iRacing - Throttle-On Braking with the F-Mazda
pdo...@mindspring.com
is in hopes that we can have a real and detailed discussion of the
issue. On the iR forums the discussion will get bogged down with a
bunch of lightweights chiming in on how accurate iRacing is, and how
stupid it is to question anything.
The issue is how a rearward brake bias combined with application of
throttle affects braking distance.
There was a post on the iR forums yesterday with a Formula Mazda setup
attached. The post was from one of the fastest guys in iRacing. IMO
the setup is ludicrous. I think it's as bad as any exploited setup in
any previous sim - GPL, rFactor, or whatever.
My previous official qual time for this week's track using my own
setups, and with substantial testing, was 1:06.8. I had one or two
laps in testing at .79 or so. Pole right now is 1:06 flat. I was
fairly happy with the car, but didn't see where I could possibly pick
up another 1/2 to 3/4 second. This has been standard for the last six
weeks of the Mazda series. I qualify around 10th or 15th for the
week, but I'm around a second off the pole time.
Then came the setup post in the iR forums. Brake bias at 48% and ride
heights at around 1/2 inch front and rear. I'll just disregard the
insanely low ride heights for now. Brake bias is the bigger issue.
So I load the setup and hit the track. First thing I notice, while
straight line braking you can't even breath on the brake pedal with
zero throttle or the car will spin. OK, well, even though I've
abandoned that style of driving, I have a good handle on the throttle-
on braking technique from years of GPL.
Next thing I notice is that you can stop more quickly by essentially
pushing both pedals to the floor than you can by balancing your brake
bias and stopping with the brake pedal only. After a few laps I'd
beat my previous best time by a tenth. With a little more practice I
entered a qual session and layed down a 1:06.4.
So great, right? I got a new setup for free that gained 0.4 seconds.
No, it's not great. I'm disgusted by it.
I always assumed the gain in lap times with those rearward brake bias
settings was the result of allowing trail braking closer to the limit
during turn entry. If that was the case you could argue that driver
skill was at play.
That doesn't seem to be the situation at all. The car just slows down
more quickly with a ton of throttle on top of braking if you use an
extremely rearward brake bias.
There has been a lot of discussion about this exact technique on the
iR forums. I just glossed over the topic for the most part figuring
it wasn't that big of a deal. Well now it is a big deal to me.
So what are the vehicle dynamics issues at work here? How will this
behavior be explained so that we can believe that a real car will stop
so quickly with the engine driving the rear wheels?
I have a feeling that it can't be explained, but I'm willing to listen
with an open mind.
--
Pat Dotson
David Fisher's Left Testicle
Going back to the low hide height: won't the bumpy tracks and bottoming out
discourage that sort of thing?
<pdo...@mindspring.com> wrote in message
news:4ef998d0-1351-404b...@y38g2000hsy.googlegroups.com...
Tim Wheatley
<snip>
> Pat Dotson
People are taking advantage of the lack of internal damage. They will
have to stop doing that or they'll not have a car beneath them for
long.
Bob Simpson
>
> Next thing I notice is that you can stop more quickly by essentially
> pushing both pedals to the floor than you can by balancing your brake
> bias and stopping with the brake pedal only. After a few laps I'd
> beat my previous best time by a tenth. With a little more practice I
> entered a qual session and layed down a 1:06.4.
>
> --
> Pat Dotson
Do you mean literally full throttle and brake travel? Sounds like a
bug and I understand your frustration, but I'm sure that iRacing is
working on it. Frankly I'm surprised things like this haven't been
ironed out yet.
Also, doesn't the very low ride height help with downforce? That
would shorten the braking distances too, I think.
I'm really looking forward to when the iRacing cars' brakes and
transmissions wear out or break when abused.
Bob Simpson
Tim Wheatley
Here you go (I never thought I would link to Wikipedia for factual
info, but anyway...)
http://en.wikipedia.org/wiki/Left-foot_braking
It covers the car stabilization of trailing the throttle under
braking, the possibility of altering brake bias to take advantage of
this more fully - but also refers to the damage caused to the car
internally when doing it.
