LE2 setup vs. AI 190cc
05-24-2006, 12:55 AM
#211
Registered User
Join Date: May 2002
Location: Saskatchewan, Canada
Posts: 1,025
Re: LE2 setup vs. AI 190cc
Originally Posted by thesoundandthefury
The only way this debate is going to be settled is for somebody to perform a double blind experiment. How this works is you have one party that you designate as the "control group," (Ai or LE, flip a coin), and you assign a person who has no idea what the variables of the experiment are to observe the results of what takes place. You then get a second person and have them do the same thing with the other companies' heads. (Once again, Ai or LE, whichever didn't get picked in the coin toss.) Neither Ai or LE are to be informed in advance that there is any testing taking place, therefore eliminating as a variable the opportunity/possibility of "doctoring" up a special set of heads.
It is double blind because neither Ai or LE know their heads are being tested, and the people observing the dyno results don't know they're dealing with Ai or LE heads and therefore have no predetermined expectations. This eliminates bias. This of course would have to be orchestrated with the precision of a CIA operation, completely under the radar screen, and by one person who was willing to buy a set of heads from both companies. For the data to remain absolutely pure, the person conducting the experiment would have to somehow find a way to supply both blind observers with the same shortblock. The only way that we could eliminate deviation in the short block assembly as a variable was if the shortblock was supplied by a third party who had no prior knowledge of the other parameters in the experiment. You can see where this really begins to become complicated, relying too heavily on the "good faith" and "trust" of all parties involved.
Taking into account the number of people who would need to be involved in the network, even if the one masterminding it all was clever and resourceful enough to pull off an experiment this elaborate and produce completely unbiased and empirical data, one question would still remain: are you going to believe it?
It is double blind because neither Ai or LE know their heads are being tested, and the people observing the dyno results don't know they're dealing with Ai or LE heads and therefore have no predetermined expectations. This eliminates bias. This of course would have to be orchestrated with the precision of a CIA operation, completely under the radar screen, and by one person who was willing to buy a set of heads from both companies. For the data to remain absolutely pure, the person conducting the experiment would have to somehow find a way to supply both blind observers with the same shortblock. The only way that we could eliminate deviation in the short block assembly as a variable was if the shortblock was supplied by a third party who had no prior knowledge of the other parameters in the experiment. You can see where this really begins to become complicated, relying too heavily on the "good faith" and "trust" of all parties involved.
Taking into account the number of people who would need to be involved in the network, even if the one masterminding it all was clever and resourceful enough to pull off an experiment this elaborate and produce completely unbiased and empirical data, one question would still remain: are you going to believe it?
This is almost as bad as a Ford vs GM event. No matter who comes out on top, there will always be a group that believes otherwise. If that wasn't the case, everyone would be driving a 03/04 Cobra
Understand?
05-24-2006, 10:05 AM
#213
Registered User
Join Date: Jan 1999
Location: Baton Rouge, la
Posts: 989
Re: LE2 setup vs. AI 190cc
Originally Posted by Javier97Z28
This is ridiculous... does the LS1 world have this problem.. can multiple head porting companies coexist there w/o bitching all the time?
but..like I said before. For 95% of the people here if your looking at LE or AI..flip a coin. You will get results your happy with from either head. It was stated before that most will not do the right supporting mods to really make a difference so chances are you will get very close results out of either head. Find which company you feel comfortable with and buy em.
05-24-2006, 10:12 AM
#214
Registered User
Join Date: Jan 1999
Location: Baton Rouge, la
Posts: 989
Re: LE2 setup vs. AI 190cc
Originally Posted by SAR2K
Anyone in this thread design F1 or Top Fuel cams?
Steve...
Steve...
....ask him he he wants some tips from someone who is obviously the "best cam guy in the world" for his next F1 cam design.
Last edited by 95Bird; 05-24-2006 at 10:15 AM.
05-24-2006, 10:51 AM
#217
Registered User
Join Date: Apr 2004
Location: Pittsburgh, PA
Posts: 341
Re: LE2 setup vs. AI 190cc
Originally Posted by A/G
Blind is to eliminate (or reduce to a aceptable minimum) the placebo effect. Machines as far as I know, do not suffer from that effect.
05-24-2006, 11:30 AM
#218
Registered User
Join Date: Aug 2005
Location: Columbus, IN
Posts: 862
Re: LE2 setup vs. AI 190cc
Originally Posted by A/G
This is NOT how a double-blind test is conducted. 'Double' involves repetiton
In a double-blind experiment, neither the individuals nor the researchers know who belongs to the control group and the experimental group. Only after all the data are recorded (and in some cases, analyzed) do the researchers learn which individuals are which. Performing an experiment in double-blind fashion is a way to lessen the influence of the prejudices and unintentional physical cues on the results (the placebo effect, observer bias, and experimenter effect). Random assignment of the subject to the experimental or control group is a critical part of double-blind research design. The key that identifies the subjects and which group they belonged to is kept by a third party and not given to the researchers until the study is over.
http://en.wikipedia.org/wiki/Double_blind_experiment
Originally Posted by A/G
though two, one from each* end of the spectrum would be ideal.
Originally Posted by A/G
You refer to not "doctoring" up or "no predetermined expectations" or "bias", but then conclude with this, "relying too heavily on the good faith and trust of all parties involved." Geez!
