rec.autos.simulators

F1 Challenge - Mechanical Failures

Nats

F1 Challenge - Mechanical Failures

by Nats » Thu, 17 Jul 2003 06:20:03

I did a bit of a quick test the other night on some races to see what
results I got on mechanical failures as follows for anyone that is
interested:

I ran sample races on autopilot Australia, Monza, Silverstone and Canada and
varied driver strength, aggression, race length, mechanical failure type
etc. The way I did this was to start the race and then drop out shortly
after the start and run the rest of the race on autopilot.

Generally no real difference could be found on effects on mech failures
whilst the game, was on autopilot, between differing driver strengths and
aggression. The races gave between 9 and 15 'did not finished's varying from
clutch/electronics (false starts), and then most race dropouts were through
suspension, engine or accident. More failures were experienced when rnning
full length races compared to 50% races on time scaled settings and of those
most failures were experienced on full 100% races run with normal mechnical
failures compared to time scaled (usually around 11-15 dropouts). However
nearly every in race dropout was due to suspension or engine failures. This
is almost certainly because I was running the first few minutes and then
exiting the race to let it run on autopilot which seems to automatically
create only suspension and engine failures except for the odd accident. So
not very promising generally and of these two types of failure they were
always one after the other ie one suspension failure then an engoine then a
suspension etc for all the drop outs.

However the good news is that if you actually drive in the race yourself,
whilst youre driving the failures that take place are generally more
imaginative. For example in the few short races I have run I have
experiences AI cars failing through brakes and fuel loss. I have also
experienced wet races where all runners fell out by dnfs which I never got
in any races on autopilot. So it seems likely that the computer AI for cars
is far more accurate and detailed whilst youre in the race yourself. However
in the few races Ive run to the end Ive experienced less numerous dropouts
that when the races are run on autopilot.

So the summary is that whilst you are racing yourself, the cars do tend to
suffer imaginative failures but generally fewer of them. However whilst the
races are being run on autopilot you tend to get more dropouts but theres
very little variation between the failures experienced. The results do
suggest though that running full length races on normal mech failure
settings show more attrition than 50% races on timescale failures.

But unfortunately not many of us will be able to spend the time to run full
races much and certainly not having a save feature whilst during the weekend
or race this is practically impossible to do unless you are a) a student b)
unemployed or c) unmarried and preferably all three. When anyone develops a
decent save feature for this game please let me know!

This also all tends to suggest that if we were ever given a replay facility
where you could watch a whole race even when you have finished due to
failure or crashing etc it would porbably tend to be pretty boring to watch
anyway. Which is a shame because it would be great fun and works well in
GP4.

--
Nigel 'Nats' Stutt

"Its life, Jim, but not as we know it."

Remove NOSPAM to respond please.

Jason Moy

F1 Challenge - Mechanical Failures

by Jason Moy » Thu, 17 Jul 2003 06:59:00

On Tue, 15 Jul 2003 22:20:03 +0100, "Nats"


>I did a bit of a quick test the other night on some races to see what
>results I got on mechanical failures as follows for anyone that is
>interested:

>I ran sample races on autopilot Australia, Monza, Silverstone and Canada and
>varied driver strength, aggression, race length, mechanical failure type
>etc. The way I did this was to start the race and then drop out shortly
>after the start and run the rest of the race on autopilot.

The way the game simulates when you drop out of a race or press the
"next session" button in qualifying is not very reliable.

If you want a reliable simulation of qualifying, you have to press
CTRL-T to accelerate time and just sit and wait for the clock to run
out.  If you want to simulate a race, drive into the pits and park
your car and, without getting out of your car, hit CTRL-T to
accelerate time and then go get a burger or something until the race
is over.  I have no idea how ISI simulates things otherwise, but the
results of accelerating time and letting the sim rapidly calculate
things are totally different, with the latter being way off in almost
all cases.

Jason

Haqsa

F1 Challenge - Mechanical Failures

by Haqsa » Thu, 17 Jul 2003 07:26:58

Did you notice any pattern in which drivers were having failures?  Because
if you look in the .gdb file for the track there is a list of all the
drivers who are going to have incidents.  Not sure if that only happens in
championship races or what, but I thought it was odd because it seems to
imply that AI DNF's will not be random.


