MS3 pre-1.4 alpha 2 released

[jsmcortina]

To announce the release of "MS3 Pre 1.4 alpha 2" - this is a new stream of development codes.
ETA on this 1.4.x code becoming release is 6-12 months.

Who should use this code?
Experienced users wanting to test the new features.

Who should not use this code?
New or casual users - please use the current release code instead - see the Downloads page linked above.

Main changes since pre-1.4 alpha 1
Many of the bug fixes are the same as applied to 1.3.x
Bug fixes:
Many after-launch features broken on 2014-08-07, STATUS2_LAUNCH was not being used.
Fix table blend behavior with boost bias table.
Fix timer rollover for boost delay.
Fix rogue pulses on MS3X secondaries for staging or dual-fuel.
MS3X non-sequential staging uses expected number of outputs.
Fix staging transition "saw-tooth" issue.
Re-do staging duty calcs to work with sequential.
Boost control "mid" output setting wasn't working.
Fix up ini bitfield errors (would have broken some settings on gen PWM, water inj)
Fix Time-After-Spark (same issue as MS2)
Fixed max VSS for launch and change logic for VSS flat-shift.
Fix MPG calc (off by factor of 10)
Fix CL idle getting "stuck" just after cranking.
Fix flex fuel temperature calculation.
Add config error for unsupported 36-2-2-2 combinations
Fix TPSdot limits in ini.
EDIS sets "Full RPM sync"
Fix rotary trailing calculation error (evident near 0deg advance)
Allow spark 45deg ATDC. (Was limited to 10ATDC in Oct'13)
Fixed zero sized extended CAN packets xmit in error.
Assign correct config error numbers for VSS2.
Fuel pump (O/L or C/L) sets correct portusage.
Allow 3-cyl single-coil
Fix to oddfire custom angles (accumulate odd_trig_offset)
Allow fuel calcs to run w/o MAP or MAF sampling.
Don't flag VVT error for on/off output.
Fix typo nitrous / 3step.
Powercycle required for nitrous.

New / changed features:
Average 1-cyl RPMs over two revs.
Experimental non-linear spk cut for launch.
Remove linearity curve and add bias curve for PID alt ctl. Switch to ideal PID.
Average 1-cyl RPMs over two revs.
Use TS generated CLT, MAT GM calibrations. Use MAT for "GM calibration"
Traction changes for perfect run and button modes.
Rename "closed/open" to "minimum/maximum" for boost and idle valve settings.
Extend vssout_scale to 32bit. (untested)
EGO "load above" setting allows negative numbers.
Implement CAN TX timeout and reset.
Change NGC6 phasing again.
Swap phase on GM7X 6cyl.
Remove "ability" to use +360 on trigger angle/offset to swap phase - not working.
VSS and SS sampled every 50ms instead of 100ms
Add 6-cyl 36-2-2-2 support. Phasing TBC.
Allow 6 cyl suby VVT (untested)
Add rough cut at Daihatsu 12+1 mode.
Add Miata bang-bang alt ctl.
CAN broadcast tweaks from MS2.
Add AEM 3.5 bar, 5.0 bar MAP sensors to known list.
Add new advanced options to launch for different limiter types.
Allow nitrous 1 during launch.
Add max RPM for A/C.
Add line-lock staging.

Minor / internal:
Use "vvt_decoder" var in init.
ini tweaks
Allow up to 24 VVT teeth in ini.
Change defaults for ITB load.
Fix a couple of ini error that TS missed.
Minor optimisation on AE code (each function call has ~87 cycles cost.)
Extend AFR table limits in ini.
ini - capitalisation and consistency
- some units standardisation (gives some warnings on old MSQ load)
- remove GridHeight
Reduce serial timeout to 25ms.
Move SPKMODE49 to part2
Clarify error 71.
Tweak CAN VSS defaults.
Add some missing I/O pin usage names.

The firmware zip is available from:
http://www.msextra.com/doc/ms3/files/al ... alpha2.zip

James

[shaodome]

YES!!!!!!!!!!!!!!!

This weekend is going to be fun. I'll report back. Thank you james.

[shauer]

Didn't see it in the list but thought I would ask anyway.

Did anything get done with extending the ego delay table values beyond 255ms?

Thread discussing it and Ken's response about a possible 1.4 change:
http://www.msextra.com/forums/viewtopic ... 25&t=54655

Thanks!

[muythaibxr]

I have not even started on this change yet. I will look at this and the other requests on my plate on the 4th.

[hardline]

I'll upload and test this tonight. Would have last weekend, but was too busy with Pikes Peak

[piledriver]

Worked fine for me so far, have yet to notice any issues.

[hardline]

Only issue I have seen so far is that the uploader did not work. I used the one from the 1.3.1 Beta 4 distro and it worked.

[jsmcortina]

Only issue I have seen so far is that the uploader did not work. I used the one from the 1.3.1 Beta 4 distro and it worked.

What exactly didn't work? What OS?
The file sizes on all four loaders appear identical between those two releases, so ought to be the same?

James

[hardline]

Said something about a serial protocol, don't remember to be honest. I did not screenshot it as I wasn't thinking. Running Windows 7 SP1 64 bit

[Suprazz]

installed on my car. Runs well. No problems with the uploader.

[hardline]

Noticed one issue last night. Was experimenting with AntiLag and roving idle. Decided it wasn't worth the effort at this time. Turned antilag off, but roving idle stayed on, even after a few key off/on resets. Had to turn antilag back on, disable roving idle, then turn antilag back off. Though that was odd that a child feature stayed enabled when the parent feature was disabled

[nismoautoxr]

I loaded this on my MS3 and MS3X and I am very pleased at the changes to launch control. in 10 minutes I was able to dial it in tight and maintain the hard limit . It took adding 7 ms fuel to do it along with using 4/5 cuts at 2* timing.. No crazy backfiring muffler destroying stuff but a very tight limiter with an indicated 14 to 1 AFR.

I did however have to try and load the firmware 3 times before I got it to go. I took screenshots of all the attempts . THis is with a laptop that has loaded multiple firmwares . OS is windows 8. FW I was coming from was pre-1.4 alpha 1. It was via FTDI cable to USB.

[nismoautoxr]

oops ...looks like i should have scrolled my screen to the bottom on the first 2 screenshots so you could see what it said . Wierdness was that it detected **something** on the 1st attempt on com4 (Im on the same port the whole time ) but it said it was garbled and it failed but it also must have initially started the erase because my can gauge read ER...maybe 4 I think which is what it does when the firmware is being erased and flashed .THen the next attempt found the MS on com5 yet Im still plugged in to the same port. Now I think that the bluetooth may have been involved in this . I had phils bluetooth plugged into the DB9 through the 1st and second attempt but removed it by the 3rd attempt . I usually pull the BT adapter off for FW flashes but it was screwed in and I was having trouble getting one side out . I dont know if it was related to the failures but I felt it only fair to include that info.

At any rate ...nice changes to launch . I had come to an impass the way it was . It was unusable for me before this. Thanks.

[jsmcortina]

Noticed one issue last night. Was experimenting with AntiLag and roving idle. Decided it wasn't worth the effort at this time. Turned antilag off, but roving idle stayed on, even after a few key off/on resets. Had to turn antilag back on, disable roving idle, then turn antilag back off. Though that was odd that a child feature stayed enabled when the parent feature was disabled

Sounds like a bug to me. I'll investigate.

James