Originally Posted by A/G
A (or each) car would be set up with a head from one or the other. After testing, heads are swapped. After testing, the heads are swapped back to starting combo. Final testing is conducted to verify original test results. That is a double test. No need to be blind. Blind is to eliminate (or reduce to a aceptable minimum) the placebo effect.
Originally Posted by A/G
Machines as far as I know, do not suffer from that effect.
Originally Posted by hsyr
This is almost as bad as a Ford vs GM event.
or
Ai=GM and LE=Ford
Yep, pretty much. You know I was thinking: wouldn't it be fun to send an F-body over to Ford to let them do some testing on it?
Yeah, I thought so...
05-24-2006, 02:52 PM
#219
Banned
Join Date: Oct 2002
Posts: 6,518
Re: LE2 setup vs. AI 190cc
I am starting to see the entertainment value of other people bitching at each other, no wonder this has so many hits.
I seriously think we could get a unbiased dyno test setup, the real key would have to be the dyno being able to make repeatable results.
Bret
I seriously think we could get a unbiased dyno test setup, the real key would have to be the dyno being able to make repeatable results.
Bret
05-24-2006, 03:13 PM
#220
Registered User
Join Date: Feb 2002
Location: Bridgewater, MA
Posts: 5,645
Re: LE2 setup vs. AI 190cc
Originally Posted by A/G
I thot we acknowledged the dyno is not what we race? Why then, make it the test results dependant upon it? Use a dyno for testing changes, and verify it on the track. It is no different that what you have posted in the past, flow numbers do not win races. Why waiver from that logical reasoning/understanding, Bret?
05-24-2006, 03:21 PM
#221
Registered User
Join Date: Jan 2005
Location: New York
Posts: 1,114
Re: LE2 setup vs. AI 190cc
Originally Posted by A/G
I thot we acknowledged the dyno is not what we race? Why then, make it the test results dependant upon it? It is no different that what you have posted in the past, flow numbers do not win races. Why waiver from that logical reasoning/understanding, Bret?
Seriously by the sounds of it, there may be people that honestly think that a dyno comparison woudn't make sense. I mean if the packages are spec'd to do the same thing it should be a great comparison, and pretty easy to see which package would be the victor of the two with the results.
05-24-2006, 03:44 PM
#222
Registered User
Join Date: Jul 2005
Posts: 123
Re: LE2 setup vs. AI 190cc
Originally Posted by RealQuick
Because its easier to dyno test a car to get the hp/tq curves and then swap heads and do it again....
Testing at the track has too many variables.
Not in an established race car it does? Are you denying the repeatability of established cars? If you caught my previous statement regarding ending test with original variable, this would hopefully address any weather conditions variables in between head change time lapses.
I like the idea of just sending a stage 2 head to Golen and for him to test them on an engine on his engine dyno.... Also verify flow numbers of each set of heads there. That's the easiest way.
As I noted and alluded in response to Bret, dyno is great for one phase of testing protocol, but needs to be verified on track. Without track verification, any gain on the dyno is pointless. Why not cut to the chase and perform what is considered the bottom line, track performance. Ever hear the question; do we race dynos, or FTM, flow benches? Again, you want ease....? Hell, you want easy and cheap? Let's stick with gossip.
Now, if you are choosing dyno because the preferred (read more accurate) choice is not available or feasible, that is different.
Last edited by A/G; 05-24-2006 at 03:53 PM.
05-24-2006, 04:44 PM
#224
Registered User
Join Date: Aug 2005
Location: Columbus, IN
Posts: 862
Re: LE2 setup vs. AI 190cc
Originally Posted by A/G
Double implies swapping of product or participants being tested.
Blind Observer+Blind Observee= double-blind
It has nothing to do with the number of heads involved in the experiment, nothing to do with the number of people observing the results, nothing to do with how many different cars they're tried on, and nothing to do with how many times you repeat the test. Not that these things don't have their place, but about all they're good for is forming statistics.
Originally Posted by A/G
In addition, your quote of that link does not address the 'double' in double-blind.
Originally Posted by A/G
It is also not about the testers, it is about the testee.
Originally Posted by A/G
If you are concerned about integrity of individuals (or team) doing the testing, whom are you gonna trust to make sure they are not cheating?
05-25-2006, 08:52 AM
#225
Registered User
Join Date: Feb 2002
Location: Bridgewater, MA
Posts: 5,645
Re: LE2 setup vs. AI 190cc
Originally Posted by A/G
What is more important ease, or validity?
Testing at the track has too many variables.
Not in an established race car it does? Are you denying the repeatability of established cars? If you caught my previous statement regarding ending test with original variable, this would hopefully address any weather conditions variables in between head change time lapses.
Not in an established race car it does? Are you denying the repeatability of established cars? If you caught my previous statement regarding ending test with original variable, this would hopefully address any weather conditions variables in between head change time lapses.
I like the idea of just sending a stage 2 head to Golen and for him to test them on an engine on his engine dyno.... Also verify flow numbers of each set of heads there. That's the easiest way.
As I noted and alluded in response to Bret, dyno is great for one phase of testing protocol, but needs to be verified on track. Without track verification, any gain on the dyno is pointless.
As I noted and alluded in response to Bret, dyno is great for one phase of testing protocol, but needs to be verified on track. Without track verification, any gain on the dyno is pointless.
Now, if you are choosing dyno because the preferred (read more accurate) choice is not available or feasible, that is different.