A True Tuner Challenge

[Motor City Hami]

Been thinking about an idea for refining the old Tuner's Challenge. This might belong in the F.I.T.T. thread, but might not. Anyway, I didn't want to clutter their thread and didn't want this to get lost 80 pages deep.

The Tuner Challenge that Adrenaline ran was really good, but had a small flaw. We built a tune for ourselves then gave it to drivers with unknown skill, style, controller, etc. The score was lap time, so as a tuner, you had to hope that your tune was a close fit for enough of the drivers to put you on top.

I recently purchased a wheel, a G27. My tunes were all created using the DS3 and are very agressive from apex to exit. I'm finding with the wheel, I cannot unwind the wheel nearly as fast or smoothly as the DS3 stick, so many of my tunes are loose on exit. I believe in my tuning method, so I immediately know the fixes to the problem and can adjust my tune with a few changes and be fast again.

What I would like to see is a tuning competition that tests the tuning method, not the lap time by the luck of enough drivers being just like me. I haven't thought through all of the details, but wanted to get this bounced around the tuning community.

The idea:
- Same as the old competition, car is selected, tune is built.
- Match a tuner with a driver (or two, or three).
- Driver(s) run laps then provide feedback on what the car is doing. Should be a specific list of questions for them to answer or maybe multiple choice plus a bottom line description.
- Tuner recommends adjustments to the driver(s).
- Rinse and repeat. Maybe give tuners the original tune and maybe three opportunities for feedback?

Not sure how to score this? Lap time(s) from the final session? Most improved?
Would the drivers for each challenge only drive for one tuner? Rotate drives with each month's vehicle?
Lot's of questions.

Any interest?

 

[XDesperado67]

Just looking at what you've posted I'd say it has possibilities, and some potential problems.

1) Starting with the potential problems, and possible solutions...
It could be divisive to the community as in effect your setting up a competition strictly to determine who is the best tuner. We run some of that risk in any competition but the variety of events and the range of choices for cars helps mitigate that to some extent. If you set it up in advance that car types and test track/s will be changing on a regular basis you probably reduce the risk, and maybe highlight tuners specialties. i.e. tuner A is currently the best with NASCARs, tuner B is a wizz with Kei cars, tuner C can make the best MR tunes, etc.
2) It looks to be rather time intensive for both the tuners and drivers. Having a long tune/test cycle helps with time management, but the longer each cycle is the more you risk people losing interest.
3) It risks being a popularity contest, if the winner is selected on purely subjective criteria. i.e. I always use tuner A's tunes because they are the best. Making sure that there are very clear firm rules for scoring the challenge.

Just some things to think about when setting this up.

 

[killerjimbag]

Here are my two cents.
I will offer up to test all the tuners set-ups.
Make one car one track sessions.
Driver A will test Tuner A,B,C,D tunes.
Driver B will test Tuner E,F,G,H tunes
After one week or set time switch and see results.
There are many tuners on here who I look at to get my final tunes.
I pick some of tuner A some of B and some of my own to get it down to what I like.
Everyone has their own driving style and most of the tuners are pretty good at doing what they do best.

 

[Johnnypenso]

Based on what you've written, there is no empirical, objective way to evaluate the tuning process. You're basically looking to see who is the best at working and reworking a tune, in conjunction with a driver. The only really objective way to measure it is through lap times and if you go with lap time improvement, and it's measured against stock, then you're right back to the fastest lap wins.

One way you might do it, is have some tuners tune up some cars, have a limited number of skilled drivers test the cars, and then have the tuners swap cars and adjust the tunes of another tuner, within a limited set of parameters. For example, you can adjust ride height and springs only by x%, Dampers by +/- 1 etc. Or assign points, like .1 of spring is 1 point, 1 on the dampers is 10 points, give them a limited number of points, and then set them loose.

Cars are then retested, lap times recorded etc. The tuner who has the tune that was improved the least by the second tuner, or ideally, no improvement or the tune goes backwards, wins.

 

[nomis3613]

Hi Motor City Hami,
It's a great idea, often a great tune is rated badly, for want of only a small tweak to better suit the driver. Here are my suggestions:
1) A competition can only be fair by having the same drivers test all cars. To keep the workload manageable for testers, split the cars into groups and use a knockout system. I think the ideal numbers are each group having 4-5 cars and 3-5 drivers (if drivers want to test more cars, they can sign up for multiple groups). The top 2 cars from each group advance to the next round. This structure would allow a winner to be crowned out of a field of 20 cars within 3 rounds, and without too much duplification a driver having to test the same car in multiple rounds.

2) Scoring based on driver feedback. Especially since we're trying to rate the improvement a tuner has made to a car. (I don't think measuring the improvement in lap time would make for a good comp, the winner could well be just the person who has found the car with the worst default setup) Sure, opinion is less scientific (and, like xDesperado says, can turn into a popularity contest), but it allows for a greater range of entries than if everyone's just shooting for a hero number. Lap times should be recorded (I'll get to that later), but the scoring will all be based on driver opinion.

3) Dual-shock or wheel? A car tuned for dual-shock will be disadvantaged if the review normally uses a wheel, and vice-versa. Getting around this is tricky, perhaps run separate comps for DS3 and wheel? Or publish each tuner and driver's DS3/wheel preference, so that cars which lost out by having heaps of judging done using the "wrong" input device can at least be noted?

