EMM for DOS, supporting up to 600MB
Felix Frey
Extender in protected mode). For that purpose one needs a memory manager. By
now I've been working with the Quarterdeck-EMM V8. It's fine but has a limit at
256MB.
Does anybody know a solution for this problem? (or any other hint)
Felix
Herman Dullink
>Extender in protected mode). For that purpose one needs a memory manager.
memory
manager certainly wouldn't help.
H
Konstantin Koll
>
> Hi, I urgently need a solution to manage up to 600MB (!) under DOS (with
> now I've been working with the Quarterdeck-EMM V8. It's fine but has a limit at
> 256MB.
>
> Does anybody know a solution for this problem? (or any other hint)
>
> Felix
Try to use a patches version of EMM386.EXE that I've created. You can
download it from http://www.deskwork.de/DPMI.ZIP
The file contains a new HIMEM.SYS and a new EMM386.EXE. I don't know it
it will work, but it"s surely woth a try.
Konstantin Koll
NEWBIE
DOS/4GW Professional Protected Mode Run-Time Version 1.97
DOS/4G from Tenberry Software is the most widely used 32-bit DOS extender,
designed to allow 32-bit DOS programs to eliminate the inherent DOS 640KB
memory limit by addressing all the memory on Intel 386 and above machines in
MS-DOS, PC-DOS, DR-DOS, the DOS boxes of OS/2, Windows, Windows NT and
Windows 95 and DOS clones.
When this program is required by 32-bit DOS software, it must be placed in
the same directory as the software's executable in order to run properly.
Requirements
MS-DOS (or DOS clones), OS/2, Windows 9x/NT
386 processor
You Can Download Here http://www.sonicspot.com/dos4gw/dos4gw.html
Or the Product Home Page http://www.tenberry.com/dos4g/
"Konstantin Koll" <ko...@ls3.cs.uni-dortmund.de> wrote in message
news:39226C46...@ls3.cs.uni-dortmund.de...
AndrewJ
>
> DOS/4GW Professional Protected Mode Run-Time Version 1.97
This is a tad irrelevent to the OP's request. He's not looking for a DOS
extender, he's looking for a memory manager. To use DOS4G/W, you also need
Watcom C to build your programs. Since Watcom is no longer being produced,
chances of getting a copy are quite slim.
Unfortunately, I don't have a solution to the problem either.
> You Can Download Here http://www.sonicspot.com/dos4gw/dos4gw.html
Incidently, the link above does not point to the "Professional" version. The
professional version costs money. DOS4G/W is the version that is distributed
with Watcom C under an unlimited use runtime license.
AndrewJ
Richard Slobod
>DOS/4GW Professional Protected Mode Run-Time Version 1.97
>
>DOS/4G from Tenberry Software is the most widely used 32-bit DOS extender,
>designed to allow 32-bit DOS programs to eliminate the inherent DOS 640KB
>memory limit by addressing all the memory on Intel 386 and above machines in
>MS-DOS, PC-DOS, DR-DOS, the DOS boxes of OS/2, Windows, Windows NT and
>Windows 95 and DOS clones.
>
>When this program is required by 32-bit DOS software, it must be placed in
>the same directory as the software's executable in order to run properly.
That's not true. You can place a single copy of DOS4GW.EXE somewhere on
your PATH; there's no need to have multiple copies sprinkled around your
harddrive.
Fabio Bizzetti
> Have you tried This
>
> DOS/4GW Professional Protected Mode Run-Time Version 1.97
>
> DOS/4G from Tenberry Software is the most widely used 32-bit DOS extender,
> designed to allow 32-bit DOS programs to eliminate the inherent DOS 640KB
> memory limit by addressing all the memory on Intel 386 and above machines in
> MS-DOS, PC-DOS, DR-DOS, the DOS boxes of OS/2, Windows, Windows NT and
> Windows 95 and DOS clones.
>
> When this program is required by 32-bit DOS software, it must be placed in
> the same directory as the software's executable in order to run properly.
>
>
>
> MS-DOS (or DOS clones), OS/2, Windows 9x/NT
> 386 processor
>
For unlimited memory support he should rather check PMode/W. I think
that DOS/4GW, other than not being free, is limited to 64MB of RAM.
BTW: does anybody know if CauseWay extender (now free) is any better
(and in which regard) to PMode/W?
Also, does anybody have a *working* email address of PMode/W authors?
Greets,
--
Fabio Bizzetti --- email: bizzetti<<at>>freemail.it
Rez
> BTW: does anybody know if CauseWay extender (now free) is any better
> (and in which regard) to PMode/W?
The head guru in the Watcom ng says Causeway is better than DOS/4GW, and
in my experience DOS/4GW apps are TONS more stable than PMode DOS apps.
Also, I've seen PMode apps leave crap in memory. Dunno how relevant that
is, but there it is anyway :)
~REZ~
NEWBIE
And the remark about it being in the apps directory Was cut and past
from the web page I Down loaded it from I keep Them For Reference
All The Files I download So I know What they are.
"Rez" <rividh.min...@earthlink.net> wrote in message
news:392F0F...@earthlink.net...
NEWBIE
Al Aab
why ?
they keep switching modes.
alternative:
(have not tried it)
programming in real mode using a trick
known among serious assebler programmers.
--
=-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-
al aab, ex seders moderator sed u soon
it is not zat we do not see the s o l u t i o n
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-+
Thomas Junge
> alternative:
> (have not tried it)
>
> programming in real mode using a trick
> known among serious assebler programmers.
...preventing usage of your application under Windows. Great solution you
propose. :-)
I used to like FRM a lot some years ago, but apparently, despite me, not
many others did :-)
Cheers,
ltn
Jukka Liimatta
> why ?
> they keep switching modes.
Yes, sometimes when using interrupts, mostly. Nothing to be alarmed of.
Really.
> alternative:
> (have not tried it)
>
> programming in real mode using a trick
> known among serious assebler programmers.
You must be referring to the so-called "flat realmode", which is a
undocumented feature in the 386 and later microprocessors. Enter protected
mode. Set addressing to flat. Return to realmode. Addressing is still flat,
it it works. Try that with a DPMI server in memory. You must boot with raw,
or himem.sys only mode -- otherwise you're F'ED up ( foo-bar ). Ditto.
Yes, I am aware my language is very professional and sophisticated. So what,
this should not really be a topic anymore, it used to be somekind of
curiosity back in 1994 or so. Extenders are way to go with MS-DOS for 32bit
programming, in short of own operating system. But that wouldn't be MS-DOS,
and you couldn't then call it an extender by definition -- it would be 32bit
shell, or whatever you want to call it.
J
AndrewJ
> why ?
> they keep switching modes.
And you think Winblowz itself isn't constantly switching modes?
Thunk thunk kludge, says windows to me.
>
> alternative:
> (have not tried it)
>
> programming in real mode using a trick
> known among serious assebler programmers.
FRM was wonderful back in the days of DOS. Y'know, back when people were
willing to reboot their machine with a boot disk, leaving only HIMEM.SYS in
memory. Nowadays no one wants to reboot their Windows machines every time they
get the urge to use that one program.
AndrewJ
Damian Yerrick
wrote:
>> DOS/4GW Professional Protected Mode Run-Time Version 1.97
>
>To use DOS4G/W, you also need Watcom C to build your programs.
>Since Watcom is no longer being produced, chances of getting a
>copy are quite slim.
Of course, the obligatory "If you want Watcom, you might also want to
look at DJGPP" post by Damian Yerrick:
If you want Watcom, you might also want to look at DJGPP:
http://www.delorie.com/djgpp/
It's a 32-bit C/C++ development system for DOS. And it's free
software.
--
Damian Yerrick
"I refuse to listen to those who refuse to listen to reason."
See the whole sig: http://www.rose-hulman.edu/~yerricde/sig.html
This is McAfee VirusScan. Add these two lines to your signature to
prevent the spread of signature viruses. http://www.mcafee.com/
RMSoft.
server probably don't like 'attatchments' or something. How I can help if I
can't
send docs in 'attachment'?
BTW, Dos Extenders are bulshits, use flatreal-mode if you want to use
memory.
--
Ivar, (c)RMSoft
http://my.tele2.ee/rmsoft
rmsoft...@mail.ee
Jukka Liimatta
> server probably don't like 'attatchments' or something. How I can help if
I
> can't
> send docs in 'attachment'?
> BTW, Dos Extenders are bulshits, use flatreal-mode if you want to use
> memory.
Great idea, have to exit windows, boot with himem.sys only, run app, reboot
back to windows. That was lame back in 1995, now it's unforgivably lame. =)
... great advice, honestly. =)
Jukka
Anatoly Muliarski
FF> Hi, I urgently need a solution to manage up to 600MB (!) under DOS (with
FF> Extender in protected mode). For that purpose one needs a memory manager. By
FF> now I've been working with the Quarterdeck-EMM V8. It's fine but has a limit at
FF> 256MB.
FF> Does anybody know a solution for this problem? (or any other hint)
Hi Felix!
There are the some ways for solution this problem.
1. Use newer version QEMM - QEMM 9.x( aka QEMM 97 )
2. Use QEMM command-line parameter USERAM=xxxxx-yyyyy for directly setting of
used RAM area( see listed below describe of QEMM parameters ).
3. Use HIMEM.SYS from Windows 98.
What DOS-extender is used? If it is PMODE/W - I wrote the small TSR which
allow to use ALL available memory( but it is worked only for RAW mode -
without HIMEM/QEMM/EMM ).
