Joe Harlan
New member
Actually that's about all it will be if the rule is written correctly
January FasTrack is up!
- Thread starter Bill Miller
- Start date
Actually that's about all it will be if the rule is written correctly
G
Guest
Guest
The sad part is the fact that the dominating E-36's has spent a rather significant amount of money on Motec's. The cars that aren't Motec equipped are usually slower than the front runners just enough that they are usually in a battle with the "non-E36" cars like the best prepped RX7's, 240Z's and GSR's.
The open ECU rule is the main culprit in this, and the problems associated with enforcing it.
The open ECU rule is inherently unfair to anyone who drives a well prepped carburated car that is prepped to the letter of the rules. There is nothing one can do when you've spent all that time and money on a carburated engine that is built to the maximum that the rules allow to gain HP the way the newer cars can just plug in a high dollar engine management like the Motec units. Many cars in IT have such antiquated FI systems that there is almost nothing that can be done to them and still be within the rules.
Maybe there ought to be claim rule on ECU's instead of trying to make Tech police them. That would slow down the dominating cars enough, and prevent people from spending piles of money just to run closer to the front and make IT more affordable.
[This message has been edited by 2Many Z's (edited November 29, 2004).]
The open ECU rule is the main culprit in this, and the problems associated with enforcing it.
The open ECU rule is inherently unfair to anyone who drives a well prepped carburated car that is prepped to the letter of the rules. There is nothing one can do when you've spent all that time and money on a carburated engine that is built to the maximum that the rules allow to gain HP the way the newer cars can just plug in a high dollar engine management like the Motec units. Many cars in IT have such antiquated FI systems that there is almost nothing that can be done to them and still be within the rules.
Maybe there ought to be claim rule on ECU's instead of trying to make Tech police them. That would slow down the dominating cars enough, and prevent people from spending piles of money just to run closer to the front and make IT more affordable.
[This message has been edited by 2Many Z's (edited November 29, 2004).]
Bill Miller
New member
George,
Fair enough, we agree.
Andy,
I take it from your comment that it is not the intent of the ITAC to run all the cars in the ITCS through the process. You guys have developed somthing that most people feel is a very good thing, yet it doesn't apply to everyone. THAT is a bad thing.
------------------
MARRS #25 ITB Rabbit GTI (sold) | MARRS #25 HProd Rabbit
SCCA 279608
Fair enough, we agree.
Andy,
I take it from your comment that it is not the intent of the ITAC to run all the cars in the ITCS through the process. You guys have developed somthing that most people feel is a very good thing, yet it doesn't apply to everyone. THAT is a bad thing.
------------------
MARRS #25 ITB Rabbit GTI (sold) | MARRS #25 HProd Rabbit
SCCA 279608
Andy Bettencourt
Super Moderator
I disagree. If we ran every car through the process, you open Pandora's box. If a car is 35 lbs. off its target, do you change it? 40? 45? 48? 50? 100? What are the boundaries that you run by that keep the vision of PCA's but enough distance from Prod-style comp-adjustments?
Are you suggesting a one-time 'clensing'? If you are, I can agree it would be nice, but only after we evaluate our target performance envelopes and make sure we have the arrow pointed in the right direction.
AB
------------------
Andy Bettencourt
New England Region, R188967
ITA project SM
www.flatout-motorsports.com
Bill Miller
New member
Andy, I think +/- 50# is probably decent enough precision.
And while I don't know if I'd call it 'clensing', that's pretty much what I had in mind. Everything gets run through to see how close it is. If you're +/- 50# of what comes out, nothing changes. If you're outside that, you get the weight set to what the process spits out.
I agree that getting a handle on the performance envelope is a good thing, but I'd like to see a time limit put on it. Say, by the end of the '06 season, all cars in the ITCS will have had their spec weights evaluated by the process, and adjusted as needed. If it can happen sooner than that, all the better.
------------------
MARRS #25 ITB Rabbit GTI (sold) | MARRS #25 HProd Rabbit
SCCA 279608
Not attacking you Bill, even though it may seem so.
Why do we (as a group) need the weight so close to ideal (+/-50#)? Our horsepower estimate figures would need to be within 2, 3, 4 HP? We can't get the ultimate IT Prep HP estimate that close...
Reign in the apparant overdogs, give us a choice of 2-4 "cars to have" in each class and we should all be happy
. Anyone who chooses one of the cars other than those 2, 3 or 4 does so because they love the car more than their desire to be competitive. I find it hard to see it any other way...perhaps my view is obstructed with my head in such a dark place...does smell like roses though
.
