Prev: large data sets [for "weighted" statistics?]
Next: Purge variable from within HPGCC program?
From: Dave Hayden on 18 Apr 2010 20:58
On Apr 5, 3:53 am, Andreas Möller <andreas_moellerNOS...(a)gmx.de>
wrote:

> So what kind of programs would the group like to appear for the HP
> 50G ?

Something that will indicate what function a key will execute when
you're in USR mode. For example, on the HP41, if you held a key down
for moment, the calculator would display the name of the function
assigned to the key. Keep holding the key for about a second and the
calc would display "NULL" to indicate that it wouldn't perform the
function (in case you pressed the wrong key).

Something like this is crucial to making USR mode useful, IMHO.

Ditto for the soft keys when shifted. I'd love it if pressing a shift
key changed the labels on the soft keys to show what would happen. For
example, in he VAR menu, pressing right-shift and a soft key will
recall the variable to the stack. Pressing left-shift stores the
stack to the variable. But how can a casual user remember whether
they're about to recall their precious program, or blow it away with
the contents of the stack? Pressing left-shift could change a label
for XYZ into ->XYZ. Pressing right-shift would change it to XYZ->.

The devil is in the details here of course. It's isn't always clear
what you should display, but I'd rather have an imperfect solution
than none at all.

Andreas, your just the right sort of guy to write this since you have
lots of knowledge of the internals of the calculator.

Dave

From: Bart on 19 Apr 2010 04:29
On Apr 17, 4:05 pm, Dueño de Monte <duenodemo...(a)gmail.com> wrote:
> On 11 abr, 23:22, TW <timwess...(a)gmail.com> wrote:
>
> > > I like the way Casio graphing calcs were bult back in the 70s. They
> > > were slim and ran on coin cells. Coin cells last for ever.
>
> > All of that and running on coin cells. . . will run for a minimum 2
> > years before needing a battery change.
>
> > TW
>
> I agree on this ...
>
> Daniel

You guys can't be serious, my 7000GB goes through a set of 2032's in a
couple of months (running a rather lengthy program several times a
day). When doing intensive graphing, my 50G is about the same on a set
of AAA's - but they are cheaper.
From: Andreas Möller on 19 Apr 2010 06:40
Hello,

> Ditto for the soft keys when shifted.

This would require to take over of the built-in key libs which can not
easily be accomplished as a bug in the deepest core of the O.S.
prevents this (search for infamous STO bug).

A complete take over of built-in functions requires a lot of overhead,
mainly for circumventing the bug, an example for this are my Multi-
Language-Packs (to my knowledge so far the only example of taking over
built-in functions).

Also long hold for a key is already used by the O.S., so one would
have to introduce something like "holding even longer" or whatever
seems to be useful for this.

This is by no way something easily which can be done in a little spare
time, it would be a major project.

If something like this is desired it would be a lot easier to do it in
the official ROM instead of "fighting" against the O.S. But again I
vote for bug fixing the ROM first before introducing new features and
from my point of a non native speaker allowing further translation of
the O.S. by moving hardcoded messages into message tables.

> The devil is in the details here of course.
As always ;-)

> That is an excellent idea IMO, something like that would have
> saved me some headaches when I first started using the 49g+.
My old SX manuals explain this deeply in detail, that was 20 years
ago.

Regards,
Andreas
http://www.software49g.gmxhome.de

On 19 Apr., 02:58, Dave Hayden <d...(a)larou.com> wrote:
> On Apr 5, 3:53 am, Andreas Möller <andreas_moellerNOS...(a)gmx.de>
> wrote:
>
> > So what kind of programs would the group like to appear for the HP
> > 50G ?
>
> Something that will indicate what function a key will execute when
> you're in USR mode.  For example, on the HP41, if you held a key down
> for moment, the calculator would display the name of the function
> assigned to the key.  Keep holding the key for about a second and the
> calc would display "NULL" to indicate that it wouldn't perform the
> function (in case you pressed the wrong key).
>
> Something like this is crucial to making USR mode useful, IMHO.
>
> Ditto for the soft keys when shifted.  I'd love it if pressing a shift
> key changed the labels on the soft keys to show what would happen. For
> example, in he VAR menu, pressing right-shift and a soft key will
> recall the variable to the stack.  Pressing left-shift stores the
> stack to the variable.  But how can a casual user remember whether
> they're about to recall their precious program, or blow it away with
> the contents of the stack?  Pressing left-shift could change a label
> for XYZ into ->XYZ.  Pressing right-shift would change it to XYZ->.
>
> The devil is in the details here of course. It's isn't always clear
> what you should display, but I'd rather have an imperfect solution
> than none at all.
>
> Andreas, your just the right sort of guy to write this since you have
> lots of knowledge of the internals of the calculator.
>
> Dave

From: TW on 19 Apr 2010 10:09
> This would require to take over of the built-in key libs which can not
> easily be accomplished as a bug in the deepest core of the O.S.
> prevents this (search for infamous STO bug).