╔═════════════════════════════════════════════════════════════════════════════╗
║ QEMM386 ║
║ Copyright (c) 1986-1997 ║
║ Quarterdeck Corporation ║
║ ║
║ QEMM is an expanded memory manager for 80386, 80486 ║
║ and Pentium computers. It also provides High RAM ║
║ and other capabilities. ║
║ ║
║ Syntax: ║
║ DEVICE=QEMM386.SYS [switches] ║
║ ║
║ QEMM switches: ║
║ ADAPTERRAM=xxxx-yyyy An adapter has RAM in the range ║
║ ADAPTERROM=xxxx-yyyy An adapter has ROM in the range ║
║ AUTO Turn on only if necessary ║
║ BOOTCOLORS:N:R:B:C:S:T List of color attributes to customize ║
║ the QuickBoot Menu (i.e. for monochrome ║
║ monitors use BOOTCOLORS:7:70:7:7:7:7) ║
║ BOOTENABLE:N Disable QuickBoot ║
║ BOOTFLOPPY:N Don't check floppy first for QuickBoot ║
║ BOOTTIMEOUT=xx Countdown in seconds for the QuickBoot Menu ║
║ BOOTVIA19:N Disable QuickBoot broadcast to drivers and TSRs ║
║ COMPAQ386S Identify this as a COMPAQ 386S ║
║ COMPAQEGAROM Relocate COMPAQ's video ROM ║
║ COMPAQFEATURES:N Disable EGAROM, HALFROM, ROMMEMORY ║
║ COMPAQHALFROM Split system ROM in half ║
║ COMPAQROMMEMORY Use COMPAQ memory reserved for ROM ║
║ DISKBUF=xx Set the size of the SCSI disk buffer ║
║ DISKBUFFRAME=xx Buffer disk access into the page frame ║
║ DMA=xx Set the size of the DMA buffer ║
║ DOS4 Alter EMS page ordering for DOS 4.00 ║
║ EMBMEM=xxxxx Limit memory available as EMBs to xxxxxK ║
║ EMS:N Do not be an expanded memory manager ║
║ EXCLUDE=xxxx-yyyy Consider the range unmappable ║
║ EXCLUDEBYTES=xxxxx:nnnn Preserve nnnn bytes inside mapped ║
║ region at linear address xxxxx ║
║ EXCLUDESTEALTH=xxxx Do not Stealth a particular ROM ║
║ EXCLUDESTEALTHINT=xx Do not Stealth a particular interrupt ║
║ EXTMEM=xxxxx Reserve xxxxxK of extended memory ║
║ FASTINT10:n Do not speed up INT 10s with Stealth ║
║ FILL:N Do not fill conventional memory ║
║ FORCEEMS Still act like EMS even if no frame ║
║ FORCESTEALTHCOPY Copy tables even when eXcluded ║
║ FRAME=xxxx Set the page frame; xxxx=seg or NONE ║
║ FRAMEBUF:y/n Buffer INT 21s into the page frame ║
║ FRAMELENGTH=x Set the page frame to be x pages ║
║ GETSIZE[:f] Report the memory used if loaded high ║
║ HANDLES=xxx Set the number of EMS handles ║
║ HMA:N Do not allow access to the HMA ║
║ HMAMIN=xx Set the minimum size of the HMA (0-63K) ║
║ IBMBASIC Maintain access to the IBM Basic area ║
║ INCLUDE=xxxx-yyyy Consider the range available for high RAM ║
║ LOCKDMA Do not allow interrupts during DMA processing ║
║ MAPREBOOT:N Do not map the "reboot" page of the system ROM ║
║ MAPS=xx Set the number of alternate register sets ║
║ MEMORY=xxxxx Use only xxxxxK of extended memory ║
║ OFF Turn QEMM OFF ║
║ ON Turn QEMM ON ║
║ PENT:VME[:nn][:N] Disable Pentium int monitoring extensions ║
║ by individual soft int (:nn 0-FF) ║
║ or completely (:N) ║
║ PENT:PSE:N Disable Pentium paging extensions ║
║ RAM[=xxxx-yyyy] Put RAM everywhere or in the range ║
║ REGION=x Specify the region to load high into ║
║ RESPONSEFILE=name Specify the file to get the Region from ║
║ ROM[=xxxx[-yyyy]] Map ROM everywhere or in the range ║
║ ROMHOLES:N Do not find holes in the system ROM ║
║ SHADOWRAM:type Specify the type of ShadowRAM ║
║ (LEAP, NEAT, NEC, OPTI, PEAK, SCAT, ║
║ TOPCAT, 386, CY82C597, or NONE) ║
║ SORT:Y Sort memory by speed ║
║ STEALTHTHUNK=xxxx-yyyy Special Stealth compatiblity page ║
║ STEALTHROM:m/f Reclaim address space of ROMs ║
║ SUSPENDRESUME Support "Suspend/Resume" feature of laptop ║
║ TASKS=xx Set the interrupt nesting level ║
║ TOKENRING:N Do not look for a token ring adapter ║
║ TOPMEMORY:N Do not look for "top memory" ║
║ TRAP8042 Trap I/O to the 8042 to disable A20 ║
║ UNMAPFREEPAGES Unmap pages from the frame when free ║
║ UNUSUALEXT The extended memory BIOS is non-standard ║
║ USERAM=xxxxxxx-yyyyyyy Specify paragraphs of additional ║
║ memory that QEMM cannot otherwise find ║
║ USEXMS:N Do not allow QEMM to get memory from XMS ║
║ VCPISHARE:Y Share the page table with VCPI clients ║
║ VDS:N Disable support for Virtual DMA Services ║
║ VIDEOFILL:N Do not fill into video memory ║
║ VIDEORAM:N Do not put RAM into video memory ║
║ VIDRAMEGA Configure for EGA feature of VIDRAM ║
║ VIDRAMEMS Configure for EMS feature of VIDRAM ║
║ VIRTUALHDIRQ:N Do not suppress INT 15/90 with Stealth ║
║ VXDDIR=path Specify the directory of the .VXD files ║
║ WATCHDOG=[0,1,2] Set the type of Watchdog timer ║
║ WINDOWS3:N Do not support Windows 3 ║
║ WINSHRINKUMBS:N Do not reclaim unused High RAM for WINDOWS ║
║ XBDA:n/y/h/l/f Control whether the extended BIOS Data Area ║
║ is relocated High or Low or Not moved ║
║ XMS:N Do not be an extended memory manager ║
║ PAUSEONERROR:N Do not pause on error ║
║ PAUSE Pause while parsing commands ║
║ HELP Display this help text ║
║ ? List all commands ║
╚═════════════════════════════════════════════════════════════════════════════╝
With best regards Anatoly Muliarski
Christian Froeschlin
>
> Some day later I was send LIM4 (EMS Specs) for you in this newsgroup, but
> server probably don't like 'attatchments' or something.
> How I can help if I can't send docs in 'attachment'?
Several possibilities:
(1) Make available via FTP or HTTP and post link
(2) Post to a binary newsgroup
(3) Use E-Mail to send the file to the person interested.
Posting large data to millions of (potential) recipients
is considered to be a waste of bandwidth.
--
__/ __/ - Christian Froeschlin -
/ /
/ _/ * If you eliminate the impossible, whatever remains, *
___/ _/ * however improbable, must be the bug ! *
piter pasma, ritz[rvl]
>
> > Some day later I was send LIM4 (EMS Specs) for you in this newsgroup, but
> > server probably don't like 'attatchments' or something. How I can help if
> I
> > can't
> > send docs in 'attachment'?
> > memory.
>
> Great idea, have to exit windows, boot with himem.sys only, run app, reboot
> back to windows. That was lame back in 1995, now it's unforgivably lame. =)
>
> ... great advice, honestly. =)
>
now you have can create a nice icon for that purpose, btw.. 'MS-DOS
MODUS'
not that hard, really.. it automagically reboots the computer (with any
configuration you can specify with autoexec.bat and config.sys) and
starts
the program (or command.com) and when finished, even restarts and goes
back
to windows. i mean, come on.. if people bother to download an amiga emu
to
watch amiga demos, it shouldn't be to much trouble to restart your pc
for
MS-DOS modus...
or are you just all afraid that the pathetic losers who can't figure out
how to configure their system right can't see your demo? HMZ.
enough with the DOS-bashing already.. and if anyone complains just a
small
bit about windows in general, all those API-freaks jump up and start
flaming
him.. please think back about all of your opinions about this OS a few
years
ago... are you all really that hypocrite?
now please flame me for remembering on which OS i coded my first plasma.
--
ritz[revolution]
ritz(at)cyberjunkie . com
a...@mail1.bet1.puv.fi
>
> now please flame me for remembering on which OS i coded my first plasma.
Flame? No, I'd like to thank you for not posting wind0ze
stuff to DOS programming groups.
Rez
>
> On Fri, 26 May 2000 14:20:32 GMT, "AndrewJ" <lumin...@home.com>
> wrote:
>
> >> DOS/4GW Professional Protected Mode Run-Time Version 1.97
> >
> >To use DOS4G/W, you also need Watcom C to build your programs.
> >Since Watcom is no longer being produced, chances of getting a
> >copy are quite slim.
Greymatter in England may still have a few copies, they had some sort of
oddball pipeline and had copies of 11.0b as late as a couple months ago.
Also there was a way to get it directly from Sybase if you called their
sales number (don't have that handy), evidently there was a loophole in
the marketing dept.
I picked up v10.6 (new enough for our group's uses) for $40 a while
back, still shrinkwrapped.
> Of course, the obligatory "If you want Watcom, you might also want to
> look at DJGPP" post by Damian Yerrick:
Canned response #17654 <bseg>
> If you want Watcom, you might also want to look at DJGPP:
> http://www.delorie.com/djgpp/
> It's a 32-bit C/C++ development system for DOS. And it's free
> software.
Has anyone worked out a method of mixing DJGPP and Causeway or DOS4GW?
Personally, I never want to see CWSDPMI again. (insert canned tirade
about its disk thrashing and memory leak :)
~REZ~
bma...@iglou.com
>> BTW, Dos Extenders are bulshits, use flatreal-mode if you want to
>>use memory.