Tony R
> It covers the car stabilization of trailing the throttle under
> braking, the possibility of altering brake bias to take advantage of
> this more fully - but also refers to the damage caused to the car
> internally when doing it.
I believe Pat is referring to lowering threshold braking times rather
than using trail braking to adjust the balance.
I guess the throttle is acting as a form of ABS on the back brakes which
are way too powerful with the bias set to 48 for normal braking with the
weight being distributed to the front they would lock up way too soon.
With a normal bias the brakes would be overpowered by the throttle and
would push the car forward increasing brake distance.
Whether the brakes would be powerful enough to stop the wheels turning
under near full or full throttle I don't know. Certainly road cars
aren't hence you see situations where people go on a wild ride with the
throttle stuck open and brakes burnt out trying to stop it.
I think this should be resolved with the brake wear and fading being
modelled. They shouldn't last long!
Cheers
Tony
pdo...@mindspring.com
> On Sep 16, 11:34 am, pdot...@mindspring.com wrote:
>
> > Next thing I notice is that you can stop more quickly by essentially
> > pushing both pedals to the floor than you can by balancing your brake
>
> bug and I understand your frustration, but I'm sure that iRacing is
> working on it. Frankly I'm surprised things like this haven't been
> ironed out yet.
Yes. With extreme rear brake bias set just right I think you can stop
equally as fast by pressing both pedals to the floor as you can by use
brake pedal only with a balanced brake bias setting. If not equal
then very close. With the slightest effort in modulating the throttle
you can brake more quickly with 50% to 75% throttle and 45% and lower
front brake bias.
> Also, doesn't the very low ride height help with downforce? That
> would shorten the braking distances too, I think.
Low ride height will increase downforce, but you shouldn't be able to
corner with the bottom of the chassis on the ground. The track this
week is extremely bumpy. The car is bottoming out all over the place
but doesn't seem to be affected. I remember the late model used to
slide up the track at Irwindale if it bottomed out in the turns. The
Mazda doesn't seem to be affected.
--
Pat Dotson
pdo...@mindspring.com
>
> > People are taking advantage of the lack of internal damage. They will
> > have to stop doing that or they'll not have a car beneath them for
> > long.
>
> Here you go (I never thought I would link to Wikipedia for factual
>
> It covers the car stabilization of trailing the throttle under
> braking, the possibility of altering brake bias to take advantage of
> this more fully - but also refers to the damage caused to the car
> internally when doing it.
I hope mechanical damage will address the issue. I honestly don't
expect that sort of thing to be implemented in lower license level
cars. Maybe the higher level cars will be introduced with mechanical
damage.
The article is a shallow look at LFB stabilization during cornering.
I understand and use LFB to turn the car all the time - even with
conservative brake bias settings. That's not the issue.
The issue I'm talking about doesn't concern cornering at all. I'm
talking straight line braking distance. There really are two separate
parts to this:
1) You ostensibly couldn't use full brakes and throttle like this in
real life due to temp and wear.
2) Even if you could use this technique in real life, there is no
reason that it would yield the same sort of performance gain in
straight line braking.
So really, mechanical damage should help somewhat at least during
races. Implemenation of damage alone will not stop exploitation of
whatever is going on with the tire modelling. I don't think temp and
wear will slow down qual times at all.
--
Pat Dotson
pdo...@mindspring.com
> heights at around 1/2 inch front and rear. I'll just disregard the
> insanely low ride heights for now.
I'm a little encouraged now on the ride height issue. The good news
is I took my original setup with ride heights over 1 inch, changed
brake bias to 48%, and went faster than I did with the setup from the
iR Forum. Got down to a 1:06.3 in testing. That's just a couple
tenths off the setup owner's qual time.
So while excessively low ride heights don't seem to negatively impact
the car as they should, low RH's don't appear to give any great
advantage either.
--
Pat Dotson
Tim Wheatley
Tyres are quite WIP I think, unfortunately.
Internal damage isn't modelled at all yet, but it will be, right down
to the rookies (this is why the lower classes get car resets, people
think it's for the crashes, hehe)
I'm wondering, are you driving this setup with the auto clutch on or
off?