Why do we (as a group) need the weight so close to ideal (+/-50#)? Our horsepower estimate figures would need to be within 2, 3, 4 HP? We can't get the ultimate IT Prep HP estimate that close...
Reign in the apparant overdogs, give us a choice of 2-4 "cars to have" in each class and we should all be happy
robits325is
New member
Originally posted by Quickshoe:
Reign in the apparant overdogs, give us a choice of 2-4 "cars to have" in each class and we should all be happy. Anyone who chooses one of the cars other than those 2, 3 or 4 does so because they love the car more than their desire to be competitive. I find it hard to see it any other way...perhaps my view is obstructed with my head in such a dark place...does smell like roses though.
Who is going to build a $50,000-$60,000 E-36 for ITS with the hope of maybe running up front? Looks like its time someone builds an E-46. Double Vanos 2.5L with no restrictor....or will that be the next witch hunt?
Ron Earp
Administrator
Well, if the "formula" or process to spec cars won't or can't be revealed to the general public we can approach it a different way.
The situation as it stands is that we have a simple transfer function that is used to spec cars for the various IT groups. We should be able to calculate that function because we know the exactly what the inputs to the function are, and obviously we know what the outputs are, they are in the GCR table.
Example:
Input(a..z) --> f(a..z) --> Output (A..Z)
Where a..z are the stock parameters of the car, i.e., weight, brake swept area, hp, displacement, etc. And, A..Z are the IT parameters for the car once the function, f(a..z), is applied.
Therefore, what we need are cars that are "properly" speced with the inputs, a..z, for each car, and the outputs, A..Z. These will serve as the correct data for development of our model. Then, we can apply a number of models and analyze the dataset until we can accurately predict A..Z for our properly spced cars. Now we have our function (obviously not simple since we have a fairly complex mutlivariate situation, but I bet we can come up with something that is quite good and, even better, is completely objective when compared to other methods).
Properly armed, we could then reclass all the cars and have a much closer, in my opinion, field since the classifcation will be scientific, documented, fit current class models, and be easy to adjust. I wouldn't mind doing it for fun, what cars should be used as initial models?
------------------
Ron
http://www.gt40s.com
Lotus Turbo Esprit
Ford Lightning
RF GT40 Replica
Jensen-Healey: IT prep progressing!
[This message has been edited by rlearp (edited November 29, 2004).]
The situation as it stands is that we have a simple transfer function that is used to spec cars for the various IT groups. We should be able to calculate that function because we know the exactly what the inputs to the function are, and obviously we know what the outputs are, they are in the GCR table.
Example:
Input(a..z) --> f(a..z) --> Output (A..Z)
Where a..z are the stock parameters of the car, i.e., weight, brake swept area, hp, displacement, etc. And, A..Z are the IT parameters for the car once the function, f(a..z), is applied.
Therefore, what we need are cars that are "properly" speced with the inputs, a..z, for each car, and the outputs, A..Z. These will serve as the correct data for development of our model. Then, we can apply a number of models and analyze the dataset until we can accurately predict A..Z for our properly spced cars. Now we have our function (obviously not simple since we have a fairly complex mutlivariate situation, but I bet we can come up with something that is quite good and, even better, is completely objective when compared to other methods).
Properly armed, we could then reclass all the cars and have a much closer, in my opinion, field since the classifcation will be scientific, documented, fit current class models, and be easy to adjust. I wouldn't mind doing it for fun, what cars should be used as initial models?
------------------
Ron
http://www.gt40s.com
Lotus Turbo Esprit
Ford Lightning
RF GT40 Replica
Jensen-Healey: IT prep progressing!
[This message has been edited by rlearp (edited November 29, 2004).]
Bill Miller
New member
Sorry Rob, but $50k-$60k cars hardly fit the stated philosophy, purpose, and intent of IT. Cars like that have no business in IT.
Daryl,
I don't take it as an attack. Let me ask you the same thing, why shouldn't we try and get things that close? We've got a lot of smart people here, as well as on the ITAC and CRB. If they can't hit a 100# window w/ the weight, something very wrong. And even if it is a little off, adjust it w/ PCA's.
And I just don't subscribe to the 'couple of cars "to have"' philosophy. People shouldn't have to choose between a car they want to race, and a car they can be competitive with. That's one of the reasons we end up w/ things like IT7. They were actually one of the 'cars to have', and got shuffled back in the grid by newer cars (and have gotten pushed even farther back w/ the new ITS transfers).
