[JaZZ]

Castleofargh...

...just do yourself (and us) a favour and listen to the DAVE extensively – and with an open mind instead of audio theories in the back of your mind! Then come back and tell that you haven't noticed any difference compared to your run-off-the-mill DAC!

 

[ZappaMan]

get thy behind me satan

 

[Triode User]

Castleofargh...

...just do yourself (and us) a favour and listen to the DAVE extensively – and with an open mind instead of audio theories in the back of your mind! Then come back and tell that you haven't noticed any difference compared to your run-off-the-mill DAC!

Sound Science is all BS.
(Blinkered Science).

 

[x RELIC x]

@castleofargh, because Rob’s DAC performance enhances the timeing, timbre, depth, and you’re just thinking in the amplitude domain (I think, that’s how I read it). Rob has posted many times that he thought -350dB would be innaudible and if someone told him otherwise he would have thought it not possible to make a difference. Well, to his ears it does so he’s reporting what he hears. It seems that by getting such performance that there are interrelated effects, not just mechanical (ear) amplitude related, that are not fully understood.

At least that’s my $0.02.

 

[514077]

Sound Science is all BS.
(Blinkered Science).

Vennum spitting forum, perhaps. At least the guys on the GeekWave forum have good reasons for hostility.

 

[Triode User]

Vennum spitting forum, perhaps. At least the guys on the GeekWave forum have good reasons for hostility.

I said blinkered science because they prefer to say that science predicts that certain differences cannot be audible. An alternative approach is to listen, realise that there are audible differences and then to try and find an explanation using science that explains the differences.

But I do like your VSF (vennum spitting forum) name. It is strangely accurate.

 

[jagwap]

Castleofargh is probably the least closed-minded of the "Sound Science" bunch. However he is right to a point: prove it.

A single listening test does not prove a thing. Sighted makes it very unlikely. There the sound science philosophy is correct.

However I agree that an open mind is absolutely essential. The seemingly bizarre things that make a difference, that later, often much later, proved to be true and scientifically measurable later is eye opening. As mentioned amplitude is only one factor. It is important, but outside of acoustic and class D amps, easy to fix. There are many other factors that also make important differences to the perception of music. This gets ignored by so many intelligent audio people.

I think science is vital to understanding the reproduction of music. But this often comes from informed and controlled listening first. In this area the Sound Science forum is off putting, as they shun the inovation from this method of discovery, as nothing is ever "blind" enough in blind listening.

However Castleofargh is the moderator because he is at least reasonable. He is asking a reasonable question. What is 320dB dynamic range giving to the signal that we need to improve the experience. I would like to know too. I'm less skeptical as I have found surprising things important of the last few decades designing audio equipment.

 

[JaZZ]

Some clarification from Rob. It's not about dynamic range (in a general sense), but noise-shaper performance.

 

[Jawed]

However I agree that an open mind is absolutely essential.

The experiment is easy to do: Hugo MScaler fed into Qutest or Hugo 2 in a listening comparison with HMS fed into DAVE.

The only mathematical difference between these two is the number of elements in the pulse array and the noise-shaper configuration (which partially depends on the number of elements).

The 1st stage WTA upsampling filter is in HMS, so it is identical in all test scenarios. The 2nd stage upsampling filter is identical in all three DACs. 3rd stage upsampling to 2048FS is based on a simple IIR filter that I believe is identical in all three DACs. So the only difference is in the noise shaper, which is what drives the pulse array, and of course the pulse array is where the resolution of the noise shaper meets the real world.

It's really that simple to conduct this experiment, but some "open minds" are all talk and no trousers.

 

[ray-dude]

The experiment is easy to do: Hugo MScaler fed into Qutest or Hugo 2 in a listening comparison with HMS fed into DAVE.

The only mathematical difference between these two is the number of elements in the pulse array and the noise-shaper configuration (which partially depends on the number of elements).

The 1st stage WTA upsampling filter is in HMS, so it is identical in all test scenarios. The 2nd stage upsampling filter is identical in all three DACs. 3rd stage upsampling to 2048FS is based on a simple IIR filter that I believe is identical in all three DACs. So the only difference is in the noise shaper, which is what drives the pulse array, and of course the pulse array is where the resolution of the noise shaper meets the real world.

It's really that simple to conduct this experiment, but some "open minds" are all talk and no trousers.

The other wrinkle is power supply, but the differences in the Hugo2 and DAVE (from Blu...still waiting on my HMS) are pretty clear. It is an experiment that is well worth conducting (I learned an awful lot going through the big Mojo/Hugo2/DAVE/Blu2/headphones/speakers critical listening tests...fantastic experience)

 

[jagwap]

The experiment is easy to do: Hugo MScaler fed into Qutest or Hugo 2 in a listening comparison with HMS fed into DAVE.

