[MvRBE10]

So it took some working on my 3D design skills to draw the DAVE.. but i made a impression of my 1st gen Mscaler in Choral housing with DAVE stacked on top.
The 2 top and bottom parts are ready for CNC fabric and i got my 1st quotes.





I'm still waiting on the optical receiver components, if it all works out i can link the two units without extra boxes





Sorry @MvRBE10 i'm not sleepwalking

With that kind of work you have to be a daywalker by now. Really nice work bro you put some effort in that

 

[GuiltyRocker]

So it took some working on my 3D design skills to draw the DAVE.. but i made a impression of my 1st gen Mscaler in Choral housing with DAVE stacked on top.
The 2 top and bottom parts are ready for CNC fabric and i got my 1st quotes.





I'm still waiting on the optical receiver components, if it all works out i can link the two units without extra boxes





Sorry @MvRBE10 i'm not sleepwalking

That's amazing, wow.

 

[OCC7N]

So it took some working on my 3D design skills to draw the DAVE.. but i made a impression of my 1st gen Mscaler in Choral housing with DAVE stacked on top.
The 2 top and bottom parts are ready for CNC fabric and i got my 1st quotes.





I'm still waiting on the optical receiver components, if it all works out i can link the two units without extra boxes





Sorry @MvRBE10 i'm not sleepwalking

This project needs its own thread/appreciation.

Wow! Good luck.

 

[Reactcore]

As to your comment about two sample filters - yes apologies but you are wrong about that, but you were right about being probably wrong! Two samples would be a linear interpolation filter, and that would be a disaster for transient reconstruction, and a disaster for sine wave or steady state signals - a HF sine wave would look like triangular waveforms, clearly very distorted.

Hi Rob,
I think you misunderstood my point.. i meant to say i thought a tap is one calculated virtual sample in between 2 original samples from the digital file..

I tried earlier to explain how i saw it in a drawing:
https://www.head-fi.org/threads/watts-up.800264/post-15974008

Its just what exactly is being tapped? visualizing makes it better to understand for us non sampling engineers

 

[Triode User]

So it took some working on my 3D design skills to draw the DAVE.. but i made a impression of my 1st gen Mscaler in Choral housing with DAVE stacked on top.
The 2 top and bottom parts are ready for CNC fabric and i got my 1st quotes.

I'm still waiting on the optical receiver components, if it all works out i can link the two units without extra boxes

Sorry @MvRBE10 i'm not sleepwalking

An interesting project for sure. But why would you go to such lengths (and costs?) to make the mscaler match the Dave aesthetic and put it so close to the Dave when many of us have the Mscaler somewhat remote from the DAVE (mine is currently 2m away from the Dave).

 

[Reactcore]

An interesting project for sure. But why would you go to such lengths (and costs?) to make the mscaler match the Dave aesthetic and put it so close to the Dave when many of us have the Mscaler somewhat remote from the DAVE (mine is currently 2m away from the Dave).

Im not so much a believer of RFI transmitting through HMS's metal housing.. but rather by cabling.. cabling can act as antennas ..even a Opto DX takes the signal by coax outside before converting to optical and back.

I wanted to build my dual data optical link inside the HMS and DAVE leaving the electrical wiring inside the shielded units.

But current HMS housing has no room for my Avago units.. hence my idea to create my own housing.

 

[Rob Watts]

Hi Rob,
I think you misunderstood my point.. i meant to say i thought a tap is one calculated virtual sample in between 2 original samples from the digital file..

I tried earlier to explain how i saw it in a drawing:
https://www.head-fi.org/threads/watts-up.800264/post-15974008

Its just what exactly is being tapped? visualizing makes it better to understand for us non sampling engineers

A tap is one original sample multiplied by the appropriate WTA coefficient. To create the new interpolated samples (I assume that's what you mean by what you call the virtual sample in between the original samples), you need to add many taps together from all of the samples stored in memory - this is done in something called an accumulator - then output the result once all the required taps have been processed.

An important point to remember is that all of the output samples are created from all of the original samples stored in memory - so with the M scaler 63,488 original samples are used, each original sample multiplied by a different coefficient just to create one interpolated sample. And there are 16 interpolated samples for each original sample, as the filter goes from 1FS to 16FS.

Yes I know it's a lot of numbers that the M scaler is working on, and it's not so easy to explain how the thing actually works.

 

[Reactcore]

A tap is one original sample multiplied by the appropriate WTA coefficient. To create the new interpolated samples (I assume that's what you mean by what you call the virtual sample in between the original samples), you need to add many taps together from all of the samples stored in memory - this is done in something called an accumulator - then output the result once all the required taps have been processed.

An important point to remember is that all of the output samples are created from all of the original samples stored in memory - so with the M scaler 63,488 original samples are used, each original sample multiplied by a different coefficient just to create one interpolated sample. And there are 16 interpolated samples for each original sample, as the filter goes from 1FS to 16FS.

Yes I know it's a lot of numbers that the M scaler is working on, and it's not so easy to explain how the thing actually works.