4) Rate the testers! Each tester should submit their score and best lap for a car. Only the score is used for the competition, the best laps will be used to choose which testers get to judge the final. This means each tester will make a good effort to test every car to its limit, and the final will be judged by the best drivers.

5) Limited Feedback cycles. I totally agree that 3 revisions to the tune is a good number. It's enough to give the tuner a good chance to suit the driver's style, but many more and it would turn into finding the right setup by process of elimination.

These are all just suggestions, take it or leave it.

Either way, I'm interested in joining the challenge as a tester.

Cheers,
Simon

 

[praiano63]

Hi Motor City Hami,
It's a great idea, often a great tune is rated badly, for want of only a small tweak to better suit the driver. Here are my suggestions:
1) A competition can only be fair by having the same drivers test all cars. To keep the workload manageable for testers, split the cars into groups and use a knockout system. I think the ideal numbers are each group having 4-5 cars and 3-5 drivers (if drivers want to test more cars, they can sign up for multiple groups). The top 2 cars from each group advance to the next round. This structure would allow a winner to be crowned out of a field of 20 cars within 3 rounds, and without too much duplification a driver having to test the same car in multiple rounds.

2) Scoring based on driver feedback. Especially since we're trying to rate the improvement a tuner has made to a car. (I don't think measuring the improvement in lap time would make for a good comp, the winner could well be just the person who has found the car with the worst default setup) Sure, opinion is less scientific (and, like xDesperado says, can turn into a popularity contest), but it allows for a greater range of entries than if everyone's just shooting for a hero number. Lap times should be recorded (I'll get to that later), but the scoring will all be based on driver opinion.

3) Dual-shock or wheel? A car tuned for dual-shock will be disadvantaged if the review normally uses a wheel, and vice-versa. Getting around this is tricky, perhaps run separate comps for DS3 and wheel? Or publish each tuner and driver's DS3/wheel preference, so that cars which lost out by having heaps of judging done using the "wrong" input device can at least be noted?

4) Rate the testers! Each tester should submit their score and best lap for a car. Only the score is used for the competition, the best laps will be used to choose which testers get to judge the final. This means each tester will make a good effort to test every car to its limit, and the final will be judged by the best drivers.

5) Limited Feedback cycles. I totally agree that 3 revisions to the tune is a good number. It's enough to give the tuner a good chance to suit the driver's style, but many more and it would turn into finding the right setup by process of elimination.

These are all just suggestions, take it or leave it.

Either way, I'm interested in joining the challenge as a tester.

Cheers,
Simon

I'm ok with this testing system. The only thing is that testers need to give the exact evaluation over each parameter ,not only a subjective comment .

I think that the way i do when i tested the cars for the Indy shootout is a good base to start.

Example for indy / road and speedway

Points attribution System , 10 points scale.

Max speed at starting line: 280-290 kmh= 8pts // 290-300 kmh= 9 pts // 300 + kmh = 10 pts


Low speed corner:
entry 10 pts = agressive entry with little overster oriented inside the corner.

mid 10 pts = quick driving direction changes

out 10 pts = little oversteer with full early power allowing round corner out.


medium speed corner:
entry 10 pts = agressive entry with little overster oriented inside the corner.

mid 10 pts = perfect control of driving line with throttle ,eagerness , front entring corner when release throttle.

out 10 pts = little oversteer with full early power allowing round corner out.


High speed corner:
entry 10 pts = Sharp direction ,lot of grip oriented inside the corner to apex when release throtle.

mid 10 pts = perfect control of driving line with throttle ,constant grip ,no floatty (snake).

out 10 pts = When full gas, no little oversteer that make loose speed, no understeer that don't allow full throttle, perfect balance.

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

Example for a tested car:

Tuner // car : acsr 421 // Ford GT No stripe ‘05

Test drive result // total points= 94 pts

Max speed on startng line: 300 km/h = 186,4 mph = 10 pts

Low speed corner:
Entry: 9 Middle: 9 Out: 9
Comments: good changes of direction, easy to drive where you want, normal little oversteer on accel.

Medium speed corner:
Entry: 10 Middle: 10 Out: 9
Comments: Steady, good grip,perfect entry under break, easy to be agressive with the help of throttle mid corner.

High speed corner:
Entry: 10 Middle: 9 Out: 9
Comments: Perfect line, very light direction.

Excelent car ,carry a lot of speed in any condition , very stable under brake, allow trail braking, gears well spreaded for this track. Very fast.
---------------------------------------------------------------------------------

Tuner // car : CSLACR // Audi R8 4.2 FSI R-Tronic '07

Test drive result // total points= 89 pts

Max speed on startng line: 294 km/h = 182,7 mph = 9 pts

Low speed corner:
Entry: 9 Middle: 9 Out: 9
Comments: perfectly balanced car

Medium speed corner:
Entry: 8 Middle: 9 Out: 9
Comments :Need more brake to be more incisive at entry.

High speed corner:
Entry: 9 Middle: 9 Out: 9
Comments: Lot of grip ,perfect line.

This car is perfect, but I can’t push it 100 % only for one reason. The brakes. I don’t agree with the 7/1 brake setting and ABS 2, the car brake well in term of distance,braking capacity , but they don’t help to the weight tranfer, to place the car playing with direction (brake and wheel),question of different driving style. To much front brake,to neutral for me. But it’s so fast, in my test the faster car road x oval combination.


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

The sum of those point for every car tested will help to classify the tested cars on a really good base and any bad or good point of the tune is showed . Easiest to check and correct later.

That what i think ,good day.