Nistune 0.10 Release

[maglito]

Excellent! Thanks Matt!

[maglito]

Just thought of something...if the innovates are daisy chained, will I be able to use a DLP for sensor input? Or maybe I can just use the DLP for sensor power and send the output to the analog inputs on the LM-2?

[Matt]

you could only use the DLP for power in this case (or just get a standard USB lead, cut the end and use the black/red to power the sensor)

[RomChip200]

Matt,

My OS is WInXP 32bits pack3.

This new 0.10.1 version is a nightmare. My previous version was 0.9.14.22, everything worked fine (incl. Innovate MTS LHS & RHS LC1 widebands)

I have Prolific USB serial cables for ages. And the driver 2.0.2.1 installed for ages too.
The 0.10.1 exhibited some problems in the Prolific driver. The Prolific drivers are known to be more than dodgy.
Well, I found out this link:
http://www.vems.hu/wiki/index.php?page= ... dowsDriver

Prolific 2.0.2.1:
COM ports are correctly shown in Windows hardware manager.
0.10.1 doesn’t find COM Ports anymore (either from the Config panel or directly from the main UI trying to connect directly) but 0.9.14.22 did fine.

I switched to Prolific 2.0.13.130 as recommended:
I got yellow marks in windows hw manager. Cannot go further.

I switched back to 2.0.0.26:
0.10.1 detects COM ports (as 0.9.14.22 …)
But when trying to connect to Innovate MTS (COM port is found), I got a blank panel, no values, if I disconnect I got an unhandled exception and Nistune crashes.
Consult connection works fine. But you have a bug in the Consult data display panel: the analog values are concentrated in the right top corner (bad spacing).

What did you do in 0.10.1 ? I’m not sure the Innovate MTS trouble is due to Prolific driver but more to Nistune 0.10.1 version.

[nissangts]

Matt,

My OS is WInXP 32bits pack3.

This new 0.10.1 version is a nightmare. My previous version was 0.9.14.22, everything worked fine (incl. Innovate MTS LHS & RHS LC1 widebands)

I have Prolific USB serial cables for ages. And the driver 2.0.2.1 installed for ages too.
The 0.10.1 exhibited some problems in the Prolific driver. The Prolific drivers are known to be more than dodgy.
Well, I found out this link:
http://www.vems.hu/wiki/index.php?page= ... dowsDriver

Prolific 2.0.2.1:
COM ports are correctly shown in Windows hardware manager.
0.10.1 doesn’t find COM Ports anymore (either from the Config panel or directly from the main UI trying to connect directly) but 0.9.14.22 did fine.

I switched to Prolific 2.0.13.130 as recommended:
I got yellow marks in windows hw manager. Cannot go further.

I switched back to 2.0.0.26:
0.10.1 detects COM ports (as 0.9.14.22 …)
But when trying to connect to Innovate MTS (COM port is found), I got a blank panel, no values, if I disconnect I got an unhandled exception and Nistune crashes.
Consult connection works fine. But you have a bug in the Consult data display panel: the analog values are concentrated in the right top corner (bad spacing).

What did you do in 0.10.1 ? I’m not sure the Innovate MTS trouble is due to Prolific driver but more to Nistune 0.10.1 version.

oh damn. Matt is it possible to have two versions of Nistune, one for XP users and the other for Vista/Windows 7? Maybe the adapation of Nistune for Vista/Windows 7 is causing all this trouble?

[Matt]

Hey guys I'm going to put the .22 version back up tomorrow night whilst I go through the issues with 0.10.1 version

I'll have to go through the individual problems with the different hardware. I do have a prolific device here which works fine with my LC-1 so that was tested

One of the main issues was that bluetooth com ports were causing Nistune to freeze up (and just about any other application like LogWorks for example) which scans all COM ports on the PC

If you look on the Innovate forums you will see issues with bluetooth ports also. It freezes up the device scan process causing the application to hang

More to follow..

