We are very excited to hit yet another amazing milestone and are opening the door to Meshtastic 1.3 to the public for the first time!
This first release, we ask that you don’t change any of the values that we don’t expose. We want to know how it behaves with the defaults we have selected. In fact, we’re not publishing a compatible version of Mesthastic Python 1.3 to prevent tweaking of the default values.
While we are making this available, please consider this only if you are an edge of your seat trail blazer.
As part of this preview, we’re releasing 4 products:
Meshtastic Device
Meshtastic Android
Meshtastic iOS
Meshtastic MacOS — Yes! This is new!
You probably can find a list of what has changed but for right now, please put yourself in the position of someone using this for the first time. What has changed isn’t important. What’s important is that the basic functionality works. Let us know how it behaves and any issues you may find.
If you’ve read this far and you are and edge-of-your-seat person, you will find 1.3 builds of our software at:
The information on how to install the software has been purposely left out. You will want to use the install scripts in the device archive, side load the Android app direct from GitHub, and use TestFlight for the MacOS and iOS builds.
If you want an easier install, this public preview isn’t for you. Hang tight, we’ll be there very soon!
Something to note – While our intention was to make the first version of 1.3 stable from the perspective of the back end API and algorithms, we are tracking one upcoming change that will not be backward compatible. We expect that change to come in about 4 weeks. When that happens, all devices will need to upgrade to that version.
Please don’t use Meshtastic 1.3 on anything “production”. This is only a technology preview.
No risk - no fun
Just installed on TBeam and TTGO lora 2.1-1.6. They found each other and after installing 1.3 version of Android-app I was able to connect to them both. So far so good.
I am an ‘edge of your seat trail blazer’, in that I only have these Meshtastic devices to muck about with and test new firmware and features (IE I have no practical purpose for them, and they’re only powered up from time to time to try new firmware and features). Isn’t this the case, to a greater or lesser extent, for the majority of Meshtastic users in truth?
But flashing new firmware is so much easier with the Windows Flasher tool. It’s so frustrating that 1.3.16 has been set to ‘populate in dropdown? False’.
Sorry mc-hamster. I didn’t intend to call into question your, and your colleagues dedication, or be an annoying whinger!
I was more frustrated because I perceived that the Windows Flasher had been purposely excluded, on the basis that ‘If you need to use the Flasher, you’re not worthy of testing this new firmware’.
Evidently, my perceptions were somewhat incorrect and it’s more of a time issue.
I was going to suggest that the Flasher tool could be run with a switch to enable Alphas to appear in the dropdown/ be able to flash them. Or, a button in the UI which says ‘Include Alpha versions - at your own risk’.
Flasher is built on top of the python api and the python api has yet to catch up. We put out a call for more volunteers for the python side of the house and last I checked, they’re getting familiar with the new world
I think another way of looking at this preview, is that we decided to get early feedback rather than waiting until absolutely everything is ready.
There are some developers waiting for that feedback and we figured it is more important to get keep them going strong rather than holding everything else back.
“After a few attempts” isn’t the experience we’re looking for. What happened during those attempts? Why did you need to reattempt? What phone model and version of Android OS do you have?
Also disconnecting from meshtasic app
showed long fast channel and QR code initially but then no more
shows no channel, only the share link works
Needed to restart bluetooth to get reconnected
no gps lock even after after 2 hours
T-Beam unit2
Critical fault #7
Meshtastic 1.3.16.97899ae
T-Beam 1.1
Found unit1 and unit1 found unit2 right after unit2’s install reboot!
but, no gps lock so no direction displayed
Critical error #7 when connecting bluetooth
Bluetooth disconnect
Samsung S7 edge android
Here’s the main list of changes we were tracking. There’s also hundreds of others, but these are the big items:
With the basic interface, the first thing you should see is the default channel is has twice the capacity as the old default and with packet level compression, sending text messages should actually be about 4x faster than before.
Unfortunately I don’t have a working android device for testing… So I haven’t even touched any options; everything has been just as is after flashing.
I am 7 installs down now. 4 different versions of tbeams and a techo. No issues and initial discovery seemed to be rather snappy. I have intentionally just left these around the house not moving while I tested messages and everything.
One thing I noticed is how quickly info packets are sent. I don’t think I ever saw a node without a position update longer than 3 minutes. It is pretty much always in the ‘seconds’ display for me which is truly amazing. My mesh is regularly hitting 25% channel utilization though which seems to be about the same or better than local meshes of about the same size before; I plan to test this much more.
I have been experiencing significantly more ‘reboot’ crashes which I was expecting but I noticed there were less instances where the device locked up, the reboot seems to happen much more reliably when there is something wrong which is awesome. On the device I would connect to through Bluetooth, I experienced a few reboots while connecting but nothing completely repeatable for now. So far the overall iOS app connection has been no less stable than before.
Message sending experience is very snappy and much much much more reliable in my experience. With high traffic there are defiantly some dropped messages and especially dropped ack packets. I think this really highlights the opportunity in foregrounding more of the channel utilization type information into the messaging view, let a user know when maybe its best to wait a second for their message to ensure a more reliable receipt. This is a much better user experience than with the old longslow, compression is doing its trick. I look forward to testing more of the range characteristics of the new channel this weekend.
Overall the mesh setup experience has been top notch and the discovery seems to be like a product new out of the box. I know some might not like the newer screen on time but I am really liking it while testing. I think it’s a nice compromise time and I don’t feel the need to just set “is always powered” on everything.
I also have been using some new 18650 cells and battery life has been over 24h so far. I don’t know these cells well yet so I can’t comment on how this version compares before.
but I noticed there were less instances where the device locked up, the reboot seems to happen much more reliably when there is something wrong which is awesome. On the device I would connect to through Bluetooth, I experienced a few reboots while connecting but nothing completely repeatable for now. So far the overall iOS app connection has been no less stable than before.