>Great idea, have to exit windows, boot with himem.sys only, run app,
>reboot back to windows. That was lame back in 1995, now it's
>unforgivably lame. =)
>.... great advice, honestly. =)
>Jukka
Net-Tamer V 1.08X - Test Drive
RMSoft.
news:8gsk0b$2c7$1...@news.kolumbus.fi...
> Great idea, have to exit windows, boot with himem.sys only, run app,
reboot
> back to windows. That was lame back in 1995, now it's unforgivably lame.
=)
>
Why you mess with windows, isn't those bewsgroups named:
"alt.msdos.programmer" and "comp.os.msdos.programmer" ????
or those names are simple for joke ?
RMSoft.
news:8gsk0b$2c7$1...@news.kolumbus.fi...
> Great idea, have to exit windows, boot with himem.sys only, run app,
reboot
> back to windows. That was lame back in 1995, now it's unforgivably lame.
=)
Also:
If you write programs for windows, then please explain why you need
Dos-Extenders. You must use win32 API's or MFC or SDK/DDK, otherwise
you make unforgivably lame. :-)
Damian Yerrick
<rividh.min...@earthlink.net> wrote:
>Has anyone worked out a method of mixing DJGPP and Causeway
>or DOS4GW? Personally, I never want to see CWSDPMI again.
>(insert canned tirade about its disk thrashing and memory leak :)
CWSDPMI is copylefted free software. Want to help
Charles Sandmann fix those bugs? He'd love your patches.
Damian Yerrick
<ri...@nospam.cyberjunkie.spam.com> wrote:
>[snip description of MS-DOS Mode in Win9x pif settings]
>i mean, come on.. if people bother to download
>an amiga emu to watch amiga demos, it shouldn't be to much
>trouble to restart your pc for MS-DOS modus...
Or better yet, download a pc emu to watch MS-DOS demos,
such as free Bochs.
>or are you just all afraid that the pathetic losers who can't
>figure out how to configure their system right can't see your
>demo? HMZ.
Or they may be running on operating systems that don't support
rebooting to MS-DOS mode (for example, Windows NT/2000 and Windows
Millennium). This will become a significant issue when WinME
is published in 2001.
Damian Yerrick
wrote:
>Some day later I was send LIM4 (EMS Specs) for you in this
>newsgroup, but server probably don't like 'attatchments' or
>something. How I can help if I can't send docs in 'attachment'?
The proper way to attach docs to a newsgroup post is to put them
on a publicly accessible pull-based service (e.g. XOOM.com Media
Sharehouse) and post a link.
>BTW, Dos Extenders are bulshits, use flatreal-mode
>if you want to use memory.
Your app will have a very limited market, as memory managers
(e.g. emm386, qemm, Windows) don't allow FRM.
Jukka Liimatta
> "alt.msdos.programmer" and "comp.os.msdos.programmer" ????
> or those names are simple for joke ?
Yes, they are perfect joke.
Jukka
(ps. I read the posting from comp.sys.ibm.pc.demos, didn't bother looking at
the header, cross-posting is stupid too ).
Fabio Bizzetti
says...
Yeah, but making demos for Windows ain't the most intelligent
thing on earth as well. ;)
Fabio Bizzetti
Bullcr_p...@hotmail.comRemoveBullcr_p says...
> >or are you just all afraid that the pathetic losers who can't
> >figure out how to configure their system right can't see your
> >demo? HMZ.
>
> Or they may be running on operating systems that don't support
> rebooting to MS-DOS mode (for example, Windows NT/2000 and Windows
> Millennium). This will become a significant issue when WinME
> is published in 2001.
Ok for Windows2000, but if Windows2000ME is based on Windows9x,
then there must be a way to use real Dos there. Win9x is a patch
upon Dos, after all.
>
>
Jukka Liimatta
> thing on earth as well. ;)
I believe you forgot to mention the great alternative.
Jukka
Fabian Giesen
> thing on earth as well. ;)
forgive me,
bust MOST PEOPLE WHO SAY HOW TRASHY WINDOWS CODING IS DIDN'T EVEN
BOTHER TO TRY IT. *ahem*
-ryg
Frans Bouma
news:MPG.139dd0ef5...@news.infinito.it...
> Yeah, but making demos for Windows ain't the most intelligent
> thing on earth as well. ;)
man you sure as hell are the real life proof why *.it is banned on #java
O.
Damian Yerrick
check_my_signature_for_my_real_email_address@I_hate_spammers.it (Fabio
Bizzetti) wrote:
I've heard you need a bootdisk to get to MS-DOS mode.
Damian Yerrick
>Yeah, but making demos for Windows ain't the most intelligent
>thing on earth as well. ;)
The Allegro library makes DirectX demos nearly as easy as DOS demos:
http://www.talula.demon.co.uk/allegro/
Jan Just Keijser
>On Sun, 28 May 2000 21:41:20 +0300, "RMSoft." <rmsoft...@mail.ee>
>wrote:
>
>>Some day later I was send LIM4 (EMS Specs) for you in this
>>newsgroup, but server probably don't like 'attatchments' or
>>something. How I can help if I can't send docs in 'attachment'?
>
>The proper way to attach docs to a newsgroup post is to put them
>on a publicly accessible pull-based service (e.g. XOOM.com Media
>Sharehouse) and post a link.
>
>>BTW, Dos Extenders are bulshits, use flatreal-mode
>>if you want to use memory.
>
>Your app will have a very limited market, as memory managers
>(e.g. emm386, qemm, Windows) don't allow FRM.
Nor do some/most newer CPU's that support SMM: I've read a c't article back in
1996 stating that FRM didn't work with a new CPU which had SMM capabilities (I
believe it was the 486sl they tested...) c't claimed it had something to do
with an SMM interrupt not restoring the 32bit GDT/IDT entries for the fs/gs
registers...
JJK
----------------------------------------------------------------
Jan Just (JJ) Keijser -- Release Engineer & Unix Hacker
http://www.sightpath.com:
Cutting-edge Solutions for Intelligent Web Content Delivery
Your mouse has moved. Windows must be restarted for
the change to take effect. Reboot now? [OK]
My views are my own...
flames > /dev/null 2>&1
---------------------------------------------------------------
Rez
> If you write programs for windows, then please explain why you need
> Dos-Extenders. You must use win32 API's or MFC or SDK/DDK, otherwise
> you make unforgivably lame. :-)
Tho a DOS app that uses a DOS extender may run under Win32 using
Windows' DPMI, so not entirely lame if the app is intended to run both
ways. Actually quite nice for we who like to run stuff both in plain DOS
and under Win32, without having to start or quit Windows to do so.
~REZ~
Rez
>
> >Has anyone worked out a method of mixing DJGPP and Causeway
> >or DOS4GW? Personally, I never want to see CWSDPMI again.
> >(insert canned tirade about its disk thrashing and memory leak :)
> CWSDPMI is copylefted free software. Want to help
> Charles Sandmann fix those bugs? He'd love your patches.
I reported the memory leak problem, but he said it wasn't possible even
tho we'd documented it on 2 systems and had a report of it happening on
a 3rd. He said it had to be something wrong with NWDOS (the system where
it's easiest to document). Well, the 3 systems ran NWDOS7, M$DOS 7,
M$DOS 6.22. Kinda eliminates NWDOS as the cause, eh?
Anyway, at that point I shrugged and went away. Am not a coder so I
can't very well fix it. If he knew our testing team.. well, coders who
work with us learn to listen, cuz when we say we've found a bug, we're
nearly always right. :)
I've seen evidence of leaks with PMode on NWDOS7 and M$DOS7, too.
I've also caught CWSDPMI writing a zero-byte file on C:\ even tho I have
disk swapping turned off via CWSPARAM (or whatever the util for that is
called). Don't remember if I reported that or not.
~REZ~
Damian Yerrick
<rividh.min...@earthlink.net> wrote:
>Damian Yerrick wrote:
>>
>> >Has anyone worked out a method of mixing DJGPP and Causeway
>> >or DOS4GW? Personally, I never want to see CWSDPMI again.
>> >(insert canned tirade about its disk thrashing and memory leak :)
>> CWSDPMI is copylefted free software. Want to help
>> Charles Sandmann fix those bugs? He'd love your patches.
>
>I reported the memory leak problem, but he said it wasn't possible even
>tho we'd documented it on 2 systems and had a report of it happening on
>a 3rd. He said it had to be something wrong with NWDOS (the system where
>it's easiest to document). Well, the 3 systems ran NWDOS7, M$DOS 7,
>M$DOS 6.22. Kinda eliminates NWDOS as the cause, eh?
>
>Anyway, at that point I shrugged and went away. Am not a coder so I
>can't very well fix it. If he knew our testing team.. well, coders who
>work with us learn to listen, cuz when we say we've found a bug, we're
>nearly always right. :)
Here's what you do: Write an "exploit" for the bug. Write a short
but complete program that makes CWSDPMI use up all of a certain
resource (but doesn't do the same on other DPMI extenders), and
specify what DOS you tested it on. This will convince Mr. Sandmann
that the bug exists.
RMSoft.
news:393420...@earthlink.net...
Yes, I know what you mean... BUT lets remember some basic truths:
If B.Gates say what its's OS is compatible with MS-Dos (written by
M.Zbikowski), then you must not believe this crazy talk. I have many Dos
programs in my computer those can't run in Win-Dos box (I was written
such programs also myself), because W-Dos-box is totally pseudo emulation
and *far* from MS-Dos. In NT it is even more *far*.
If you write in asm then you must know this fact.
And also you don't need to care about Dos at all if you are win programmer.
Gaffer
> >thing on earth as well. ;)
>
> The Allegro library makes DirectX demos nearly as easy as DOS demos:
> http://www.talula.demon.co.uk/allegro/
The OpenPTC library makes DirectX demos *easier* than DOS demos:
(arse)
>B)
a...@mail1.bet1.puv.fi
> On Tue, 30 May 2000 18:33:58 +0200,
> check_my_signature_for_my_real_email_address@I_hate_spammers.it (Fabio
> Bizzetti) wrote:
>
> >Ok for Windows2000, but if Windows2000ME is based on Windows9x,
> >then there must be a way to use real Dos there. Win9x is a patch
> >upon Dos, after all.
