PDA

View Full Version : Sep 2015: EFILive Public Pre-Release



Blacky
September 2nd, 2015, 03:04 PM
Please download and install both the V7.5 and V8 software packages.

Both the V7.5 and V8 software were updated in this release, you must update both software packages.

Download V7.5 Software (http://download.efilive.com/Software/V7.5/EFILiveV7.5.7.293_Setup.exe) (101 MB)
Download V8 Software (http://download.efilive.com/Software/V8/EFILiveV8.2.2.286_Setup.exe) (64 MB)



Release Notes:
Whats New (http://download.efilive.com/Software/whatsnew_286.htm)

IMPORTANT
The Auto-Lock implementation for E39, E92 and E98 controllers requires that tuners and customers upgrade their AutoCal and FlashScan firmware to V2.07.86 or later. The upgrade is required to prevent *.ctz and *.coz tune files created using this pre-release (or later) from being used with older software or firmware.

Tuners: If you update your software and firmware and attempt to deliver new tune files to your customers, then your customers must also update to the following versions (or later):



[*=1]V7.5 Software: V7.5.7, Build 288
[*=1]V8 Software: V8.2.2, Build 281


or they will not be able to flash those new tune files. Existing/old tune files will continue to work normally with the new software.

Existing tune files will continue to work normally with the new software.


Known Issues
Please post new issues, questions or clarifications (large or small) in this thread.


As new issues are verified they will be posted in the Known Issues (https://forum.efilive.com/showthread.php?25938-Known-issues-with-the-September-2015-Public-Pre-Release) thread.


Version Numbers:

After installing the latest release you should have the following software, boot block, firmware versions installed:


EFILive V7 software: V7.5.7 Build 293
To view the installed version, select Help->About (in the V7 Scan or Tune software).

EFILive V8 software: V8.2.2 Build 286
To view the installed version, select Help->About (in the V8 Scan and Tune software).

Boot Block: V2.07.07 (Feb 07, 2014)
To view/update the Boot Block, click the [Check Firmware] button in the V8 Scan and Tune software.

Firmware: V2.07.91 (Sep 08, 2015)
To view/update the Firmware, click the [Check Firmware] button in the V8 Scan and Tune software.


Regards
The EFILive Development Team

joecar
September 12th, 2015, 07:58 AM
When closing the tunetool I see several popups saying this:

---------------------------
EFILive Tune - Digital tools for onboard vehicle electronics
---------------------------
System Error. Code: 1400.

Invalid window handle.
---------------------------
OK
---------------------------


To reproduce, you have to view some number of tables, then close tunetool.

JUSTASBC
September 18th, 2015, 08:50 AM
The black box feature is still not working. I have only tried it on LS1B. I am sort of surprised that people don't use or want this feature. Why lug around a laptop with you to read or flash in tunes with you all the time. You would be surprised how many potential customers turn into customers just because of that feature. Before I could meet someone when I have just my V2 with me and tell them I will download their tune for free, and adjust the tune at my leisure. Next time I bump into them just say if you want it, it is all done, it will only take a few minutes.

joecar
September 18th, 2015, 09:00 AM
I have it working on LS1B...

did you also update the V2 firmware...?
did you also update the BBx config...?

JUSTASBC
September 18th, 2015, 09:37 AM
Firmware I believe is updated it is currently 2.07.91. I'm not sure how to update the BBX config. I do have everything ticked for LS1B. I can still log data, I just can't upload and download tunes. I get error $0333. The only way I can upload or download a tune is to use V8 and tick Try Alt Keys., and Assume Lock May Be Faulty.

Blacky
September 18th, 2015, 11:19 AM
Firmware I believe is updated it is currently 2.07.91. I'm not sure how to update the BBX config. I do have everything ticked for LS1B. I can still log data, I just can't upload and download tunes. I get error $0333. The only way I can upload or download a tune is to use V8 and tick Try Alt Keys., and Assume Lock May Be Faulty.

Sounds like the LS1B controller has a damaged seed/key combo. When managing remote customers with an LS1 with a faulty seed/key, you can bypass the faulty seed/key combo when flashing the controller using BBX by setting the [Remote Control] options as shown below. Those settings are stored in the file and are only used for that single file. Because those settings are stored in a tune file they can't be used for reading a controller.
18841

If you want to fix the seed/key (and I recommend that you do), then you need to use V8 pass-thru reading/flashing:



Read the tune from the PCM (with the "Try Alt Keys" and Assume Lock May be Faulty" options checked) or open a file that matches the OS in your PCM.
"Save As..." the file with the name "Unlock.ctz".
Check the "AutoLock" option on the [Security Restrictions] tab page.
"Save As..." the file with the name "Lock". You muct have your FlashScan connected when saving any file with any security restriction active.
Cal-flash the Lock.ctz file (with the "Try Alt Keys" and Assume Lock May Be Faulty" options checked). That will set the seed/key to a custom seed/key combo.
Cal-flash the Unlock.ctz file (with the "Try Alt Keys" and Assume Lock May Be Faulty" options un-checked). That will reset the seed/key to a stock seed/key combo.


