Maker Pro
Maker Pro

Engineering Mathematics

J

Jonathan Kirwan

It depends exactly what finite resources the 'Dozy operating system is
running out of. My copy of Win98SE has no trouble at all managing 192MB
of ram and would probably go slightly faster with 256MB if I could be
bothered.

It's running out of those I posted about:

16-bit User heap 64K
32-bit User window heap 2M
32-bit User menu heap 2M
16-bit GDI heap 64K
32-bit GDI heap 2M

How these are specifically accounted for by the summarized display of the
program called RSRCMTR.EXE, I can't exactly say. But it is typically controls
added to sheets that cause such memory leaks (when they aren't properly returned
to the system, as the visual page changes and is updated.)

In any case, these limitations have nothing whatever to do (generally) with
management or mismanagement of the totality of memory, such as the 192M or 256M,
etc. That memory isn't about the above memory limitations. Those regions I
listed above have their own special limitations that are unrelated to the
overall RAM available in the system.
The typical sorts of things that Windows trashes resources for are
multiple (redundant) loading of fonts when drawing graphs. You can force
an error reported as "not enough memory" in Excel for example by drawing
around 50 charts in one spreadsheet (no matter how much ram is free).

Interesting. I haven't tried that, of course.

In the case of Mathcad, I've tried to keep to VERY SIMPLE processes of
documenting my thought processes for applications I write, without getting fancy
about using features. Font types for text I type are limited to one, with
perhaps some italics. Mathematics is as Mathcad uses, so whatever that means.
Other than that, basic stuff.

Doesn't matter. Unusable because I can just simply watch that resource meter
memory bar drop and drop and drop. And when I close the program entirely,
saving my work, very little comes back in reply. When I restart, I restart from
a much lower starting position and I have far less time left "on the clock," so
to speak. So I have to reboot.

Doing that every hour or half-hour or hour-and-a-half just isn't my idea of a
workable program. And, sadly, I'd applied this program and patches onto four
machines, all with varying versions of Win98 on them with varying drivers. But
after long talks with Mathcad folks about these problems, I finally broke down
and reformatted a disk and completely rebuilt Win98 as a clean system, applied
Microsoft's "critical updates" onto the system, did nothing else, and then ran
Mathcad. Same results.

The several copies of Mathcad I have now sit on a shelf.
I suspect the problem here is similar. It is a bit careless to ship
software with gratuitous and large memory leaks. A better choice of OS
may help.

As I also said earlier, my work habits preclude that. I didn't get into
details, but the issue is that I have several machines -- one at my electronics
workbench that includes several ISA slots in it for old hardware cards I need to
support, one in my library area, and a laptop. I keep all my projects on
separate disks -- they are cheap to buy and I use removable drive enclosures for
the drives so that I can just slap them in. So, if I work on a product under
Linux for project A today and work on a product under Win98 for project B
tomorrow, I simply grab the project A disk and boot today and grab the project B
disk tomorrow and reboot on it.

This allows me to work on clean operating system environments for all projects,
without the required installations of software for one project causing me
problems for a different and unrelated project, allows me to operate under any
number of operating systems, etc. But the upshot of all this is that I place
copies of Win98 on many different mountable disks. Microsoft doesn't support
this mode of operation with XP, for example. I'd be calling and begging from
them every time I needed to configure a new project using their OS.

So Win98 it is.
There are registry tweaks to fix some of these flaws....

Well, Mathcad support had me doing some of those. No help there.
Anyone with the tools to say if v12 has serious memory leaks or not?

And specifically under Win98 or Win98SE???

Much obliged if anyone can say!

Jon
 
J

Jonathan Kirwan

Speaking of physics, and yeah, Hawking is some sh*t, what do you
think of that "Theory of Everything." Should I spand (waste?) any
time reading any of that?

If you mean M-Theory, I'd say what I've studied so far seems promising. You'll
need some grounding in catastrophe theory and Lie algebras and groups, but there
is a pair of excellent books on both subject areas by Robert Gilmore, which I
can heartily recommend.

Jon
 
A

Active8

If you mean M-Theory, I'd say what I've studied so far seems promising. You'll
need some grounding in catastrophe theory and Lie algebras and groups, but there
is a pair of excellent books on both subject areas by Robert Gilmore, which I
can heartily recommend.

