new $EE tuning thing!

[steveo]

in scan9495 error 51 eeprom programming dtc shows up. when comparing before and after flashes 0x3848 Shift Time (sec) Vs. %TPS Vs. Shift, Normal Mode, High Alt. table has 3858 as FF instead of what was originally there.

ah, is that an automatic transmission bin? it shouldn't have done that. i'll look into it.

in scan9495 error 51 eeprom programming dtc shows up.

i guess i screwed the checksum up.

i think read flash is not working by that i mean that eehack will pull the bin but not to prompt to save it

thanks i'll look into that too

appreciate the testing!

[jthompson122183]

ah, is that an automatic transmission bin? it shouldn't have done that. i'll look into it.



i guess i screwed the checksum up.



thanks i'll look into that too

appreciate the testing!

yes auto bin.

[steveo]

i would recommend against using the optimizer for the time being...obviously needs some work.

would be curious if the patch works though

[steveo]

i think this patch "fixes" the optimizer by removing a routine that makes compression work more often, but it still has a benefit. the optimizer, right now, should only patch out the automatic transmission tables if the automatic transmission flag is not set, and the maf tables if the speed density flag is set.. since this is write-time, as long as you're the kind of person that keeps track of their bins locally, you should never know it's happening. it skips some 0xFF bytes if they show up at opportune times, though, so very small speed-up on auto/maf bins too.

i can't seem to reproduce where the automatic transmission bin had its shift tables filled with 0xFF, are you sure that's not a manual trans bin? might have been simply corrupt due to the optimizer being broken.

[jthompson122183]

i think this patch "fixes" the optimizer by removing a routine that makes compression work more often, but it still has a benefit. the optimizer, right now, should only patch out the automatic transmission tables if the automatic transmission flag is not set, and the maf tables if the speed density flag is set.. since this is write-time, as long as you're the kind of person that keeps track of their bins locally, you should never know it's happening. it skips some 0xFF bytes if they show up at opportune times, though, so very small speed-up on auto/maf bins too.

i can't seem to reproduce where the automatic transmission bin had its shift tables filled with 0xFF, are you sure that's not a manual trans bin? might have been simply corrupt due to the optimizer being broken.

16209481 from your site is the bin im using

[steveo]

i got a FAST dual wideband donated to me, can't wait to hook that up, and my sockets/chips should be here any day. should be a great excuse to test my fast-rate logging.

[steveo]

identified a problem that decreased wide band resolution (due to an integer conversion that I missed). also threw in a better wb config interface that lets you specify x, y for the afr point, so you don't have to understand linear math to get two-point configs other than 0v/5v correct. im going to provide presets for common widebands too. lt1+wideband+eehack=awesome in the next release....