>
> I've heard you need a bootdisk to get to MS-DOS mode.
Right. If W2K/WME somehow infested your hard drive you
definitely need a boot disk. Not just any kind of boot disk,
but an OS installation boot disk.
Fabio Bizzetti
says...
> > Yeah, but making demos for Windows ain't the most intelligent
> > thing on earth as well. ;)
>
Hmm.. big respect to Linux, but I don't want to swap a big
brother for another (where with "big brother" I don't mean
something like Bill Gates, but something like an OS). I think
that big part of the fun in coding demos was to program the
hardware by yourself.. always thought so, and will probably
always think so.
Otherwise let's follow that new-age philosophy and let's all
code Java s*it, no?
Just my opinion of course.. but it's too bad that (due to the
nature of the PC) too few people in this platform have the
"hardware bashing" mentality so common in the 8bit, 16bit and
32bit (Amiga and Archimedes) machines.. a beautiful philosphy
IMHO. Good coders get formed coding down to the metal.. the
others (e.g. Java programmers) often don't even know how a
computer works inside. I'm one of those (and I think I've the
titles to speak, due to my past and current work) that thinks
that hardware ignorance doesn't help software skills.
Greets,
--
Fabio Bizzetti --- email: bizzetti<<at>>freemail.it
Fabio Bizzetti
> > Yeah, but making demos for Windows ain't the most intelligent
> > thing on earth as well. ;)
>
> bust MOST PEOPLE WHO SAY HOW TRASHY WINDOWS CODING IS DIDN'T EVEN
> BOTHER TO TRY IT. *ahem*
I did, and unfortunately do as well... and unfortunately++ will
have to do too, until the world finally gets rid of Microslow.
Greets,
--
Fabio Bizzetti --- email: bizzetti<<at>>freemail.it
>
> -ryg
>
>
>
Fabio Bizzetti
> "Fabio Bizzetti" <check_my_signature_for_my_real_email_address@I_hate_spammers.it> wrote in message
> news:MPG.139dd0ef5...@news.infinito.it...
> > thing on earth as well. ;)
>
Oh well, I take it as a big compliment, brother! :))))
Cheers ;D
--
Fabio Bizzetti --- email: bizzetti<<at>>freemail.it
>
> O.
>
>
>
Fabio Bizzetti
Bullcr_p...@hotmail.comRemoveBullcr_p says...
> Fabio Bizzetti wrote:
>
> >Yeah, but making demos for Windows ain't the most intelligent
> >thing on earth as well. ;)
>
> http://www.talula.demon.co.uk/allegro/
I don't contest that Windows is relatively "easy" (it probably took
me much less time to write a DIB/DirectDraw/D3D wrapper than to write
half of my Dos routines), but here the point is not "easyness".. but
rather fun in programming, and "elegance" of the final results.. and,
we may add, coding challenge.
That is, using the S.E.U.C.K. is probably easier than to write your
own shoot'em up, but it's less rewarding, and the final results
is much poorer as well. Unfortunately the competition is not fair,
because Windows has a lot of drivers.. and the hardware manufacturers
don't even let you take a look at those h/w programming docs that
would let you write better drivers for your own OS (I think that Dos,
as a OS, megasucks, but at least it allows you to program the hardware
by yourself: this is the only quality of Dos, and it's a huge one.
Ok, you can do that also in Linux.. but it's *still* Linux, i.e.
another OS, not your self-made and self-tuned one).
I also add (about API's, e.g. 3D API's): where is the innovation,
if all you've to do is to use this function or that function?
Everybody "old" enough to have lived the 8bit or Amiga eras know
how much funnier and better was the coding life at those times..
there was always a programming challenge. Now it has been all
ruined by the BUSINESS. Or not?
IMHO the solution would be to divide the programming world into
two distinct cathegories:
1) utilities and non-too-much-hardware-demanding games
2) much-hardware-demanding games and demos
For 1) I'd be happy to use a widespread standard (e.g. Unix,
and thus Linux). For 2) instead, you'd just have access to the
hardware (shutting down the OS in a legal way, if necessary),
and get fun with it. Of course the anarchy in gfx/snd boards
on the *PC* would make necessary also some degree between
these two scenarios. All that is necessary is anyway the
*mentality* to do it, and thus to e.g. release in the public
domain the hardware programming docs of the various chips
(they "protect" them as if they were industrial secrets.. but
they megasuck after all (once you sign an NDA and see how much
normal are those chips after all), and anyway you go to the
patents office (www.uspto.com) and get all the "industrial
secrets" you may want to know about those chips.. but they
anyway won't give you the *hardware programming docs*, meaning
only (to me at least) that Bill Gates years ago kissed their
asses for not letting everybody code the drivers by themselves,
and thus force us all to buy that Winslows crap).
About this 1) / 2) "modes", as some of you may know, the Amiga
and its OS were following this philosophy, and it worked great
(don't any of you "ex" Amiga coders here have big nostalgia?).
IMO (and I know quite well, I was the designer of the AgaEXTENDER
and other stuff, and have been in contact with the managers and
engineers there) the Amiga died because the marketing guys didn't
want to invest in new hardware (new chips).. and not the other
way round (i.e. that the philosophy of the Amiga created
compatibility problems that didn't allow expansion), as some
were saying around to justify some very lame choices of the
new Amiga Inc. owners.
Anyway, enough off topic for today?
Regards,
--
Fabio "Maverick" Bizzetti --- email: bizzetti<<at>>freemail.it
Frans Bouma
> In article <8h0vn6$4m1$1...@news1.xs4all.nl>, per...@xs4all.nl says...
> > "Fabio Bizzetti" <check_my_signature_for_my_real_email_address@I_hate_spammers.it> wrote in message
> > news:MPG.139dd0ef5...@news.infinito.it...
> > > thing on earth as well. ;)
> >
>
> Oh well, I take it as a big compliment, brother! :))))
I know you would :)
hehehehe
O.
Rez
> > I've heard you need a bootdisk to get to MS-DOS mode.
>
> Right. If W2K/WME somehow infested your hard drive you
> definitely need a boot disk. Not just any kind of boot disk,
> but an OS installation boot disk.
I know someone who managed to get W2K to agree to dual boot without
nuking the boot sector every time it shuts down, using some workaround
with a linux boot manager. Will have to get more details -- I'd think it
should work to dual-boot to DOS, too.
~REZ~
Rez
>
> On Tue, 30 May 2000 20:07:17 GMT, Rez
> <rividh.min...@earthlink.net> wrote:
>
> >Damian Yerrick wrote:
> >>
> >> >Has anyone worked out a method of mixing DJGPP and Causeway
> >> >or DOS4GW? Personally, I never want to see CWSDPMI again.
> >> >(insert canned tirade about its disk thrashing and memory leak :)> >> CWSDPMI is copylefted free software. Want to help
> >> Charles Sandmann fix those bugs? He'd love your patches.
>>I reported the memory leak problem, but he said it wasn't possible even
>>tho we'd documented it on 2 systems and had a report of it happening on
>>a 3rd. He said it had to be something wrong with NWDOS (the system where
> >it's easiest to document). Well, the 3 systems ran NWDOS7, M$DOS 7,
> >M$DOS 6.22. Kinda eliminates NWDOS as the cause, eh?
> >
> >Anyway, at that point I shrugged and went away. Am not a coder so I
> >can't very well fix it. If he knew our testing team.. well, coders who
> >work with us learn to listen, cuz when we say we've found a bug, we're
> >nearly always right. :)
>
> Here's what you do: Write an "exploit" for the bug. Write a short
> but complete program that makes CWSDPMI use up all of a certain
> resource (but doesn't do the same on other DPMI extenders), and
> specify what DOS you tested it on. This will convince Mr. Sandmann
> that the bug exists.
LIS I'm not a coder, but I've been watching this happen for two years
now with our DJGPP-built DOOM source mods, and our 3rd example above was
with Emacs. It does NOT happen using DRDOS's DPMI and the same app (the
disk thrashing also goes away). I believe I pointed all this out to CWS.
Also that the problem is in GO32V2 as well, but is not in GO32v1. You'd
think that alone would be enough to isolate it! (I haven't tried the new
version of CWSDPMI yet.)
Funny how since I stopped using CWSDPMI, the system crashes in DOS went
away too. I am of the "never reboot unless you need to change configs"
school of computing <g> and pre-CWSDPMI it wasn't unusual for the
DOS/Win system to be up for 5-6 weeks at a crack w/o a reboot.. and then
only to change boot configs.
~REZ~
Jukka Liimatta
>
> Hmm.. big respect to Linux, but I don't want to swap a big
> brother for another (where with "big brother" I don't mean
> something like Bill Gates, but something like an OS). I think
> that big part of the fun in coding demos was to program the
> hardware by yourself.. always thought so, and will probably
> always think so.
Linux - almost there. XF4/DRI finally shipped, and I even got drivers for
GeForce. Infact, sir, I have programmed multiplatform for a while now. Stuff
I write compiles as-is for BeOS, Linux and Windows.
I find the Windows most appealing at the moment simply thanks to sheer
number of installed copies ( which also have decent drivers for API's we
use ).
MS-DOS definitely is not an option for me anymore. This is what I had in
mind. The same comment applies to Linux and BeOS: "using flatmode addressing
is a poor advice, unless it's a lot of rebooting you are after". Let's not
get carried away from the topic at hand ..
> Otherwise let's follow that new-age philosophy and let's all
> code Java s*it, no?
No, I don't prefer Java. I don't feel it's time is ripe yet for prime time
for what I do. For some other people it might be a great alternative,
though.
Why you bring java up in this thread? This is about how stupid using flat
realmode addressing is in this date. I know it must feel good when you think
you're doing something new and unique. The fact is, this is old hats and
long forgotten those who gone that road already.