I really don't think weight transfer is happening so much in this type
of stiff suspension cars... I think the suspension takes the strain
and the body doesn't roll forward much at all... If you can use the
throttle to help with this, it would explain why you can benefit a
little... But you're right, it should be looked at. Can you email me
(my email address on RAS is real - gmail spam filter is very good)
giving a clear explanation of what you're seeing and I'll forward it
on, that way you won't have to post in the forums. I get the feeling
I'm perhaps not understanding full implications as you seem more
concerned than I would be... I'm no physics prof. :)
Larry
of my RC Mazda setups have had the bias set between 48 and 52 or so.
I am also a power-on braker. Bear in mind we're not talking about
pedal-to-the-metal throttle here. We're talking about braking and a
judicious amount of throttle, both combined, to get the best turn-in out of
the car and keeping it balanced at the same time. I'm better at balancing
the car with both feet than one foot (brake OR throttle).
For me, I think the habit started in the Skippy. The skippy is notoriously
ass-happy under decelleration, and trail/power-braking is what helped me
survive that car.
In closing, I should also add that I am definitely NOT the right person to
take RC tips from LOL
-Larry
<pdo...@mindspring.com> wrote in message
news:4ef998d0-1351-404b...@y38g2000hsy.googlegroups.com...
Steve Blankenship
news:aae08b65-d075-409b...@73g2000hsx.googlegroups.com...
On Sep 16, 2:08 pm, Bob Simpson <simpsonthesec...@hotmail.com> wrote:
> On Sep 16, 11:34 am, pdot...@mindspring.com wrote:
>
> > Next thing I notice is that you can stop more quickly by essentially
> > pushing both pedals to the floor than you can by balancing your brake
>
> Do you mean literally full throttle and brake travel? Sounds like a
> bug and I understand your frustration, but I'm sure that iRacing is
> working on it. Frankly I'm surprised things like this haven't been
> ironed out yet.
Yes. With extreme rear brake bias set just right I think you can stop
equally as fast by pressing both pedals to the floor as you can by use
brake pedal only with a balanced brake bias setting. If not equal
then very close. With the slightest effort in modulating the throttle
you can brake more quickly with 50% to 75% throttle and 45% and lower
front brake bias.
Shot in the dark here Pat, but it might be a way of effectively reducing the
braking force at the loaded wheels sufficiently that you do a better job of
staying on the edge of the traction circle under braking. Here's my
(questionable) rationale. The brakes in just about all the iR cars seem too
powerful to me, as you can lock them at speed much too easily - that's been
a much-discussed topic, of course. By moving the bias back, you reduced
total possible force applied at the fronts, while simultaneously getting
dynamic, right-foot-controlled abs at the rear. That could well make it
easier to threshold brake, even in a straight line. With iRacing's tire
model we don't get that immediate drop-in-grip-feel right past the limit
that we've gotten used to from ISI's tires, so I'd bet more wheel locking is
going on under braking with your original setup than you'd expect. It'd
take telemetry to really prove it, but that's what I'd be expecting the
traces to tell me.
I'm a left-foot-braker too, and have noticed in working on setups for
ISI-based sims lately that I'm lowering the braking force to avoid excessive
locking in order to save the tires, and have concurrently been moving my
brake bias forward to get my stopping distances down. If I run 100% brakes
in say, GTR2, I find I'm moving the bias back to keep from locking the
fronts on turn-in, but if I drop the braking force a bit I can move the bias
forward and get shorter stopping distances without lockups. So that sort of
agrees with my suspicion about iR's killer brakes.
Might be an interesting experiment with iRacing to alter your brake pedal
calibration values with DXTweak to take out the last part of the travel,
effectively lowering the braking force (leaving the original, full-travel
in-game calibration alone). Find a value that lets you just barely lock
them up and see how it affects braking and lap times...
Might all be 100% off of course, but food for thought anyway.
SB
pdo...@mindspring.com
I am coming around to the idea that this behavior might not be far
from reality. Most of the stopping power comes from the front
wheels. If you adjust the bias to minimize potential for front wheel
lockups braking will be easier. It's a lot like lowering brake
pressure in rFactor, except that in that case you don't really have to
alter your driving style.
