rec.autos.simulators

GPL: Why do FPS change for same settings ?

rob

GPL: Why do FPS change for same settings ?

by rob » Wed, 23 Dec 1998 04:00:00

Are you in the same position within the grid in each case?


>Hi All, As my subject says, I've experienced differing frame rates for
>exactly the same configuration. Eg. One day turn on short novice race at
>Silverstone 5 opponents, on grid get 22FPS.  Next day exactly the same but
>get only 16-17FPS

<snip>

rob.

Marcel Offerman

GPL: Why do FPS change for same settings ?

by Marcel Offerman » Sat, 26 Dec 1998 04:00:00


> My only theory that I can come up with is the old problem of Win95 not
> handling memory over 64Mb too well, I don't know whether this is still a
> problem with OSR2. Any thoughts would be appreciated. Dave.

This has nothing to do with Win95, but is a hardware limitation of
(older) motherboards. Memory over 64 MB can't be cached, which makes
programs that run in that memory slower. There's nothing Win95 OSR2 can
do to fix that (except not use it :-).

Greetings, Marcel

Rob Adla

GPL: Why do FPS change for same settings ?

by Rob Adla » Wed, 30 Dec 1998 04:00:00

  You say this is a hardware limitation with older motherboards. Does this
mean this was rectified with the new 100 mhz motherboards (BX?)?

--

     Rob.



>> My only theory that I can come up with is the old problem of Win95 not
>> handling memory over 64Mb too well, I don't know whether this is still a
>> problem with OSR2. Any thoughts would be appreciated. Dave.

>This has nothing to do with Win95, but is a hardware limitation of
>(older) motherboards. Memory over 64 MB can't be cached, which makes
>programs that run in that memory slower. There's nothing Win95 OSR2 can
>do to fix that (except not use it :-).

>Greetings, Marcel

Wolfgang Prei

GPL: Why do FPS change for same settings ?

by Wolfgang Prei » Thu, 31 Dec 1998 04:00:00


>  You say this is a hardware limitation with older motherboards. Does this
>mean this was rectified with the new 100 mhz motherboards (BX?)?

The problem occurs exclusively with some of the older "classic"
Pentium (non-II) chipsets by Intel, most notably the TX chipset. The
BX chipset for Pentium II is not affected, since the cacheing takes
place on the processor itself. A P II with 512k of L2 cache and a 16
bit TAG RAM (I believe) can cache up to 65536 x 512k = approx 32
GigaByte. This limit shouldn't pose a serious problem for another year
or two. :)

A Celeron A still can cache up to 8 GB with its 128k L2 cache,
provided the TAG RAM is indeed 16 bit wide.

BTW, the 64MB limit was neither a technical necessity nor a bug. The
newer chipsets for the old Pentium were designed by Intel that way for
marketing reasons. The message was: "if you need more RAM, buy a
Pentium II". Chipsets by other manufacturers (VIA, Ali, etc.) weren't
limited like that.



>>> My only theory that I can come up with is the old problem of Win95 not
>>> handling memory over 64Mb too well, I don't know whether this is still a
>>> problem with OSR2. Any thoughts would be appreciated. Dave.

>>This has nothing to do with Win95, but is a hardware limitation of
>>(older) motherboards. Memory over 64 MB can't be cached, which makes
>>programs that run in that memory slower. There's nothing Win95 OSR2 can
>>do to fix that (except not use it :-).

--
Wolfgang Preiss   \ E-mail copies of replies to this posting are welcome.


Charli

GPL: Why do FPS change for same settings ?

by Charli » Thu, 07 Jan 1999 04:00:00

One odd thing I found was that if I use Direct Input for the joystick (with
"Poll with Interrupts Enabled") I get a much better frame rate than if I use
Generic Input.  I thought that I wasn't changing anything important, but I
had been having problems with Direct Input so switched to Generic and it
took me a while to find the cause of the loss in frame rate.

Good Luck
Charlie

Michael E. Carve

GPL: Why do FPS change for same settings ?

by Michael E. Carve » Fri, 08 Jan 1999 04:00:00


% One odd thing I found was that if I use Direct Input for the joystick (with
% "Poll with Interrupts Enabled") I get a much better frame rate than if I use
% Generic Input.  I thought that I wasn't changing anything important, but I
% had been having problems with Direct Input so switched to Generic and it
% took me a while to find the cause of the loss in frame rate.

I would hazard to guess that this is because the Win95 routine for
polling the gameport is less precise (and thus less intensive) than
Papyrus' generic input.  The polling range produced by Direct Input is
much lower than the range when using generic.  I assume this is because
Win95 polls the gameport less often than does the generic.  Thus, more
cycles devoted to video refresh via direct input and more cycles spent
polling the game port in the generic mode. [Of course this is just an
educated guess/hunch.]

--
**************************** Michael E. Carver *************************
     Upside out, or inside down...False alarm the only game in town.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=<[ /./.  [-  < ]>=-=-=-=-=-=-=-=-=-=-=-=-=-=


rec.autos.simulators is a usenet newsgroup formed in December, 1993. As this group was always unmoderated there may be some spam or off topic articles included. Some links do point back to racesimcentral.net as we could not validate the original address. Please report any pages that you believe warrant deletion from this archive (include the link in your email). RaceSimCentral.net is in no way responsible and does not endorse any of the content herein.