We've been over this many times. :-)

Infamous implies that it is well known, notorious, or something like
that. As far as I can tell, you are the only one who has ever talked
or complained about it. This hardly qualifies as "infamous" by any
stretch of the definition. If I remember properly, it also is only a
"bug" if you try to expect it to do something that it was never
defined to do. In that case it would be a missing feature, not a bug.

TW
From: Andreas Möller on 19 Apr 2010 11:59
Hello,

I am sorry Tim, but I can not let that uncommented.

> If I remember properly, it also is only a
> "bug" if you try to expect it to do something
> that it was never defined to do.

According to Charles Patton the system has been designed for this:
http://groups.google.com/group/comp.sys.hp48/tree/browse_frm/thread/0c6f788d0ede1711/dc89bbd77ec6b03c?rnum=11&q=Library+Trouble&_done=%2Fgroup%2Fcomp.sys.hp48%2Fbrowse_frm%2Fthread%2Fc6f788d0ede1711%2F3384f3a3bc46fbc7%3Flnk%3Dgst%26q%3DLibrary%2BTrouble%26#doc_dc89bbd77ec6b03c

Quoting explicitly from the above
"... these aspects of the machine (Hash tables for converting text to
and from commands, and Message tables for defining standarized
messages) must, at the very least, have "hooks" in RAM."

In the original system (48 series) these "hooks" remain under normal
operations, e.g. storing something into a port which is not a library,
whereas in the 49 series the alternative "hooks" are zeroed out. And
unfortunately there are no "hooks" for FLASHPTRs, so overtaking them
is extremely cumbersome and in this case I agree with you that the
system has not been designed to overtake FLASHPTRs, which is a pity in
my beliefs. So in that case it would be a missing feature, IMHO.

But just because it is not used as much as the designer thought it
would be used is not an excuse for leaving the bug. And maybe it is
not used that much in the 49 series because it is buggy ;-)

> We've been over this many times. :-)
Yes. But that does not fix the bug nor its impact ;-) And it is of
this bug that makes programming system enhancements for the 49 series
so much more difficult than for the 48 series. A good example is
Raymond’s SpeedUI for the 48 series. If someone wants to do something
like this for the 49 series (GUISLV is an attempt) the O.S. treats
this like an organic body would treat a virus.

> Infamous
I get a whole bunch of German meanings when I put this in my
dictionary (notorious, mean, lousy, ill-famed, discredited,
disreputable, dingy, sleazy, dubious, opprobrious, characterless,
dishonoring, unholy, wicked, etc.). Notorious is among them, but that
is not what I meant. From my limited understanding of the English
language both words, "notorious" and "infamous" have a negative
meaning whereas the negative image of the latter one is bigger as it
is the opposite of fame in a positive way.

The best German term for it would be infamous-notorious = berühmt-
berüchtigt, which does not exist as an English word in the
dictionaries I used.
And berühmt-berüchtigt translates to either infamous or notorious, but
that’s not exactly what it means ;-)
Quoting my English teacher: "Language is so limited."

From the Compact Oxford English Dictionary of Current English:
infamous
• adjective
1 well known for some bad quality or deed.
2 morally bad; shocking.
— DERIVATIVES infamously adverb infamy noun (pl. infamies).

> As far as I can tell, you are the only one who has ever
> talked or complained about it.
As far as I know I have also been the only one complaining about the
carry bug in the Saturnator. A bug is not defined/declared by the
number of complains AFAIK, but on the impact it has on the system

When I found this bug I was shocked about it and that there was
nothing to easily circumvent it ;-) Also I call it "infamous" because
the bug has been in there for years and no attempt was ever made to
fix it. But maybe you can give me a better English term for it.

I do however understand, that it is extremely difficult to fix and
that this is not a priority and since it can be fixed from the outer
now it will probably be left as this.

And as you know I’d rather focus on extending the message tables at
the moment. Fixing this bug might be a left over as an intellectual
exercise to find out where a mistake was made while porting the
software from the 48 to the 49.

Happy programming,
Andreas
http://www.software49g.gmxhome.de
First  |  Prev  | 
Pages: 1 2 3 4 5 6 7
Prev: large data sets [for "weighted" statistics?]
Next: Purge variable from within HPGCC program?