> I did a bit of a quick test the other night on some races to see what
> results I got on mechanical failures as follows for anyone that is
> interested:

> I ran sample races on autopilot Australia, Monza, Silverstone and Canada
and
> varied driver strength, aggression, race length, mechanical failure type
> etc. The way I did this was to start the race and then drop out shortly
> after the start and run the rest of the race on autopilot.

> Generally no real difference could be found on effects on mech failures
> whilst the game, was on autopilot, between differing driver strengths and
> aggression. The races gave between 9 and 15 'did not finished's varying
from
> clutch/electronics (false starts), and then most race dropouts were
through
> suspension, engine or accident. More failures were experienced when rnning
> full length races compared to 50% races on time scaled settings and of
those
> most failures were experienced on full 100% races run with normal
mechnical
> failures compared to time scaled (usually around 11-15 dropouts). However
> nearly every in race dropout was due to suspension or engine failures.
This
> is almost certainly because I was running the first few minutes and then
> exiting the race to let it run on autopilot which seems to automatically
> create only suspension and engine failures except for the odd accident. So
> not very promising generally and of these two types of failure they were
> always one after the other ie one suspension failure then an engoine then
a
> suspension etc for all the drop outs.

> However the good news is that if you actually drive in the race yourself,
> whilst youre driving the failures that take place are generally more
> imaginative. For example in the few short races I have run I have
> experiences AI cars failing through brakes and fuel loss. I have also
> experienced wet races where all runners fell out by dnfs which I never got
> in any races on autopilot. So it seems likely that the computer AI for
cars
> is far more accurate and detailed whilst youre in the race yourself.
However
> in the few races Ive run to the end Ive experienced less numerous dropouts
> that when the races are run on autopilot.

> So the summary is that whilst you are racing yourself, the cars do tend to
> suffer imaginative failures but generally fewer of them. However whilst
the
> races are being run on autopilot you tend to get more dropouts but theres
> very little variation between the failures experienced. The results do
> suggest though that running full length races on normal mech failure
> settings show more attrition than 50% races on timescale failures.

> But unfortunately not many of us will be able to spend the time to run
full
> races much and certainly not having a save feature whilst during the
weekend
> or race this is practically impossible to do unless you are a) a student
b)
> unemployed or c) unmarried and preferably all three. When anyone develops
a
> decent save feature for this game please let me know!

> This also all tends to suggest that if we were ever given a replay
facility
> where you could watch a whole race even when you have finished due to
> failure or crashing etc it would porbably tend to be pretty boring to
watch
> anyway. Which is a shame because it would be great fun and works well in
> GP4.

> --
> Nigel 'Nats' Stutt

> "Its life, Jim, but not as we know it."

> Remove NOSPAM to respond please.

Nats

F1 Challenge - Mechanical Failures

by Nats » Thu, 17 Jul 2003 08:40:01

Hmm I will look at that file.

Furthermore I have found that more interesting accidents seem to happen when
the other cars are set to FULL visual effects but that of course could be
just a coincidence so I will try a few more races with that setting as Im
sure that must be just a display enhancement.

On an unrelated point there is a setting in the .PLC file that is really
nice. Try setting environment map to 1 to really get a sense of speed. It
enhances the track graphics greatly to show the texture of the termac
brilliantly. In fact its worth lowering the circuit graphics a notch to get
smooth racing with this setting on to really appreciate the driving
experience.

--
Nigel 'Nats' Stutt

"Its life, Jim, but not as we know it."

Remove NOSPAM to respond please.

> Did you notice any pattern in which drivers were having failures?  Because
> if you look in the .gdb file for the track there is a list of all the
> drivers who are going to have incidents.  Not sure if that only happens in
> championship races or what, but I thought it was odd because it seems to
> imply that AI DNF's will not be random.



