I’ve got some test logs and configurations from people experiencing odd displays/logging on MS1-Extra versions. In particular this seems to effect the hi res rpm calculation and I currently think it’s related to installations where the spark table isn’t being used. I’m going to load one of the supplied configurations on a spare ECU and see if I can reproduce the problem.
The launch of the iPhone 5 with it’s new processor and screen size has left me in a bit of a quandry. To fully support the iPhone 5 (which I want to do) I’ll need to drop support for the iPhone 3G and iPod Touch gen 2 and earlier and any devices running iOS version prior to iOS 4.3. The stats I have suggest this shouldn’t be a problem BUT if it is a problem for you, please let me know.
Obviously I’m looking to support the iPhone 5’s new screen size in the next release.
I now have a working MegaJolt Lite Jr that has been lent to me for testing and development so I’m beginning work on the connection and logging classes to support it. Not sure yet which versions of MegaJolt it will be compatible with initially but hopefully that will become apparent as development progresses and I work my way through the tech notes.
The outstanding MegaSquirt version support (MS2 B&G code) is also on my list of things to add. I may end up adding them both as part of a single update or add and release each one as a subsequent release. I’m not sure how many users there are of the MS2 firmware but no doubt there will be some.
As always I’m open to any other suggestions and/or bug reports for the current release versions or the test versions. Feedback from users always gives a boost to the development effort, if only because it lets me know I’m not the only one using the app ! Version 1.1 now has over 200 users since it was released so hopefully some of those users are using it for logging and display !
V1.1 of the MobiSquirt app is now available on the App Store. This version contains a number of improvements, new features and bug fixes.
The most obvious changes from v1.0 are…
- This version adds support for logging of MS2-Extra firmware revision ECUs
- The app is now a universal app, running on both iPhone/iPod and iPad devices with new view layouts tailored to the iPad.
- The design of the main gauges views have, in my opinion, been much improved over the last version. Users who prefer the original look can still use that design via a switch on the settings screen.
- An AFR gauge option has been added that uses configurable lookup .ini file that can be customised by the user to suit their lambda sensor. This gauge does not currently work on log playback as the data to reproduce it isn’t available in the log files. This may change in a future version.
- The architecture used to handle the low level communications with your MegaSquirt ECU has undergone a complete rewrite. This improves initial connection establishment speeds and should result in a more stable connection.
You can view/download the app from the App Store here
I’ve submitted v1.1 of the MobiSquirt app to the App Store. Hopefully it will get through the review process and become available within the next couple of weeks.
I’ve run out of devices to test against now. I’m trying to track down a MegaJolt ECU I can test/develop against next and will continue to build up a library of raw data and pages files for my test environment so I can perform regression testing against them.
Anyone using the latest test versions or, when it becomes available, v1.1 or greater on the app store, can contribute raw data and pages file to have their setup included in the regression testing. More details in the user manual troubleshooting page.
As I’ve not received any substantiated error/bug reports for the current test release I’m anticipating submitting this to the app store at the end of this week.
There seems to be a problem with some MS1-Extra installs but I haven’t been able to reproduce it and none of the people reporting the problem have submitted the required files so I’m assuming it’s a local problem with their WiFi setup that they have since resolved.
I’ll be reviewing access to test versions for subsequent releases as the lack of any kind of feedback from “testers” is starting to make testing more difficult rather than easier. This in turn has started delaying development as I’ve ended up chasing bugs based on insufficient information.
If anyone does have a problem with the current test release please let me know after following the steps in the troubleshooting guide
This version is being treated as a release candidate and will be released to the App Store unless any problems are reported. If any problems are encountered please let me know via the forum or email so they can be resolved prior to release.
There are a few changes in this build with a couple of bug fixes. The most obvious change is a UI change with the gauges views having undergone a major overhaul. The new gauge design also allows a new visual indicator when a gauge is in a wanting or alarm condition. The original gauge style can still be used by turning it back on in the settings view.
The new gauge control classes should allow some level of customisation in future releases. I’ll document the files it uses at some point but if anyone has an idea for any gauge designs in the meantime, send them to me and I’ll look at implementing them into a future version.
I’m not getting any reports of problems with the current test version so I’m getting closer to releasing a new version to the app store. I’m adding a new gauge display layer to the app that will make it easier to customise the look of the gauges. This has come off the back of my own endeavours to try and offer a better dashboard look to the gauges view on the iPad.
Other than the graphics improvements the current test version is now being considered stable and working as far as it’s functionality is concerned. If anyone has any issues with the current test version then please let me know sometime soon !
I’ve had some positive feedback around the previous test version and have been working on the iPad views for this new version, particularly clearing up some issues around the log replay views for the iPad.
I think the iPad version is now functionally at the same point as the iPhone version so assuming nothing comes up from the testing of this version it’s looking good for a hybrid iPhone/iPad release to the app store in the next couple of weeks.
Build 261 has been released on TestFlight for evaluation by the registered testers. There are some testers who are signed up who currently don’t have any devices registered, please make sure you register your devices – once you are signed up simply visit the testflightapp.com from your device browser and log in, you will be given the option to register the device at that point. You can’t load or run the test versions of the app on devices that aren’t registered.
As an interim update I’ve added a new test build to TestFlight which registered testers can access. If you’d like to be added to the list for future test releases you can sign up to TestFlight here… http://bit.ly/vS78u9 – Don’t forget to add your devices so they can be included in the next test build.
The iPhone aspects of the app should be considered as a “Release Candidate” but the iPad views still need a bit of work. The main areas that require testing are the new connection functions and the MS2-Extra support. As per my previous post the connection architecture has undergone a major rebuild and on my test setups I’m seeing massive improvements in the time taken to establish a connection.
The MS2-Extra support was almost working in the previous release version but the logging wasn’t implemented properly and many of the gauges were hit and miss. This latest test version should resolve those problems and both record logs correctly and replay them on the iPhone/iPod platforms.
If the response to this test build is positive and the iPad version looks like it will drag on for more than a couple of weeks then I’ll release an interim version to the app store.
I finally completed the work on the MS2 Extra logging and I’m almost ready to send out a new test version.
I have been undergoing a bit of a code review too. In particular I’ve been re-looking at the architecture I implemented to handle the connections to different versions of MegaSquirt. At the time I first implemented the various classes and their structure it seemed to make sense but the downside seems to be a, sometimes lengthy, delay in establishing and initialising a connection.
As a result of this review I’m in the process of changing the architecture around connection handling and it’s looking very promising with connections being established much more quickly and reliably – typically in around 20% of the time the previous way of doing things took. There are a couple of outstanding things to resolve but on the whole it’s looking good for a new test release and subsequent app store submission in the next few weeks.