rec.autos.simulators

where is a web page with info on this wheel? pix, prices

--
Pat Dotson
IMPACT Motorsports

If you have experience with current FF devices, and don't
think they are any good, please tell me why.

--
Pat Dotson
IMPACT Motorsports

I have tried FF joysticks and whilst they are fun, the feedback isn't
exactly subtle and has noticable lag. To be worthwhile for a sim, FF has to
provide the 'seat of the pants' feel with comparable resolution to that we
get audibly and visually - i.e. 25 - 30 times a second. One off effects for
hitting walls or even kerbs are not going to make the *driving* any better.

Getting FF latency down to around 0.04 seconds will require rather better
components than could be fitted to a $199 wheel.

But there again, some people like the standard Per4mer and find 10fps
acceptable. No doubt they will also find the current state of the art FF
acceptable.

Cheers,
Richard

PS: Ever wondered why there isn't a FF TSW or ECCI? <g>

I haven't seen one of those wheels so I'll trust your judgement.

I agree wholeheartedly about the "one off" effects - we need to feel the
road.  According to the research I've read, lag times of less than 25
milleseconds are imperceptible.  That's 40 updates a second, which is
well
within the capability of a serial port.  Maybe the lag you've seen on
joysticks is due to poor programming?

Are you talking about latency across the communications bus?  0.04
seconds
shouldn't be a problem, especially if USB can be used.  As far as the
latency between the I-Force processor and the control motors, the
information
I've read implies that it can maintain a 1000 Hz loop (1000 times a
second).

If there is a market for it, then why not include it?  My argument
against leaving it out is that including FF support will speed up
development of the hardware.  No one is going to produce the hardware
if there is no software to support it!  So what if the currently
available controllers have too much lag?  It will be that much shorter
a period of time until better controllers are available if the software
support is there.

Maybe because there is a big difference between wiring together some
springs and potentiometers vs. implementing motors, processors, and
software?  There is a big leap in the engineering involved - perhaps
they aren't up to the challenge :)

BTW, Thrustmaster does supposedly have an I-Force 2.0 compliant
wheel in the works.

I do find it hard to believe that good FF can be had for the price
of some of these wheels.  I'm not going to write them off without
trying it myself, though.

--
Pat Dotson
IMPACT Motorsports

Until serious sim makers such as Papyrus and others adopt FF and there
is a way to implement sideslip and centrifugal force then it will most
likely remain as it is today, a silly arcade game gimmick :(.

Trev

Pat,
i've already answered your question - see above!

You can edit the effects in F1RS and POD using iForce Studio.

This is why they should include the capability, even if it isn't
perfect.

I think that the effects produced in Hard Drivin' were incredible.  It
came out, what, 10 years ago?  It can surely be reproduced on a PC
by now.

Maybe you mean something in addition to a wheel?  I don't think that
platform-
type motion simulators will be accepted for a long time.  I've kind of
been kicking around ideas for a seat-cover type of thing that would
apply
forces to the body.  If you are turning right, the left side of the
seat stiffens, if you step on the gas the back of it stiffens, etc.
That will require another level of software support beyond what's
already
available, so implementing it would be difficult right now.

BTW, how about a sponsorship? :)

--
Pat Dotson
IMPACT Motorsports

Yes, the API provides everything that's needed to get the job done
pretty darn well. The problem is building the device that gets the job
done well.

Not at all; that problem is easy, provided you are willing to stick a
small microprocessor in the unit, or less flexibly, build a small
analog computer to compute the error term. For the computational power
you'd need, you'd be looking at no more than the 386 or 486 embedded
range. You could probably get by with an HC11 or large PIC, but
embeddable 486's aren't all that expensive...

At this point (assuming a digital control loop), you're better off
installing optical shaft encoders [no wear, EXCELLENT tracking], which
adds to the cost, but makes the end product much better. You now have
drift free, multiple rotation (> 360 degrees) steering input. (And if
you write the controller code right, it can be "self-calibrating"
[drive it full left deflection, full right deflection, and "bingo"
it's calibrated].

The real problem is how do you make parts (encoders, populated circuit
boards, motor and power supply) that cost >= $500 "fit" in a consumer
price-point item.

In other words, for pure force feedback, you'd need to either:

Have a servo motor that did "everything" so there was no mechanical
spring or damper in the unit; that's a fairly powerful motor, and
you'd need a reasonable control loop to make it feel right.

Or, you need a spring in the system and a servo motor powerful enough
to overcome it (when the front wheels are airborne for example, or
they are driven to the steering stop by impact). Again, you need a
powerful motor and an even faster control loop that can "back out" the
effect of the spring to make it feel right.

Agreed all round.

It's not computational power, it's mechanical design, component cost
and reliability.

---Jim

Usually not enough force (force is also a function of power supply; PS
without sufficient "stiffness" also can't drive the servo motors well;
"stiff" power supplies with high continuous current aren't cheap, even
when powered by the mains [as they would certainly have to be])

I disagree somewhat; why bother implementing something that gives your
customers comparatively little benefit and that you think is a joke
given the state of the world. It would be akin to modelling rain and
water effects "correctly" just to give Intel something to chew on as
they design the next generation of chips.

Jump-starting consumer force-feedback is not Papyrus' job.

Agreed. Any idea what a Hard Drivin' cabinet with everything working
goes for? :-)

Sure it could, but not for $40 in parts... (Factoring in the "factor
of five" from parts cost to final product retail.)

---Jim

I really liked the Key ;-)    I would guess an unit like that could cost
more than 10000$ these days, especially with collectioners. I have a dream
to have one in my house, along with a life-size replica of Han Solo in
Carbonite (that's another fantasm !).

Fran?ois Mnard <ymenard> Good race at the Brickyard, (-o-)

Official Mentally retarded guy of r.a.s.
Member of the r.a.s. Ego-maniac club
Excuse me for my English (I'm French speaking)
Excuse me for being provocative (I'm dumb speaking)

--"People think it must be fun to be a super genius, but they don't realise
how hard it is to put up with all the idiots in the world."--

The other thing to remember is that Atari has a good patent on FF combined
with driving games, so anyone producing FF wheels has the added cost of a
license.  Not a lot, but in manufacturing the costs add up quickly...

-- Doug

The cost of licensing is a consideration, true but the main cost factor
is our requirements for quality components such as motors, gear trains
etc. Since we have the best warranty of any manufacturer of racing
controllers, we demand the highest possible precision in component parts
and these dont come cheap!

The toy FF wheels out right now will probably last about as long as the
warranty and when they fail, it isnt going to be a simple cheap broken
spring, bungee or whatever and may cost as much as the wheel did just to
get it going again.

Thomas Ent has a rep for absolute quality and service and we wont
endanger that by using inferior components.

Trev

Current fast PC's can now equal or better the performance of the 6
processors (including several fast DSP's) that were in the final version of
Race Drivin', we have a model that can do it.  But I doubt that anyone will
be coming out with a FF wheel that uses a huge DC torque motor like the
sit-down versions of the game -- it's about 6 inches in diameter and 9-10
inches long!

Probably not at home (too big)...but there are several arcade examples.
We will be in Chicago soon, with the NASCAR Silicon Motor Speedway:
        www.lbet.com

-- Doug

                Milliken Research Associates Inc.