SBGK wrote:
I just recompiled mqn using vs 2015 and that made a difference to the
sq, another factor to measure.
Would love to see the results of your measurements.
To try to judge the real from the false will always be hard. In this
fast-growing art of 'high fidelity' the quackery will
Archimago wrote:
I actually think the ultimate in tweakability is Bug Head Emperor -
MEGA tuning opportunities: 'http://orya.world.coocan.jp/bughead'
(http://orya.world.coocan.jp/bughead/)
Look at the description on that page and self-disclosure.
Well, at least the poor guy is
Archimago wrote:
SBGK, I don't think you ever answered the question of what methodology
you use to evaluate sound quality... Some details would be interesting
and useful. I've asked before and I believe a very fair question.
I mainly use my earholes and feedback from people nice enough
Julf wrote:
Would love to see the results of your measurements.
Well, the render loop takes 9 uops with no port pressure or register
stalls, that's the only measurement I've done recently, seems better
than the previous 13 and 11 uop versions.
The problem with digital is it's fairly easy to
SBGK wrote:
but these are not concerns of yours, so back to your squeezebox.So can we
take it from that childish dismissive comment that you no
longer have a squeezebox product?
LMS Version: 7.9
TranquilPC T2-WHS-A3 - WHS 2011
2x Touch, 3x SB3
SBGK wrote:
I mainly use my earholes and feedback from people nice enough to try it
and who I respect
It is well known that hearing involves far greater complexity than
merely using one's earholes.
I suspect that you need to up your game to include the functions of the
most powerful
Archimago wrote:
Yeah. Plus he admits to having some kind of mental issue and finds
programming a source of comfort. That's cool. No need to stress the guy
out.
Indeed. But he does seem to exhibit the rather frequent combination of
unverified, non-scientific miracle audiophilia and other
SBGK wrote:
Most science was observed or theorized before actual measurements were
made, so I don't think I'm being unscientific in my discoveries. Just
the measurers need to step up their game and measure rather than sniping
from the sidelines.
The above is a truism (true, but
Julf wrote:
And the investigation would follow the scientific method. First you
would verify that there really is a difference, by isolating possible
other causes (such as confirmation bias and the placebo effect), and by
independent verification / replication. Then you would formulate an
arnyk wrote:
How is hearing a difference in a DBT really any different from hearing a
difference in any other context? It isn't.
Well, if you have drunk the cool-aid, the artificiality, forced
conditions and pressure of DBT makes you less sensitive to differences -
just like the bad
SBGK wrote:
Now what about answering my questions to you about how you can measure
something when you can't hear any differences, most people would hear a
difference and then investigate why.
Measuring differences is pretty easy. At a sufficiently microscopic
level everything is changing
SBGK wrote:
Now what about answering my questions to you about how you can measure
something when you can't hear any differences, most people would hear a
difference and then investigate why.
And the investigation would follow the scientific method. First you
would verify that there really
Julf wrote:
Well, at least the poor guy is honest - However, I am not able to show
scientific basis.
Yeah. Plus he admits to having some kind of mental issue and finds
programming a source of comfort. That's cool. No need to stress the guy
out.
I actually saw this program a year back and
SBGK wrote:
Well, the render loop takes 9 uops with no port pressure or register
stalls, that's the only measurement I've done recently, seems better
than the previous 13 and 11 uop versions.
So how does the speed of the render loop correlate with sound quality?
Or perhaps an easier question
SBGK wrote:
I mainly use my earholes and feedback from people nice enough to try it
and who I respect
eg - here is some feedback for the vs 2015 recompiled version, of course
this way of working is impossible in your world, but it seems to work,
some versions are identified as
SBGK wrote:
Most science was observed or theorized before actual measurements were
made, so I don't think I'm being unscientific in my discoveries.
I was talking about the scientific method, so confining myself to
empirical science. And you are definitely not following scientific
methodology.
arnyk wrote:
It's also false because it forces a false dichotomy between observations
and measurements when in fact they are the same thing.
I think the way SBGK uses the word observation is in the sense of
acquisition of information employing the senses, while you are using
it in the
Julf wrote:
I think the way SBGK uses the word observation is in the sense of
acquisition of information employing the senses, while you are using
it in the scientific meaning of the word.
A few useful reminders on 'observation'
(https://en.wikipedia.org/wiki/Observation):
Absolutely
Are we not making things to complicated in these cases archimago shows
with two methods that the output of the dac is the same hence no need
for DBT or other complex measures. The same output is the same the tiny
dissimiliarities that dont make the curves overlap exactly is the random
noise . Two
Julf wrote:
Well, if you have drunk the cool-aid, the artificiality, forced
conditions and pressure of DBT makes you less sensitive to differences -
just like the bad vibrations caused by the presence of a sceptic in a
room makes fairies less likely to show up. :)
This effect of controls
Mnyb wrote:
Are we not making things to complicated in these cases archimago shows
with two methods that the output of the dac is the same hence no need
for DBT or other complex measures.
That is the whole point about the scientific method (especially
post-Popper) - any scientific theory
Mnyb wrote:
Are we not making things to complicated in these cases archimago shows
with two methods that the output of the dac is the same hence no need
for DBT or other complex measures. The same output is the same the tiny
dissimiliarities that dont make the curves overlap exactly is the
Mnyb wrote:
A very interesting part is that piece of software you tested , people
testify to have all kinds of better sound experience with it. But none
ever heard the only verifiably thing it's does regarding audio , bodge
24/48 ;) just like you pointed out in your blog ( not as explicit as
Archimago wrote:
Nonetheless, I do want to hear from folks like SBGK to make sure
verifiable claims are looked at. If claims are unverifiable; perhaps
better yet the beliefs unfalsifiable, then I think we can say clearly
we are not dealing with the scientific domain.
Yes, that is a very
Wombat wrote:
Again, i think it is simply the absense of any capacitor at the output.
The 1000yF are for the electronics, i doubt they are in the signal
path.
I didn't look around much but this review lists no capacitor in the
signal path as feature.
A very interesting part is that piece of software you tested , people
testify to have all kinds of better sound experience with it. But none
ever heard the only verifiably thing it's does regarding audio , bodge
24/48 ;) just like you pointed out in your blog ( not as explicit as I
just did )
utgg wrote:
Maybe just showing the DC decay resulting from AC/capacitively coupled
output. I.e. the impulse injects a very small amount of DC which decays
over a few ms - perhaps no further overshoot/ringing involved after the
initial 'rebound'. Should be visible with a lot more gain on the
Archimago wrote:
Excellent work utgg! Indeed it looks like as you said, a D/C offset
which decays over at least 40ms. Here's a zoomed in look at the tracing
of the impuse
Result of those big 1000uF capacitors seen in the 'hardware teardown'
28 matches
Mail list logo