Ah ofc.. theres always 16 new created samples in case of a 44.1k or 48k signal.
Even mojo creates that samplingrate internally i guess.

Leaves it difficult for me to grasp how past and future samples further than the 1st side ones can make the right 'predicion' as music is so random.. or are you looking to obvious non signal dependant aspects of original samples?

 

[ecwl]

Ah ofc.. theres always 16 new created samples in case of a 44.1k or 48k signal.
Even mojo creates that samplingrate internally i guess.

Leaves it difficult for me to grasp how past and future samples further than the 1st side ones can make the right 'predicion' as music is so random.. or are you looking to obvious non signal dependant aspects of original samples?

I think the challenge here is that the Nyquist-Shannon sampling theorem is not intuitive. So it's often hard to wrap our minds around it other than to "trust the math".

This is how I try to conceptualize the math. The problem we often have is that the digital samples we have were taken from the original analog waveform.
So if we have two sequential samples and we are trying to reconstruct the analog waveform, intuitively we would say, well, technically there is an infinite number of ways to connect these two dots so anything goes. Moreover, why would the samples occurring 1 second before these two samples or 1 second after these two samples have anything to do with the "correct" analog waveform between these two samples.

But in reality, that's not true because your analog waveform first of all is bandwidth limited to 20Hz to 20kHz. So while technically, there are seemingly still infinite ways to connect the two samples, you're restricted to waveforms that would be part of a 20Hz-20kHz waveforms.

Now let's first imagine we are dealing with a pure tone at a specific frequency. How do we know whether it's a 20Hz tone or a 20kHz tone? Well, the samples before and the samples after would tell us.

This is how I conceptualize why the longer the tap length, the more accurately we can re-construct the original analog waveform. It's still very non-intuitive because you can always increase the tap length so at what point, e.g. 1s, 5s, 10s? does the samples from a long time ago or a long time later no longer influence the analog waveform reconstruction in-between those two specific samples?

But so far, I definitely prefer 1 million taps of WTA over shorter WTA filters over other non-WTA filters.

 

[ruthieandjohn]

I have purchased the M-Scaler to go with my Chord Hugo TT, and it should arrive this week. I am very interested in whether it will accept the same inputs as my TT AND output in the same format that my TT Noel receives, or if I will have to cast about for adapters and such.

‘’I am also interested in how the sound with M-Scaler on vs off compares, not only for 44.1Ksource material, but also HiRes.

 

[Somafunk]

Monty explains the D/A & A/D below in a very easy to follow video

 

[Reactcore]

I am very interested in whether it will accept the same inputs as my TT

The auto input switching is really convenient

Be sure you also try optical in and out at same time with nothing else connected to both units

 

[Powersquat]

I have purchased the M-Scaler to go with my Chord Hugo TT, and it should arrive this week. I am very interested in whether it will accept the same inputs as my TT AND output in the same format that my TT Noel receives, or if I will have to cast about for adapters and such.

‘’I am also interested in how the sound with M-Scaler on vs off compares, not only for 44.1Ksource material, but also HiRes.

If memory serves me correctly, the Hugo TT doesn't have the dual BNC connections to take full advantage of full upscaling, I believe only the Qutest, TT2 and the Dave are equipped with dedicated dual BNC inputs, there is a workaround to fully upscale to the Hugo 2 by the use of an adaptor, but I'm really not sure about the TT.

 

[alxw0w]

If memory serves me correctly, the Hugo TT doesn't have the dual BNC connections to take full advantage of full upscaling, I believe only the Qutest, TT2 and the Dave are equipped with dedicated dual BNC inputs, there is a workaround to fully upscale to the Hugo 2 by the use of an adaptor, but I'm really not sure about the TT.

Yes TT1 can only handle single BNC at 384khz max. No double BNC option.

M scaler fot TT1 is a waste of money I wrote it some time ago. It's better to buy TT2 and then add m scaler.

 

[HughF]

Yes TT1 can only handle single BNC at 384khz max. No double BNC option.

M scaler fot TT1 is a waste of money I wrote it some time ago. It's better to buy TT2 and then add m scaler.

>>> M scaler fot TT1 is a waste of money I wrote it some time ago. It's better to buy TT2 and then add m scaler.

I respectfully disagree - for my use in my system. I started with TT1 which suited my system very well and added HMS - the improvement was major, the sound became very enjoyable indeed.
When I moved to TT2 several months later (still using HMS) I was more impressed sonically but did not enjoy the actual musical experience very much more and not enough to justify the £4K cost (less p/x for the TT1).

If you ALREADY have a TT1, adding the HMS I consider VERY worthwhile IF your system is similar to mine - highly transparent but not ruthless. I listen to speakers via a pre/power amp rather than headphones as you do - maybe you benefitted more from the extra power of the TT2 driving your headphones?

In other systems, moving to a TT2 from a TT1 may indeed be a better choice compared to adding an HMS but not in mine. An HMS will always be useful later whether you stay with TT1 or move to TT2 / DAVE.

I hope this helps anyone with a TT1 pondering HMS or TT2 as their next move - especially if using loudspeakers,

Hugh