- 3.3% growth last month
- One ANR from a very old version. That doesn’t count 😉
- Crashes about 1.5 per day, more than half of them caused by an uncatchable error related to Android 8.1
Month: June 2019
Thank you all for installing and running no less than 5000 CanZE instances on Android! A humbling number for a total amateur project.
Copied from the comments section……..
A new beta was just released today.
You can download this beta by following this link : https://testflight.apple.com/join/nhMX6f5n
WARNING : Version 4 of iOS CanZE need both WiFi and Cellular network.
To allow device connecting booth WiFi with dongle and Cellulat to CanZE Server, follow this tutorial :
- Connect to the WiFi Dongle
- Go to settings App
- Go to WiFi settings
- Select Dongle’s WiFi network
- Tap the (i)
- Note IP address and subnet mask
- Switch from Automatic to Manual
- Enter IP address and subnet mask
- IMPORTANT, leave router blank
Although the last picture of Raplh’s Zoe Plugin has a make-like flair, he didn’t stop there and even installed some led indicators using a NeoPixel strip.
In the last 14 days almost 50% of all crashes came from one single user on Android 8.1 and Cubot J3 hardware. If you are that user, could you please leave a reply? We’d love to figure out what’s going on!
Thanks to Ralph for sending me pictures of his Zoe plugin, which is basically a CanSee dongle that exactly fit’s into the car. But take a look at the pictures yourself.
As mentioned a few times, the last few month lots of effort has gone into quality instead of features. And it has payed off. ANRs (App not responsive) and crashes levels are at or better than the average app. It seems dongle crashes have gone down too. Given the massive diversity in devices (dongles and phones/tablets), we consider this a great improvement, though admittedly you, the individual user, probably hardly noticed. A non-crash is and should be indeed a non-event.
This process isn’t finished. First of all it is something that we continuously monitor, especially after new features or fixes. Secondly and as announced earlier, we are in the process of sunsetting unmaintained languages as well as Fluence/Kangoo support to further cleanup code.
With this major effort behind us, this slowly gives space for functional development again. At this moment there are three issues that need our attention:
- Resetting DTCs doesn’t work anymore (this seems to be a regression);
- Setting TPMS IDs seems either buggy or not working at all (see next paragraph);
- We need to make a version where all data is retrieved through ISO-TP commands. This requires lots of research but will enhance speed and stability on ELM type dongles a lot. For CanSee dongles, free frames are actually faster, so we try to use them as much as possible there.
Now features like TPMS are hard to implement and maintain. Being early adopters we don’t own cars with TPMS. It’s a bit hit and miss and politely asking for access to other people’s cars. And that issue is bound to get a lot worse soon: the new ZOE (nicknamed Gen 2) will be released in a few month. It will feature CCS, have a somewhat larger battery and I expect it will feature the same motor as the R110. Especially CCS will have significant impact on what we do with CanZE, notably of course in the charger sections.
Edit. A friendly dealer in the UK has tried CanZE on a brand new ZE50 model. No response whatsoever. QED
We have no access to Renault’s designs, nor the Clip tool, nor a car to play with. Speaking for myself, I have no intention of buying a Gen 2 anytime soon. My 2013 Q210, which has done over 100.000 km now is trotting along quite nicely, showing 97% battery SOH (!!) and I sincerely hope it will for at least another 5.5 years.
We’ll have to see how we can keep up. With almost 5000 CanZE users on the current models, there is no reason for any immediate measures.
Dutch Q210 driver “OlafH” just had his August 2014 ZOE pass 200.000 kilometers (about 125.000 miles). SOH 91%. Who would have thought.
- Growth slowing down to 2.4% last month
- ANRs have gone to zero!
- Crashes have gone down further, but a nasty regression slipped in just a few days ago. Put here is a 180 days graph for your perusal.
The problem is known and will be fixed soon. If you ignore that spike, you will see we went from hovering around 20 to around 3. Be aware that this graph runs across all app versions and there are always a few hundred un-updated versions out there.