Unlike DOS, Windows is a multi-tasking environment. Therefore, if you
have many programs running at once, each will run slower. (You can, if
you want, apportion the CPU time unequally among them.) However, as far
as I can tell, the total time it takes for all of them to do their
calculations isn't inherently slower with Windows than DOS.
Windows-based antenna simulation programs are no different than other
Windows applications.

When running in "DOS mode" (under Windows systems prior to XP -- it's
not available in XP), you are running in a true, single-tasking DOS
environment. This is the mode you have to boot separately into when
starting the computer. But if you choose the "DOS prompt" (or "command
prompt") while running Windows, you're really running in the full
Windows environment, and emulating DOS as just another Windows task.
Programs running in this mode can't run any faster than a normal, native
Windows program, since they're also subject to the time sharing of the
multi-tasking system. There might actually be some additional overhead
from the emulation process.

The first Windows version of EZNEC, v. 3.0, ran calculations about 20%
faster than the DOS version, possibly due to a compiler change. There
was certainly no major slowing down of the calculations due to the
different operating system. The current version of EZNEC, v. 4.0, runs
up to several *times* faster than that due to code changes.

Roy Lewallen, W7EL

wrote:
. . .
On a side note with respect to some computors slowing down. Are the Windows
based computor programs
with respect to antenna modelling also subject to overload or slowing down
problems or are they some how
immune to the described problems?
. . .

Roy
Believe it or not I understand what you have posted so the info you have
provided will be good news to antenna modelers.
I really do not know how my personal antenna program works except I have a
small disc inserted that when I start the computor I have to press the F12
key when prompted and somehow it does not display the normal windows entry
but
goes straight into a DOS managed antenna modeler program. I suspect that the
disc somehow partitions the Dos emulator from the windows program and makes
it a seperate entity in a similar way that previous widows programs did
before they were bundled together.When I model closely coupled antenna
elements designs where one has to make it 80 segments a shot a slow down of
the coprocessor would be unacceptable
Regards
Art

"Roy Lewallen" wrote in message
...
Unlike DOS, Windows is a multi-tasking environment. Therefore, if you
have many programs running at once, each will run slower. (You can, if
you want, apportion the CPU time unequally among them.) However, as far
as I can tell, the total time it takes for all of them to do their
calculations isn't inherently slower with Windows than DOS.
Windows-based antenna simulation programs are no different than other
Windows applications.

When running in "DOS mode" (under Windows systems prior to XP -- it's
not available in XP), you are running in a true, single-tasking DOS
environment. This is the mode you have to boot separately into when
starting the computer. But if you choose the "DOS prompt" (or "command
prompt") while running Windows, you're really running in the full
Windows environment, and emulating DOS as just another Windows task.
Programs running in this mode can't run any faster than a normal, native
Windows program, since they're also subject to the time sharing of the
multi-tasking system. There might actually be some additional overhead
from the emulation process.

The first Windows version of EZNEC, v. 3.0, ran calculations about 20%
faster than the DOS version, possibly due to a compiler change. There
was certainly no major slowing down of the calculations due to the
different operating system. The current version of EZNEC, v. 4.0, runs
up to several *times* faster than that due to code changes.

Roy Lewallen, W7EL

wrote:
. . .
On a side note with respect to some computors slowing down. Are the
Windows
based computor programs
with respect to antenna modelling also subject to overload or slowing
down
problems or are they some how
immune to the described problems?
. . .