> > I did a bit of a quick test the other night on some races to see what
> > results I got on mechanical failures as follows for anyone that is
> > interested:

> > I ran sample races on autopilot Australia, Monza, Silverstone and Canada
> and
> > varied driver strength, aggression, race length, mechanical failure type
> > etc. The way I did this was to start the race and then drop out shortly
> > after the start and run the rest of the race on autopilot.

> > Generally no real difference could be found on effects on mech failures
> > whilst the game, was on autopilot, between differing driver strengths
and
> > aggression. The races gave between 9 and 15 'did not finished's varying
> from
> > clutch/electronics (false starts), and then most race dropouts were
> through
> > suspension, engine or accident. More failures were experienced when
rnning
> > full length races compared to 50% races on time scaled settings and of
> those
> > most failures were experienced on full 100% races run with normal
> mechnical
> > failures compared to time scaled (usually around 11-15 dropouts).
However
> > nearly every in race dropout was due to suspension or engine failures.
> This
> > is almost certainly because I was running the first few minutes and then
> > exiting the race to let it run on autopilot which seems to automatically
> > create only suspension and engine failures except for the odd accident.
So
> > not very promising generally and of these two types of failure they were
> > always one after the other ie one suspension failure then an engoine
then
> a
> > suspension etc for all the drop outs.

> > However the good news is that if you actually drive in the race
yourself,
> > whilst youre driving the failures that take place are generally more
> > imaginative. For example in the few short races I have run I have
> > experiences AI cars failing through brakes and fuel loss. I have also
> > experienced wet races where all runners fell out by dnfs which I never
got
> > in any races on autopilot. So it seems likely that the computer AI for
> cars
> > is far more accurate and detailed whilst youre in the race yourself.
> However
> > in the few races Ive run to the end Ive experienced less numerous
dropouts
> > that when the races are run on autopilot.

> > So the summary is that whilst you are racing yourself, the cars do tend
to
> > suffer imaginative failures but generally fewer of them. However whilst
> the
> > races are being run on autopilot you tend to get more dropouts but
theres
> > very little variation between the failures experienced. The results do
> > suggest though that running full length races on normal mech failure
> > settings show more attrition than 50% races on timescale failures.

> > But unfortunately not many of us will be able to spend the time to run
> full
> > races much and certainly not having a save feature whilst during the
> weekend
> > or race this is practically impossible to do unless you are a) a student
> b)
> > unemployed or c) unmarried and preferably all three. When anyone
develops
> a
> > decent save feature for this game please let me know!

> > This also all tends to suggest that if we were ever given a replay
> facility
> > where you could watch a whole race even when you have finished due to
> > failure or crashing etc it would porbably tend to be pretty boring to
> watch
> > anyway. Which is a shame because it would be great fun and works well in
> > GP4.

> > --
> > Nigel 'Nats' Stutt

> > "Its life, Jim, but not as we know it."

> > Remove NOSPAM to respond please.

Redmis

F1 Challenge - Mechanical Failures

by Redmis » Thu, 17 Jul 2003 10:29:50

Somebody told me that you can set mechanical failures to '2002 Season' - not
sure what that means exactly - does it mean that engine blowups will occur
for the AI drivers on the exact same laps as they did for the real drivers
last year? ;-)

Haqsa

F1 Challenge - Mechanical Failures

by Haqsa » Thu, 17 Jul 2003 11:35:21

I think so.  I haven't tried it, but I'll bet that's why that data is in the
.gdb file.


Nats

F1 Challenge - Mechanical Failures

by Nats » Fri, 18 Jul 2003 04:57:51

Yes the cars fail as they did during that season. So if you want to
experience a certain failure just play a car that experienced it during that
race. I make try looking for a hydraulics failure just for fun!

--
Nigel 'Nats' Stutt

"Its life, Jim, but not as we know it."

Remove NOSPAM to respond please.

> I think so.  I haven't tried it, but I'll bet that's why that data is in
the
> .gdb file.



> > Somebody told me that you can set mechanical failures to '2002 Season' -
> not
> > sure what that means exactly - does it mean that engine blowups will
occur
> > for the AI drivers on the exact same laps as they did for the real
drivers
> > last year? ;-)


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.