On 2014-10-15 21:32:26 +0000, Brian Reay said:

Oregonian Haruspex wrote:
On 2014-10-11 16:27:04 +0000, gareth said:

Relying on someone's NEC program to number crunch to
produce the answer 42 is not the same as understanding
what is going on as an aid to didactification of novices.

And having the formulas in front of you aren't much help when they can't
account for the local ground, or nearby objects.

The only way to figure it out is to do it. The math is only helpful for
spherical cows in a vacuum.

He always preached that 'real hams' should write their our software and he
claims to be a software engineer.
Perhaps he could take a break from his busy Freecell playing schedule and
develop his own antenna modelling package. I am sure another slight delay
in his 18 year project to build a receiver can stand another slippage, it
will make a change from the much reused excuse of 'gear hobbing' problems.
This Rx must have more gears than my Rolex.

Writing software is generally one of the most frightfully boring tasks
that one can possibly do. I will say, though, that since I finally
bought Mathematica I am having more fun than I have in years with
software. This is mainly because it's easy to get from point A to
point B without screwing about with finding libraries, accounting for
memory allocation, and all the crap that traditional software
development entails.

Oregonian Haruspex wrote in
:

Writing software is generally one of the most frightfully boring tasks
that one can possibly do.

True, but the payoff is amazing at times. I'm coding a phase modulation
synthesiser based on Yamaha's DX7, with months full on, then slightly less
months full-off, as the only way to get free of it, and to come back at it
and see it as another person would, because I work entirely alone. Now, it si
tough, for sure, never mind 'Doctor's hours', try 'Edison's hours', sometimes
missing out whole days and nights of rest to see somethign through. Nothing
has ever imposed discipline on me like my want to make this happen, I taught
myself more than a childhood of schooling, by magnitudes, not multiples.

Now, the payoff... WHen I got past basic principles of audio and MIDI on a
PC, logarithms and bitshifts and lookup tables foe speed, etc, designed my
own realtime interpolator for MIDI data and got the pitch control engine
taking linear signals for log domain calculations, the response of pitch over
ten octaves is swift and clean, NO digital zipper noise whatsoever despite a
mere 7 bits avalaible to direct the sweep, regardless of speed. MIDI is
usually scorned for failure to acheive this, but I did it, and you'd be hard
pressed to find a commercially available synthesier that can do this. I have
a good 'analog' simulation, and a realtime variable non-linear compression
and expansion method capable of extremely realisting imitations of horn and
string sounds. This instrument has polyphony and multitimbrality enough to
allow composition for a small symphony orchestra.

It's too off-topic for me to go on here any further, but I hope this is
enough to convey the reality: that writing software, while almost
insufferably tedious at times, can lead to long moments of exhilaration like
orbital flight, it feels like achieving the building of a space shuttle in a
back yard. To be able to play a moderately realistic piano, knowing that
every part of its existence except the host machine and the coding language
used, is beyond parallel, at least for me. I think if Bach or Beethoven had
been sent a time machine with a message to the effect that they could have
had this, they'd have got in and things might have been very diferent for
music. On the other hand, it is because of what they did do that this is
possible at all...

On 2014-10-16 07:08:04 +0000, Lostgallifreyan said:

Oregonian Haruspex wrote in
:

Writing software is generally one of the most frightfully boring tasks
that one can possibly do.

True, but the payoff is amazing at times. I'm coding a phase modulation
synthesiser based on Yamaha's DX7, with months full on, then slightly less
months full-off, as the only way to get free of it, and to come back at it
and see it as another person would, because I work entirely alone. Now, it si
tough, for sure, never mind 'Doctor's hours', try 'Edison's hours', sometimes
missing out whole days and nights of rest to see somethign through. Nothing
has ever imposed discipline on me like my want to make this happen, I taught
myself more than a childhood of schooling, by magnitudes, not multiples.

