WANTED: Beta testers for the next release of OpenRocket

[cfb_rolley]

Has anyone tried editing the shape of a freeform fin set using the new beta? If so please report your experience, and be sure to include what OS you're running.

In my own testing it's been extraordinarily slow, whereas in 22.02 it was fast and fluid. Bad on Mac but unusable on Windows.

I’m on windows 10, and yeah it’s a bit slow. I didn’t have issues with 22.02. That said I typically just type in the numbers rather than dragging the nodes, and dragging is where it’s extremely noticeable.

 

[lhelseth]

Neil,
On a 16 GB Mac Air (Ventura 13.2.1; Intel processor) with 23.0.beta.01 the Dark Mode layout is different from "Default" or "Light", with the three tabs (Rocket design, Motors & Configuration and Flight simulations) shifted to the top left instead of centered as in the other background modes. Also, I was able to edit the fins fine, though slightly more sluggish than with 22.02. 23.0.beta.01 rendered the freeform fins fine in Tools/Photo Studio and the Flight simulations updated accordingly.

 

[SolarYellow]

Does this beta have the dual reporting of stability factor in calibers and percent that was discussed several months ago?

 

[neil_w]

Yes, can be set in preferences to show both.

 

[rocketgeek101]

Can confirm the freeform fin editor is sluggish on my Mac (macOS Ventura 13.5.2) with the beta version of OR. With v22.02 it is nice and snappy.

 

[JoePfeiffer]

Seems fine under Linux

 

[SiboVG]

There was indeed a culprit for the sluggish shape editor. Pushed a fix to improve it.

 

[SiboVG]

Many programs offer multiple flavours of dark mode. Some with more contrast than others. Since the contrast could be tweaked according to @SiboVG I wonder about offering the user a second darker/higher contrast option in addition to the current dark mode.

Sure, we could add a high-contrast dark mode. Something like this?

 

[rocketgeek101]

Sure, we could add a high-contrast dark mode. Something like this?
View attachment 605739

Looks great!

 

[OzHybrid]

OBJ export of fin is outputting square fins when the aerofoil profile has been selected.

 

[OzHybrid]

OBJ export of bodytube and fin with 6mm fillet. Fin only making tangential contact with body at root. Gap for rest of length on sides. No fillet. No Aerofoil profile.

 

[rharshberger]

Hopefully going to have my new PC up and running this week and can try the new beta out on it, its a i9-13900k, 32gb DDR5 6000, RTX 4060Ti, with SSD.

 

[SiboVG]

OBJ export of fin is outputting square fins when the aerofoil profile has been selected.
View attachment 607420
View attachment 607422

Yes, it's simply not yet supported right now. Feature request for a future release.

 

[SolarYellow]

There are some good fin can generators for free design software out there that will do that pretty easily using the same basic inputs applied in OR. A very low priority IMO.

 

[Buckeye]

There are some good fin can generators for free design software out there that will do that pretty easily using the same basic inputs applied in OR. A very low priority IMO.

I agree. OR should not be a full-featured CAD software (or graphic artist software.) There are better tools available to do those things. With the limited resources, I say stick to the core business: model rocket parametric design and flight simulation.

However, I did read somewhere that RockSim is putting correct airfoils on the fins for rendering and CAD export.

 

[SolarYellow]

... RockSim is putting correct airfoils on the fins for rendering and CAD export.

What is "correct?" Which airfoil profile is chosen? If beveled, is it flat with beveled LE/TE, or just a diamond? What bevel angle. Does it taper in thickness from root to chord? All this can be determined by the user in a fin can generator.

It would be nice to render, and maybe even output in 3D, a simple semicircular radius on fin LE/TE (as well as tip and corners) if the user specifies. And fillets would be cool. But as I said, not a priority.

Like Buckeye, I use OR more as a performance prediction tool than as a cosmetic design tool. I really appreciate @K'Tesh's work on the aesthetic side, but it's not what I care about most. I can mind sim aesthetics a whole lot more effectively than I can performance.

 

[Buckeye]

What is "correct?" Which airfoil profile is chosen? If beveled, is it flat with beveled LE/TE, or just a diamond? What bevel angle. Does it taper in thickness from root to chord? All this can be determined by the user in a fin can generator.

Who knows? It was a blurb from TVM. Probably just something that "looks" like the airfoil selected from the drop down box. If you want fancy stuff, then a CAD software or the fin can generator you mentioned is the way to go.

 

[SolarYellow]

Yeah, I was just spitballing to illustrate the complexity of doing anything more than the most simplistic cosmetic result.

 

[neil_w]

I confess I would like to see more accurate rendering in the 3D views, but with freeform fins it can be quite difficult to even interpret how to apply the shapes. Trapezoidal fins would be easiest, but generating the mesh is still an interesting math problem. @SiboVG can correct me if I'm wrong, but I believe that the 3D rendering and the OBJ output go hand-in-hand; solve one solve both.

But to really control the fin profile properly would require a bunch of extra parameters for the fins, which would have little or no impact on simulation, just visuals and 3D export. And so it is a low priority item.

But as always, if anyone wants to jump in and help us implement it....