Jon

I don't recall that it was called M-Theory but it's the one where
all forces can be approximated by the relative spins of the bodies
whether they be atomic particles, planets, stars, or galaxies.
 
J

Jonathan Kirwan

I don't recall that it was called M-Theory but it's the one where
all forces can be approximated by the relative spins of the bodies
whether they be atomic particles, planets, stars, or galaxies.

Huh? That's something described in a way that is different from anything I've
heard. Scratch my comments, then.

Jon
 
A

Active8

Huh? That's something described in a way that is different from anything I've
heard. Scratch my comments, then.

Jon

Yahoo groups has a forum on Theory of Everything and some long and
short files on the subject. You'll find a simple question I asked
under "Active8_1" that no one answered. I'm thinking it's all
bullsh*t, but WTF do I know about theoretical physics? Wee little.
 
D

ddwyer

Tuurbo46 said:
Hi

What math package do all you clever Fourier, Z and laplace people use. I
have Matlab 6.5, but im not very good at it. Do you enter sums into command
window, or is there a special toolbox? Also i have Mathmatica. Which is
the most user friendly?
As I had no other I got good at Fourier on EXCEL works well if you
follow the rules.
 
P

Pooh Bear

Active8 said:
This might come as a surprise to you, but winders 98 can only handle
(manage) 64 MB of ram, IIRC, so if you have more ram in your system,
it's wasted.

Certainly not the case for 98SE - and very highly doubtful for the original 98
either.

A tool called Cache Manager shows 88MB of memory used typically after booting my
system on 98SE.

I've certainly seen performance improvements by adding ram up to about 220MB (
128+64+32 )

Graham


I've only used Mathcad btw. Going back to the DOS versions too !
 
K

Kevin Aylward

Active8 said:
Speaking of physics, and yeah, Hawking is some sh*t, what do you
think of that "Theory of Everything." Should I spand (waste?) any
time reading any of that?

Which "Theory of Everything" would that be?

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.
 
A

Active8

On Wed, 25 Aug 2004 06:28:17 GMT, Kevin Aylward wrote:

Which "Theory of Everything" would that be?
The one that turns up in a google search and has a Yahoo group
dedicated to it. I read the short paper only so far to figure out
that they're claiming that all the forces can be explained by the
spins of particles relative to each other. So if the sides nearest
each other are spinning in the same direction, i.e., on part CW and
the other CCW, there's an attraction.

The paper also claimed that all the old physisicts like Einstein,
Bohr, etc. were headed in the right direction at first, and that
they all came up with good stuff, but they missed a road sign on the
way that would have led them to the right unifying law. But they
also claim that we'd never have found this unifying law if those
guys hadn't missed the sign in the first place.

You'd have to check it out. I'm smelling psuedoscience, but you'd be
more qualified to make that call.
 
K

Kevin Aylward

Active8 said:
On Wed, 25 Aug 2004 06:28:17 GMT, Kevin Aylward wrote:


The one that turns up in a google search and has a Yahoo group
dedicated to it.

This don't nail it down for me. I would have you actually go and look in
the groups. A simple google search gives masses of data with seems
unrelated to what you say below.
I read the short paper only so far to figure out
that they're claiming that all the forces can be explained by the
spins of particles relative to each other. So if the sides nearest
each other are spinning in the same direction, i.e., on part CW and
the other CCW, there's an attraction.

This makes no sense to me. TOE that I am vaguely familiar with are like,
string theories. They having n'out to do with spins. Its good vibrations
dude..

You need to point me the paper.
The paper also claimed that all the old physisicts like Einstein,
Bohr, etc. were headed in the right direction at first, and that
they all came up with good stuff, but they missed a road sign on the
way that would have led them to the right unifying law. But they
also claim that we'd never have found this unifying law if those
guys hadn't missed the sign in the first place.

You'd have to check it out. I'm smelling psuedoscience,

It probably is. I'll know when I see the paper.
but you'd be
more qualified to make that call.

A 10% improvement dont count for much.

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.
 