Regards
Paul

CR500
September 19th, 2015, 01:57 AM
So was this issue related specifically to JUSTASBC?

I have an HPDE coming up and I'll need to BBX my LS1B. But I also have some data files I need to go through before that track day to modify the tune and issues with B276 makes it tough. So I'm stuck, do I struggle through B276 or update to B286?

Thanks

joecar
September 19th, 2015, 09:54 AM
So was this issue related specifically to JUSTASBC?

I have an HPDE coming up and I'll need to BBX my LS1B. But I also have some data files I need to go through before that track day to modify the tune and issues with B276 makes it tough. So I'm stuck, do I struggle through B276 or update to B286?

Thanks
Yes, specific to him.


To update the software/firmware, the steps are fairly easy:
1. install V7,
2. install V8,
3. start V8 S&T and click Check Firmware, the rest is fairly obvious (if the firmware needs updating you will see a button "Update Firmware", click it),
4. and it also shows you if the V2/BBx config files need updating (you will see a button to update the files).

To configure V2 BBx: Setting-up-a-FSV2-from-scratch (https://forum.efilive.com/showthread.php?20699-Setting-up-a-FSV2-from-scratch)

CR500
September 19th, 2015, 12:01 PM
Good deal.

I'll try B286 tomorrow and see if it works with my calc pids. I'm already set up for BBX so as long as it doesn't get screwed up I'm good to go.

JUSTASBC
September 20th, 2015, 01:55 AM
Paul,

Thank you, I will try that later today. I never had an issue until I updated everything earlier this year.

JUSTASBC
September 20th, 2015, 07:39 AM
Unfortunately that still didn't fix it. I still get error $0333

Blacky
September 20th, 2015, 09:16 AM
Unfortunately that still didn't fix it. I still get error $0333

Can you please locate the trace files that were created when you performed the two cal-flashes?
The trace files are stored in the folder called: \My Documents\EFILive\V8\Trace
The file names will be something like this: 20150827_121612_LS1_W.htx
Where the two sets of numbers in the file name are the date and time the flash was attempted, in the format: YYYYMMDD_HHMMSS.
The "W" means the trace file is from a cal-flash.

Please send the files to paul@efilive.com with a link to this thread so I know what they are for.

Regards
Paul

Tre-Cool
September 23rd, 2015, 01:59 AM
not so much as a bug, but after wiring in a couple of widebands into the aircon pressure 0-5v pin and egr, when i went to update the bbx there was no gm.acp voltage pid.
I even tried editing the options file manually and added in:

PCM.ACP, N:::0:1:0:N, 0:N:0:N:0:10:0

Can you add this in please.

Blacky
September 23rd, 2015, 02:11 AM
not so much as a bug, but after wiring in a couple of widebands into the aircon pressure 0-5v pin and egr, when i went to update the bbx there was no gm.acp voltage pid.
I even tried editing the options file manually and added in:

PCM.ACP, N:::0:1:0:N, 0:N:0:N:0:10:0

Can you add this in please.

In V8, the GM.ACP PID has been renamed to ACHPRS (Air Conditioning High Side Pressure Sensor). USe that PID when using BBL. If/when you open the file in V8 it will automatically use the name ACHPRS, if/when you open the file in V7 Scan Tool it will automatically use the name GM.ACP.

For future reference the PID names that have changed between V7 and V8 are listed in the file: \Program Files (x86)\EFILive\V7.5\Configuration\sae_generic.txt
under the heading: "Conversion of BBL PID names back to V7Parameter names"

An easy way to "convert" from V7.5 PID names to V8 BBL names is to select the V7 PIDs that you want in the V7 Scan Tool, then highlight all the PIDs you want to log in V8 (click the first one, then hold the Shift key down and click the last one), and select Edit-> Copy for export to V8. Then in the V8 [F5: BBX] tab page, right click in the green section and select Paste. The V7.5 PID names are automatically converted to their V8 BBL equivalent names.

Regards
Paul

Biodiesel66
September 23rd, 2015, 10:10 AM
When BBL E86A controller. BBL with PID DPFPRV added. I get Error: Unsupported PID node: ECM PID: DPFPRV is not supported, plese de-select it.

