a magicdata thread? could it be useful?

I have just updated the guide about compiling the magicdata (inside the track editing tutorial) as I think a good md is almost vital for a decent racing and I particularily sux at filling it. It's a proposal with tips picked here and there, so it's open to any suggestion though.
There are many useful tips around the forum but almost all are scattered through topics and lost in the fog of the search function. I've spent hours to find an old post and not a keyword pointed on it so I was thinking about packing all the tips and experiments we may enterprise in a single place easier to search.
If you like the idea let it flame on otherwise just let it fade off

Anyway, I'll start with some re-experiments I'm making with tyre compounds.

53 ; hard tyre
52 ; soft tyre
Resistance of the two tyre compounds. Hard tyres should have an higher value than the soft tyres.

I'm trying to understand if these values are percentages and/or what's their weight. In example Hard Tires do last 53% of the total laps, while Soft Tyres 52% etc...?

100 ; >= 50 tyretype 1 else 0
Tyre type selector. 0= all with hard tyres. 100= all with soft tyres.

Is this intended as percentage too? For me it reminds the brake balance. In example
50: means half grid may use softs, the other half may use hards
0: all hards
100: all softs
75: 2/3 of the grid with softs, the rest on hards
25: 2/3 of the grid with hards, the rest on softs

I haven't seen much variety of values about them. most of the md have 53/52 and 0 or 100.

ideas?

-------------------------------




Edited 1 time(s). Last edit at 04/29/2009 08:15PM by R_Scandura.

Great idea!
I think a good MD is very important to have a realistic (and so funny) race.
I experimented with MD values a lot too, i'm ready to contribute to this thread.
Very interesting the % of soft/hard tyres issue.
I'll test, even though i think it should be quite difficoult to undertand who uses hard tyres and who soft ones.
Bye

i remember there was a tool that allowed to see the condition of each car by reading the savegames. Don't remember if it was Gp3 Master but as it came to mind i'll have a look if i find it here. if this theory could be confirmed would be a great step. All AI cars race as one in gp3 so would be a interesting variable to add.

-------------------------------




Edited 1 time(s). Last edit at 04/29/2009 09:16PM by R_Scandura.

Wasn't that GP3Wears?

Going back to this: 100 ; >= 50 tyretype 1 else 0... My reading of that line is that if the value is 0-49 then hard tyres will always be used, and if the value is 50-100 then soft tyres will always be used. I am not sure there is any way to be completely sure about this unless there is a program that shows the actual tyreset number for AI cars.

Some comments also on this:
53 ; hard tyre
52 ; soft tyre

I used different values (i.e. 55/54, 54/53 etc) and ran a series of demo races. At no point did I see any difference in best lap times or lap time drop-off/increase into a race. More strange was that I also used fifferent values for the "tyre wear" line (from 3000 up to 13000) and again there was no difference in lap times.

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
The test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function. -- F.Scott Fitzgerald

Theory: 52/53/54/55 stands for super-soft/soft/medium/hard? (not exactly in this order...)

GPGSL-3: Fox race driver and team manager - 2011-2012 Constructor Champion - 2012 Driver Champion

Season 9 - #18 Erick Von Draxeler - #35 Marcus Bongardt

ErickVonDraxeler schrieb:
-------------------------------------------------------
> Theory: 52/53/54/55 stands for
> super-soft/soft/medium/hard? (not exactly in this
> order...)


Don't think so, because back in the days GP 3 was created I think there wasn't this choice of slicks (or grooved tyres, for that matter). In reality, I reckon Bridgestone and Goodyear will have used different tyre types for different races, but I don't think that has been included in GP 3.

Yes, the tool is Gp3Wears. When you open the saved game, you can see the tyre wear of CC and yours: sometimes it showed that the same driver used in race both the soft and hard tyres...and it showed this also for me. Strange, as 1998 rules didn't allowed that, and the MD was set to have everyone uses the same compound.
I used it to set in the MD the correct tyre wear (for human driver only) in ovals, and noticed differences between one value and another in ovals and in other tracks too.
But, as always, i'll test.