> Just my opinion of course.. but it's too bad that (due to the
> nature of the PC) too few people in this platform have the
> "hardware bashing" mentality so common in the 8bit, 16bit and
> 32bit (Amiga and Archimedes) machines.. a beautiful philosphy
So what. Nothing to do with the topic.
> IMHO. Good coders get formed coding down to the metal.. the
> others (e.g. Java programmers) often don't even know how a
> computer works inside. I'm one of those (and I think I've the
> titles to speak, due to my past and current work) that thinks
> that hardware ignorance doesn't help software skills.
Been there. Done that. Don't miss any of that. Fabio, start writing software
for Playstation2 -- sounds like perfect match for you.
Jukka
Jukka Liimatta
> if all you've to do is to use this function or that function?
What do you propose to fix the situation? And why should the situation be
"fixed" in the first place? Progress goes on - it is influenced by, duh,
people who have influence. Ranting in usenet is hardly what I call
"influence", it's entertainment. =)
If you want to do something, you need a great innovation. Then you need to
back it up, and make it well known enough to be considered worth supporting
for by the major players in the industry. Considering people cannot agree
which is better, OpenGL or Direct3D.. good luck...
> Everybody "old" enough to have lived the 8bit or Amiga eras know
> how much funnier and better was the coding life at those times..
> there was always a programming challenge. Now it has been all
> ruined by the BUSINESS. Or not?
I don't feel anything has been "ruined" per-se, we can still use the old
programming techniques we accustomed to, and skip the hardware accelerated
rendering if we feel it best serves in the end. What's the problem? If it is
a problem that less and less people use MS-DOS, it's a funny way to define
"the problem".
Sure thing, MS-DOS would theoretically give "more performance" to individual
application. However, the way most people use computers novadays is that
they have many programs running. I am currently running PIRCH, Outlook,
IE5.5, DeveloperStudio and have some notes open in notepad in other monitor.
I could also be playuing CD, or MP3 file, or downloading something off the
net. I know for a fact that MS-DOS is not operating system that would best
use the resources of my hardware, even if these services would cut a small
pie of the processor power towards this end. All things considered, I feel
that the loss of processor power is acceptable.
> IMHO the solution would be to divide the programming world into
> two distinct cathegories:
Divide? Why? Each programming team is it's own entity, and has it's own
goals. That's more than enough division to my taste.
> 1) utilities and non-too-much-hardware-demanding games
> 2) much-hardware-demanding games and demos
What purpose would this "division" serve? I don't quite franky get the
point -- is this somekind of a system how to score respect points from the
other programmers?
I have nothing to comment on what was cut below this point.
J
Jukka Liimatta
> > definitely need a boot disk. Not just any kind of boot disk,
> > but an OS installation boot disk.
>
> I know someone who managed to get W2K to agree to dual boot without
> nuking the boot sector every time it shuts down, using some workaround
> with a linux boot manager. Will have to get more details -- I'd think it
> should work to dual-boot to DOS, too.
I use the NT boot manager as master boot manager, it lists:
* Windows2000
* Linux
Choosing "Linux" pops up the LILO. I used a small utility to add LILO to the
bootmenu, I fount it at tucows.com, I think -- it was a while ago,
apologies.
J
Fabio Bizzetti
rividh.min...@earthlink.net says...
> > definitely need a boot disk. Not just any kind of boot disk,
> > but an OS installation boot disk.
>
> I know someone who managed to get W2K to agree to dual boot
> without nuking the boot sector every time it shuts down,
What's the difference between Windows2000 and a virus?
Windows2000 takes much more time to load..
Fabio Bizzetti
says...
> > I also add (about API's, e.g. 3D API's): where is the innovation,
> > if all you've to do is to use this function or that function?
>
> What do you propose to fix the situation? And why should the situation be
> "fixed" in the first place? Progress goes on
Yes, but future is not always "progress".
Or MsDos 4.0 is better than MsDos 3.11?
Or Windows2000 is better than AmigaOS?
> > Everybody "old" enough to have lived the 8bit or Amiga eras know
> > how much funnier and better was the coding life at those times..
> > there was always a programming challenge. Now it has been all
> > ruined by the BUSINESS. Or not?
>
> I don't feel anything has been "ruined" per-se, we can still use the old
> programming techniques we accustomed to, and skip the hardware accelerated
> rendering if we feel it best serves in the end. What's the problem? If it is
> a problem that less and less people use MS-DOS, it's a funny way to define
> "the problem".
>
> Sure thing, MS-DOS would theoretically give "more performance" to individual
> application. However, the way most people use computers novadays is that
> they have many programs running. I am currently running PIRCH, Outlook,
> IE5.5, DeveloperStudio and have some notes open in notepad in other monitor.
>
> I could also be playuing CD, or MP3 file, or downloading something off the
> net. I know for a fact that MS-DOS is not operating system that would best
> use the resources of my hardware, even if these services would cut a small
> pie of the processor power towards this end. All things considered, I feel
> that the loss of processor power is acceptable.
Exactly. What I was ranting about is that we should get both
the multitasking for utilities/applications, but also *retain*
the FREEDOM to code OUR computer, that we payed big buck$ for,
having all the docs to do it, in the way we most prefer.
Refer to my 1) and 2) "modes" below.
That is, if I need/want a little program to turn on some relay
switches to control some stuff in my house, tell me why should
I be forced to ignorance (why dont they even give anymore a
lousy Basic interpreter with the computers?), and instead pay
somebody else (a smartass probably) to make this program for me?
Somebody that e.g. signed a NDA with the big brother, and thus
is allowed to program.. but under the control of the big brother.
I dont pretend that they throw at me manuals with all the docs
to program the computer by myself.. but at least that they dont
DENY them when I ask for them. Nowadays all this stuff is secret
*because Microsoft wanted to have a monopoly on hardware support*,
and they probably had the mean$ to convince the hardware
manufacturers to not disclose any of this information anymore.
That is, we're slaves of corporations that have access to all
these informations. If it wasn't so, even e.g. Linux would have
many more sndcard/gfxcard drivers than it has. What is missing
is not the will (or the ABILITIES) to write good drivers, but
the hardware programming docs. It's a mafia.
> > 1) utilities and non-too-much-hardware-demanding games
> > 2) much-hardware-demanding games and demos
>
> What purpose would this "division" serve? I don't quite franky get the
> point -- is this somekind of a system how to score respect points from the
> other programmers?
I dont know from where your posting originates (there's
crossposting, you know), but if you know what's a demo, and
the gold times of the demo scene (expecially on the Amiga,
since the PC never had such a significative demo scene), then
you'll very well know the importance of this distinction, and
to disable multitasking and all the like *in certain specifical
circumstances*.
The world is not monochromatic: just because you got a car
capable of running (e.g. in case of emergency, or just of
fun), it doesn't mean that you've ALWAYS to run. But it doesn't
either mean that you don't have to run, never, never.. ever.
Same for programming.. I spend 95% of my computer use under
multitasking, using utilities.. I would never want to e.g.
use a monotasking OS for that.
But, still, when I need to concentrate all the power and
"syncronization" resources of my hardware on a specific
realtime application, I want to be able to take over the
machine and program it for this application in the *only*
proper way, which is the opposite of the normal scenario.
What's wrong in having a choice?
Regards,
--
Fabio Bizzetti --- email: bizzetti<<at>>freemail.it
Gaffer
> > others (e.g. Java programmers) often don't even know how a
> > computer works inside. I'm one of those (and I think I've the
> > titles to speak, due to my past and current work) that thinks
> > that hardware ignorance doesn't help software skills.
you ignorant fuck, dont you dare classify java programmers as clueless
idiots.
its possible to call assembler routines from inside java, yet still provide
fallback java routines so the code runs everywhere. simple rule you should
follow fabio, if you dont know anything about the subject, shut the fuck up
before you make yourself look very stupid to anybody who does.
the only idiots nowadays are people who use the wrong tool for the job. if
you use assembler to do something that is easier done in c/c++/java you are
a damned fool.
conversely, if you truly understood java and scripting languages. you'd see
that they can really play a useful role in a *real world* project.
the best of both worlds? object management and high level in java, plus
tight c/asm inners? i think so, but thats only my opinion...
Frans Bouma
> In article <8h3teo$gte$1...@news.kolumbus.fi>, ju...@twilight3d.com
> says...
> > > I also add (about API's, e.g. 3D API's): where is the innovation,
> > > if all you've to do is to use this function or that function?
> >
> > What do you propose to fix the situation? And why should the situation be
> > "fixed" in the first place? Progress goes on
>
> Yes, but future is not always "progress".
>
man... how can you compare windows2000 with an OS that didn't even
have memory protection.
(forgive me, I programmed amiga for years)
> Exactly. What I was ranting about is that we should get both
> the multitasking for utilities/applications, but also *retain*
> the FREEDOM to code OUR computer, that we payed big buck$ for,
> having all the docs to do it, in the way we most prefer.
do you care if that 'computer' is the plain hardware or
a blackbox? if you call a function and that function does
exactly what you want with the hardware, what's the difference
with jamming a number in some memory address (where for example
a videoregister is located).
I used to enjoy programming on the metal (hell, I liked my
32buffer MFM trackloader with blitter MFM decoding on the amiga,
but it's rather silly actually... to code your own MFM decoding before
you can use the data..)
> I dont know from where your posting originates (there's
> crossposting, you know), but if you know what's a demo, and
> the gold times of the demo scene (expecially on the Amiga,
> since the PC never had such a significative demo scene), then
> you'll very well know the importance of this distinction, and
> to disable multitasking and all the like *in certain specifical
> circumstances*.
yeah, but deep down you know this is not needed.
if you need real time feed back, you should take a real time OS.