I thought this PID was corrected and working?

joecar
September 23rd, 2015, 10:21 AM
When BBL E86A controller. BBL with PID DPFPRV added. I get Error: Unsupported PID node: ECM PID: DPFPRV is not supported, plese de-select it.

I thought this PID was corrected and working?It's there (for E86A and E86B)... enter DPFPRV in the Search box...

which V7/V8 build versions do you have installed...?

Biodiesel66
September 23rd, 2015, 12:22 PM
It's there (for E86A and E86B)... enter DPFPRV in the Search box...

which V7/V8 build versions do you have installed...?

That 's not the issue. The PID does no work. I did not say I could not add it. What part of the question don't you understand? Did you ever BBL a controller and get the above message on the V2 screen? ECM PID: xxxxxx is not supported, plese de-select it. That means remove the PID from the Config.txt file for the selected controller using EFIlive Scan &Tune. Re program the changed setting to V2 and start BBL.

It's a question for Paul or anyone else BBL an LML controller with DPFPRV PID selected along with other working E86 PID.

Using the version that I posted for in this thread.

Blacky
September 23rd, 2015, 12:56 PM
When BBL E86A controller. BBL with PID DPFPRV added. I get Error: Unsupported PID node: ECM PID: DPFPRV is not supported, plese de-select it.

I thought this PID was corrected and working?

After gettingthat message on your FlashScan (or AutoCal), can you please save a trasce file on the device before unplugging from the vehicle?
How to save a trace file: https://support.efilive.com/kb_article.php?ref=8552-EOAJ-5912
Under the heading: "FlashScan/AutoCal Read/Flash/Log Trace Files".

Please send the trace file to me at paul@efilive.com.

Regards
Paul

Biodiesel66
September 24th, 2015, 01:42 AM
Trace file sent

Blacky
September 24th, 2015, 02:01 AM
Trace file sent

There is either a mis-configuration in either EFILive's PID configuration for that PID, or GM is being inconsistent in its use/definition of that PID. It is defined as a 16-bit wide field in EFILive's software. However the trace shows that your ECM is only returning an 8-bit field. If all E86 controllers return an 8-bit field, then EFILive's config is wrong and we'll fix it asap. I.e. if all E86 controllers cannot log that PID using EFILive BBL then we have probably got it wrong and need to fix it. If one or more E86 controllers can log that PID using EFILive's BBL then we'll have to make a special hybrid definition of both 8 and 16-bits.

Have you successfully logged that same PID on any other E86 platform? If so what operating system number was that?

Regards
Paul

Biodiesel66
September 24th, 2015, 06:02 AM
There is either a mis-configuration in either EFILive's PID configuration for that PID, or GM is being inconsistent in its use/definition of that PID. It is defined as a 16-bit wide field in EFILive's software. However the trace shows that your ECM is only returning an 8-bit field. If all E86 controllers return an 8-bit field, then EFILive's config is wrong and we'll fix it asap. I.e. if all E86 controllers cannot log that PID using EFILive BBL then we have probably got it wrong and need to fix it. If one or more E86 controllers can log that PID using EFILive's BBL then we'll have to make a special hybrid definition of both 8 and 16-bits.

Have you successfully logged that same PID on any other E86 platform? If so what operating system number was that?

Regards
Paul

No.

JUSTASBC
September 30th, 2015, 03:04 AM
Just to update this. I can still not flash in or read a tune with my V2 in my truck. It appears this might be something to do with my truck or the tune. I am able to flash in tunes in another LS1B without problems. I'm stumped on this one.

joecar
September 30th, 2015, 03:25 AM
Just to update this. I can still not flash in or read a tune with my V2 in my truck. It appears this might be something to do with my truck or the tune. I am able to flash in tunes in another LS1B without problems. I'm stumped on this one.
Did you disable the modules that want to claim the bus during PCM flashing...?

Can your PCM be bench flashed...?

JUSTASBC
September 30th, 2015, 03:38 AM
I'm not sure how to do that. Everything worked fine before I updated earlier this year.

joecar
October 1st, 2015, 03:35 AM
Just to update this. I can still not flash in or read a tune with my V2 in my truck. It appears this might be something to do with my truck or the tune. I am able to flash in tunes in another LS1B without problems. I'm stumped on this one.When you try to read/flash, can you copy the console text log and paste it here...?

Blacky
October 1st, 2015, 08:20 AM
Just to update this. I can still not flash in or read a tune with my V2 in my truck. It appears this might be something to do with my truck or the tune. I am able to flash in tunes in another LS1B without problems. I'm stumped on this one.

How are you attempting to read the PCM in your truck?



