MSA TORA GT General Discussion

I would love to see this 1.59 in the GT500.

Fastest I can manage is a 2.01.675??????

I ran a 59.6 in the build on the SummerGT storefront and it isn't close to 100%. Go grab that

The GT500 tune aint there?

Can you please gift it over?

Where can I get a number plate? I would really appreciate a gifted one so I can then make the design for my team mates and gift it to them.

Yes it is. Page two second car down in the storefront.

Another thing, guys you cannot scream I'm 5 seconds off the pace when you don't run at the fastest pace. If that's the case we are balancing drivers and not cars. Good luck with that

Thanks worm i have found it.

and i am not moaning about ther drivers being quick, i just dont think my tune correct for the car and i am getting annoyed I personally can't get it work to the same speed as other people that i am normally as quick as.

All we are not balancing drivers just balancing cars

The best way to balance would have been to get five or six drivers to try out a car, then average the laptime and do that for each car. Work out uncertainties in time for each car ([Maximum Reading - Minimum Reading]/Number of Readings), and use those results (in the form of Average Laptime +- Time Undertainty). Then its just a case of each car falling within each others uncertainty.

Possibly too late for this now though.

How would we determine this "time uncertainty". You balance the cars for the fastest drivers so they don't run away more than normal.

I think the cars should not be equally quick at every track or that's exactly what will have to happen with the balancing. The cars would need PI's for each track!!

No one should expect the GT500 to win at Catalunya surely, not up against a Porsche or similar. And on the faster tracks or tracks with longer straights the pure power combined with torque from the GT500 should see that as a front running car.

For any changes my opinion, for what it's worth, would need data from multiple tracks, say four, made up of a good mixture of fast, twisty, long etc.

I just built a GT500 and put on a quick 10 sec tune and ran a 1:51 at Cat no probs, so it's not that far off to be honest. With a bit of effort it could surprise people I feel!!

Ash, could you test those cars on 3 different tracks? If you have the time that is.

Speed
Handling
Medium

Not going to say which tracks come under which as I'm sure you already know.

If changes are to be made, I want them tested on different tracks. No point giving a car more PI and only testing it on Catalunya...could end up getting to Sebring and being completely demolished by a rocket on wheels....also different builds would be good too. I'm sure your builds are already decent but you never know.

i have done this slim and will be putting up the times i have got;

Cat - handling
Road atlanta - Power
Motegi - medium

Already on it!


Every car will shin in its own way at each track no matter who the driver is!

VVV worm wrote:How would we determine this "time uncertainty". You balance the cars for the fastest drivers so they don't run away more than normal.

The whole idea of this is to come up with a value of speed of a car around a circuit that was representative of a sample of drivers, rather than just the personal times of one driver driving all the cars. Read this carefully: With this method, each car would have a) Its own average lap time around a track and b) A threshold of time for which the average lap time may vary depending on who was driving. It can be done easily with very simple maths. Each drivers lap time would be converted into either seconds or mph for the number crunching to be straight forward.

-Take a sample of drivers, say 8.
-Give them 20 minutes in a car each with the same tune.
-Take in all of their times, label them T1 through T8.
-Find the average of the times, Average T (T1+T2+T3+T4+T5+T6+T7+T8, all divided by 8 )
-Find the error in the average (Slowest T-Fastest T, all divided by 8 )
-You then get an Average T +- Error in Average T
-The error in average T basically gives you a threshold for which other cars must lie within for their own Average T.

With enough time and plenty of careful procedure, you could have a car list that was in theory, bullet proof. It would take a lot of trial and error to get the cars within the thresholds, but atleast you have something to work towards that was representive of more than one driver, rather than one drivers individual times.

Again, this sort of method should be kept for when Forza 4 comes out. I'd be more than happy to do it if I had a group of test drivers.

main issue = people.

i.e. sandbagging, not being able to build cars as well as others, not being able to tune as well as others.

To many factors the maths can not resolve alone I'm afraid. The only way around these factors is to enforce and make people use specific builds and tunes, which imho takes away the majority of what makes FM3 and TORA races attractive. Otherwise, just run stock series!!

I would probably suggest not messing with it now. People who have made their choices would effectively have to go back to the drawing board. The cars were deemed "close/balanced" and a line was drawn under it. People then chose based on those parameters. If the car you're in is slow - you made a poor/untested choice OR have not yet found the magic setup or driving style that makes that car work best. Pick something else that DOES work for you. Don't ask for the car to be changed.

Agreed with Magic. It won't work nicky. Driver A doesn't have a snowballs chance in hell of tuning or running the same laps as Driver B and you balance the cars for faster drivers not slower ones. Everyone moans "I'm slow because of my car" when it ain't the damn car.

Weren't you a part of the testing group and contributed nothing? I'll let you answer that question.

@Chris +1

Whenever one of you gets a chance shoot me over the number plate, I have the design finished but am missing that final piece.

VVV worm wrote:Weren't you a part of the testing group and contributed nothing? I'll let you answer that question.

Worm...just incase you missed this bit.

nickyf1 wrote:Again, this sort of method should be kept for when Forza 4 comes out.

To start there was as good as no structure to the testing that I was aware of, and within a week everyone was just told to download your tunes and post up times. Essentially if you were busy in that first week you had no input into the direction of the series...which was hardly a great motivation. Personally I never bothered posting up any times after that because I couldn't be arsed dealing with you and your poor attitude. Was it the same for anyone else? I'll let you answer that question.

