[ardbeg1975]

Somebody wake me when this ASR topic quiets down.

 

[coopdog]

Thats slightly out of context, he is refering to his design choices (qv Gausian v triangular dither) of his upscaler. Not as a comparison to other DACs. If you enjoy the Topping, good for you, you have saved a lot of money. No one is critising you for your choice. Why take offense if others choose differently?

I was one of those that chose with his wallet…..I owned a DAVE and M-Scaler……I have no problem with people choosing whatever equipment they please….where is usual triangular dither used? In other dacs? Is Gaussian dither used only in Watts designs?

 

[coopdog]

Thats slightly out of context, he is refering to his design choices (qv Gausian v triangular dither) of his upscaler. Not as a comparison to other DACs. If you enjoy the Topping, good for you, you have saved a lot of money. No one is critising you for your choice. Why take offense if others choose differently?

FYI

This was just posted at ASR:

That aside, his assertion is wrong about superiority of Gaussian noise relative to Triangular Dither. Here is the paper that is considered the bible of dither:
AES paper, DIGITAL DITHER 2412 (C-8)
Stanley P. Lipshitz
John Vanderkooy
Audio Research Group
University of Waterloo
Waterloo, Ontario,

As you see, triangular dither is superior in both performance and level of noise it adds to the system. This is why it has become the de facto standard in digital audio processing.

As you see at the bottom, the reason to use Gaussian is because it is simpler/cheaper to implement. Not that it is better. In a $6,000 device, I expect the better dither to be used, than the cheaper one that raises the noise floor yet again

 

[anyoldears]

Probably the best place to discuss it then is on ASR

 

[coopdog]

Probably the best place to discuss it then is on ASR

I thought that this was the M-Scaler thread. Wouldn’t the best place to discuss the merits of the M-Scaler be the M-Scaler thread? Are you only allowed to post gushing reviews, or can you also discuss the possible negatives of the M-Scaler?

 

[bpcans]

I thought that this was the M-Scaler thread. Wouldn’t the best place to discuss the merits of the M-Scaler be the M-Scaler thread? Are you only allowed to post gushing reviews, or can you also discuss the possible negatives of the M-Scaler?

What negatives with the M Scaler are you talking about?

 

[audiojr]

Hi guys, have some questions with regards to running the m scaler with battery packs. Wondering whether can this be done on a power supply that outputs 12v/2a? Thanks.

 

[ra990]

Hi guys, have some questions with regards to running the m scaler with battery packs. Wondering whether can this be done on a power supply that outputs 12v/2a? Thanks.

Yes, anything under 15v should be ok.

 

[audiojr]

Yes, anything under 15v should be ok.

Is the current of 2a sufficient? As the supplied adapter is rated 15v/4a.

 

[ra990]

Is the current of 2a sufficient? As the supplied adapter is rated 15v/4a.

Try it and see if it works. Since it's under the current of the stock psu, it should be safe.

 

[Baten]

So noise is higher, but so what? You still won't hear it, but you will hear the benefits of Gaussian against the usual triangular dither. That is, you will hear better depth and detail resolution, unless you are deaf, or not blessed with good hearing, or not interested in the SQ and musical performance, in which case why on earth would you be interested in an expensive up-sampler?

So you're saying Gaussian dither is actually better and you will hear better depth and detail resolution? Funny how the literature on the matter does not agree, I guess there's no Chord/Watts paper on this available for reading anywhere...?

"You must be deaf or not blessed with good hearing" isn't the kind of argument I'd expect from an actual expert.

 

[GoldenSound]

While I have no doubt Amir takes pleasure in revealing bad measuring gear that is expensive, I highly doubt he purposefully produces bad results. If the MScaler had better measurements, he would have reported them, while likely coming to the same conclusion (that he hears no difference). At the very least, he has produced data about the MScaler that never existed before. Owners and fans may not like what the results were, but let's face it, they showed that the MScaler has some issues. Rob Watts has already stated he's working on the next version of the product, so I hope that addresses some of the issues we now know exist with the current product.

I don't like posting there because everyone thinks they have a doctorate in audio science and almost none of them have any actual experience with the products they criticize. They worship objective measurements there, we value subjective and real experience here.