I'm still surprised that the 100% brake and throttle thing seems to
exceed the performance of a balanced brake bias / brake pedal only
approach.
It might come down to controller quality though. I've ordered one of
Leo Bodnar's G25 pedal USB adapters which is supposed to improve the
response of the pedals. Maybe manual braking will be more effective
with smoother controller response.
One thing I am thankful for is that all this has highlighted where I
was losing time. I've been struggling at around a second off of
Huttu's pace week in and week out. This week I'm 0.3 seconds off his
qual time with only one day of using rearward brake bias with throttle-
on braking, and that's with only one day practicing the techinque.
Maybe I can catch him next week! :)
--
Pat Dotson
vigi
wrote:
>One thing I am thankful for is that all this has highlighted where I
>was losing time. I've been struggling at around a second off of
>Huttu's pace week in and week out. This week I'm 0.3 seconds off his
>qual time with only one day of using rearward brake bias with throttle-
>on braking, and that's with only one day practicing the techinque.
The only thing is, he doesn't use that technique.
pdo...@mindspring.com
> On Thu, 18 Sep 2008 07:45:44 -0700 (PDT), pdot...@mindspring.com
By "technique", I meant generally a rearward brake bias with throttle-
on braking - Not the full brake and throttle deal. I was running more
like 60% front brake bias and it was killing me.
BTW, I ended up outqualifying the original poster of the rearward
brake bias setup over at the iR forums. In the end I gained 0.7
seconds from my original qual time, down from 1:06.8 to a 1:06.1.
--
Pat Dotson
Tim Wheatley
I cannot believe Greger... He just replied to your thread over at iR
and said he did a 1:06.1 braking with his right foot when he was
seemingly bored...
Asgeir Nesoen
braking with your heel as you throttle with your toe.
The traditional american way of scrubbing off speed is just to stomp the
brakes, and pick your exit gear as the speed gets low enough to do so.
As a sidenote I would think that the historical reason for this is that
the americans have always raced heavier cars, and ditto strong
transmissions, with ensuing more awkward shifting. In europe, and
britain in particular, race car drivers have used the throttle while
braking to ensure that the car brakes in a straight line without
oversteer over a wide range of racing conditions. The use of the
throttle while braking would be limited to, I'd guess, about 10-15%
engine power. We're talking a very fine addition of power to the rear
wheels. If you have to brake over a hill, you'd use additional throttle,
if you have to brake at the bottom of a hill, you'd use none.
In a non aero-downforce racing car there is no more efficient way of
braking than being as close as possible to the adhesion limit of each
tyre. If you cross the slip-angle/grip curve efficiency is impaired, as
being to far on the "safe" side would.
You can take out a racing car on a scratchpad and try and see how short
braking distance you can get by stomping the brakes while free coupling
with the clutch and compare it to doing the same thing while
downshifting and throtteling without using your clutch. You will always
find that the distance will be shorter without downshifting and throtteling.
Using the throttle while braking is only performed in order to control
the over/understeer of your car (and of course matching up the input
shaft speed with output shaft speed), never to increase the braking
efficiency. If you depend on your downshifting to control the braking
power, you'd wear out the clutch/box in 10 laps, and you'd achieve no
consistency whatsoever since it is next to impossible to change gears in
a way that would ensure constant braking.
I would think that iRacing need to have a look at more than the internal
wear to fix this, given that "this" being the case as described here. I
cannot come up with any racing class on the planet where it is efficient
using engine-braking to this extent.
--A--
Asgeir Nesoen
from a well-controlled physics environment (suspension movement being
controlled in a predictable and "calculatable" way), to a unknown state
where you cannot calculate the grip reduction etc. In practice, I'd
think it would be impossible to real-time simulate this situation, since
the tiniest of details and time spans would cause devastating effect on
grip and balance.
Bottoming out is something that would never take place under normal
racing conditions because the racing would become so erratic, so I would
guess that reducing grip heavily on the tire bottoming out would suffice
to prevent racers from taking advantage of low ride height while not
having to pay the bill for it.
If a light race car bottoms out it should become uncontrollable, and if
a simulation fails to implement this it lacks a very important part of
simulating race cars IMV.
--A--