------------------
MARRS #25 ITB Rabbit GTI (sold) | MARRS #25 HProd Rabbit
SCCA 279608
lateapex911
Super Moderator
You are absolutely right!!!!!!!!
For $50K, you should be guaranteed a win! As a matter of fact, for that kind of K, you should have driver do the dirty work so you don't have to break a sweat......
????
Whitchhunt?? ooookkkkkaaaaaaaaaaaaay...
------------------
Jake Gulick
CarriageHouse Motorsports
ITA 57 RX-7
New England Region
[email protected]
If - again BIG IF - the system is going to look at on-track performance, I agree with what I think is the current ITAC thinking: We can't hope to get closer than 100# or so becaue the variance due to driver skill and other considerations, even at the top of the results chart, is too great. Heck, the same difference resulting from 100# might be accounted for in "good day" vs. "head cold" alone, for the same driver.
It's kind of like those opinion polls that we heard so much about this fall. A margin of error (or conficence interval, more accurately) of +/- 3% is achievable with a random sample of a thousand or so people. Pollsters get a rapidly decreasing return in improvement at about that point, where even dramatically increasing the sample size doesn't make a heck of a lot of difference in the confidence interval.
When the results are in, 3% is considered close enough so if the candidates come in at 47% to 50%, it's technically a tie. If some objective Reality (capital R) is that they are one point apart, we will NEVER know given the limitations of our methodology...
...so similarly, if two reasonably skilled shoes pedal the same ITB car around VIR and turn mean lap times that are 2.0 seconds different (about 1.5%) and, let's say, that much time can be attributed to the fact that one of them doesn't know the track very well yet, then how can we hope to adjust weights of different cars to get them closer than that?
Before you go saying that Kirk has gone all soft on this issue, remember that my much-maligned proposal suggested using ONLY a formulaic determination for weight, based on physical attributes of the car, and letting the chips fall where they may.
That plan presumed essentially what I suggest here - that we could get them "close enough" without resorting to considerations of lap times, finishing positions, etc. that are really not a direct indicator of the input variables under consideration - weight, and now restrictor size.
If you are suggesting (Bill, et al.) that we should try for accuracy greater than 100#, how will we know when we are there? There just isn't enough data to be sure.
Finally, it sure would be cool if people would stop mixing metaphors here. The issues of ECUs, spec weights, and restrictors are certainly related in the real world but they need to be argued as logically distinct questions, or you are shooting your case down before it's even made.
K
[This message has been edited by Knestis (edited November 29, 2004).]
Andy Bettencourt
Super Moderator
The same 'type' of people who would build a 10/10ths ANYTHING Rob. Why is it the E36 is the only car that can run up from when built and driven to the max? What right do you have to WIN a Regional Championship with a car that ISN'T?
It's SOOOO far from a witch hunt its silly to even use the term. Facts are facts.
AB
------------------
Andy Bettencourt
New England Region, R188967
ITA project SM
www.flatout-motorsports.com
Andy Bettencourt
Super Moderator
For an ENTIRE class...not the difference between the top 2...
AB
------------------
Andy Bettencourt
New England Region, R188967
ITA project SM
www.flatout-motorsports.com
seckerich
New member
One reality that keeps getting missed with these continuous rants about the "perceived advantage is only at a certain track" is that I can't think of one track that ends a race in a corner. I can run fast laps, sit on pole, and have a BMW blow by me when they drop the green. Then I spend the whole race passing the same car as they hold me up in the areas I am faster and pull away bigtime on every straight. I have run VIR with York, Chet, and some of the other fast BMW's from the areas and was only 1 to 1.5 seconds slower in a mazda but got drilled at the start. Same for slower tracks like Kershaw. If you want to compare numbers that matter you need HP and TORQUE if you want real comparisons. Races end and start on straights and this is where the cars kill the others in the class while giving up very little in the corners. This is reality.
Steve Eckerich
ITS 18 Speedsource RX7
Southeast
Steve Eckerich
ITS 18 Speedsource RX7
Southeast
Did not. I know it was a little rambling but re-read that again. The 3% bit was an analogy.
We sure spend a lot of time here picking ammunition for our arguments by reading selectively out of other's posts.
Is a 1% variance in lap times "close enough?" I don't know. What portion of that amount of time would the addition of 100# account for? One of the sophisticated lap time simulators out there might actually give us some data to play with.
K