(like os/9?). True, a demo on amiga for example killed the
OS in the first 2 lines of code and then started working.
but is this truely needed? today, on the PC (in windows) I
don't do that. it hangs in a messagepump and is doing it's job
nicely. It's not the experience you get from a realtime OS,
but again, if you need realtime response, don't KILL the os,
USE a realtime OS.
> Same for programming.. I spend 95% of my computer use under
> multitasking, using utilities.. I would never want to e.g.
> use a monotasking OS for that.
> But, still, when I need to concentrate all the power and
> "syncronization" resources of my hardware on a specific
> realtime application, I want to be able to take over the
> machine and program it for this application in the *only*
> proper way, which is the opposite of the normal scenario.
use a realtime os for that. killing the current OS you're under
is still not the way to go, because perhaps today the hardware
is pretty simple, but this will change.
> What's wrong in having a choice?
nothing. you have a choice, you just seem to be picking the
most worse option
O.
Roy Jacobs
> man... how can you compare windows2000 with an OS that didn't even
> have memory protection.
>
> (forgive me, I programmed amiga for years)
Why must you force this down everyone's throat every other posting?
Yes, you've been a scener for yonks. Yes, you know how an Amiga works.
--
sagacity
3state
Fabio Bizzetti
says...
> > > IMHO. Good coders get formed coding down to the metal.. the
> > > others (e.g. Java programmers) often don't even know how a
> > > computer works inside. I'm one of those (and I think I've the
> > > titles to speak, due to my past and current work) that thinks
> > > that hardware ignorance doesn't help software skills.
>
> you ignorant fuck, dont you dare classify java programmers as clueless
> idiots.
Get a life, Gaffer. I still recall the thread (1 year ago or so.. I
can repost it here for others pleasure) where you, with your stupid
arrogant usual attitude, were flaming me for not "understanding" CPU
cache issues, and then I proved you a PITYFUL LAMER and completely
ignorant in the matter. But you had/have a big mouth, don't you????
Get a life Gaffer.. until you're in time. You're gonna lose your face
again.
> its possible to call assembler routines from inside java, yet still provide
> fallback java routines so the code runs everywhere. simple rule you should
> follow fabio, if you dont know anything about the subject, shut the fuck up
> before you make yourself look very stupid to anybody who does.
Hey, piece of idiot, stick your fucking slow-as-hell Java in your ass,
and don't provoke me.. I'm gonna make you look again like the fool you
really are.
> the only idiots nowadays are people who use the wrong tool for the job. if
> you use assembler to do something that is easier done in c/c++/java you are
> a damned fool.
Here the only fool is you, Gaffer.
> conversely, if you truly understood java and scripting languages. you'd see
> that they can really play a useful role in a *real world* project.
Like realtime demos that (ideally should) push the hardware to the
limit? You newborn imbecile, I've been in the scene when you were
still sucking the milk from your mother, and demos have always been
a mean to prove the coder's skills, the graphician's skills, the
musicians's skills, AND the designer's skills. It's SMARTASS lamers
like you that lacking all and any qualities, would like to reduce
a demo to just a lame animation with no code. But you aren't a
true scener.. you're JUST A PITYFUL LAMER.
Now, tell me, you fucking arrogant ignorant LAMER, is Java the ideal
tool to exploit the CPU power as much as possible? By the way, this
thread was more or less about microcontrollers, you fucking PITYFUL
LAMER, would you code an industrial microcontroller (or a DSP) in
Java?????????????????????? You're just a TROLL, and a GIANT LAMER.
Get a life, moron... troll. Kids like you shouldn't even be allowed
to code in BASIC V2.
> the best of both worlds? object management and high level in java, plus
> tight c/asm inners? i think so, but thats only my opinion...
In any case, you're an idiot. You already proved it last year with
that thread about how CPU caches work, and how to optimize for them.
We're all still laughing for it after 1 year.. good thing you did at
that time disappearing for a lot of time.. hahaha. Lucky for you I'm
good and don't wanna push the shame on you.. I've better things to
do with my (precious) time.
But if you show your fucking arrogant LAMER face around in my threads
again I'll spend some time to make you look again like the fool and
ass that you are. Hey, I hope now you got (even only in theory) how
CPU caches really work.. hahaha.. it's incredible how big is your
mouth when you've to flame.. it's as huge as your ignorance and
presumption in that matter.
DAMMIT HOW MUCH OF A LAMER YOU ARE, GAFFER!!!!! YOU BEAT 'EM ALL.
> Gaffer
Have a nice day, kid.
Fabio Bizzetti
Hi Frans,
I'm gonna reply to your posting.. but after this time, I won't have
time for this (kinda pointless in my opinion) discussions anymore.
I'm very busy using my time in a *concrete* way, I think it's the
best thing (me, you, or anybody else) could/should do, rather than
wasting time in Usenet.
Anyway.. here are my final thoughts about what you said, please
read them (they're opinions after all), but you've to know I've
no time to continue this thread.
In article <8h5cp5$9ld$1...@news1.xs4all.nl>, per...@xs4all.nl says...
> "Fabio Bizzetti" <check_my_signature_for_my_real_email_address@I_hate_spammers.it> wrote in message
> news:MPG.139fb358a...@news.infinito.it...
> > In article <8h3teo$gte$1...@news.kolumbus.fi>, ju...@twilight3d.com
> > says...
> > > > I also add (about API's, e.g. 3D API's): where is the innovation,
> > > > if all you've to do is to use this function or that function?
> > >
> > > What do you propose to fix the situation? And why should the situation be
> > > "fixed" in the first place? Progress goes on
> >
> > Yes, but future is not always "progress".
> >
> > Or Windows2000 is better than AmigaOS?
>
> man... how can you compare windows2000 with an OS that didn't even
> have memory protection.
>
> (forgive me, I programmed amiga for years)
Well, Windows95 is supposed to have memory protection too, yet it
crashes much more than AmigaOS (I use to run a lot of tasks at the
same time in both PII-W95 and 68060-AOS3.1 systems).
As for comparing Windows2000 with anything, we'll do it when they
release a "stable" version.. but then again we'll probably be able
to compare it with the next release of AmigaOS (that will feature
memory protection, if Amiga Inc. doesn't go bankruptcy again..).
> > Exactly. What I was ranting about is that we should get both
> > the multitasking for utilities/applications, but also *retain*
> > the FREEDOM to code OUR computer, that we payed big buck$ for,
> > having all the docs to do it, in the way we most prefer.
>
> do you care if that 'computer' is the plain hardware or
> a blackbox? if you call a function and that function does
> exactly what you want with the hardware, what's the difference
> with jamming a number in some memory address (where for example
> a videoregister is located).
Exactly the same difference between "buy a faster CPU" or "optimize
the code better so you dont have to buy a faster CPU".
> I used to enjoy programming on the metal (hell, I liked my
> 32buffer MFM trackloader with blitter MFM decoding on the amiga,
> but it's rather silly actually... to code your own MFM decoding before
> you can use the data..)
>
> > I dont know from where your posting originates (there's
> > crossposting, you know), but if you know what's a demo, and
> > the gold times of the demo scene (expecially on the Amiga,
> > since the PC never had such a significative demo scene), then
> > you'll very well know the importance of this distinction, and
> > to disable multitasking and all the like *in certain specifical
> > circumstances*.
>
> yeah, but deep down you know this is not needed.
> if you need real time feed back, you should take a real time OS.
> (like os/9?). True, a demo on amiga for example killed the
> OS in the first 2 lines of code and then started working.
> but is this truely needed? today, on the PC (in windows) I
> don't do that.
In fact a C64 still scrolls 2D bitmaps in a more smooth way than
99% of Windoze programs.. what a shame really.
Yet, my Windoze/DX code (in a way or another, read: polling the
VBL bit in the VGA hardware when the equivalent DirectX function
reports "success" but in reality doesn't do anything.. and if
this ain't lame, then I really don't know..) syncronizes perfectly
with the VBL of the videocard, on almost all videocards around.
In substance, the PC hardware is not that bad.. it's Windoze
that isn't up to the job. And in this case I see nothing wrong
in "patching" the OS with my own code to do what it's supposed
to do (and even reports "success", falsely!) but it doesn't do.
> it hangs in a messagepump and is doing it's job
> nicely. It's not the experience you get from a realtime OS,
> but again, if you need realtime response, don't KILL the os,
> USE a realtime OS.
Or code that same OS in a better way? Even if this means hacking
here and there from time to time (hell, if it works, better than
nothing. At least you won't have to ask the users to install the
unknown RtOS (unknown not because it's not good, but because the
money of Micro$oft didn't want the world to know it) just to run
your little program.. and if after working wonderfully in a semi
hacked Win9x, it fails in Windows2002, then either 1) fix it so it
works also on the new OS or 2) the culpage is anyway of Microsoft
that FORCED (with unethical or even highly illegal methods) the
market to use an OS that wasn't capable of providing the necessary
power and flexibility, and provoked the bankrupcty of much better
products, under illegal commercial actions (Anti-Trust anybody?)).
> > Same for programming.. I spend 95% of my computer use under
> > multitasking, using utilities.. I would never want to e.g.
> > use a monotasking OS for that.
> > But, still, when I need to concentrate all the power and
> > "syncronization" resources of my hardware on a specific
> > realtime application, I want to be able to take over the
> > machine and program it for this application in the *only*
> > proper way, which is the opposite of the normal scenario.
>
> use a realtime os for that.
No, I use what it's *practically* available, and if it's Win9x
and it isn't up to the job, I hack that piece of shit to the death.
Simple as that... blame Microsoft either because:
1) Through illegal activities they killed the competition of much
better OS's around, that would have made the same job (i.e. use
the hardware properly) with no need to hack the OS.
OS's exist to let you use the hardware, after all. If they fail
in this simple thing, you're allowed to do it yourself, in the
most "clean" possible way, of course.
2) After having forced (through illegal mean$) the market to use
their own OS's, and kill (again, illegally) the competition, they
didn't make their O$ powerful and flexible enough.
