route_66
A-Spec points are calculated by (I think this is the consensus from Famine's thread)
1. PWR (not just power)
2. Weight (which explains the Dodge Ram - same PWR but higher weight = more points)
3. Tires (worse = better)
4. Downforce (lower = better)
With all due respect, I read Famine's thread a couple days ago and didn't find much in it. AFAICT, there's a bunch of conjecture based on limited, anecdotal data that's not properly isolated. There's just not enough data to reasonably speculate on a trend. I found some suspicious experiments ("proving" that suspension settings didn't change A-spec points by only twiddling the ASM and TCS nobs??? Ehhh...bad way to go about it).
Don't get me wrong, they were valuable experiments, but they're just not enough to draw conclusions from.
Perhaps he has something there, but I haven't seen it yet.
Tires are obvious. So is downforce. So is adding power mods, getting an oil change and adding ballast. So is basing it on the entire field. Power and weight most definitely have something to do with it, but HOW is still a mystery. Is it formulaic enough to reverse engineer the formula? Not so sure...
I have plenty of anecdotes from data I've gathered than seem to counter his conclusions.
EDIT - Example from something I just tried: M3 1570kg @344hp vs. Audi S4 1660kg @346hp against the same lineup. You'd think that the S4 would get higher ASpec points yes? Well... actually, they have the EXACT same ASpec points provided you do one other thing..
ADD S3 tires to the M3!!!
I've at least a half dozen focused experiments I'd like to try if I get around to them...
A couple that come to mind.
1) Try out different configs in a couple of the "one car" races and see what happens. (e.g. S7 and A3). Is there much variance in the percentage change of various power mods (these DO change and will severly limit the experiment) Are the point differences consistent across different car levels? Can we conclude anything or not?
2) Try out all the stock cars of a STATIC lineup (e.g. the Vette or M3 race). Is there an observable trend in the data? Now play with configs much like #1. Is there a consistent trend?
Anyone take linear algebra and still remember it???
3) Maybe as part of #1. Take a car. Get points. Get WR1. Get points. Increase ballast weight so it has the same original weight. Get points. Do points match the original points?
4) Take 3 cars against the SAME lineup that give different A-Spec points (spread out betwen about 50 and 175). Like #1, try some different (yet the same amongst all 3 cars) mods. Note A-Spec point changes along with PWR variances. Are there any trends to be observed?
NOTE: Points SATURATE at 10p, so crossing that threshold will simply confuse the data. Don't do it!
One thing I'm hoping to get out of these... Does each car have a PRE-EXISTING database entry on which A-Spec points are based? In other words, does the game calculate power/weight on the fly? Or does it start with an initial value and augment that value based on the mods? If so, that could explain a lot of the inconsistencies.
In particular... some observations that are counter-intuitive.
- Take 3 cars against say... 3-5 lineups with a range of A-Spec points. (Good place to try this would be the high end races). Write them all down. Say Car #1 is 50p against lineup #1 while car #2 is 150p against lineup #1. Car #1 is 20p against lineup #2. Chances are Car #2 is 120p against lineup #2. In other words, it's a simple addition/subtraction operation. Doesn't make a lot of sense.
- Tires also seem to be +/- operations (although the amount differs based on the car).
I have a few others written down in my notes, but those come to mind.
And as for that RAM. IMO, it's a bug. Maybe it's data entries are screwed up? I don't have proof yet but there's plenty to indicate that it is. I think it's best to consider it an outlier and not worry about it... kind of like the difference between relativity and Newtonian physics
e.g. in the Sunday Cup, I gathered the points between it, a Ford Lightning and a Volvo Estate noting the PWR,weight and HP of each. Try it for yourself. It's ridiculous.
RAM experiment:
1) Take RAM, reduce weight, add power. Take another truck that's heavier and has less power (once the mods on the RAM are don). I'm willing to bet it that it will *still* get FEWER A-Spec points! If so, this would also tend to indicate that the values are pre-calculated and the RAM value is screwed up!
EDIT: Need to learn to type slower... or preview more carefully