VVV worm wrote:Agreed with Magic. It won't work nicky. Driver A doesn't have a snowballs chance in hell of tuning or running the same laps as Driver B and you balance the cars for faster drivers not slower ones. Everyone moans "I'm slow because of my car" when it ain't the damn car.

Weren't you a part of the testing group and contributed nothing? I'll let you answer that question.

I was part of the testing group, yes. I didn't actually say i wanted to join, I just happened to post in the original GT Testing Group thread, so, understandably, I was mistakenly added to it, as I did not clearly state that I didn't want to take part. I also had lots of project work and studying to be done at the time, which meant I could not contribute as much as I would have liked to.

Now, read very carefully the rest of my post, because clearly nobody has taken the time to understand what I am proposing.

I came up with my idea because of exactly the reason you say it wont work. The whole idea of it is to include how much the time of a car around a track would vary depending on the driver (the uncertainty) onto the average fastest lap time of a testing group driving identically prepared cars. I said that in the post before I explained how it works.

I do acknowledge that there are many other factors that go into this, but if followed by the book, and done in a ''fair test'' kind of way, where only one variable is changed (in this case the driver) and the rest is kept the same (car, tune, track, build).

This would be the only way to find out, with an attached number, how quick a car actually was around a track! Think about it.

Make of the method as you will.

nickyf1 wrote:where only one variable is changed (in this case the driver) and the rest is kept the same (car, tune, track, build)

Howdee. That's the issue Nicky. Right there. The Summer GT and most Champs in TORA have tuning and building skills as part of the challenge. Removing these removes, in my opinion, a great deal of what attracts and holds people to racing in TORA on FM3.

Surely your method just ends up balancing the drivers and not the cars, which remain the same in build and tune for each track.

If I've followed this correctly, what is the point in having different cars please?

Cheers for extra clarification. I am not intentionally jumping all over your ideas in a negative way to be an arse, but merely to state where I see and or feel there are issues. In the hope something good could emerge for future consideration.

VVV Magictap wrote:

nickyf1 wrote:where only one variable is changed (in this case the driver) and the rest is kept the same (car, tune, track, build)

Howdee. That's the issue Nicky. Right there. The Summer GT and most Champs in TORA have tuning and building skills as part of the challenge. Removing these removes, in my opinion, a great deal of what attracts and holds people to racing in TORA on FM3.

Surely your method just ends up balancing the drivers and not the cars, which remain the same in build and tune for each track.

If I've followed this correctly, what is the point in having different cars please?

Cheers for extra clarification. I am not intentionally jumping all over your ideas in a negative way to be an arse, but merely to state where I see and or feel there are issues. In the hope something good could emerge for future consideration.

Thats absolutely fine.

This would just be for testing only, not during the season.

During testing, you get your testing group to download one tune for one car. So thats 8 drivers all driving the same car with the same tune on the same track. If you give them 20 minutes on a track, you can collect each drivers time.

This time will reflect the ability of the car and driver. The time of each driver will vary depending on skill. Finding the average time of the eight drivers will tell you how fast the car is on average of those eight drivers.

As each drivers time will be different due to driver skill, you can work out how much the slower or faster the average time is depending on how skilled the driver is. After obtaining this value, you end up with ''Average Time, Plus or Minus the Average Variation in Time''.

So you end up with, for that particular car, a number, expressed as ''What it is capable of on average and how much this time will vary depending on the driver''.

From this, you can repeat the process for other cars, and find a number for these other cars. After running through the entire car list once, you will be able to observe the threshold of time which each car is capable of on that one circuit, and therefore you can see how close the cars are, with driver skill being taken into account.

After the first run through, a graph could be made to show all of the time thresholds overlapped on top of each other. If any cars threshold is significantly far away from the other cars thresholds, you can make PI adjustments to suit.

Quite complicated, and rather hard to explain... I tried my hardest there

The purpose of the method is to establish the PI's for each car before the season starts. Once the car list is released with these PI's, people can tune and build them up to their hearts content.

nicky I understand what you are saying but it doesn't work like that. You can't put 8 guys in a tune and say "go" then average it. If say, Magic is in the group and runs 2 seconds quicker in every car but finds one that he can drive another second faster than the rest but no one else can his time gets lost. Well come car selection time you now have Magic knowing what car he is choosing because he is now 3 seconds quicker than the field. Guess what car all the faster members choose?

The only way that works is if you take your fastest 8 members, give them a list, and tell them to have at tuning and running cars. That is the only way that works.

its troff wrote:

VVV worm wrote:Weren't you a part of the testing group and contributed nothing? I'll let you answer that question.

Worm...just incase you missed this bit.

nickyf1 wrote:Again, this sort of method should be kept for when Forza 4 comes out.

To start there was as good as no structure to the testing that I was aware of, and within a week everyone was just told to download your tunes and post up times. Essentially if you were busy in that first week you had no input into the direction of the series...which was hardly a great motivation. Personally I never bothered posting up any times after that because I couldn't be arsed dealing with you and your poor attitude. Was it the same for anyone else? I'll let you answer that question.

Alright don't get your knickers in a twist mate.

All Worm has done is help TORA with balancing the car's, as far as I'm aware everyone get's on with Worm and you should be thankful for what he has done. I'm sure Worm and the rest of VVV can easily go and help somewhere else, if that's what you want.