V7 pass-thru
V8 pass-thru
FlashScan BBR standalone


I recommend using V8 pass-thru to try and read the truck and if/when it fails, please locate the matching LS1 trace file that gets created in this folder: \My Documents\EFILive\V8\Trace and send it to me at paul@efilive.com. Include a link to this thread so I know what its about.

Regards
Paul

5.7ute
October 7th, 2015, 04:36 PM
Paste & multiply by custom percent is messed up. It pastes and multiplies by percentage of the actual cell Value and not the BEN percentage.
For instance a cell set at 100 with a BEN of 1.08 applied with a custom percentage of 90% will become 97.1276. Lowering the value not raising.
This occurs in all VE tables.

Blacky
October 7th, 2015, 06:49 PM
Paste & multiply by custom percent is messed up. It pastes and multiplies by percentage of the actual cell Value and not the BEN percentage.
For instance a cell set at 100 with a BEN of 1.08 applied with a custom percentage of 90% will become 97.1276. Lowering the value not raising.
This occurs in all VE tables.

Use "Paste error correction" when pasting BEN factors. The error correction correctly applies the BEN factor.

Regards
Paul

Blacky
October 7th, 2015, 06:57 PM
The "Paste error correction" options are not yet documented in the V7 manual. I'll try and explain them here.

"Paste and multiply (X%)" means that the data in the cell is modified to be:
[cell value]*[clipboard value]*X/100
You can set a custom % value at the bottom of the Properties->[Options] tab page.

The options named "Paste error correction" are used when pasting data that represents an error correction ratio. An error correction ratio is a value that lies between 0 and 2 with 1 as the neutral point. Values between 0 and 1 are negative corrections, values between 1 and 2 are positive corrections. It is most useful if/when you are logging data using a calculated PID (i.e. BEN factor PIDs) where the calculated PID outputs a ratio between 0 and 2 (lambda is one such ratio).

For example when calibrating the VE tables, it is useful to generate a scan tool map the same shape/size as the VE table, filled with BEN factors which are logged based on the ratio of the commanded AFR v's recorded wideband AFR. A BEN factor of 0.9 represents a required negative correction of 10%. And conversely a BEN factor of 1.25 is a positive correction factor of 25%.

Before the "Paste Error Correction" option existed, customers would just use the "Paste and multiply" options to paste the BEN factor error corrections into the VE table. That worked well when the user did not attempt to apply any percentage weight to the correction. However, some customers requested the ability to apply the percentage weight to the BEN factors when pasting them into the calibration to prevent overshooting the optimal result. So the "Paste error correction" options were added along with a different (but correct) way to apply the percentage weight. The percentage weight is intelligently applied to the correction factor with respect to the direction of the correction. Error corrections are always modified towards 1, correction values less than 1 are increased by the percentage weight and correction values greater than 1 are reduced by the percentage weight.

"Paste error correction (X%)" means that the data in the cell is modified to be:
[cell value]*(1.0+(([clipboard]-1.0)*X/100)).

Pasting an error correction value of 0.9 (without any % correction, i.e. with X=100) into a cell that already contains the value 123.45 would result in a new cell value of
123.45*(1.0+((0.9-1.0)*100/100))=111.11
So (as everyone does and as I did when I first designed the software) you may ask what's the difference between "Paste error correction" and "Paste and multiply". In the above example, nothing. But watch what happens when you start to apply percentage weights to the paste operation.

Say you wanted to use a 50% modifier to only apply half the weight of the clipboard correction data (i.e. X=50). What we are really asking for is to reduce the calibration value by 50% of the 0.9 error correction value that is on the clipboard. Remember 0.9 on the clipboard is an error correction of negative 10%. In other words what we are ultimately asking for is a reduction of 5% of the value in that cell.

Using the Paste and multiply way: [cell value]*[clipboard value]*X/100
123.45*0.9*50/100 = 55.55 (which is not a 5% reduction).

Using the Paste error correction way: [cell value]*(1.0+(([clipboard]-1.0)*X/100)).
123.45*(1.0+((0.9-1.0)*50/100))
123.45*(1.0+(-0.1*50/100))
123.45*(1.0+ -0.05)
123.45 * 0.95 = 117.28 (which is a drop of 5% from the original value).
Notice how in the above example the error correction of 0.9 (i.e. negative 10%) was effectively halved to 0.95 (negative 5%) because the requested percentage weight was 50%.

Note:
The "Paste error correction" options ignore any values coming from the clipboard that are not in the range 0..2, the corresponding cell in the calibration is not modified.

