Couldn't you look at schemes like MP3 as a form of (software) adaptive quantisation?
jan
exactly.
I found this official document: http://www.etsi.org/deliver/etsi_en/300100_300199/300163/01.02.01_60/en_300163v010201p.pdf
the encoding is primitive and there is nothing, and I do mean nothing that is inherently better about it. of course, you can make an Audio R8 run worse compared to a Lada.
initially I simply thought that for low-level signals the resolution was better compared to RedBook. looks like it's the other way round.
what do you mean? MP3 is just that. DCT transformed data is altered so that it's better compressed with Huffman coding while trying to keep the quantization noise masked. adaptive quantization, only more intelligent.
Last edited:
I often put some square waves through things I am building . Until recently it was rather boring . That was until I finished some valve amps my brother rejected . Suddenly even 1 kHz is a nice result . 2 kHz starting to look a lost cause . There is no mystery . I had to accept it as the best it can be . So I said to myself . Good enough for CD , good enough for my amp .
The question I am asking is . 5 kHz , full power . straight sides and top . Job done ? We have used these tests for years . Do we get full value from them ? Douglas Self points out the so called ringing on old 10 kHz tests was the output choke . What I am trying to get is a way of relating slew rate which isn't V/uS/w . Pick a value and say ideally you can do this . A short cut , no more than that . 5 kHz square I guess would be no small deal . 400 kHz bandwidth I guess ( 1/79 F 79 perhaps as far as we need go ) . Like I seem to be able to see the sound of an amp on scope the squareness can be seen .
On slew rates I tend to agree with this .
Slew Rate etc.
Speakers
If I may pull this discussion in the direction of speakers for a moment.
Recently I've been working with a big Open Baffle speaker that consists of 2x15" woofers and a fullrange driver on each baffle. While keeping the woofers the same I've swapped out the fullrange driver. The acoustic crossover point is circa 250Hz.
To date I have had in the fullrange position:
For me the challenge will be to find what, in the measurements, relates to the big subjective difference.
If I may pull this discussion in the direction of speakers for a moment.
Recently I've been working with a big Open Baffle speaker that consists of 2x15" woofers and a fullrange driver on each baffle. While keeping the woofers the same I've swapped out the fullrange driver. The acoustic crossover point is circa 250Hz.
To date I have had in the fullrange position:
- SABA 8" green cones
- Voice of Music vintage Danish 8"
- Wild Burro 7" prototype
- JBL 6" ceiling speaker
- Vifa 3" fullrange
For me the challenge will be to find what, in the measurements, relates to the big subjective difference.
Actually, that was more of a private joke, what you cannot know is the discussion Nige and I had, in which he said that under some circumstances in audio, a 741 op amp would be enough. I disagreed, and that comment was a hark back to that discussion.
741 . Surprising how many times it still appears in technical papers .
741 , CCS from output to -VE rail , 2N4403 input . I asked my friend John to supply one of his best mics and see how good it sounds via this mic amp . After screwing up his face he said probably very good . The next question was . Microphone more important than chip ? Answer , it's got to be . Before anyone says . Yes we know it is bad to do this . How many here would be fooled ? 100 % is my guess ? It almost certainly would sound great . Without any reference we probably would hear nothing wrong . Digital processing should knock it back more . John is good at getting mics to work . That also would be a factor ( distance , type , suit singer )
32-bit 20kHz sine generated in CoolEdit, converted to 16 bit with 0.5 bits dithering, triangular pobability distribution function and "Noise Shaping 44.1kHz". noise floor at ~ -142dB. I did not use averaging, because it looks even better then.
and for a bit of fun, here are the measurements of a EUR 200 DAC, the Musical Fidelity VDAC-II:
Musical Fidelity V-DAC II D/A processor Measurements | Stereophile.com
the same CCIF IMD test, even better tan the 15k Meridian:
and for a bit of fun, here are the measurements of a EUR 200 DAC, the Musical Fidelity VDAC-II:
Musical Fidelity V-DAC II D/A processor Measurements | Stereophile.com
the same CCIF IMD test, even better tan the 15k Meridian:
Attachments
There, you see, Barleywater, that's what I'm talking about, Nige just can't get over it.