[nissangts]

Hey guys I'm going to put the .22 version back up tomorrow night whilst I go through the issues with 0.10.1 version

I'll have to go through the individual problems with the different hardware. I do have a prolific device here which works fine with my LC-1 so that was tested

One of the main issues was that bluetooth com ports were causing Nistune to freeze up (and just about any other application like LogWorks for example) which scans all COM ports on the PC

If you look on the Innovate forums you will see issues with bluetooth ports also. It freezes up the device scan process causing the application to hang

More to follow..

Thank you very very very much Matt . I sincerely hope you get all the bugs worked out.

[RomChip200]

Matt, what's your Prolific driver version ?

It seems you optimized connection times over COM ports in 0.10.1 (when it works )... scanning all the ports (when entering config panel) is shorter than in previous versions.
Did you shortcut some stuff ?

[RomChip200]

Another bug very annoying b/c it occurs very often (COM connections unstable):
if spurious disconnection occurs when log player is ongoing, the file is not saved on the disk, even if you press "yes I really want to save this file"

[DVS JEZ]

so far i dont have any bugs with 10.1 using acer with vista 32bit
but i cant find any of my tuned files, in either documents/nistune or program files/ nistune.
not even in nistune software itself either, did i miss an install step somewhere?

about to try connect to hcr32, will post up how it goes.
Edit: ok the hcr32 was a bit of a fail, takes 2 attempts to connect, then once connects and try to resync- read ECU spazes out and bottom left of screen reads "Check sum error detected (total 4 frames) and then alot of the timing tables, enrichment tables etc are red, this is with a factory ecu non nistuned, i connected to the same car this morning with the existing software no probs.

[Matt]

I have Prolific USB serial cables for ages. And the driver 2.0.2.1 installed for ages too.
The 0.10.1 exhibited some problems in the Prolific driver. The Prolific drivers are known to be more than dodgy.

Okay here is what I am running:

Windows XP Home (32 bit)
ATEN (Prolific driver) 2.0.0.18 (16/7/2003)

Windows 7 Home (64 bit)
ATEN (Prolific driver) 3.3.7.131 (15/10/2009)

Dare say its doing timeout on the COM port not letting you select it? I am using a different scan function to search for com ports (rather than opening each one which freezes up on bluetooth ports and odd devices). Perhaps that driver doesn't work with this function? I can try switching versions here based on your feedback

if I disconnect I got an unhandled exception and Nistune crashes.

The crash probably come from the MTS DLL itself since I noticed that here also whilst running in the debugger. What was the method of disconnect (you clicking Wideband button or external disconnection like pulling out the plug or power disconnected to wideband)

When selecting Innovate in the list, we no longer use the normal COM port scan methods, we use the innovate ones since I have no choice but to use them

oh damn. Matt is it possible to have two versions of Nistune, one for XP users and the other for Vista/Windows 7? Maybe the adapation of Nistune for Vista/Windows 7 is causing all this trouble?

Vista/7 updates were limited to the location of the wideband tables, address file and ROM pack only

These other updates with COM ports were for XP also since various machines freeze up when connecting to consult in scan mode (or going into File-Configuration) because we were opening each port and checking it. I can roll back the com scan change if these issues persist

I sincerely hope you get all the bugs worked out.

So do I since I literally spent hours reading and trying different ways of enumerating serial devices in windows to come up with a better solution and then testing on three different machines and monitoring time outs etc

Matt, what's your Prolific driver version ? It seems you optimized connection times over COM ports in 0.10.1 (when it works )... scanning all the ports (when entering config panel) is shorter than in previous versions. Did you shortcut some stuff ?

You will see versions above in this post

Yes as mentioned I have optimised scanning. This is used to create the COM port list in File-Configuration as well as Consult auto scan.