 

[JoePfeiffer]

I confess I would like to see more accurate rendering in the 3D views, but with freeform fins it can be quite difficult to even interpret how to apply the shapes. Trapezoidal fins would be easiest, but generating the mesh is still an interesting math problem. @SiboVG can correct me if I'm wrong, but I believe that the 3D rendering and the OBJ output go hand-in-hand; solve one solve both.

But to really control the fin profile properly would require a bunch of extra parameters for the fins, which would have little or no impact on simulation, just visuals and 3D export. And so it is a low priority item.

But as always, if anyone wants to jump in and help us implement it....

As long as our fin airfoil models are as crude as they are, I don't want us to be rendering or exporting them in a way that gives the user the illusion they're better than that.

 

[SolarYellow]

Yes, I would rather see the team focus on improving the simulation than on cosmetics.

 

[cwbullet]

I am pretty darn impressed so far. I am not seeing any major bugs.

 

[neil_w]

As long as our fin airfoil models are as crude as they are, I don't want us to be rendering or exporting them in a way that gives the user the illusion they're better than that.

Fair enough.

 

[OzHybrid]

Yes, it's simply not yet supported right now. Feature request for a future release.

https://www.rocketryforum.com/threa...ext-release-of-openrocket.181945/post-2481157

I would love to help with #2. I do a lot of printing and would love to be able to export nose cones and tailcones specifically.

You'll be able to export everything, including entire rockets.

But Neil Promised..........

 

[neil_w]

https://www.rocketryforum.com/threa...ext-release-of-openrocket.181945/post-2481157
You'll be able to export everything, including entire rockets.

But Neil Promised..........

Depends on your definition of "everything", I suppose.

 

[OzHybrid]

As long as our fin airfoil models are as crude as they are, I don't want us to be rendering or exporting them in a way that gives the user the illusion they're better than that.

It's still useful for loading into an existing fin can generator for a shape comparison between the fin can generator and the OR export. Even if it's just producing the basic shape. At least there is a fin can object for comparison rather than having to use a fin co-ordinate export for a complex fin.


To get the fin can to centre properly, had to put a ring around. Looks like the OR export of a 3 fin can is using the "square" projection of the fin can to calculate the centre. A 4 fin can should be ok. So might be something to look at for exporting. Without a proper centre it's painful to fix. This was an export of the OBJ fincan file. Imported into Cura. In Cura the centre was out. Put a transition and minimal ring on in OR reexported. Opened in Cura. OK, but transition dissapeared. Probably too thin. Centred. Exported as STL. Opened in SCAD. Gave it a colour, overlayed it with my fincan generator.

 

[SiboVG]

I confess I would like to see more accurate rendering in the 3D views, but with freeform fins it can be quite difficult to even interpret how to apply the shapes. Trapezoidal fins would be easiest, but generating the mesh is still an interesting math problem. @SiboVG can correct me if I'm wrong, but I believe that the 3D rendering and the OBJ output go hand-in-hand; solve one solve both.

Well... Not entirely. For some of the custom, more complex 3D renders such as nose cones, transitions, and fins, the OBJ exporting takes inspiration from the 3D rendering code. But, the 3D rendering uses special (really outdated) OpenGL helper functions and is in general just really embedded into the OpenGL environment. So, in practice the OBJ exporting code is basically written from scratch. Another testament of why it took so long to complete this feature...

The way 3D geometries are created in the OBJ exporting code is more or less the standard in low-level 3D engines (defining custom vertex, normals, and texture arrays). So, if at any point we would like to switch to a different 3D engine, we could reuse some of the OBJ exporting code.

 

[NTP2]

OK folk, have at it.

Beta installers are here: https://drive.google.com/drive/folders/139X6fEj9Z58LEuvfxeIKgXWQUMGw6S2V.
Full release notes are here: https://github.com/openrocket/openrocket/blob/unstable/ReleaseNotes.md

Report feedback here, or via PM, on the beta channel on Discord, or file a Github issue. If you're already part of that Discord channel, please direct your communications there.

Primary focus is on three new features:
1) Export OBJ file for 3D printing or CAD import. Select one or more objects and do File -> Export -> Export as Wavefront OBJ.
2) Export / Import .CDX1 files for RASAero
3) Dark mode and settable font size for UI (look in Settings)

Of course, we welcome feedback in any other part of the program as well.

If you report an issue, please provide sufficient detail for us to be able to reproduce it. Let us know which operating system you're running, what you were doing when you encountered the problem, and is appropriate include the file that triggered the issue.

Thanks in advance!

Ok I see now thanks.

 

[lakeroadster]

I just downloaded 23.09. Multiple bugs:

Doesn't show Optimum Delay in Flight Simulations​

Doesn't show Apogee, Max. Velocity or Max. Acceleration in pictorial representation view​

3D Figure, 3D Unfinished and 3D Finished views don't work​

I'm switching back to 22.02

 

[NTP2]

I just downloaded 23.09. Multiple bugs:

Doesn't show Optimum Delay in Flight Simulations​

Doesn't show Apogee, Max. Velocity or Max. Acceleration in pictorial representation view​

3D Figure, 3D Unfinished and 3D Finished views don't work​

I'm switching back to 22.02

Not for me, have you submitted a bug report?