Now, the payoff... WHen I got past basic principles of audio and MIDI on a
PC, logarithms and bitshifts and lookup tables foe speed, etc, designed my
own realtime interpolator for MIDI data and got the pitch control engine
taking linear signals for log domain calculations, the response of pitch over
ten octaves is swift and clean, NO digital zipper noise whatsoever despite a
mere 7 bits avalaible to direct the sweep, regardless of speed. MIDI is
usually scorned for failure to acheive this, but I did it, and you'd be hard
pressed to find a commercially available synthesier that can do this. I have
a good 'analog' simulation, and a realtime variable non-linear compression
and expansion method capable of extremely realisting imitations of horn and
string sounds. This instrument has polyphony and multitimbrality enough to
allow composition for a small symphony orchestra.

It's too off-topic for me to go on here any further, but I hope this is
enough to convey the reality: that writing software, while almost
insufferably tedious at times, can lead to long moments of exhilaration like
orbital flight, it feels like achieving the building of a space shuttle in a
back yard. To be able to play a moderately realistic piano, knowing that
every part of its existence except the host machine and the coding language
used, is beyond parallel, at least for me. I think if Bach or Beethoven had
been sent a time machine with a message to the effect that they could have
had this, they'd have got in and things might have been very diferent for
music. On the other hand, it is because of what they did do that this is
possible at all...

I appreciate the sentiment and I have in fact been excited about
writing software. I am enjoying myself tremendously now, though with
Mathematica the goal isn't exactly to create self-contained
applications, but is more akin with exploration. I am an amateur
musician myself and I really enjoy tinkering with MIDI. To my ear the
zipper noise as you change envelopes and whatnot is quite
synth-dependent. Anyway good luck and have fun!

Oregonian Haruspex wrote in
:

To my ear the
zipper noise as you change envelopes and whatnot is quite
synth-dependent.

True, some are ok with it. Kurzweil for example. There's usually a payoff
though, a sluggishness in response that can be treacly or even intolerable if
wanting the kind of response to glide a violinist wants. Mine's not perfect
either, but I got a better compromise than anything I'd heard in anything I'd
owned since 1983 when I first got one. (DX7). I just tried to eliminate
every vice that ever irked me in an instrument, note stealing in voice
allocation, for example... it has to happen, but making it sound less
conspicuous turned out to be a demanding puzzle to solve efficiently.
Likewise the non-linear slopes in just about everything, that is at the core
of why a DX7 is inherently 'natural' when so many others are not. That is fun
though, very interesting discoveries there..

Oregonian Haruspex wrote in
:

screwing about with finding libraries

I hated that part too. I looked for MIDI library code for years, on and off,
then when I finally bit the bullet and learned the core Win32 API and C, and
callback functions, etc, I found I had full control. I ended up with ALL that
I needed, within about a month. Finding libraires is overrated. I like to
build mine. I started learning about pointer iteration and dereferencing and
buffer overflows by building my own fast and safe string concatenation
function based on Theo de Radt (of OpenBSD, known for high security coding).
In many cases like these, I found that all the 'learning' in the world will
never cut to the chase as effectively as deducing what is to be done, what is
closest to doing it that is within reach, and adapting it directly. If done
thoroughly, by good effort to explore the possibilities that arise instead of
copying chunks like a script kiddie, the learning is also thorough, and the
code will become your own. Real satisfaction comes from that, enough to
counterpoise the neat reach.

"Oregonian Haruspex" wrote in message
...

Writing software is generally one of the most frightfully boring tasks
that one can possibly do.

Au contraire, it is one of the most mentally stimulating, have an idea,
jot down a prototype, and within the initial surge of excitement
of creativity, you get success, which then goes on to feed
your excitement.

Get a circuit idea, and it can be a couple of hours before you can check it
out.

Do some machining, and it can take a whole week just to work out how
to hold the workpiece in readiness for cutting!

Oregonian Haruspex wrote:

snip

Writing software is generally one of the most frightfully boring tasks
that one can possibly do. I will say, though, that since I finally
bought Mathematica I am having more fun than I have in years with
software. This is mainly because it's easy to get from point A to
point B without screwing about with finding libraries, accounting for
memory allocation, and all the crap that traditional software
development entails.

I would say that it highly depends on what the code is for and does.

The code for the Philae lander comes immediately to mind.

--
Jim Pennino