interpolation filter needs to know the future - and this creates the pre-ringing. An ideal sinc function interpolation filter has an infinite amount of pre-ringing, yet will perfectly reconstruct the bandwidth limited signal. The reason an impulse test gives pre-ringing is because an impulse is an illegal signal, because it is not bandwidth limited. If you use bandwidth limited impulses then you will get no pre-ringing, and a sinc function filter will always perfectly reconstruct the original analogue signal.

So the idea that pre-ringing is bad and unnatural is something that simplistically appears correct but is based upon a severe misunderstanding of sampling theory. But this misunderstanding is common within the whole audio business, and has led to everyone using NOS, apodizing or minimum phase filters - all of which create severe transient timing problems, giving a soft unnatural bloom to bass.

The second reason - attenuating ADC aliasing in the 20-22.05 kHz band - does not work in practice. If you filter this band post reconstruction (in my test case I did it at 705 kHz after WTA1), there are little or no sound quality benefits from attenuating ADC aliasing. This was a little odd, as I did expect some improvements when I tested out such a filter. I guess once the aliasing damage is done within the ADC decimation filter, then it is impossible to repair that damage.

 

[801evan]

It is absolutely nothing like the PGGB files I heard early on. IMHO PGGB is very much the wrong approach, and when I heard the test files I said they sound like apodizing filters, which seriously damage transient time reconstruction. Problem is some like the soft bloated bass from apodizing, as they prefer the timing problems. And when the website went live they claimed it was an apodizing filter. It doesn't matter if it's 10 taps or 10 billion taps - apodizing is the wrong approach.

OMG....I've been looking for such a response to bring reason to the sonic issues I've heard on PGGB and HQplayer. PGGB sonic issues does remind me of the apodizing filter on a Topping.

Hence why HMS is superior to HQplayer and PGGB.

 

[adrianm]

And many would agree with you. But enquiring minds want to know: does the M-scaler really sound any different? How has this been demonstrated?

If you have to ask this you're 10 steps behind the current discussion. You'd have to be deaf not to hear it sounds different. How much effort is to decide for yourself? "Enquiring minds" that are skeptical and won't accept other's impressions would be better served finding out for themselves instead of wasting everyone's time.
The issue is does it sound subjectively/objectively better or worse. To me, the software is objectively as good or better than anything else out there, but the hardware comes with big issues, (Noise and jitter ) which detract so much from the overall improvement, that I ultimately found it a considerably worse experience than HQP + Src-Dx.
Granted, it's a better experience than Dave alone, but attenuating those shortcomings is an expensive proposition and ultimately, not worth the effort and mess, since there are now better solutions out there.
Hopefully the new M-scaler will be better, but for the prices Chord are throwing out there to test the waters, I think they can just keep it. Even though I don't listen without up-sampling anymore, there's limits to what it brings to the table.

 

[GuiltyRocker]

There are two intents with apodizing or slow roll off filters - the first is to remove pre-ringing and the second intent is to attenuate aliasing from common half band ADC decimation filters within the 20 kHz to 22.05 kHz band. As I have talked about before, removing pre-ringing creates huge problems for transient reconstruction, in order to correctly reconstruct the timing of transients the interpolation filter needs to know the future - and this creates the pre-ringing. An ideal sinc function interpolation filter has an infinite amount of pre-ringing, yet will perfectly reconstruct the bandwidth limited signal. The reason an impulse test gives pre-ringing is because an impulse is an illegal signal, because it is not bandwidth limited. If you use bandwidth limited impulses then you will get no pre-ringing, and a sinc function filter will always perfectly reconstruct the original analogue signal.

So the idea that pre-ringing is bad and unnatural is something that simplistically appears correct but is based upon a severe misunderstanding of sampling theory. But this misunderstanding is common within the whole audio business, and has led to everyone using NOS, apodizing or minimum phase filters - all of which create severe transient timing problems, giving a soft unnatural bloom to bass.

The second reason - attenuating ADC aliasing in the 20-20.05 kHz band - does not work in practice. If you filter this band post reconstruction (in my test case I did it at 705 kHz after WTA1), there are little or no sound quality benefits from attenuating ADC aliasing. This was a little odd, as I did expect some improvements when I tested out such a filter. I guess once the aliasing damage is done within the ADC decimation filter, then it is impossible to repair that damage.

Excellent, doesn't MQA also use this approach of apodizing?

 

[801evan]

Excellent, doesn't MQA also use this approach of apodizing?

Whoa, that would crazy if true coz MQA has a very different response and is much better than pggb/ HQplayer.

After googling, anti-mqa guy said it is not apodizing. Meanwhile, I find it to be the best format.

 

[auricgoldfinger]

I saw this post on Audiophile Style from the developer of PGGB:

https://audiophilestyle.com/forums/...ath-and-magic/?do=findComment&comment=1206191