Don't even get him started on the NE series, we won't be hearing the end of that.
I'm sorry, I still have no idea what your point it.
Slew rate is related to max output value, it's a large signal property.
Your square waves have lots of harmonics that are not present in music, so slew rate for those square waves is of no interest for music reproduction.
Are you trying to build the ultimate amp?
I thought you also referred to CD and what that required in terms of amplifiers? Or what??
I mean, I'm all for an informed discussion, but incoherent ramblings I can't handle so well.
jan
Last edited:
I will think it through . Try to find a way of relating slew rates to simple square wave testing .
I understand your frustration with this idea . It's only when playing with the amps of old it even came into my head .
When I translated this to a class A transistor amp it seemed mostly output stage problems ( FET's ) . Driving the VAS didn't seem too troubling . At 30 kHz things seem to come apart . My problem I found was I could improve what I saw on scope but it sounded harsher to me . The optimum seemed a good 10 kHz square-wave . One does get a correlation between slewing and square-wave testing . One sees the asymmetry of a single VAS .
I understand your frustration with this idea . It's only when playing with the amps of old it even came into my head .
When I translated this to a class A transistor amp it seemed mostly output stage problems ( FET's ) . Driving the VAS didn't seem too troubling . At 30 kHz things seem to come apart . My problem I found was I could improve what I saw on scope but it sounded harsher to me . The optimum seemed a good 10 kHz square-wave . One does get a correlation between slewing and square-wave testing . One sees the asymmetry of a single VAS .
It's called dare think outside the box . Usually to provoke others to say something they understand but probably think too unimportant to share . I found in mechanical engineering progress can be made by saying what if . Most good mechanical engineers have all the answers . They seem willing to take chances . 25 % of the time at a rough guess the chance works . 75% it doesn't . If that's 1000 extra RPM all well and good . My engineer Peter did the valves after the Piper Alpha disaster for 450 Bar ( was 150 bar and wasn't enough ) . Peter told me the valve was a stab in the dark which in the end worked . 450 bar by usual understanding was not possible . 25 years today it happened . Sad to say experts thought 150 bar enough to protect the rig .
I have to say I can't get my head around the difference between mechanical and electronic engineers ? The mechanical guys are so optimistic .
Maybe this is a good time to flag the 'Alternative Science Respectability Checklist'. It does take effort, sweat and time, but in the end you get the respect you deserve.
jan
jan
Have you looked at polar pattern vs frequency in any detail? When you say "EQed to a very similar FR," that can only be true at one point in space.
edit: I would also strongly urge a Heyser plot of the complex impedance.
very good article.Maybe this is a good time to flag the 'Alternative Science Respectability Checklist'. It does take effort, sweat and time, but in the end you get the respect you deserve.
jan
I especially like this portion:
So here’s what any working theoretical cosmologist would do (even if they aren’t consciously aware that they’re doing it): they would glance at the introduction to the paper, looking for a paragraph that says “Look, we know this isn’t what you would expect from elementary perturbation theory, but here’s why that doesn’t apply in this case.” Upon not finding that paragraph, they would toss the paper away.
it sums it up nicely.
and this:
If you try to brush those issues under the rug, rather than confronting them straightforwardly, your credibility suffers greatly.
wonder caps anyone?
Last edited:
...and this
Also, one last thing. Don’t compare yourself to Galileo. You are not Galileo. Honestly, you’re not. Dude, seriously.
if I am to take a quick guess, I would estimate that there are at least 10 instances of variations of "when bacteria were discovered no-one believed they really existed" in this thread
Also, one last thing. Don’t compare yourself to Galileo. You are not Galileo. Honestly, you’re not. Dude, seriously.
if I am to take a quick guess, I would estimate that there are at least 10 instances of variations of "when bacteria were discovered no-one believed they really existed" in this thread
so much of modern day science is really just patronising bullying hiding behind a thin veneer of academia and a bit of skill with words... your little article exemplifies it quite clearlyMaybe this is a good time to flag the 'Alternative Science Respectability Checklist'. It does take effort, sweat and time, but in the end you get the respect you deserve.
jan
bullying; so many "academics" just can't resist it!
Science Needs Natural Philosophers | holoscience.com | The Electric Universe
- Status
- Not open for further replies.