NEWS: Hugo M Scaler announced

Interesting. I wonder would there be a market for this. Personally it reminds me of upscaling an image in photoshop, the detail isn't really there, but it tricks you into thinking it is. I reckon it's a bit of a rip off
 
I think the idea is that you push the sampling frequency high that the DAC reconstruction filter has an easy job, and you're not left with any in-band amplitude or phase non-linearities.

It's not snake oil.

Nick
 
Last edited:
I think the idea is that you push the sampling frequency high that the DAC reconstruction filter has an easy job, and you're not left with any in-band amplitude or phase non-linearities.

Nick

Care to expand or point me at an article that explains it? Genuinely curious.

Thanks

G
 
Can someone explain this? How can you upscale a 16bit 44.1khz signal? to add in more detail? - The SPDIF signal going into it will be what's on the disc, what is it "filling" in.. I am very suspicious about it and I am a chord electronics fan!
 
Can someone explain this? How can you upscale a 16bit 44.1khz signal? to add in more detail? - The SPDIF signal going into it will be what's on the disc, what is it "filling" in.. I am very suspicious about it and I am a chord electronics fan!
Yep I'm suspicious as well. You can't add back in what isn't there. Or maybe it contains an infinite number of Digital Musical Monkeys that can then recreate the original ;)
 
For £1,500, the TEAC NT-505 can also offer upsampling - plus MQA decoding. Thus, unlike the Chord, it can properly decode/upsample a USB fed signal up to full-fat MQA. That’s, typically 24/352. However, the TEAC NT-505 can upsample to 32/768.

TEAC | NT-505
 
Yep I'm suspicious as well. You can't add back in what isn't there. Or maybe it contains an infinite number of Digital Musical Monkeys that can then recreate the original ;)
It's good to be suspicious.. and you cannot add back something which is not there.. good points.
However if you have two samples taken in adjacent time intervals, and they have different values, which is highly likely, you can interpolate, and guess what the sample inbetween should be, and insert that.
Is it cheating? Yes and no. It is reasonable to assume that had the intermediate sample been taken , it would have been close to the inserted one.
We are very happy for our Full HD TVs to do that with DVD data or 4K TVs with BluRay.

Now xxGBHxx, the maths of digital filtering is rather heavy, so rather than, get outside my comfort zone, could the following explanation work for you.?
.. the same 16 bit CD quality data is fed into a high speed DAC and converted into the analogue voltage perhaps 32 times with the same data. Now every time a conversation takes place there is a little momentary glitch. By doing the conversion 32 times faster, and having a low pass filter on the output it is much easier to filter out these gliches.
 
It's good to be suspicious.. and you cannot add back something which is not there.. good points.
However if you have two samples taken in adjacent time intervals, and they have different values, which is highly likely, you can interpolate, and guess what the sample inbetween should be, and insert that.
Is it cheating? Yes and no. It is reasonable to assume that had the intermediate sample been taken , it would have been close to the inserted one.
We are very happy for our Full HD TVs to do that with DVD data or 4K TVs with BluRay.

Now xxGBHxx, the maths of digital filtering is rather heavy, so rather than, get outside my comfort zone, could the following explanation work for you.?
.. the same 16 bit CD quality data is fed into a high speed DAC and converted into the analogue voltage perhaps 32 times with the same data. Now every time a conversation takes place there is a little momentary glitch. By doing the conversion 32 times faster, and having a low pass filter on the output it is much easier to filter out these gliches.
Yep totally agree with the analogy with Blu Ray upscaling, & see that a quality unit upscaling will allow the DAC to give a better approximation of the Analogue.
In the end, like many things, as you get closer & closer to a good equivalent of the actual original, the cost increases exponentially for small improvements, & I guess it is up to the individual as to how much they are willing to spend for these ever more & more minute improvements
 
Well this is it.. I think I'll need to a home loan of this to see how much better it is than a standard transport - DAC combo.
 
Going to watch this:-

Might be helpful!

The caption should read CanJam rather than CamJam 2018.

It was 33 minutes long, but it was worth every minute for anyone who's really interested in digital audio. It's not really about the 768k upscaling so much as the million tap filtering, which has almost historical significance.