Hmmm, maybe the difference is small, and more important is the weight of fuel. Is there a magic value that controls the fuel penalty? I was playing around recently to try and get a 1/3 race with consistant times from start to finish but no matter what I did there was no change. I feel that tyres today are generally much more stable than in 1998 (or 2000 in my case of playing GP32k) so it would be better to have steady laps rather than a big increase in times.

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
The test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function. -- F.Scott Fitzgerald

yes Alan, there is the cc-shuffler that plays with weight somehow. I set it at 9000 and there is a great challenge with cars running 3 stops. as you lose weight, your laptimes improves sensible til the pitstop.

I did some more tests on the hard/soft tyres and got some more info.

They are not percentages. Probably the top is 64 as at 70 the game ran crazy and crashed.

Another thing I noticed is that the two values start from equal but are not independant. I used for test the Castle Combe track. I reversed the values and the behaviour of the tyres reversed too. Also changing the combo (55/54 53/52 55/53) the difference reflected on both compounds and not the one modified. Looks like they do a reference to each other somehow. But probably the value of hard must be greater than the soft with minumal differences.

I did not a test pointed on the fast lap but on tyre life. I tried to run a 10 laps stint human and then AI. I noticed a bigger decadence on hards so the value may work as a 'coefficient of friction'. Setting values like 56 for hards made a decadence of 2 seconds in few laps.

Here is a sample of my tests. I played 'wear vs fuel weight' to take it into account too and noticed a sort of balancing between the tyre decandence and the fuel loss. so with softs you have a slower decadence that will be also tied by the shuffler, while on hards the decadence overcome the shuffler

Combo 55/54
100=Softs_____0=Hards
51.902________51.758
52.177________51.925
52.246________52.105
52.239________52.225
52.095________52.218
52.100________52.260
52.111________52.386
52.029________52.380

Combo 55/53
100=Softs_____0=Hards
51.787________51.947
52.009________51.935
52.174________52.099
52.167________52.286
52.198________52.308
52.024________52.348
52.248________52.355
52.228________52.417

looks like a greater torque between them increased the wear but touched the laptimes. i'll try some other combo later

-------------------------------




Edited 5 time(s). Last edit at 04/30/2009 02:21PM by R_Scandura.

Good stuff. I would be nice to have a better understanding.

About the CCshuffler, your figure of 9000 is very dependent on the track. I tested on 2k with Belini Bahrain and found a figure of about 5200 kept steady lap times as tyres dies and fuel burnt away. A lower value than 5200 makes the car faster through a run and a value higher makes the car slower. I then tested the default Buenos Aries track (md converted to md2 of course ) and to keep steady lap times the CCshuffler needs to be set to 0. The default value in the exe is 2086 and this sees a drop of over 3 seconds in a 24 lap run which is 1/3 race distance. Using value=0 the times dropped only 0.3s. I didn't think about testing a negative value here...

Anyway, the point is, it is no good just putting in 9000 and expecting a certain behavious. I guess this is closely related to tyre wear and hard/soft tyre values.

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
The test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function. -- F.Scott Fitzgerald

i don't know about belini tracks. they are set to be in their own world with specific files he does, grip, physics x year etc. I found 9000 seems very useful on normal tracks to my usage. Of course i play with other parameters to level it but til it makes a satifying output i'm fine. I only report it for the chronicle, you're free to try your specific case

-------------------------------

This is what i noticed about tyres.
- tyre wear value in the MD is only for human (of course)
- soft tyres allow you to do better lap (at least in qualifyng, and in general with 4 laps fuel load) of about 0.500 of a second (when the combo is 54/55 or 53/54 or 52/53 and so on...), so they are different (of course)
- soft tyres have a shorter life (of course)
- but, this is my experince, CC tyre wear is something pre-determined, as CC adjust their setup to track conditions (i mean, wiht great bump value in the MD, CC had no stability/grip problem, i just couldn't stay on the ground...)
- so the tyre wear for human is affected only by the value you have set in the MD (tyre consumption) and it is aq bit different for the two compounds

The last: all this with the original physic file, as Gp3 has been developed for that physic. I fear that custom physic is just another world to discover.

Erick is right about the tyre types. 52/53/54/55 are different tyre types. In GP3 1.13 each one has its own grip and wear. In GP32K each one has its own grip, but the wear depends on what type you put to be the hard and the soft one.

