ANNOUNCEMENT: The OpenRocket 22.02 Beta Period is now finished

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I'm probably a bit late in this request, but I've found it a bit awkward to specify body tube sizes - specifically motor tubes, but the problem can appear elsewhere:

1. Add (or modify existing) internal tube.
2. There are (at least for me..) some default values. I change the units to mm for the internal dimension.
3. Specify a motor tube internal size - say 24mm
4. As it can be awkward to accurately measure O.D. I like to measure the thickness. Click on "wall thickness". Oops. The internal dimension goes back to whatever default was there. Enter the wall thickness.
5. The internal size now changes again. Re-enter the internal size. But it won't "stick".
Aargh.

It seems to want the external dimension, which is usually the one that I don't have. So I have to calculate the O.D. manually. Which seems odd that it won't self calculate. And after a few beers, this requires a calculator, as my math abilities start slipping.

It's not a killer, just a bit awkward.

Hans.
So you want some sort of smart behavior where the thickness setting checks whether the last radius edit was with the outer radius or the inner radius? If the last edit was the inner radius, setting the wall thickness will keep the inner radius constant, and change the outer radius. And vice versa for the outer radius.

Something like this?

View attachment Screen Recording 2023-01-10 at 02.30.43.mov
 
So you want some sort of smart behavior where the thickness setting checks whether the last radius edit was with the outer radius or the inner radius? If the last edit was the inner radius, setting the wall thickness will keep the inner radius constant, and change the outer radius. And vice versa for the outer radius.

Something like this?

View attachment 556187
Interesting. That's way cool.

What I keep thinking about is how you enter the data in an HP financial calculator. There are 5 interrelated variables for compound interest, all you have to do is enter any 3. It calculates the other 2.

Hans.
 
I'm finally trying the new beta and not on my MacBook Pro but on a new Education Series Lenovo Thinkpad 11e running Windows 10 that I got from the Lenovo Black Friday sales. It's going to be my new take-it-to-the-field computer and so far has been quite satisfactory and a big step up from the 11 inch Ideapad it has replaced.

Anyway, today I wound up pulling up a model I probably did ten years ago to tweak and all seems well except for one irritating thing that OR maybe has always done and maybe this is intended:

When one defines a new configuration (get another motor to try) all the existing ones are flagged as out of date, even though they've not been touched. So....they all have to be rerun to get rid of that shiny red ball on the left in the list. If this is normal, please, someone, tell me why. I can see if I change something in the design (changed the 'chute, so had to override the mass) that all the existing simulation runs would be invalidated. But why, when all I did was create a new config and select a motor for it are all the others turned red?

As I say, this may well have been the way it was "since the beginning" as I haven't done the same thing in 15.03 recently enough to remember, but it still bugs me.

And apologies to all of the folks who have been doing an amazing job working on this, if this has been asked and answered and I just hadn't searched well enough to find it yet.

Just in case it matters, here is the .ork file I've been tweaking today.
 

Attachments

  • SkyTrax_RTF_twoA2.ork
    343.2 KB · Views: 0
Not sure how much a stress tester I am, but I have some pretty complex models. I'll try to load up the best/worst of them and poke around a bit.
I've poked around some of my more complex designs (including the infamous shrouded cluster which started as an unshrouded cluster, which tickled out NaN errors in the past) in both x86-64 Linux and M1 Arm MacOS and haven't seen anything unusual. Bear in mind both that I build from source and have done since Q3 of '21 or so so am unlikely to notice any deeply-rooted bugs (slowly-boiled frog theory) and also that I mainly look for reasonableness in sims and am less concerned with some of the minutia of display, interface, etc. (I've been an open-source zealot for nearly thirty years and have grown so accustomed to minor fit-and-finish glitches that I probably don't even notice them most of the time.)

Disclaimer aside, it looks great to me, and compared to the last release version, y'all have a home run on your hands! Kudos to each and every one of the developers. You're all giving an amazing gift to the rocketry community.
 
I'm finally trying the new beta and not on my MacBook Pro but on a new Education Series Lenovo Thinkpad 11e running Windows 10 that I got from the Lenovo Black Friday sales. It's going to be my new take-it-to-the-field computer and so far has been quite satisfactory and a big step up from the 11 inch Ideapad it has replaced.

Anyway, today I wound up pulling up a model I probably did ten years ago to tweak and all seems well except for one irritating thing that OR maybe has always done and maybe this is intended:

When one defines a new configuration (get another motor to try) all the existing ones are flagged as out of date, even though they've not been touched. So....they all have to be rerun to get rid of that shiny red ball on the left in the list. If this is normal, please, someone, tell me why. I can see if I change something in the design (changed the 'chute, so had to override the mass) that all the existing simulation runs would be invalidated. But why, when all I did was create a new config and select a motor for it are all the others turned red?