Paraphrasing very heavily, the story seems to be that Chord's digital designer Rob Watts realised 35 years ago that the theoretical performance of CD audio could be realised if you had a good enough reconstruction filter. The filter should be a Sync filter, but it would need a million taps to realise the full 44.1/16 performance. Hardware limitations have prevented that until now. Chord use a large Xilinx Artix-7 FPGA with 740 DSP cores to do the digital filtering. The output of that filtering is fed to a suitable (very good) DAC for conversion. Or the filtering can be performed in an integrated transport, the Blu MkII, launched last year.

Rob was quite dismissive about what had been achieved by digital audio to date (including his own designs) saying that it was some way short of what should theoretically be achieved. In particular, current reconstruction filters were never able to get transient timing right, and this wasn't good enough. Now however, the M scaler can filter sufficiently well to realise the theoretical 44.1/16 envelope, and this is first time that this has been done. He says the improvements are considerable, and improve everything.

This does sound like the real thing.

Nick
 
Thanks Nick, you succinctly put together what was said. I guess proof will be in the pudding come autumn!
 
If as it likely appears, this is the real deal, it is very good news in the long run for everyone. In the short term Watts and his team stand to make a good profit, as wealthy patrons buy the newest and best. In the medium term the technology will be licenced out and in the longer term,it will be either public domain or pirated.
3000 quid is a substantial sum of money but not exorbitant, .. I have seen plasma TVs at 40k quid dropping to 20k ,then 10k and then almost free fall as LCD TVs came on stream. .. and they were being purchased at the 10k and 5k levels in quantity. Many musical instruments are in that ballpark.. pianos, branded guitars,as indeed are turntables and cartridges
 
Last edited:
What is guaranteed is those that spend £3000 on this will tell you it is wonderful, whilst the rest of us argue about its efficacy.
 
What is guaranteed is those that spend £3000 on this will tell you it is wonderful, whilst the rest of us argue about its efficacy.
.. yes but There will be one difference between that and the afore mentioned RA products. .. it will be possible to measure on a scope whether the transients are in or out of phase... What
measurement will not tell us however, is whether it matters
 
This scaler is the same one that's inside the Blu2 CD player, which I've heard being used with a Chord Dave DAC. Adding the scaler in gave a surprising uplift in sound quality. I currently use a Dave, so I'm very keen to try this device in my own system.
 
I got a response from the man him self to the same question I posed here on head-fi :)

"Interesting post - I am actually not adding in any more information - the information is all contained in the original bandwidth limited sampled signal. but what we have is sampled data, and what we need is a continuous un-sampled signal - and we need to do a huge amount of processing to do this without error. I will give you an example. Imagine a sine wave. You can state it's a sine wave; give its frequency and its amplitude. So the information content is fixed; but if you want a waveform of infinite length, and precision, then you would need an infinite amount of processing to create the infinite number of points. And with the sampled data, we can convert it to a continuous signal - with exactly the same information content - and recover the original bandwidth limited continuous signal - if and only if you do an infinite amount of processing and use an ideal sinc function interpolation filter. So I am not trying to create new information - actually we are converting from a sampled bandwidth limited signal to a continuous with exactly the same information. The problem with conventional filters is they are the ones that are adding extra information, as the interpolated signal is different from the original. What I am trying to do is merely reduce these errors, which are audible as it degrades the timing of transients - something which is essential from human psychoacoustics."

Chord Electronics Hugo M-Scaler
 
Ok, there’s something here that’s potentially very good news. I’d have to hear it to say for sure, but the theory looks sound.
 
Same, the difference in the Mojo to the Qutest was huge so hoping the MScaler opens things up massively. Time to stock up on CDs :p
 
Picture I took from Canjam unveiling along with Hugo TT2

322FF0FF-3D49-4C67-83E6-0AD72B304759.jpeg
 
While I’ve no idea if it makes any difference or not, I’d think this ought to be part of the input stage to one of their own dac’s rather than a stand-alone bit of kit..?
 
Last edited:
…… or part of the output stage of one of their transports.

I thought that, too. Apparently the reason for not integrating it with the DAVE DAC is that the power consumption is too high.

Nick
 

The latest video from AVForums

Is 4K Blu-ray Worth It?
Subscribe to our YouTube channel
Back
Top Bottom