Measurements did exist before: https://goldensound.audio/2022/03/14/chord-dave-measurements-with-mscaler/
https://goldensound.audio/2022/03/14/chord-dave-measurements-with-mscaler/

Additionally a lot of this debate could be avoided if Amir had not bottlenecked the device in testing. The problem isn't that Amir is showing some less than ideal measurements, that's his job to talk about the good and the bad.
But the issue comes in when a device is tested in unrealistic or problematic ways and then poor measurements result, leaving us unsure if it's the device, the testing methodology, or some of both.

Why use it only in 2x mode when the device is capable of much more and almost no one will be using it in that way?

Even though the APx555 cannot accept the full 16x / dual BNC input, you can either at least use 4x direct to the AP, and you don't see the same quantization noise limitation as Amir showed:

About 130dB of attenuation here compared to the ~80dB Amir showed.

Or given he had a chord DAC there, he could use that as a middleman to show the inferred attenuation (limited by the analog device of course but it'll still show more than 80dB).

(Note these graphs are not normalized to 0dB like Amir's are, so subtract 60dB, though I should probably do that in future to avoid confusion)

The jitter performance is a genuine concern though, but as Rob pointed out, the primary intent is to use these with Chord DACs which are highly resistant to jitter. When using the MScaler and DAVE for example you get practically perfect jitter performance:

But regardless for those using it with any other DAC it is something to keep in mind and I do hope the next gen MScaler doesn't have this issue.
If using the MScaler with a non-chord DAC though you do have the option of getting an SPDIF reclocker/DDC such as a DI20HE or ifi iPurifier. The audio content won't be affected so you still get the benefit of the MScaler's upsampling, but the DDC will reduce the jitter drastically.


Additionally the added noise with the topping D70S seems to be being blamed on the MScaler, when in reality it would seem more appropriate to be blaming it on the DAC? This behaviour isn't observed on the Hugo 2 (and also didn't happen when I tested it on my May) so perhaps the D70S has issues with galvanically isolated/ungrounded sources? Further testing would have been ideal to check rather than implying that the MScaler is at fault.

 

[Baten]

Additionally the added noise with the topping D70S seems to be being blamed on the MScaler, when in reality it would seem more appropriate to be blaming it on the DAC?

Isn't it already ascertained that it is because of the dithering, when in passthrough and 2x mode? So no, it is definitely the M-Scaler's doing.

 

[GoldenSound]

Isn't it already ascertained that it is because of the dithering, when in passthrough and 2x mode? So no, it is definitely the M-Scaler's doing.

I was referring to this one

Harmonic distortion goes way up (which should not be affected by digital data, the MScaler does not add harmonic distortion to the digital content) and a random spike at ~330hz shows up.
These could be jitter related (though they're high enough in level that I doubt it, again, further testing would be needed to confirm) but can't rule out a grounding related issue. As mentioned, the MScaler provides galvanic isolation so if your DAC requires a grounded source you might see funky behaviour.

In regards to the quantization noise, as I mentioned, using it in 4X oversampling showed no concerns with quantization noise and I got significantly better attenuation than what Amir showed, both when measuring via a DAC and when looking at the digital output directly.

 

[Baten]

I was referring to this one

Harmonic distortion goes way up (which should not be affected by digital data, the MScaler does not add harmonic distortion to the digital content) and a random spike at ~330hz shows up.
These could be jitter related (though they're high enough in level that I doubt it, again, further testing would be needed to confirm) but can't rule out a grounding related issue. As mentioned, the MScaler provides galvanic isolation so if your DAC requires a grounded source you might see funky behaviour.

While this output might be totally attributed to the Topping's internal workings (leaving my judgment out on that one), the 96dB SINAD is a dead giveaway that it is a product of dithering, wouldn't you agree? If you measured a lot devices, chances are you've seen this kind of output before when driver or software is truncating to 16-bit? (for example, different situation obviously).

In regards to the quantization noise, as I mentioned, using it in 4X oversampling showed no concerns with quantization noise and I got significantly better attenuation than what Amir showed, both when measuring via a DAC and when looking at the digital output directly.

Yes, 4x and up will push any problems away far enough that it's not an issue. If however this "problem" does occur in both 2x and passthrough mode as mentioned, the very least that can be said is that passthrough should not affect the digital output. It should not be doing this. I rest my case on that part, at least