Regards

savinaclaudio Wrote:
-------------------------------------------------------
> This is what i noticed about tyres.
> - tyre wear value in the MD is only for human (of
> course)
> - soft tyres allow you to do better lap (at least
> in qualifyng, and in general with 4 laps fuel
> load) of about 0.500 of a second (when the combo
> is 54/55 or 53/54 or 52/53 and so on...), so they
> are different (of course)
> - soft tyres have a shorter life (of course)
> - but, this is my experince, CC tyre wear is
> something pre-determined, as CC adjust their setup
> to track conditions (i mean, wiht great bump value
> in the MD, CC had no stability/grip problem, i
> just couldn't stay on the ground...)
> - so the tyre wear for human is affected only by
> the value you have set in the MD (tyre
> consumption) and it is aq bit different for the
> two compounds
>
> The last: all this with the original physic file,
> as Gp3 has been developed for that physic. I fear
> that custom physic is just another world to
> discover.


No, cc tyre wear is affected only by the tyre choice. Human players are affected by the tyre choice plus the consumption value in the MD. It can be checked easily.

RRRReméééédiooo Wrote:

> No, cc tyre wear is affected only by the tyre
> choice. Human players are affected by the tyre
> choice plus the consumption value in the MD. It
> can be checked easily.

A bit misunderstendig: of course the tyre wear depends first of all by the tyre type (as i wrote: soft tyres have a shorter life but allowe you to have better lap time, as they have more grip).
CC tyre wear is affected only by the tyre choice: yes (as i wrote: CC tyre wear is something pre-determined, as CC adjust their setup to track conditions: of course depending on what tyre type they use).
Human player's tyre wear is also affected by the value in MD.
So we agree.
Bye.

About something completely different:

Next year refueling in the race is history. Can we simulate this with the magicdata or is this impossible? Every car needs to start with tanks full. This can be done by just not having pitstops but there probably will be tirestops. Any sugestions?

i think it's not possible to simulate only tyre changes.
The only way eventually would be if cars will mandatory pit when the tyres are over the wear limits. in this case you could set for no stops with the fuel load and let the AI stop when tyres are gone. You'd have fresh tyres and fuel load would remain untouched. Never went into this with cars pitting for excessive wear, so probably only a theory...

anybody?

-------------------------------




Edited 1 time(s). Last edit at 05/02/2009 03:46PM by R_Scandura.

savinaclaudio Wrote:

> A bit misunderstendig: of course the tyre wear
> depends first of all by the tyre type (as i wrote:
> soft tyres have a shorter life but allowe you to
> have better lap time, as they have more grip).
> CC tyre wear is affected only by the tyre choice:
> yes (as i wrote: CC tyre wear is something
> pre-determined, as CC adjust their setup to track
> conditions: of course depending on what tyre type
> they use).
> Human player's tyre wear is also affected by the
> value in MD.
> So we agree.
> Bye.


I need to read things more slow...I've misunderstood everything.

R_Scandura Wrote:
-------------------------------------------------------
> i think it's not possible to simulate only tyre
> changes.
> The only way eventually would be if cars will
> mandatory pit when the tyres are over the wear
> limits. in this case you could set for no stops
> with the fuel load and let the AI stop when tyres
> are gone. You'd have fresh tyres and fuel load
> would remain untouched. Never went into this with
> cars pitting for excessive wear, so probably only
> a theory...
>
> anybody?

it could be possible by increasing the fuel flow so it's
faster than tyre change, I'll look into it one day

pk


belini on youtube
[uk.youtube.com]

RRRReméééédiooo Wrote:
-------------------------------------------------------
> savinaclaudio Wrote:
>
> > A bit misunderstendig: of course the tyre wear
> > depends first of all by the tyre type (as i
> wrote:
> > soft tyres have a shorter life but allowe you
> to
> > have better lap time, as they have more grip).
> > CC tyre wear is affected only by the tyre
> choice:
> > yes (as i wrote: CC tyre wear is something
> > pre-determined, as CC adjust their setup to
> track
> > conditions: of course depending on what tyre
> type
> > they use).
> > Human player's tyre wear is also affected by
> the
> > value in MD.
> > So we agree.
> > Bye.
>
>
> I need to read things more slow...I've
> misunderstood everything.

We agree, and this is what matter