A

Active8

On Wed, 25 Aug 2004 16:08:12 GMT, Kevin Aylward wrote:

This makes no sense to me. TOE that I am vaguely familiar with are like,
string theories. They having n'out to do with spins. Its good vibrations
dude..

I know that much about strings. This is different.
You need to point me the paper.

Aw **** it, Kevin. I'll just gift wrap it for you. Yahoo sucks. One
sec... Earthlink sucks too.

http://home.earthlink.net/~mcolasono/toe/short_TOE.htm

That's a short one. If it doesn't work,

You might have to sign up for the group to get more. Sorry if I was
too vague. I forgot the Aufbau part

http://groups.yahoo.com/group/the_Aufbau_Laws_unify_everything/

the group name is

the_Aufbau_Laws_unify_everything. I had no prob finding all this
from google.

A 10% improvement dont count for much.
I don't know f'all.
 
R

Rich Grise

Active8 said:
On Wed, 25 Aug 2004 16:08:12 GMT, Kevin Aylward wrote:



I know that much about strings. This is different.

Aw **** it, Kevin. I'll just gift wrap it for you. Yahoo sucks. One
sec... Earthlink sucks too.

http://home.earthlink.net/~mcolasono/toe/short_TOE.htm

That's a short one. If it doesn't work,

You might have to sign up for the group to get more. Sorry if I was
too vague. I forgot the Aufbau part

http://groups.yahoo.com/group/the_Aufbau_Laws_unify_everything/

the group name is

the_Aufbau_Laws_unify_everything. I had no prob finding all this
from google.


I don't know f'all.
I took a look at the paper, and the guy's ideas are pretty close to
mine, but with way different terminology. We're all just big quantum
wave functions.

Cheers!
Rich
 
K

Kevin Aylward

Active8 said:
On Wed, 25 Aug 2004 16:08:12 GMT, Kevin Aylward wrote:



I know that much about strings. This is different.

Aw **** it, Kevin. I'll just gift wrap it for you. Yahoo sucks. One
sec... Earthlink sucks too.

http://home.earthlink.net/~mcolasono/toe/short_TOE.htm

Its complete and utter nonsense. Its not worth commenting on. i.e there
are 10000's of other crank theories (http://www.crank.net) just like
this drivel. The author is *completely clueless about physics.




Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.
 
M

Martin Brown

Jonathan Kirwan said:
It's running out of those I posted about:

16-bit User heap 64K
32-bit User window heap 2M
32-bit User menu heap 2M
16-bit GDI heap 64K
32-bit GDI heap 2M

How these are specifically accounted for by the summarized display of the
program called RSRCMTR.EXE, I can't exactly say. But it is typically controls
added to sheets that cause such memory leaks (when they aren't properly
returned
to the system, as the visual page changes and is updated.)

This sounds familiar and was the case with the Excel chart problems
(which MS confusingly report to end users as "not enough memory"). ISTR
it tends to be one of the two smallest regions that race to run out
first.
In any case, these limitations have nothing whatever to do (generally) with
management or mismanagement of the totality of memory, such as the 192M
or 256M,
etc. That memory isn't about the above memory limitations. Those regions I
listed above have their own special limitations that are unrelated to the
overall RAM available in the system.


Interesting. I haven't tried that, of course.

In the case of Mathcad, I've tried to keep to VERY SIMPLE processes of
documenting my thought processes for applications I write, without
getting fancy
about using features. Font types for text I type are limited to one, with
perhaps some italics. Mathematics is as Mathcad uses, so whatever that means.
Other than that, basic stuff.

Doesn't matter. Unusable because I can just simply watch that resource meter
memory bar drop and drop and drop. And when I close the program entirely,
saving my work, very little comes back in reply.

That is a bit strange. Normally when the process dies its resources are
released back to the OS. A better OS might help - like NT4 for instance.
There may even be some third party tweak products for Win98SE that tidy
up messy resource handling (but I have never tried them).

This is a long shot, but one of the Matrox graphic cards utilities is
alleged to cause GDI object leaks in other unrelated software. It might
explain why the lost memory does not get returned to the pool.
When I restart, I restart from
a much lower starting position and I have far less time left "on the clock," so
to speak. So I have to reboot.

