***The Official Forza Motorsport 3 Thread***

Vigoro said:
But i can see that the PI ratings dont mean all that much, as i went and done a bit of a test using the second group on the top gear lap times list, couldnt use the Zonda F though as it isnt on there.

Veyron - S700 - 52.312
MC12 - S663 - 50.925
Enzo - R3 721 - 51.484
LP670 SV - S648 - 52.205

So although the MC12 isnt the highest PI it won, knowing the MC12 is based on the Enzo if i remember rightly, why is the Enzo classed as an R3 car, when other competitors are in S. It makes no real sense.
Click to expand...

T10 have said how they get the base PI rating, taking AI drivers around an imaginary course that has lots of different bits of each track (or something like that).. so comparing on 1 track is going to show differences.. put a Veyron on Le Sarthe, and surely with the humungous straights, it'd be much quicker..

I think people are expecting too much from a single PI number, without even turning a wheel in FM3, just going by how it's calculated you can immediately list a whole set of situations under which it doesn't show the real performance on any one given track/situation..

To be perfect, you would have to assess each car with every permutation of 'component' using some AI and real drivers on every course over a wide range of race distances, and Have a PI rating for every car/track/race/upgrade combination..

Or does anyone have a better idea?
 
Mr Latte said:
Here we go again, attitude, so anything ibert says I will ignore, already he just dismisses what I say, strikes it off but no surprise as he will always think I have an agenda.

The issue again let me say is not to find the fastest car on a track thats the problem with the game. The issue is to find 6-8 cars that can be matched to be competitive together. Thats what will enchance online races.

By all means look at other more normal cars on the TG track and you could use a few of the Tracks in FM3 like Sebring or Tsukuba for a testbed. Its to the benifit of everyone if as players we find a range of cars that do compete well or even with certain mods/upgrades can compete better together. Put it to the test, create a track discussion thread and follow it through with various people . Im game if you lot are.

TG Times

Heres a good range of cars

Audi R8 1.24.4
TVR Sagaris 1.24.6
Mitsubishi Evo FQ400 1.24.8
TVR Tuscan 1.24.8
Noble M400 1.25.0
Lotus Exige S 1.25.1
BMW M3 Saloon 1.25.3
Lamborghini Gallardo Spyder 1.25.7
Lotus Evora 1.25.7
Lamborghini Gallardo 1.25.8

Do any of these in FM3 remotely compare well together, if not how do you mod them to do so?
Click to expand...

There are plenty of similar cars you can pick in FM3 that perform very closely together, we do it all the time in MP, pick 'stock' cars and all hit a track, the only rule is you have to pick something different each, and it's never been an issue.

I also am in a second season of a touring car type series, you have a limit of A550 / 400BHP and a few 'must have' fixed upgrades and about 25 cars to choose from, that series has some 'favourites', but it's quite diverse, Golfs/Civics/Hyundai/Astra all have won races, and the top 3 last season where all different cars..

So in answer to what I think your point is, you don't have to look far to see things being closely matched in FM3, you just make it thatway if you want it to..

Compared to RL, well I don't doubt it'll diverge a bit, even the TG times are done in different conditions each time, temperature/weather/tyres, etc, etc.. so their times are not particularly going to relate to much, and only serve as an indicator.

All games are going to have flaws in performance measuring, what cars are 'close' on one track will differ on another track, you either have to make it work for you or against you, the choice is always yours.
 
Last edited:
Demon, its encouraging to hear that...

Well as a FM3 player that would like to enjoy the game more I would recommend to those of you that are aware of more well matched cars to perhaps then create a thread listing such cars.

Do we have big enough fans here for FM3 to do that to help the community here? Perhaps some of you are aware of such a thread on other FM3 forums thats listed such well matched cars to particular tracks...
 
^^been trying to find something like that for ages, it's so annoying when I'm in 2nd place but 1st place is like 4000 yards ahead and there is literally nothing I can do :( oh, and its usually an s5!
 
Sp00n said:
^^been trying to find something like that for ages, it's so annoying when I'm in 2nd place but 1st place is like 4000 yards ahead and there is literally nothing I can do :( oh, and its usually an s5!
Click to expand...

What class? S5 can be competitive in A to C so need to know. :p
 
Sp00n said:
^^been trying to find something like that for ages, it's so annoying when I'm in 2nd place but 1st place is like 4000 yards ahead and there is literally nothing I can do :( oh, and its usually an s5!
Click to expand...

This isn't quite the same issue, no doubt that S5 is highly modded/tuned.. what you need to do is find a good 'tune' for another car that can rival the S5..

for example, you could start at Official FM.NET Tuning marketplace there are several tuning shops operating out of there with plenty of good cars..

However if you want a random car you like to compete with a tuned S5, that might not be possible!
 
