Quick IAC question
#21
Race Director
Member Since: Feb 2002
Location: Compound in the Grove, Ga.
Posts: 11,329
Received 910 Likes
on
583 Posts
2020 C3 of the Year Finalist - Unmodified
2019 C4 of Year Finalist (performance mods)
2018 C4 of Year Finalist
2015 C4 of the Year Finalist
St. Jude Donor '16
2020 C3 of the Year Finalist - Unmodified
#22
Race Director
Thread Starter
Member Since: Oct 2001
Location: Canton GA
Posts: 12,740
Received 81 Likes
on
32 Posts
Cruise-In III Veteran
Cruise-In IV Veteran
Cruise-In VI Veteran
St. Jude Donor '03-'05-'06-'10-'11-'12
#24
(1) Grab a copy of the ANHT ($8D) listing, which is widely available on the 'net.
(2) Work through the calibration section byte at a time, comparing the LT5 listing with the ANHT. They are somewhat similar. To fully determine what a particular calibration byte does, you'll have to trace the calls to it in the program. If they are called in a similar-looking segment of code, you can be pretty sure it is the same variable.
(3) Fill in the major items, by looking at the addresses from "public" definitions out there, TunerCat, Tuner Pro.
(4) Once you feel reasonably confident you've identified 90% of the variables, you could always send it off to someone who knows and have them check it for you.
I hope you are patient. That is basically how I did it back in 2002-3. It can take literally hundreds of hours doing it alone. Once you've finished the calibration section, then you can move on the program code, and understand not only "what" is there, but "how" it's used. The nice thing is, GM used a library of very similar code routines in many of the P4 ECMs, so there are definitely similarities with other well-hacked calibrations of that era.
You'll find the LT5 cal has nearly 600 constants, and over 70 tables.
Todd
#25
Le Mans Master
If you're serious about this, here are few tips.
(1) Grab a copy of the ANHT ($8D) listing, which is widely available on the 'net.
(2) Work through the calibration section byte at a time, comparing the LT5 listing with the ANHT. They are somewhat similar. To fully determine what a particular calibration byte does, you'll have to trace the calls to it in the program. If they are called in a similar-looking segment of code, you can be pretty sure it is the same variable.
(3) Fill in the major items, by looking at the addresses from "public" definitions out there, TunerCat, Tuner Pro.
(4) Once you feel reasonably confident you've identified 90% of the variables, you could always send it off to someone who knows and have them check it for you.
I hope you are patient. That is basically how I did it back in 2002-3. It can take literally hundreds of hours doing it alone. Once you've finished the calibration section, then you can move on the program code, and understand not only "what" is there, but "how" it's used. The nice thing is, GM used a library of very similar code routines in many of the P4 ECMs, so there are definitely similarities with other well-hacked calibrations of that era.
You'll find the LT5 cal has nearly 600 constants, and over 70 tables.
Todd
(1) Grab a copy of the ANHT ($8D) listing, which is widely available on the 'net.
(2) Work through the calibration section byte at a time, comparing the LT5 listing with the ANHT. They are somewhat similar. To fully determine what a particular calibration byte does, you'll have to trace the calls to it in the program. If they are called in a similar-looking segment of code, you can be pretty sure it is the same variable.
(3) Fill in the major items, by looking at the addresses from "public" definitions out there, TunerCat, Tuner Pro.
(4) Once you feel reasonably confident you've identified 90% of the variables, you could always send it off to someone who knows and have them check it for you.
I hope you are patient. That is basically how I did it back in 2002-3. It can take literally hundreds of hours doing it alone. Once you've finished the calibration section, then you can move on the program code, and understand not only "what" is there, but "how" it's used. The nice thing is, GM used a library of very similar code routines in many of the P4 ECMs, so there are definitely similarities with other well-hacked calibrations of that era.
You'll find the LT5 cal has nearly 600 constants, and over 70 tables.
Todd
Where can we find the ZR-1 listing and which Mask ID is it of? Or does it matter? Of the 600 constants and 70 tables, how many did you find were really relevant to tuning beyond the basic SA and VE stuff? As you and I have discussed previously, I'm interested in PID controls, IAC/TF, sensor voltages and filter coefficients.
Thanks for the help.
#26
Todd,
Where can we find the ZR-1 listing and which Mask ID is it of?
Or does it matter? Of the 600 constants and 70 tables, how many did you find were really relevant to tuning beyond the basic SA and VE stuff?
As you and I have discussed previously, I'm interested in PID controls, IAC/TF, sensor voltages and filter coefficients.
Thanks for the help.
Where can we find the ZR-1 listing and which Mask ID is it of?
Or does it matter? Of the 600 constants and 70 tables, how many did you find were really relevant to tuning beyond the basic SA and VE stuff?
As you and I have discussed previously, I'm interested in PID controls, IAC/TF, sensor voltages and filter coefficients.
Thanks for the help.
Dom,
I doubt you will find a complete ZR-1 listing in public domain. My post above was to give tips on how one might go about *creating* their own ZR-1 listing from a binary file. Believe me, it takes a lot of work, and I have not seen anyone else show the initiative to attempt it on their own.
There are two official documents in private hands...1) an '89 pre-production calibration listing, maybe 100 pages. This one has made the rounds throughout the early tuning community, several have bootleg copies.
2) the motherload...a complete 750-page GM Engineering printout of the 1990 Level 1 development release calibration. The whole program. This is rarer than a golden unicorn, and there aren't copies runnning around out there. I am grateful to a good friend for even letting me put eyes on it.
Unfortunately (for me), neither of these documents were available to me when I started disassembling the LT5 code. I did it the old-fashioned way -- lots of hours slogging through the code. Had 99% of it nailed myself, then confirmed and corrected my errors with this documentation.
As far as relevance to tuning, it all depends on what youre trying to do. But I've done cals with 300 byte changes, or just a handful. Just depends on what I was working on.
Todd
Last edited by tpepmeie; 01-06-2008 at 12:30 PM.