As I say, this may well have been the way it was "since the beginning" as I haven't done the same thing in 15.03 recently enough to remember, but it still bugs me.

And apologies to all of the folks who have been doing an amazing job working on this, if this has been asked and answered and I just hadn't searched well enough to find it yet.

Just in case it matters, here is the .ork file I've been tweaking today.
This is not intended behavior, just something that slipped under the radar I guess. I'll take a look at it, thanks for the report!
 
And thank you for the kind comments, this is what drives us to improve the program! Contrary to what some people appear to think, complaining in a rude way of everything that's wrong with the (free) software, maintained purely by people's good will, demotivates us. Constructive feedback, with a small 'thank you' goes a long way.
 
@SiboVG and all the others who are working on OR: Thank you again!

I just pulled up 15.03 here on my Mac, loaded a model and added a new configuration. That did not set all the prior ones to red...though there was a brief moment where it looked as if it might — as if for a fraction of a second they were all turned red and then to green. Is it possible it's just whatever process that sets that flag in 22.02beta5 is just jumping to the wrong spot?

(it has been a VERY long time since I've written any code so that may make no sense in the context of what the OR code really looks like).
 
In my beta.05, it just reruns all the sims automatically whenever I change over to the sim tab. They will be red for a second or two, then turn green once rerun.
I think because you have the "Run out-dated simulation" checkbox in the preferences dialog checked? @BEC do you have this checkbox checked or unchecked?

1673813502060.png
 
@BEC do you have this checkbox checked or unchecked?

Let me look at 22.02 on the Lenovo and see.....

OK. I did NOT have that checked on 22.02 (though I do now) and DO have it checked on 15.03 on my Mac.

That said, my original question really still stands. Why are all those other simulations even flagged as "outdated"? All I did was add another configuration to the list, so I could select another motor.

But the checkbox, which is a workaround (in my mind anyway), will suffice for now I think.
 
Let me look at 22.02 on the Lenovo and see.....

OK. I did NOT have that checked on 22.02 (though I do now) and DO have it checked on 15.03 on my Mac.

That said, my original question really still stands. Why are all those other simulations even flagged as "outdated"? All I did was add another configuration to the list, so I could select another motor.

But the checkbox, which is a workaround (in my mind anyway), will suffice for now I think.
You're right that it's not supposed to do that, but I'm working on a fix for it right now.
 
I'm going to duck for cover to avoid the rocks that will likely head my way, but I'm going to ask anyway.

It's my understanding that OR is written in Java, and thus should be more or less platform independent. However I think I read that certain packages, like probably graphics, are specific to different platforms . Is that anywhere close to accurate?

So.... Perhaps there could be an OR "lite" that could run on Android. Strip out some of the design features, just have the ability to run sims in the field if motor choices or launch conditions change. Currently, I bring my Chromebook to launches for this purpose. I only need it occasionally, but it's nice to have when I actually do need it.

Yes, I know there are some generic sim programs like Rocket Reviews or Thrust Curve. However it would be nice to be able to sim the actual design and account for things like wind.

Hans.
 
There used to be an Android version, it eventually fell into disrepair and was discontinued.

The simple truth is this: right now, there are two people doing almost all of the coding for OR. There is simply no bandwidth for creating and maintaining an Android version right now.

If some enterprising Android developer wants to jump in and give it a shot, they would be absolutely welcome to do so.
 
There used to be an Android version, it eventually fell into disrepair and was discontinued.

The simple truth is this: right now, there are two people doing almost all of the coding for OR. There is simply no bandwidth for creating and maintaining an Android version right now.

If some enterprising Android developer wants to jump in and give it a shot, they would be absolutely welcome to do so.
If I were younger (meaning more ambitious) and my coding skills weren't so rusty, I'd attempt digging into it.

Thanks anyway. My newest Chromebook is quite small and works well for this purpose. So good 'nuff.

Hans
 
It's my understanding that OR is written in Java, and thus should be more or less platform independent. However I think I read that certain packages, like probably graphics, are specific to different platforms . Is that anywhere close to accurate?
The language itself is the same, but the framework surrounding it is not. In particular the parts the user interacts with are quite different. An OpenRocket for Android could easily reuse some parts from under the hood, like the simulation models, but the GUI would have to be rewritten.

While the framework is being kept more or less compatible between Windows, Mac and Linux, the requirements for a smartphone are so different (small touch screen, aggressive power management, etc..) that Google threw this out of the window and came up with it's own solution.

Reinhard
 
I have written in Java for the past 20+. Virtually ALL server side backends and the for past 4 years, all on Kubernetes and AWS. Now I write in not-so-good Kotlin, also a JVM language and wonderful!
To the OR devs, I might have some bandwidth this next year to help out. But not with Android. Everything I have ever dealt with that was 'Android' was terrible compared to the iOS version (assuming there was an iOS version). Lest ye think I am an iOS lover, while it's true I love my iOS devices, I am loath to deal with coding for iOS after a few years doing that and dealing with Apple's greedy little hands and certification process.