Last edited:
I'll bet that when GT5 comes out (around the time of the second coming, yeah?) there'll be exactly the same problem on the leaderboards with the same cars dominating things.
 
vonhelmet said:
I'll bet that when GT5 comes out (around the time of the second coming, yeah?) there'll be exactly the same problem on the leaderboards with the same cars dominating things.
Click to expand...

This isnt the place to discuss it regards GT yet they did show it possible to use actual GPS technology from a real car/track recorded and inputed into the games replay engine. So I presume if they wanted to, although not to say they have. They could for example taking TG use the TG licensce using GPS data from their TG track to record laps from guests, cast or indeed the STIG's drives implemented as a ghost car/replay within the next GT game. The point they demonstrated with the Toyota/GPS feature was it was possible to use real car performance data and implement it into the game. I very much doubt this will mean all its cars will perform close to the real life counterparts but the technology is interesting.

It could well have the dominant cars problem as FM3 too. If it does it wont be defended by me. Theirs no excuse for developers using 100s of cars not being able to ensure many of them perform similar. Its a fundemental gamplay mechanic for good races imo. GT5P did have some good matches from what I remember but its about a year since I played it online.

Demon
Regards real cars at TG and different temps weather varibles thats covered by the 1 second margins in the cars listed. How do such similar cars on various tracks compare, point is if you have in most clases 1-2 cars that easily goes much faster then thats a bit pap and why leaderboards usually have limited cars. The leaderboards are likely to give the answers to what are the cars that are the most superior and should then be the ones that wont offer the best races unless everyone is in those superior cars.
Id certainly appreciate input regards cars that create good races please...

[EDIT]
Lets say I as to run a LAN with mates what 8 cars or indeed tracks would be a great race in the various classes in FM3?
Maybe some of you are aware of a certain track that shows more promise towards different cars mixing well together. It seems looking at Sebring both in Full & Short youd have a hard time getting more than 5 different cars in the top 50 even upto the top 100 times in many of the car classes. I havnt looked but maybe thats also the case with other tracks as well and if so doesnt it then prove the PI issue or general car physics arent well matched as a strong point but perhaps even a crippling factor of the game? Yes Im sure their are cars that dont have such differences when compared together. However has any FM community created a good list of those cars? By all means someones welcome to point out a track and leaderboard that indeed is really well mixed with different cars or point out groups of cars the make the game more enjoyable when raced together.

Leaderboards

One person stated these are common cars in the game that seem to be quickest.
Porsche RS Spyder #16 in R1

Lola Mazda, Koenigsegg CCGT in R2

Dodge Viper 08, Lola Mazda in R3

Dodge Viper '08 in S

Audi S5, Tom's MR-S, Subaru WRX '05 in B and A

Celica '03 D and C

Golf MkII F and E
 
Last edited:
I've just done my first endurance race - did the F class 34 laps of Road Atlanta with a slightly upgraded Bora. Was a minute ahead of second place by the end of it, and was about to lap third place. The middle was so boring with no one else around. Lapping people brought some life back!

Should make level 45 over the weekend when I finish up a few more events, then might go back to the R2 championship.
 
muggs said:
its probably already been explained in here but i cba searching, so can some explain this please.....btw im not here to bash f3, i have f3 and paid a lot of money for a porsche wheel that can play it. i just want to know what you guys think

Click to expand...

Missed your post earlier.
FM3 to me feels best in something close to the below settings for the PWTS but your preference may differ.

Personally Id say try:

Rotation
270 or 210 mode.
I dont like its 900 degree mode at all, maybe its because of what you show in the videos that is part reason why it doesnt feel just right.

Drift Mode
"off" or "1". I like the loose weightless feel the drift mode gives but some are quicker without it. In some cars it requires being a bit more sensative in steering.

FFB
Strength for best laptimes is usually easier to detect the wheels feedback with a bit less resistance so try about 66%.

ABS
Depends on the pedals you use. If normal try having this lower as the wheel will shake with little pressure and on such pedals its easy to lock up even with little pressure. If you have CSP and a stiff pedal then try raising the ABS to about 65% or more before the wheel shakes letting you detect the lockup limits of traction. Pedals with pointimeters are different to load cells as its travel Vs pressure used.

Firmware 669 Ultimate Edition
Ive been just recently testing the new as yet unreleased 669 firmware and well it makes the wheel even better. No date has yet been announced for the firmware to go public though.
 
Last edited:
Demon said:
T10 have said how they get the base PI rating, taking AI drivers around an imaginary course that has lots of different bits of each track (or something like that).. so comparing on 1 track is going to show differences.. put a Veyron on Le Sarthe, and surely with the humungous straights, it'd be much quicker..

I think people are expecting too much from a single PI number, without even turning a wheel in FM3, just going by how it's calculated you can immediately list a whole set of situations under which it doesn't show the real performance on any one given track/situation..

To be perfect, you would have to assess each car with every permutation of 'component' using some AI and real drivers on every course over a wide range of race distances, and Have a PI rating for every car/track/race/upgrade combination..

