rec.autos.simulators

Recently I have converted my TSW Sport to be switchable between separate
and combined throttle and brake. I suddenly found that I was locking up
the brakes much too easily when I switched to separate axes and so did
some further work. I have discovered some very interesting stuff about
controllers (TSWs in particular), GPL and probably other sims as well. I
intend to put all of this on a web page soon but here are some tit-bits.
I have also discovered a few ways to reduce pot bounce (jittery or
spikey pots). I have probably got some of the electrical physics wrong
but I believe that I am correct about the effects. Please correct me
where I am in error - I would love some feedback - good or bad on this.

It can make a big difference to both brake and throttle sensitivity by
switching between separate and combined pedals but this difference IS
ADJUSTABLE!!!! I have discovered that by fiddling with the overall pot
ohmage and "at rest" resistance you can make your car behave quite
differently!! I would rate the net effect nearly as important as car set
up in say GPL (which is the only sim that I have tested this on).

Basically with combined pedals you are passing through two pots in
series. Depending on the "at rest" point of each pedal pot you are
varying a higher electrical resistance across a relatively wider band
than with separate pedals. And this band is only linear if both pedals
have the same amount of travel. I restricted my range of travel on the
brake by sticking a squash ball inside the pedal mechanism to make the
brakes less soggy but I over-compensated for this by opening up a
tightening***inside the TSW Sport to allow me to be more sensitive
with my braking. I believe that you can muck about with any reasonable
quality wheel in this way.

So as soon as you switch to separate pedals you presumeably make both
pedals linear - WRONG - THE LINEARITY IN GPL DEPENDS ON THE AT REST
POSITION OF THE POTS!!!!! (Oh yes it does, Steve Blank.) By monkeying
with the "at rest" position of the pots you can make the brakes more or
less linear and thus reduce or increase the chances of locking up. You
can do the same with the throttle and I guess that, here, different
settings will be better suited to different tracks. In case you think I
using logarithmic pots - I am not.

I am compiling some basic pot setup tips which I shall post shortly. It
will be broadly based on the raw data figures - see TMS (an invaluable
Thrustmaster utility) or, if you can't get hold of this, GPL's
configuration screen is adequate, but takes much longer to load and
unload. I have also been testing with 50K and 100K pots which, as you
might expect affects the brake throttle behaviour as well. How about
non-linear pots! Well they're pricey but could be good for the brake pot
(possibly unfair - I'm sure the virtual FIA might object to the use of
these). A standard pot has a total travel of about 330 degrees which is
about the same as the steering wheel - so there's not a lot you can do
with this.

The other things I have found out are to do with pot bounce which
obviously has a major negative effect on the car. Low raw numbers
decrease pot bounce. If you have bounce in your wheel adjust the "at
rest" settings on your pots - but bear in mind that it will affect the
car in other ways. The other things you can do to reduce pot bounce are:

1) Clean the pots. Use a hydrocarbon fluid based electrical contact
cleaner.
2) Replace the pots if they are old or damaged.
3) Use 50K pots rather than 100K - again this will affect car
performance.
4) Use a game port with an adjustable speed and turn it up to the top
(usually 10) - ignore what TMS says - don't centre it, just turn it up
high.
5) Make sure both wheel and pedals are grounded. Thomas has ignored the
twisted ground wire that runs around all the others wires going to the
15-pin gameport on my TSW. You should run this to a loop that is screwed
to the pot rod just like a washer for each of your pots. I got this idea
by looking inside my Thrustmaster Formula 1 - a big omission by Thomas.
Am I right here - it seemed to make a big difference.

Cheers,
Paul

<<So as soon as you switch to separate pedals you presumeably make both
pedals linear - WRONG - THE LINEARITY IN GPL DEPENDS ON THE AT REST POSITION OF
THE POTS!!!!! (Oh yes it does, Steve Blank.) By monkeying with the "at rest"
position of the pots you can make the brakes more or less linear and thus
reduce or increase the chances of locking up. You can do the same with the
throttle and I guess that, here, different settings will be better suited to
different tracks. In case you think I
using logarithmic pots - I am not.>>

     Hmmmm; interesting stuff.......where to begin.  I'll try this, with the
caveat that I'm an electrical dunderhead.  So feel free to enlighten me.
     Assume your pedals both have equal travel, mechanically.  Regardless of
the raw resistance readings inherent in the pots, they will have a certain
minimum and a maximum value at the ends of pedal travel, both roughly the same
for separate axis pedals if the pots are affixed equally.  If by "monkeying"
with the "at rest" positions, you mean adjusting the static position of the pot
shaft to use a different portion of its possible travel, then yes, you will
alter the raw resistance numbers.  However, when you calibrate the pedals in
Windows or a game, the pertinent software driver "learns" how to interpret
those numbers.  The driver matches its own min/max/center numbers, etc. to the
relevant resistance values.  The game will then use the numbers from its
assigned driver, not the raw resistance readings.  I use a CH utility called to
test my controllers, as it allows you to see the readings from the gameport or
the Windows driver readings, with or without using the calibration settings.
It's surprising how different they can be on some controllers.
     As for linearity, if you use linear-taper pots you will always get a
linear response from a controller, unless the game's software modifies the
input, ala GPL's linearity slider.  My previous suggestion about holding down
on the pedals a bit to set the beginning actuation point would indeed result in
a steeper response line.  It would still be a line rather than a curve, though,
and one that began later in the pedal's mechanical travel.  The purpose of
doing this is to compensate for the weak springs in some pedals, as foot weight
can tend to actuate them unintentionally, and also to balance the response of
both pedals to suit the user.  There's no rule here; it's just something I
experimented with and found helpful.  You can have earlier actuation with
shallower response line or later actuation with steeper response line, with
equal or different profiles on each pedal.  I expect ten different drivers
would end up with ten different settings(at least :-)), but it is a useful
"off-the-menu" setup tool.
     I have seen past reference on r.a.s. to using audio-taper pedal pots for
less touchy response , but haven't tried it myself.  In the States you can get
'em at RadioShack for around $2 apiece, about the same as their linear-taper
ones.  They stock pots in resistance readings from 50Kohm up to 1Mohm, but
they'd need modifying to use in most pedals or wheels I've seen, as they don't
have the typical half-round shaft end.  As for lower raw numbers being more
stable, I'd agree with that.  When I was using a TM GP1 with my CH pedals, the
steering readings, from a 200K pot, were much jumpier than those from the 100K
pedal pots.  Furthermore, some games (F1RS, for instance) couldn't handle the
difference in values and never would recognize the GP1 and CH pedals
simultaneously.  The Logitech wheel I'm using now has a built-in A/D converter,
which eliminates the gameport-related pot noise you notice in GPL as shaky
driver arms.

