Kevin Blown 95 TA

Steve, this looks really cool and I'm going to try it as soon as I get the chance.

Very cool that you put in all this work on a platform not too many care about any more.

steveo_

there were a few stupid mistakes in the last couple releases. i need to slow down and test everything a bit better.

please download latest version for bug fixes, this time i tested 'er out properly (and added a few new features as a bonus)

fbody_brian

I had to add
# HACK For linux
QMAKE_CXXFLAGS += -std=c++11

In the .pro file under your "HACK for xp, to get it to compile under my linux install. If you can add it and still compile under windows maybe you can add it and leave it there.

steveo_

i will try right now, thanks. is it actually workin' on linux for you?

fbody_brian

Almost!
lol

I need to figure out what I need to do for the serial port. I'm currently getting this message in the console:
Baud rate of serial port /dev/ttyUSB0 is set to 8193 instead of 8192: divisor 2929.687500 unsupported

and it just sits there unresponsive.

steveo_

ahhhh i was hoping that **** wouldn't happen! the oddest thing about this is that it's 8192 baud, and nothing else is. no serial drivers assume 8192 baud even exists.

would you be interested in testing a patch or two?

for my 'other' linux aldl software i used raw USB commands to the ftdi chipset using libftdi, which does all the actual chipset commands in userland. it had the side effect of having great synchronus timing with no real driver in the way so i could nail insanely tight packet retrieval (exactly matching the ECM's delays). the ftdi chip is programmable to any baud rate in 1bps intervals.

EDIT: what version of qt are you running??

fbody_brian

Running 5.5
I'll try any patches you want to. Right now even if i try to manually set the baud rate it still does the same thing.

chipset is "Future Technology Devices International, Ltd FT232 USB-Serial (UART) IC"

works great under wine in datamaster, so I know that the capability is there. Is this an issue with the serial configuration from QT?

Edit:
FYI i also have libftdi installed, don't know if that's relevant or not.

steveo_

i have the same chip. i'll get a virtual machine going and see what i can figure out.

not sure.

the configuration is straightforward, lines 261-267 in datastream.cpp. there's very little control outside of that.

nah, the qserialport lib doesn't support it anyway. you'd know if you were using libftdi, you'd have to disable your in-kernel ftdi driver.

fbody_brian

Well, it seems that the unix portion of qtserialport checks to make sure that the baud rate is exactly what it is supposed to be, no wiggle room. so even though a baud rate of 8193 is acceptable, it kicks out the warning. If I try to set the baud rate manually with:
setserial /dev/ttyUSB0 spd_cust spd_cust baud_base 24000000 divisor 2929.687500
which would give me 8192, it changes the divisor to 2929, as the divisor register on the UART accepts an integer, not
a floating point number. Giving us the 8193 baud rate.
It seems that the qt code should take this into account, but it doesn't.

strangest thing is the way it checks, because even if I set the baud rate in eehack to 8193, I get this:
Baud rate of serial port /dev/ttyUSB0 is set to 8193 instead of 8193: divisor 2929.329834 unsupported

steveo_

with 8193 baud you should still be getting data, as far as i know. i'll look into it a bit more.

fbody_brian

oh, I know. I believe the issue is with QT. it's throwing up that error, because of the way the check is run. it's using floats and the numbers returned by the serial port driver are integers.Possibly rounding errors?

I'm really thinking about changing the relevant code in qserialport_unix.cpp,

Prototype007

iTrader: (2)

I'm just now seeing this thread for the first time. Many thanks for your time and effort on creating this program.

Can't wait to try it out when I get back home to my laptop.

steveo_

new release:

http://fbodytech.com/files/eehack-3.0.zip

3.0:
** "More awesome than DataMaster_EE in every way unless you have a 4L60E"(tm)
** Add 4 line graphing.... whaaat!! (Actually wasn't that hard)
** Increase accuracy of RPM in tests and analyzer (use RPM16 everywhere)
Add reset BLM and clear DTC functions
Make some things more visible
All records are now timestamp sequential, even when loading multiple logs.
Misc. bug fixes, more analyzer tuning

Chas91Z28

iTrader: (1)

steveo, this looks like awesome work!
Does this only work with the $EE? Not the $EEB? Also, I'm assuming this is tested on the 16188051 ECM for the Camaro, but should also work for the 16181333 ECM for the Corvette, right?

steveo_

next release will have an option to work with 'vette body control module, this one does not silence it. i'ts never been tested, please test it..

$eeb uses the same datastream so it should work, but i'd appreciate you testing it out if you run $eeb (which there is no good reason to,afaik)

steveo_

new release http://fbodytech.com/files/eehack-3.2.zip

please test with 'vette to make sure it works! see settings page if you have a 'vette

has some more new features too.

Chas91Z28

iTrader: (1)

Glad you could tweak it to work with the Corvette. I'll give it a try in the next few days, and I'll post how it goes.

Nophix

Awesome work! I'm going to give it a shot on my '94 Formula. I've been tuning with TunerProRT and WinFlash. This looks so much cleaner, and being able to make tweaks without flashing every time will save me an ECM or 2.

Thank you!

junkLT1

interesting, glad to see new things coming for us

steveo_

version 3.3 is a big one, has FLASH READ capability!

please help me test it!

also a new custom command mode for guys that want to play with their own raw ALDL commands for educational or research purposes.

http://fbodytech.com/files/eehack-3.3.zip