Doing that every hour or half-hour or hour-and-a-half just isn't my idea of a
workable program. And, sadly, I'd applied this program and patches onto four
machines, all with varying versions of Win98 on them with varying drivers.

Bad luck. I am surprised that the manufacturer doesn't fix this fault.
Although it is quite possible that it is very specific to Win98 (and
older OS's) and certain graphics card drivers
The several copies of Mathcad I have now sit on a shelf.


As I also said earlier, my work habits preclude that.

It might still be worth testing it out on an old NT license or maybe
Win2000 to see if memory management is better on those. I suspect that
the problem isn't seen on modern OS's or it would have been fixed.

Regards,
 
A

Active8

Its complete and utter nonsense. Its not worth commenting on. i.e there
are 10000's of other crank theories (http://www.crank.net) just like
this drivel. The author is *completely clueless about physics.
Great, I can disregard it. I'm perfectly fine believing the universe
is a figment of my imagination. It explains everything.
 
K

Kevin Aylward

Active8 said:
Great, I can disregard it. I'm perfectly fine believing the universe
is a figment of my imagination. It explains everything.

Well, I don't go that far. I take the view that the universe exists,
even without consciousness to observe it. It gets way too complicated if
this aint the case. Like, how did the roads get built before *I* was
there to see them getting built.

There is really a subtle point about the "observer" effecting the object
being measured, (observer, being the physical equipment, not conscious
observer). It don't imply that without the "observer" the measured
object doesn't exist, only that the measurement is not would it would
have been. For example, too 1kg masses brought near each other, no
longer have a combined mass of 2 kg under GR. One can only measure mass
wrt another mass, and it turns out the relationship is entangled.

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.
 
J

Jonathan Kirwan

It might still be worth testing it out on an old NT license or maybe
Win2000 to see if memory management is better on those. I suspect that
the problem isn't seen on modern OS's or it would have been fixed.

I think you are right. I believe it, for just the reasons you say, too.
Otherwise, they'd have to address the problem.

I should mention that the tech support folks I spoke with, and I'm recalling one
particularly long discussion just now, admitted that this was a common thread to
reports from the field regarding Win98 and that he felt badly because he was
pretty sure they weren't going to address the Win98 problem in any thorough
fashion.

Sadly, as I have already mentioned, my work arrangements don't permit my using
an operating system that I cannot place on dozens of hard disks. And I cannot
afford to pay for a new license every time I create a new, mountable project
hard disk. (I have two desktops and one laptop, for all of which I use any
number of loadable hard disks.) XP, for example, precludes this business model.

I haven't tested out the idea of doing this with Win2000. If it requires going
begging to Microsoft every time I install it on a new disk, like XP does, then
that's out, too.

Jon
 
A

Active8

On Thu, 26 Aug 2004 15:36:49 GMT, Kevin Aylward wrote:
Well, I don't go that far. I take the view that the universe exists,
even without consciousness to observe it. It gets way too complicated if
this aint the case. Like, how did the roads get built before *I* was
there to see them getting built.

Doesn't matter. It's an imaginary road.

<snap>
 
K

Kevin Aylward

john said:
[clp]
There is really a subtle point about the "observer" effecting the
object being measured, (observer, being the physical equipment, not
conscious observer). It don't imply that without the "observer" the
measured object doesn't exist, only that the measurement is not
would it would have been. For example, too 1kg masses brought near
each other, no longer have a combined mass of 2 kg under GR. One can
only measure mass wrt another mass, and it turns out the
relationship is entangled.
What's that then?.
Why not 2kg?. >< 2kg?.

Its what the complicated sums
(http://www.anasoft.co.uk/physics/gr/index.html) tell you is the case:)

Roughly:

Consider a mass at a location near another mass then, move it away a
bit. It now has a different relative potential energy. This difference
in energy is manifested in a change in mass, as mass is equivalent to
energy. Its like, winding up a watch increases its mass!
Can't mass be measured by
inertia?.

The issue is how is mass defined?

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.
 
P

Paul Burke

May I put in my usual plug for The Solipsist Society here? Sadly,
membership hasn't been growing lately, and meetings have been rather
sparsely attended, indeed often inquorate.

Paul Burke
 
Top