Cheers,

Steve B.

remove "edy" from address for email

Yes, this is what I mean.

I cannot argue with your logic - it seems impeccable - BUT I have tested this to my
own amazement and can assure you that if you set the raw numbers on the brake pot
as low as possible you will have an initially less responsive brake pedal - not
exactly a dead zone but tyre lockups will cut in later. Why this is I have no idea.
I have not tested on the generic input of GPL just direct input. Also 50K pots
appear to exacerbate the effective dead zone.

I'm fast coming to the conclusion that potentiometers are not the ideal solution. I
have no idea how optical "pots" work nor have I tried them but perhaps these
represent a significant improvement.

Papyrus are the only people who can answer this.

It's a good idea.

I shall try them.

Oh, for the ideal wheel! I am getting sick and tired of opening my wheel up every
so often to adjust, clean or replace pots. My ongoing battle to eliminate pot noise
entirely is a long way from won. 50K pots do help but I don't want a less
responsive throttle. In separate mode I am coming to the conclusion that a 50K
brake and a 100K throttle maybe best. I haven't done enough work on the steering
but this is just as crucial.

Cheers,
Paul

And another interesting factette - I left my internet (serial to USB  modem)
connection open and went into TMS. The pots were bouncing around like mad flies and
the port speed had dropped considerably. I quit my connection, and reentered TMS
and the port speed was back to normal and the pots were bouncing only slightly - as
normal. Ooo-err. Time for a USB wheel with optical pots, methinks. Maybe I'll get
the Microsoft one - any other USB wheels with optical pots? Which is best?
Cheers,
Paul

to my own amazement and can assure you that if you set the raw numbers on the
brake pot as low as possible you will have an initially less responsive brake
pedal - not exactly a dead zone but tyre lockups will cut in later. Why this is
I have no idea.<<

Never been accused of impeccable  logic before!   But I'll bow to testing; it
sounds pretty interesting.  I wonder if it's due to the mismatch between pot
values, since you're tweaking only one.  That would parallel my experience with
the TM wheel/CH pedal setup.

Second that.  I'm pretty happy with my current setup, but as soon as an
all-digital, pot-free alternative surfaces ( at a sub-stratospheric price), I'm
in there!

Happy Tweaking,

Steve B.

remove "edy" from address for email

   Strange;  I just looked at mine and the internet connection has no effect at
all on my pot readings.  Perhaps Logitech's A/D converter makes the difference;
it really does away with all the pot noise.
    There aren't too many offerings out there in  USB/Optical wheels.  Saitek
use MS guts, but I tried one of theirs and didn't care for it at all.  The MS
is probably the only such choice for now, but I expect that won't last long.
Pot-based controllers are on the endangered species list already.  They won't
be missed!

Steve B.

remove "edy" from address for email

% 3) Use 50K pots rather than 100K - again this will affect car
% performance.

Just one thought on this one...  The pot's ohmage and peformance is also
tied to one's gameport quality and bus speed.

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

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

Hi Steve,
I'd be very keen to know the exact spec of the wheel, converter and gameport you
are using. It sounds like you are much more on top of the pot noise problem than I
am. I have tried the MS wheel/pedal combo in a store demo and wasn't too impressed.
This latest discovery that the modem connection affects my wheel/pedals is most
disturbing but does, at least partially, explain why I can usually get better times
in practice than I can on VROC.
Cheers,
Paul

Hi Michael,
I'm using Thrustmaster's ISA ACM gameport. Obviously the ISA bus is a slower than
PCI which is slower again than USB. What products could I get to accomodate my
existing gameport based wheel/pedal combo? What other steps can I take to eliminate
pot noise and the seeming interference I am getting between my modem and my game
port.
Thanks in advance,
Paul

Hi Paul -
I'm currently using a Logitech wheel, the Wingman Formula (non-FF), plugged
into an Ensoniq AudioPCI card.  The A/D converter is an integral component of
the wheel.  I was unaware it had one until one of the guys at Logitech (who've
been very helpful!) mentioned it in response to a query of mine about the pot
stability.  You'd think the marketers would trumpet that sort of thing.
Apparently, there's also some averaging of the pot values done at a firmware
level, further reducing noise.  I had suggested it might be marginally
overdamped, but I'm probably just revisiting the old A/D - CD/LP argument of
years ago.  We know how that turned out.  I've seen a number of mentions of the
wheel's stabilty with other soundcards, so I don't think that's at issue.  The
Logitech's not a perfect setup out of the box, but it's pretty easy to tweak
into a nice package.

Cheers,
Steve

you are using.<<

remove "edy" from address for email