The only mathematical difference between these two is the number of elements in the pulse array and the noise-shaper configuration (which partially depends on the number of elements).

The 1st stage WTA upsampling filter is in HMS, so it is identical in all test scenarios. The 2nd stage upsampling filter is identical in all three DACs. 3rd stage upsampling to 2048FS is based on a simple IIR filter that I believe is identical in all three DACs. So the only difference is in the noise shaper, which is what drives the pulse array, and of course the pulse array is where the resolution of the noise shaper meets the real world.

It's really that simple to conduct this experiment, but some "open minds" are all talk and no trousers.

Easy perhaps, but a little expensive for all of them. Also doing it double blind will be necessary to convince the nay sayers.

 

[514077]

Easy perhaps, but a little expensive for all of them. Also doing it double blind will be necessary to convince the nay sayers.

That's because people who are "wiser than thou' will try to place limits on your ability to impose your own objectivity on the experiment. I WILL NOT do a doubleblind test of anything, as I'm only concerned with proving something to my noble self. The only reason for a doubleblind test is for the smug performer of said test, to get his/her jollies out of humiliating the testee into giving the wrong answer.

 

[Jawed]

Easy perhaps, but a little expensive for all of them.

Seems like it would be at the cost of a visit to a CanJam and persuading some people there who have the components to help with a comparison in a quiet room, e.g. after hours in a hotel room.

 

[castleofargh]

Castleofargh...

...just do yourself (and us) a favour and listen to the DAVE extensively – and with an open mind instead of audio theories in the back of your mind! Then come back and tell that you haven't noticed any difference compared to your run-off-the-mill DAC!

what could I expect to learn by listening to the Dave? that I like it(or not), that I'm impressed(or not). but I'm not sure how that will tell me anything about human hearing thresholds and what we need for some subjective impression to be right. which is what I'm contesting here.

Sound Science is all BS.
(Blinkered Science).

I'm sorry you got traumatized to the point where anytime you see "Sound Science" written under my nickname, you completely forget what's going on and get the urge to bark at it like this.
be strong!

@castleofargh, because Rob’s DAC performance enhances the timeing, timbre, depth, and you’re just thinking in the amplitude domain (I think, that’s how I read it). Rob has posted many times that he thought -350dB would be innaudible and if someone told him otherwise he would have thought it not possible to make a difference. Well, to his ears it does so he’s reporting what he hears. It seems that by getting such performance that there are interrelated effects, not just mechanical (ear) amplitude related, that are not fully understood.

At least that’s my $0.02.

agreed, I'd be on board with "something goes on and I don't know what or why, but when I do that I think it gets better". but look at what I quoted in my last post. all the questions and uncertainties seem to also have moved down to -350dB.

However Castleofargh is the moderator because he is at least reasonable.

I wish that was true, but I ended up with the tag after I went running my loud mouth and asking for more proactive moderation instead of always having Currawong the Hun coming to evenly chop heads once everybody had long escalated into a situation of no return. so basically my qualifications are that I was whining a lot.
but thank you for the ego boost. even if it's not deserved, I take it.

The experiment is easy to do: Hugo MScaler fed into Qutest or Hugo 2 in a listening comparison with HMS fed into DAVE.

The only mathematical difference between these two is the number of elements in the pulse array and the noise-shaper configuration (which partially depends on the number of elements).

The 1st stage WTA upsampling filter is in HMS, so it is identical in all test scenarios. The 2nd stage upsampling filter is identical in all three DACs. 3rd stage upsampling to 2048FS is based on a simple IIR filter that I believe is identical in all three DACs. So the only difference is in the noise shaper, which is what drives the pulse array, and of course the pulse array is where the resolution of the noise shaper meets the real world.

It's really that simple to conduct this experiment, but some "open minds" are all talk and no trousers.

there is no way that trying 2 different playback systems and hearing a difference is going to convince me that it's important to have noise shaping down 350dB for good "depth perception". how would I even start to see causality with all the potential variables involved? to me that's like people having two amplifiers, preferring the one with a higher max output and concluding that you need at least that much power for the headphone to sound right. effectively hearing a difference is only a small part of actually being able to draw that sort of conclusion.

even if the 2 setups were indeed strictly identical aside from the noise shaping, I would then agree that this particular noise shaping ends up causing an audible impact on that particular gear. I couldn't deny that of course. but I'm not sure I would be willing to extrapolate about other gears or how much noise shaping is having an impact on them(if they can even handle the concept of values so low). or how much noise shaping is needed to achieve audible transparency, proper depth impression, or any audible concern like those.

 

[x RELIC x]

@castleofargh, did you read the post by Rob that @JaZZ linked? -350dB is in the digital domain and the analogue output dynamic range is obviously not in the same ballpark. When Rob got -350dB digital performance the perception of depth changed dramatically to him due to reducing small signal errors (he surmises). Read Rob’s posts and you may get some insight to his observations.