> killing the current OS you're under
> is still not the way to go, because perhaps today the hardware
> is pretty simple, but this will change.
Also "software patches" are a solution, don't you know.
Personally, I've my own OS (mostly a "wrapper" under Windows, with
some exceptions; and a nearly-full OS under Dos), with my own
executable format, programming language (HLA, if you were on the
Amiga you may recall it, I've made a couple of commercial games
with it) and compiler/debugger.
So the same program works unmodified under Winslows, Dos, or any
other OS (e.g. Linux, BeOS, etc..) if I just write the main loader
( + callable functions, i.e. a game/demo OS in practice ) for that
other x86 platform. I use this mostly for realtime hi-performance
stuff, where I want to have a choice under which system make it run,
but still minimize debugging problems, and increase portability by
not using directly any of that system's functions. It works wonders.
> > What's wrong in having a choice?
>
> nothing. you have a choice, you just seem to be picking the
> most worse option
You say?
I'm gonna release soon a ~200/300 *micro*seconds latency realtime
DSP program with tens of effects at the same time, for Dos and
(semi-hacked) Windoze.
Now reach this latency under normal DirectSound (which has about
one HUNDRED times more latency, still with much more overhead,
often with interruptions ("clicks") in the sound output, etc..),
or ask that troll of Gaffer to rewrite my assembly-coded (btw,
Gaffer, the language is spelled A S S E M B L Y, not ASSEMBLER),
ask him to rewrite them in... JAVA!!! And then let's see how much
fast and realtime are them.. ROTFLOL =)
Sorry, but all I can do is laugh in front of some Gaffer-style
lameness. He's the biggest troll of the world.
Let's cut the crap anyway, I've much better things to do with
my time than to convince you or anybody else. Nothing personal
of course, mate.
Regards,
--
Fabio Bizzetti --- email: bizzetti<<at>>freemail.it
> O.
>
>
Frans Bouma
news:Pine.BSF.4.21.000601...@turtle.stack.nl...
> >Otis wrote:
> > man... how can you compare windows2000 with an OS that didn't even
> > have memory protection.
> >
> > (forgive me, I programmed amiga for years)
>
> Yes, you've been a scener for yonks. Yes, you know how an Amiga works.
why do you always have to throw your useless mud on my postings.
*deep sigh*
If you have a problem with me, we'll discuss it at TO. It's very obvious
your reply more to my postings than to other people's postings.
I almost never mention I programmed on the amiga, it's just a statement
so mr Fabio doesn't come on me I don't know what I'm talking about.
or are YOU mr fabio? seems like it.
Ah well... now I'm again explaining why I say something to a complete
stupid fool who doesn't want to hear it, because next time he's forgotten
it anyway. grow up, kid.
O.
Fabian Giesen
> nature of the PC) too few people in this platform have the
> "hardware bashing" mentality so common in the 8bit, 16bit and
> 32bit (Amiga and Archimedes) machines.. a beautiful philosphy
argh
you can't compare that AT ALL
amiga had custom chips that were in *every* model
which is a property that certainly DOESN'T apply to pcs
it's hard enough to write a soundsystem that works on many cards
(and maintaining such a system), but for example 3d acceleration
WOULDN'T BE POSSIBLE w/out apis like direct3d or opengl... why?
because the average demo coder doesn't have access to thousands
of graphics cards and chip descriptions... and writing a demo
which only works on one computer is more than simply lame.
(although kinda common practice, still)
look at all those demos around '92, '93.. many of them tweak
vga registers in a way that just doesn't work for todays cards
anymore, and it didn't work for most graphics cards back in then
either, just on for example ET4000s or similar, which were quite
widespread.
but i personally DON'T want to write a demo that ppl will not be
able to watch in 1-2 years... and i guess i'm not the only one.
-ryg
AndrewJ
> crossposting, you know), but if you know what's a demo, and
> the gold times of the demo scene (expecially on the Amiga,
> since the PC never had such a significative demo scene), then
> you'll very well know the importance of this distinction, and
> to disable multitasking and all the like *in certain specifical
> circumstances*.
Well, I wasn't around for the golden age of the demoscene, but I've read
stories... sniff.
This paragraph, however, makes me question your knowledge of the PC scene. The
PC never had a significant demoscene??? Haven't you ever heard of the Future
Crew, Triton, Renaissance, and all the other kick ass PC demo coders? Granted,
many came from the Amiga-scene, but they just raised the standards in the PC
scene. And then came Winblows '95, which heralded the end of the scene (in my
eyes at least).
Anyway, sorry to butt in, but I just had to, ya know. ;)
AndrewJ
Fabian Giesen
> have to do too, until the world finally gets rid of Microslow.
then make a start getting rid of "microslow" and delete ms-dos...
which btw wasn't coded by m. zbikovski, he was just a coder at ms.
the original msdos was 98qdos by someone whose name i don't remember
anymore... but i doubt't you'll care either
-ryg
Fabio Bizzetti
> "Roy Jacobs" <saga...@stack.nl> wrote in message
> news:Pine.BSF.4.21.000601...@turtle.stack.nl...
> > >Otis wrote:
> > > man... how can you compare windows2000 with an OS that didn't even
> > > have memory protection.
> > >
> > > (forgive me, I programmed amiga for years)
> >
> > Why must you force this down everyone's throat every other posting?
> > Yes, you've been a scener for yonks. Yes, you know how an Amiga works.
>
> why do you always have to throw your useless mud on my postings.
>
> *deep sigh*
>
> If you have a problem with me, we'll discuss it at TO. It's very obvious
> your reply more to my postings than to other people's postings.
>
> I almost never mention I programmed on the amiga, it's just a statement
> so mr Fabio doesn't come on me I don't know what I'm talking about.
> or are YOU mr fabio? seems like it.
Dont get paranoid now please, dude.
> Ah well... now I'm again explaining why I say something to a complete
> stupid fool who doesn't want to hear it, because next time he's forgotten
> it anyway. grow up, kid.
Flamewars suck.. time to learn how to use that killfile feature.
Greets,
Fabio Bizzetti
That's the usual lameness.. heard thousands times "hey, you hate
Microsoft? then why do you love your wonderful MsDos? Don't you
know it has been made by Bill Gates's wife?".
As I said THOUSANDS times.. but, wait, it's really to stupid simple
I shouldn't have to repeat it, nor even say one single time.. anyway:
MsDos is PLAIN CRAP, the crappiest "OS" on earth, even more than
Windoze.. *BUT* .. BUT.. but being so much excessively crap, it lets
you code lowlevel things by yourself, and thus the final product
depends by your abilities, and can be very good (or very lame, I
don't question this).
Windows pretends to solve all the problems by itself instead, while
it only adds layers of lameness upon Dos; which at least lets you
bypass it and do it yourself (tm).
Is it too hard to understand? It has been repeated lotsa times:
the only quality of Dos (and it's a good one indeed) is that it's
so poor that it's not even a real OS.. and you have to (and thus
can) program the hardware by yourself.
Anyway, all these discussions are pointless. They would have
been ok in ~1993, when things like VESA were a possible way to
the future.. but now that Microsoft "suggested" to the hardware
manufacturers that every h/w board has to be coded through the
Windows API functions, all this talking is useless.
It seems the future is set now... and we see it's not as brilliant
as it could have been.
End of posting.. before my gfx board driver crashes my whole PC
(protected memory OS, eh? =) ).
RMSoft.
news:8h657v$m5n$11$1...@news.t-online.com...
> it's hard enough to write a soundsystem that works on many cards
> (and maintaining such a system), but for example 3d acceleration
> WOULDN'T BE POSSIBLE w/out apis like direct3d or opengl... why?
> because the average demo coder doesn't have access to thousands
> of graphics cards and chip descriptions... and writing a demo
> which only works on one computer is more than simply lame.
> (although kinda common practice, still)
Video-card have also Bios for this reason, also LinearFrameBuffer and
flatreal-mode to access this buffer. Also multiplatform programs quality
isn't
as good. Maybe in future hardware cost less than software and you can buy
computer for every program separately and don't need multitasking. :-)
(this called embedded systems :)
>which btw wasn't coded by m. zbikovski, he was just a coder at ms.
>the original msdos was 98qdos by someone whose name i don't remember
Example I use only MS-Dos6.22 and don't care about this "qdos" at all.
Theoretically is possible what all OS'es are written by someone _other_.
Dos is good for accessing computer hardware, but it's services and internal
structures are also more logical than other OS'es, especially in asm point
of view.
(Linux is for c++)
----
Ivar, (c)RMSoft
http://my.tele2.ee/rmsoft
rmsoft...@mail.ee
Jukka Liimatta
> you code lowlevel things by yourself, and thus the final product
> depends by your abilities, and can be very good (or very lame, I
> don't question this).
You just cannot go and do everything yourself. If you do, you would still be
writing 2D device drivers over and over. Idea with drivers is to let
hardware manufacturers worry about writing drivers for their shit, and your
programs when well written, will take automatically advantage of the
drivers.
This is definitely better than re-inventing the wheel over and over again.
By all means, re-invent the wheel and feel proud of the achievement.
J
Fabian Giesen
which doesn't even support lfb vesa modes. great, really.
> flatreal-mode to access this buffer. Also multiplatform programs quality
flatreal mode even gives problems with many processors (i once had such a
processor) and vesa 2.0 is extremely fucked up because every video card
manufacturer seems to understand the standard different or simply isn't
compliant.
same for most vesa engines out there, most ignore the bytesperscanline field
for example.
> isn't
> as good. Maybe in future hardware cost less than software and you can buy
> computer for every program separately and don't need multitasking. :-)
> (this called embedded systems :)
times are over where you need every spare cycle of processor time,
even for embedded systems.
> Theoretically is possible what all OS'es are written by someone _other_.
> Dos is good for accessing computer hardware, but it's services and
internal
"doesn't hinder you accessing your computers hardware directly" would be a
better expression. take a look at directdraw/direct3d. they are also good
for
accessing computer's hardware, and not only my own, but also OTHERS. which
*is* a feature imo.