For each port I call GetDefaultCommConfig() twice. After this if the port exists then I call a thread which does a CreateFile() to open it up. If it is a bluetooth port the second part will get stuck (like it does normally). We give the thread 200ms to respond with port number otherwise it times out and continues on

Therefore any ports which take longer than this time will timeout but not hang the software. If some devices are taking longer than this then I will have to increase the timeout which will extent the scanning time potentially but prove more reliable

if spurious disconnection occurs when log player is ongoing, the file is not saved on the disk, even if you press "yes I really want to save this file"

okay will look into this

but i cant find any of my tuned files, in either documents/nistune or program files/ nistune.
not even in nistune software itself either, did i miss an install step somewhere?

This is that annoying Vista/7 issue so the reason for moving to My Documents. Try seeing if you can find them in the older version (uploaded today).

I only did the following:
1. Updated the drivers
2. Move C:\Program files\Nistune\ROM Pack to My Documents (were you tunes in this folder?)
3. Move C:\Program files\Nistune\Logs to My Doucments
4. Remove C:\Program files\Nistune\Address and put updated copies in My Documents

Can you remember where they might have been saved? Have you searched for .BIN files on your machine as windows puts them in a different (locked folder) if saved inside program files so they dont really go there

ok the hcr32 was a bit of a fail, takes 2 attempts to connect, then once connects and try to resync- read ECU spazes out and bottom left of screen reads "Check sum error detected (total 4 frames) and then alot of the timing tables, enrichment tables etc are red, this is with a factory ecu non nistuned, i connected to the same car this morning with the existing software no probs.

I've noticed the double connect here also. Its the com scan thing above timing out (so will disable the timeout for consult scanning for now and increase the timeout for File-Configuration scanning list generation)

As for the checksum error detected. This error message was added back in but I've been testing many ECUs including factory here (previously it was disabled). Was the car running when this was happening? Just tried it here with HCR32 ECU with 04U02 on the bench and displays FACTORY after the part # and is able to connect and sync okay

Okay I've re-released 0.9.14.22 version today whilst I work through the issues with 0.10.1 version. This is up on the website. I will put together a 0.10.2 to fix addressed issues so far

[DVS JEZ]

yes it displays FACTORY after the ecu #.
i have reloaded 13.22 but i definately can't find my tunes except for 1 that i had saved in a different folder outside of program files last week.
i will try the hcr32 again in the morning and let you know

[TM_S13]

are you sure it ain't supposed to be the 0.9.14.22 version instead? or is it a miss-type error ?

[Matt]

yes typo. 0.9.14.22

[Matt]

0.10.2 - Updated version up tonight

Replaced GetDefaultCommConfig with SetupAPI function calls for comms port autoscan
Added comms port description to drop lists
Consult threading to callback updates to fix initial connectivity issues
Innovate dual AFR chaining fix
Allow configuration directory change back to Program Files
Drag/drop cell move crash issue fix with Grid Control
Added logging recognition for map knock flags on playback
Added divide by zero checks throughout to prevent potential crashes

GetDefaultCommConfig() was not reporting all the COM ports each time. SetupAPI seems to work well but I am concerned that if you dont run with higher level privledges that the COM ports may not appear at all but most people run as admin rights on their PC so I think it should be okay. Comms scanning is still quick now, even with bluetooth ports. SetupAPI calls also give me the device name so I'm able to display that like Innovate software does when it lists its devices

Please let me know if map highlighting is still an issue. I went through that code looking for potential problems and added some divide by zero checking. As I cant repeat that one here it is proving difficult

Innovate dual chaining should work again - but I only tested with my simulator so will do full testing once my serial patch cable arrives

I tested Zeitronix for nanook - this appears to be working fine now with the COM port. Even just selecting old logs and playing those back displays correct boost, AFRs etc from the log file. Was a problem with knock flag in the log not recognised so I've added that back in. Still to test and complete the knock map recording/playback tomorrow

Havent looked at the log disconnect and file not saving yet. So far my log files seem to be there whilst doing some random testing ... still to look into that

These are my remaining items left for this release

- Get AFR vs RPM/speed graph working again (would like to put boost in here also whilst I am it so its comparable to what you see on the dyno)
- AFM trigger voltage (check)

Anything else extra please let me know