Michael L. Dawley wrote:
>I rem'd out the power.exe and have about 21K free extended memory
>now, but I have given up trying to load EPPPD high, it's just too
>big. I really did not understand how this worked before your
>explanation. Now I see what is going on, and if I'm going to use
>EPPPD I'll have to leave it in conventional memory.
Still I would suggest that you use ctmouse (or ctmousep if it's a PS/2
mouse). Take a look at my mem/c>mem.txt (slightly edited)
MSDOS 11,517 (11K) 11,517 (11K) 0 (0K)
HIMEM 1,168 (1K) 1,168 (1K) 0 (0K)
EMM386 3,248 (3K) 3,248 (3K) 0 (0K)
COMMAND 4,864 (5K) 2,800 (3K) 2,064 (2K)
DOSMAX 256 (0K) 0 (0K) 256 (0K)
ANSI 4,208 (4K) 0 (0K) 4,208 (4K)
RAMDRIVE 1,264 (1K) 0 (0K) 1,264 (1K)
FILES 3,280 (3K) 0 (0K) 3,280 (3K)
FCBS 272 (0K) 0 (0K) 272 (0K)
WKBUFFER 528 (1K) 0 (0K) 528 (1K)
LASTDRIV 2,304 (2K) 0 (0K) 2,304 (2K)
STACKS 3,024 (3K) 0 (0K) 3,024 (3K)
INSTALL 144 (0K) 0 (0K) 144 (0K)
SMARTDRV 75,136 (73K) 0 (0K) 75,136 (73K)
CTMOUSE 6,384 (6K) 0 (0K) 6,384 (6K)
KEYB 6,944 (7K) 0 (0K) 6,944 (7K)
Ledigt 689,216 (673K) 636,448 (622K) 52,768 (52K)
Konventionellt 655,360 18,912 636,448
�vre 158,576 105,808 52,768
Reserverad 0 0 0
Ut�kat (XMS) 65,836,176 33,925,264 31,910,912
Totalt minne 66,650,112 34,049,984 32,600,128
Totalt under 1 M 813,936 124,720 689,216
St�rsta exekverbara programstorlek: 636,432 (622K)
St�rsta lediga �vre minnesblock: 48,928 (48K)
MS-DOS �r resident i det h�ga minnesomr�det.
Ok, I didn't actually edit as much as might be needed - I hope you know
what's on every line anyway (and those strange chars are Windows fault -
they look less atrange otherwise <g>).
And before anyone asks - the loss of XMS is due to the fact that SmartDRV
has the maximum it can have in DOS (it's halfed when I start Windows).
Many of the strange items loaded (stacks, lastdriv, wkbuffer etc.) are from
dosmax. (And before anyone says something - I've tried to get Arachne
online without dosmax loaded).
And the RAMDrive is for a minimum one that has a copy of my autoexec.bat
(to boot a little faster) and command.com (to exit programs a little faster).
Another interesting thing is this litle item in your memmory:
EMM386 4,144 (4K) 4,144 (4K) 0 (0K)
while mine is:
EMM386 3,248 (3K) 3,248 (3K) 0 (0K)
Probably due to that you use EMS:
> Total Expanded (EMS) 24,444,928 (23,872K
> Free Expanded (EMS)* 21,643,264 (21,136K
>
> * EMM386 is using XMS memory to simulate EMS memory as needed.
> Free EMS memory may change as free XMS memory changes.
Why are you using EMS instead of XMS?
Do you have like this in config.sys?
DEVICE=C:\DOS\HIMEM.SYS /TESTMEM:OFF
DEVICE=C:\DOS\EMM386.EXE noems highscan
dos=high,umb
Otherwise change to this, it should give you more free memmory (and XMS
instead of EMS).
And you also wrote:
>I really did not understand how this worked before your explanation.
Perhaps this is an issue we (more experienced users) should be more aware
of? (A small how-to perhaps - or a start of the SURVPC FAQ? Of course what
I wrote is far from anything that might be usefull for someone who doesn't
understand anything. Ah well just a thought).
And why do you run share.exe ? The only time I run this is when I have the
Novell Metwork started... (If you know that you need it let it be,
otherwise "rem it out" and when you find that a program doesn't work
anymore just un-rem the share line and try the program again.)
Also setver.exe is only if you want to fake another DOS version (might be
needed to run ex. xcopy.exe)
Also it's better to put as much as possible in the upper memmory IMO. So if
I could run Arachne I would put EPPPD in the upper memmory by moving
keyb.com down (the keyboard driver for you who don't know) removing dosmax
and moving ctmouse down.
I still think it would be a very good thing if ctmouse and ctmousep could
be added to the Arachne distribution package. How many people even know
that a mouse requires a driver? It wouldn't add much to the total size of
Arachne anyway.
That addition will probably be removed by adding apm.exe and unarj.exe
together - I've saved ca 14K on it sofar (but since it isn't working
properly who cares <g> (I lost the source code and needed to start all over
again, and strangely enough nothing worked directly this time - not even
compiling it as it was).