Or does anyone have a better idea?
Click to expand...

This is pretty much spot on. A prime example I can give is the Subaru Imprezza tuned to B class. The going trend with people who use manual clutch is not to bother with a sequential gearbox or uprated clutch as your shift times are determined by you and not the game as would be with auto clutch. This means you can use the PI elsewhere to improve the car even further but you can not change the gear ratio's. This is why the subaru is execellent because it has great gear ratio's standard for a B class setup. Even though at the same PI my mitsubishi evo 6 can handle just as well if not better it has horrible standard gear ratios which makes it suffer.
 
Last edited:
adam cool dude said:
The going trend with people who use manual clutch is not to bother with a sequential gearbox or uprated clutch as your shift times are determined by you and not the game.
Click to expand...

nah mate thats not true, i use manual and the first thing i do to a car is upgrade the gearbox to make the shifts quicker because they are well to slow. if i change fast on a non sequential car without uprated gearbox/clutch the gears crunch quite a bit if i remmeber correctly.
 
cheers mr latte you are a star, and i cant believe you got the 669 firmware mate, im running g25 pedals and i do use 900 degrees, but its not a true 900 anyway, its more 720. i will try those settings, but tbh i havent played it for a while, and when i found out about the auto steering i wasnt exactly rushing to get back on. the funny thing is i think the physics are pretty realistic, especially when oversteer comes in, and the forcefeedback is brill. actually i now realise it was a mistake to let the auto steering bug hhave a i'll effect on me, ill get back into it tonight.
 
Sp00n said:
all of the above :(

I dont like the idea of buying tunes, I would lvoe to make my own but I just cant compete no matter how long/how much money I spend!
Click to expand...

:) I don't buy tunes. I just experiment myself - I'd say 1 in 5 of my experiments results in a competitive car.

Class A - few choice upgrades to a Diablo SV makes it stonkingly quick around any track, handling is ~ 7.1!

Also loving my A-class supercharged MX5 (new shape one), very quick and not as feisty as you'd imagine for such an epic power --> weight ratio.

Class B - Lotus Evora. Batter any S5 round a tight track. Failing that, try and E92 M3...great balance around the bends and good speed overall.

Class C - Any Honda (generally lightweight) :p
 
muggs said:
cheers mr latte you are a star, and i cant believe you got the 669 firmware mate, im running g25 pedals and i do use 900 degrees, but its not a true 900 anyway, its more 720. i will try those settings, but tbh i havent played it for a while, and when i found out about the auto steering i wasnt exactly rushing to get back on. the funny thing is i think the physics are pretty realistic, especially when oversteer comes in, and the forcefeedback is brill. actually i now realise it was a mistake to let the auto steering bug hhave a i'll effect on me, ill get back into it tonight.
Click to expand...

No probs mate.
I dont play online much, prefer to decide who I get to play with mainly. Also for me Ive found the career to be a bit hit and miss. Some races my wee girl could win. I need to find the harder races and keep at it, any hints? Most of my time playing is just enjoying the wheel/system and setting up multiplayer with all AI cars as opponents.
 
@ Mr Latte,

The simple answer is, ignore 'leaderboards', that's another issue (I'll touch on that in a minute)..

If you want close racing, just pick the 'stock' cars and be sensible..

i.e. Hot hatches, take the Focus ST, Leon Cupra, Astra VXR, Golf GTi, you know in real life these are close enough, and unsurprisingly on track they are also close enough.. well as close as real life, there's a 2 second or so spread on the TG Track for example..

If you stick to what are normally grouped together in real life, by and large you will find that taking normal untuned versions in FM3 gives you close racing on most tracks..

However, if you mix car 'types' (e.g. EVO Vs Gallardo), then clearly you will expect that the 'track' will suit one car or the other, so although they are matched on the TG Test Track, since there is no TG Test track in FM3, you are going to have to get your mates to try various tracks etc to see which tracks make them more 'even'..

But then, cars are never identical, one will always be a little quicker then the other, even a 0.3 second difference per lap will be 3 seconds over 10 laps which is a fair distance!

All the above isn't restricted to FM, it applies to all racing games..

Now, leaderboards.. this is the domain of so many variables, they are but a token effort, in this case using PI to define the leaderboard validation criteria, but it doesn't matter what they used, with the amount of upgrades available and natural differences in cars, you can only ever expect that some cars will favour certain tracks/classes, plus people are lazy, and as we've proved in some cases, you can build alternative top 10 cars for some classes, if people pick up on that car, suddenly you've got 2 top 10 cars.. hell we even managed a #13 in a Supra as well after some tuning..

I'm not even remotely trying to defend FM3 either, I can't see how any driving game is going to succeed at the impossible either.. matching cars so evenly that there become no 'favourites'..
 
You must log in or register to reply here.
Back
Top Bottom