Hope that helps to clear up any concerns over the different paste methods.
Regards
Paul

5.7ute
October 8th, 2015, 01:15 PM
Thanks for clarifying that Paul. I hadn't noticed the paste error correction before. So used to doing it the old way.

IdahoRob
October 14th, 2015, 03:47 AM
Any time line on when this will become the public release? Thank you,

GMPX
October 14th, 2015, 08:40 AM
I'm not sure as I don't know what feedback has come in via support, but I recently had to make a change to accommodate the 2016 LML so that will need to be included in the final release to save everyone having to update again. Paul might jump in and give a more precise time frame.

Blacky
October 14th, 2015, 08:42 AM
The plan is before end of October.
Regards
Paul

5.7ute
October 19th, 2015, 10:56 AM
Is the chart full screen view (F10) in V8 implemented yet? I get two different types of failures depending on what laptop I use. Windows XP has a black screen & only shows the chart as I move the cursor, Windows 10 sticks in a loop tracing the charts then goes to a black screen, touching any key restarts the loop.

Blacky
October 19th, 2015, 10:58 AM
Is the chart full screen view (F10) in V8 implemented yet? I get two different types of failures depending on what laptop I use. Windows XP has a black screen & only shows the chart as I move the cursor, Windows 10 sticks in a loop tracing the charts then goes to a black screen, touching any key restarts the loop.

It is a bit dodgy on some platforms. It has been reworked for the next update due out end of this week.
Regards
Paul

auspeed
October 21st, 2015, 11:10 PM
i keep getting this error when updating the config files

18961

auspeed
October 21st, 2015, 11:13 PM
after a refresh it now shows all files are updated.... all good

Blacky
October 22nd, 2015, 06:45 AM
i keep getting this error when updating the config files

18961

It may not be "all good" just yet. Error $0194 usually means that the internal memory is full. So it may be one config file did not get completely updated. The update option may show files as "up to date" even if they did not copy across correctly. That is because it uses the date/time stamp of the file to determine if it is the latest version of the file. It does not check the contents of each file.

If an error $0194 occurs during the update process, I strongly recommend removing one or more controllers (that you don't need to work with) from the list of supported controllers that get loaded onto your FlashScan. You can do that using the [F5: BBX]->[F2: Scan] and [F3: Tune] options. Either delete the controller's entry completely, or just uncheck the entry. Then click the [Program] button at the bottom of the screen.

Regards
Paul

Chuck CoW
October 23rd, 2015, 06:38 AM
Say you wanted to use a 50% modifier to only apply half the weight of the clipboard correction data (i.e. X=50). What we are really asking for is to reduce the calibration value by 50% of the 0.9 error correction value that is on the clipboard. Remember 0.9 on the clipboard is an error correction of negative 10%. In other words what we are ultimately asking for is a reduction of 5% of the value in that cell.

Using the Paste and multiply way: [cell value]*[clipboard value]*X/100
123.45*0.9*50/100 = 55.55 (which is not a 5% reduction).

Using the Paste error correction way: [cell value]*(1.0+(([clipboard]-1.0)*X/100)).
123.45*(1.0+((0.9-1.0)*50/100))
123.45*(1.0+(-0.1*50/100))
123.45*(1.0+ -0.05)
123.45 * 0.95 = 117.28 (which is a drop of 5% from the original value).
Notice how in the above example the error correction of 0.9 (i.e. negative 10%) was effectively halved to 0.95 (negative 5%) because the requested percentage weight was 50%.

Note:
The "Paste error correction" options ignore any values coming from the clipboard that are not in the range 0..2, the corresponding cell in the calibration is not modified.





I tried to beak the habit of paste and multiply...

When you are close to your target, paste and muiltiply works good.

When you are trying to apply a ben factor of approx 10% or more, the multiply function

overshoots the target and it wastes tons of time. (ie: Wideband is only accurate when close to your target)

When a wideband says you're 10+% lean or rich, that applied correction overshoots the target every time.

When using the multiply with 50% weight it makes for a more accurate adjustment and ultimately less re-tuning time.

I really like to use that feature lately.

Chuck CoW

Move'n Up
October 28th, 2015, 05:34 AM
Paul, is the update release still on target for this week?

GMPX
October 28th, 2015, 06:37 AM
'maybe' , another build was done yesterday for some internal testing.

Blacky
October 28th, 2015, 07:01 AM
Paul, is the update release still on target for this week?

Yes.

Move'n Up
October 28th, 2015, 08:09 AM
Thx, Good to know!

Blacky
October 28th, 2015, 08:24 PM
October update posted here:
https://forum.efilive.com/showthread.php?26072-October-2015-EFILive-Public-Pre-Release

Regards
Paul