> structures are also more logical than other OS'es, especially in asm point
> of view.
erm?
i don't call hundreds of deprecated functions for cp/m compatibility that
never
were used anyway (except in dos 1.0) logical... same goes for the internal
structures of msdos which aren't logical at all imo. and don't forget that
dos
isn't reentrant, which made programming tsrs really a pain in the ass.
that's not what i would call logical structures.
> (Linux is for c++)
don't think so... take a look at www.linuxassembly.org. linux kernel
services
are provided via int 80h, and glibc is just a wrapper you *can* use, but
don't
have to.
-ryg
Damian Yerrick
wrote:
>Frans Bouma <per...@xs4all.nl> wrote in message
>news:8h5cp5$9ld$1...@news1.xs4all.nl...
>> have memory protection.
>
>You simple *can't* protect memory - if someone protect something, then
>it can always unprotect.
Not if the instruction that unprotects is itself protected (may be
executed only at kernel privilege level).
>but some people use c++/java/basic
>to do something that is easier done in assembler.
It's assembly with a Y. An "assembler" is a compiler for "assembly"
language. But when I was first coding on the PC, I wrote in QBasic
because I didn't have the Internet to download a proper assembler and
C compiler. I currently use DJGPP http://www.delorie.com/djgpp/
--
Damian Yerrick
"I refuse to listen to those who refuse to listen to reason."
See the whole sig: http://www.rose-hulman.edu/~yerricde/sig.html
This is McAfee VirusScan. Add these two lines to your signature to
prevent the spread of signature viruses. http://www.mcafee.com/
RMSoft.
> which doesn't even support lfb vesa modes. great, really.
Simple don't use such cards. I have only 1MB video-card and it's more
than I need.
> times are over where you need every spare cycle of processor time,
> even for embedded systems.
But not for me and I don't believe what I am the *only one* man in the world
(ONLY ONE?????? - No, I don't believe this, it's too flattering :).
And I am just oriented to this kind of market, exactly as you are oriented
to
other kind - in my opinion this is normal as free choice and free market.
> take a look at directdraw/direct3d. they are also good for
>accessing computer's hardware, and not only my own, but also OTHERS. which
>*is* a feature imo.
If it is standard software feature, then it can't be direct hardware access.
In multitasking systems can't be exist any *pure* "direct hardware access".
> i don't call hundreds of deprecated functions for cp/m compatibility that
> never
> were used anyway (except in dos 1.0) logical... same goes for the internal
> structures of msdos which aren't logical at all imo.
I use those functions and I have some other understanding about _logic_
also.
>and don't forget that dos
> isn't reentrant, which made programming tsrs really a pain in the ass.
"pain" ? - but never for _good_ programmers.
Eliminating reentrancy problem is exactly the same as making
"Perpetume Mobile", believe me, it's impossible.
In PM systems this problem is fixed _very_ banal and brutal - nothing
more simpler than just don't allow hooking (tsr's don't exist at all).
> that's not what i would call logical structures.
It's just plain logic.
>
> > (Linux is for c++)
>
> don't think so... take a look at www.linuxassembly.org. linux kernel
> services
> are provided via int 80h, and glibc is just a wrapper you *can* use, but
> don't
> have to.
Yes, I know this, but I was hear also that using int80h directly isn't so
safe if ther are some other version of Linux. More common use is making
procedure calls, but this is dependent from libraries.
Best wishes.
-------
Jukka Liimatta
> than I need.
Tell that to people who did buy such cards, and are trying to run your
ILL-WRITTEN program. Use the VBE api correctly, but don't blame hardware
when it is obviously not it's fault. Sometimes it was, but mostly not.
> If it is standard software feature, then it can't be direct hardware
access.
> In multitasking systems can't be exist any *pure* "direct hardware
access".
The drivers have direct hardware access. This is why it's up to you to know
the best means of using these API's -- a good programmer still has a lot of
work left, just in less trivial things. You can concentrate on the bigger
picture -- or, twiddle with the bits and io ports. I had my share of that in
the past, sure, felt like I was 0wn1ng the microsoft l4meR coders, but more
and more I learn, the more I learn to appriciate other programmer's work
aswell. Funny thing this respect... it's like a learnt skill.
J.
Rez
> processor) and vesa 2.0 is extremely fucked up because every video card
> manufacturer seems to understand the standard different or simply isn't
> compliant.
Some so-called VESA 2.0 cards require a driver to have full VESA 2.0
support -- they don't have it in hardware. So far I've caught S3Trio and
and Matrox G400 at this nastiness. (Matrox G200 has full support in
hardware). MBF's hires mode makes a good quick video card test, since it
depends on full VESA 2.0 in hardware to work right without a secondary
driver like UNIVBE or W9* video driver support (S3, etc.)
ftp://ftp.cdrom.com/pub/idgames/source/mbf.zip and /mbfsrc.zip (GPL'd)
(this is a doom source mod, needs a DOOM IWAD to run -- shareware
version works fine with it)
~REZ~
Malte Clasen
> Yes, but future is not always "progress".
I'd say progress means the difference between two chronological ordered points in
evolution, and evolution is time-based, to you can apply future to it: If time
runs forward (and it does always), every change means progress.
c u,
Malte
--
Malte Clasen (The Update / CoPro)
mailto:Malte....@gmx.de
http://www.copro.org/malte/
"Man drückt einfach auf 'nein' und der Rechner geht!" (c)kb
Matias Lahti
>Anyway, all these discussions are pointless. They would have
>been ok in ~1993, when things like VESA were a possible way to
>the future.. but now that Microsoft "suggested" to the hardware
>manufacturers that every h/w board has to be coded through the
>Windows API functions, all this talking is useless.
One word: BULLSHIT. I do remember the problems with hardware
'back in the day'. No sound from games, since I had a GUS. If
there would've been general apis then, i would have gotten sound.
Fabio Bizzetti
mat...@fisherman.temple.brainlez.org says...
I've got nothing against API's.. when they don't suck.
--
Greets,
Fabio Bizzetti - developer of DspRT and maintainer of the DspRT mailinglist
DspRT: an extremely powerful realtime DSP and electronic modeling program.
DspRT WEB page: http://www.infinito.it/utenti/bizzetti/dsprt
DspRT mailinglist: Subscribe or read at http://www.egroups.com/group/dsprt
Damian Yerrick
check_my_signature_for_my_real_email_address@I_hate_spammers.it (Fabio
Bizzetti) wrote:
>I've got nothing against API's.. when they don't suck.
Here's something that doesn't suck: Allegro.
http://www.talula.demon.co.uk/allegro/
If you use Watcom or DJGPP, Allegro works very well.
Fabio Bizzetti
Bullcr_p...@hotmail.comRemoveBullcr_p says...
> On Thu, 8 Jun 2000 03:19:33 +0200,
> check_my_signature_for_my_real_email_address@I_hate_spammers.it (Fabio
> Bizzetti) wrote:
>
> >I've got nothing against API's.. when they don't suck.
>
> Here's something that doesn't suck: Allegro.
> http://www.talula.demon.co.uk/allegro/
> If you use Watcom or DJGPP, Allegro works very well.
Yes, but on the other hand, I doubt it has a mode for ~zero latency..
I sure don't get fun using my time writing drivers for soundcards,
but if this is unavoidable to get the performances that I want..
Damian Yerrick
check_my_signature_for_my_real_email_address@I_hate_spammers.it (Fabio
Bizzetti) wrote:
>In article <e10ujso3cfovgul02...@4ax.com>,
>Bullcr_p...@hotmail.comRemoveBullcr_p says...
>> If you use Watcom or DJGPP, Allegro works very well.
>
>Yes, but on the other hand, I doubt it has a mode for ~zero latency..
>I sure don't get fun using my time writing drivers for soundcards,
>but if this is unavoidable to get the performances that I want..
I see your point, but think of how small a 6 millisecond latency is.
It's like sitting 2 meters (6 feet) farther away from the speaker,
given sound velocity ~= 300 m/s.
Fabio Bizzetti
Bullcr_p...@hotmail.comRemoveBullcr_p says...
> On Thu, 8 Jun 2000 18:59:25 +0200,
> check_my_signature_for_my_real_email_address@I_hate_spammers.it (Fabio
> Bizzetti) wrote:
>
> >In article <e10ujso3cfovgul02...@4ax.com>,
> >Bullcr_p...@hotmail.comRemoveBullcr_p says...
> >> If you use Watcom or DJGPP, Allegro works very well.
> >
> >Yes, but on the other hand, I doubt it has a mode for ~zero latency..
> >I sure don't get fun using my time writing drivers for soundcards,
> >but if this is unavoidable to get the performances that I want..
>
> I see your point, but think of how small a 6 millisecond latency is.
> It's like sitting 2 meters (6 feet) farther away from the speaker,
> given sound velocity ~= 300 m/s.
6ms is excellent by Windows standards (I know you're referring
to a Dos product, though). But is it just input latency? Or the
sum of input through computer, then output?
Anyway, I sure don't want to sound picky dude (and you certainly
wouldn't deserve it), so don't get me wrong please, but I think
that not few people (e.g. guitarists) are able to hear (and feel
it fastidious) a 12 ms (6+6) latency.
But regardless of that, Allegro is a very good library.. too bad
that business forced us all to the worst OS ever made (Winslows).
I don't want to start a flamewar with Windows lovers, but I suggest
them to try out any other OS (e.g. BeOS, Linux, etc.. etc..) and
push their multitasking to the limit, before even breathing that
Windows is good (or decent.. or even just "acceptable").
Anyway, back to my work.. I leave on 21th June, till 13Th July,
and I want to release a good amount of informations (WEB pages)
and demo samples about DspRT before I leave. The first demo
executable will follow later, when I come back and work on the
protection and other minor stuff, like drivers for some other
soundcards.