But maybe I can help out in normal OR. I'll have to check out the code and see. There are a lot of UI things I'd love to change (like the entire choose an engine and launch UI). Have to see...
 
I have written in Java for the past 20+. Virtually ALL server side backends and the for past 4 years, all on Kubernetes and AWS. Now I write in not-so-good Kotlin, also a JVM language and wonderful!
To the OR devs, I might have some bandwidth this next year to help out. But not with Android. Everything I have ever dealt with that was 'Android' was terrible compared to the iOS version (assuming there was an iOS version). Lest ye think I am an iOS lover, while it's true I love my iOS devices, I am loath to deal with coding for iOS after a few years doing that and dealing with Apple's greedy little hands and certification process.

But maybe I can help out in normal OR. I'll have to check out the code and see. There are a lot of UI things I'd love to change (like the entire choose an engine and launch UI). Have to see...
We'd love that! The Android part is very low priority, there's still a bunch left to be done in "normal OR". Hope to see you soon.
 
A minimal UI that took an .ork file and could sim it on various motors would be useful for mobile, but beyond that, you'd need a tablet format more than a phone format to make design, etc. useful.
That's kind of what I had in mind when I suggested it. Graphics and design features could be minimal or non-existent. I'd just like the ability to run sims on existing .ork files.

But, as I mentioned, my 11" Chromebook is not that inconvenient to take to launches. And I use my 15" Chromebook to do any actual design work.

Hans.
 
It's my understanding that OR is written in Java, and thus should be more or less platform independent. However I think I read that certain packages, like probably graphics, are specific to different platforms . Is that anywhere close to accurate?

Not anywhere close. Android isn't Java, its very close, but its not the same Virtual Machine and the UI framework doesn't really either work well or even work on Androids - it wasn't designed for it, and Android did not take into account because it has its own UI complete specific.

Download Android Studio and here is the link to the GitHub repo for the Core (https://github.com/openrocket/openrocket/tree/unstable/core) and see if you can get it to compile in an Android application. Here's my https://github.com/thzero/openrocket.gradle for building OpenRocket in gradle, which is important since Android really relies on Gradle to do builds and not the Ant that OR does - this is out of sync with latest beta changes and haven't had change to do any testing.

It all depends on the resolution of the laptop/tablet and what you can stand. I use OR on my Surface Pro 3 (now running Ubuntu) and its a small 12" screen but I also run at a high resolution so yes everything is small, but it works.
 
But maybe I can help out in normal OR. I'll have to check out the code and see. There are a lot of UI things I'd love to change (like the entire choose an engine and launch UI). Have to see...
Get rid of the Jakarta libraries for loading/saving XML!
 
Cannot launch system graphics editor
The editor used can be modified in the Preferences dialog.

1674295638841.png

1674295800392.png

1674295895547.png

1674296259886.png

HELP!!!!
 
Cannot launch system graphics editor
The editor used can be modified in the Preferences dialog.
HELP!!!!

Your issue has been reported as #1991.

In the meantime, until it is fixed, you must select the command line option and enter the name of (or navigate to) your graphics program (e.g., Windows is "mspaint").
 
Last edited:
Cannot launch system graphics editor
The editor used can be modified in the Preferences dialog.

View attachment 558499

View attachment 558500

View attachment 558501

View attachment 558502

HELP!!!!

You configured Windows to use a link (.lnk) file on your desktop as default image editor. The correct file is the executable (.exe). As far as I can tell, Windows 10 provides no graphical way to edit this, but the registry editor can be used.
Navigate to the key
Computer\HKEY_CLASSES_ROOT\SystemFileAssociations\image\shell\edit\command
and change its value to (including the quotes). This assumes you've installed paint.net in the default location
"%programfiles%\paint.net\paintdotnet.exe" "%1"

However, this leads to the next issue. Clicking the "Edit" button when no texture has been defined yet will create a temporary .png file for the texture. OR simply creates a completely empty file (0 Bytes in size) with a .png extension. This is not a valid file as far as paint.net is concerned. Microsoft Paint on the other hand will simply treat that as a new file but when it is saved that will result in a Windows Bitmap (.bmp) but with a .png extension.

So in a nutshell: OR can not handle .lnk files and OR will generate invalid temporary .png files with unpredictable results depending on the image editor.

Tested version 22.02.beta.05 with Windows 10 an mspaint and paint.net

Edit: What is the intended behavior when the "Edit" button is clicked while the "Use default" checkbox is selected? I tried to make it work manually, but was not successful (texture was not loaded after saving) and now I'm just wondering wether the "Edit" button should be grayed out in this state.

Reinhard
 
Last edited:
Back
Top