Sad state of Rocksim and Openrock

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

gdjsky01

Kim's Rocketeer
TRF Supporter
Joined
Jan 20, 2011
Messages
6,385
Reaction score
2,062
I can crash them both within a few minutes. Especially ROCKSIM which is, IMO, unforgivable given I have paid for it since v5. I guess I just love misery. Literally I was designing a 2.6 inch based RedTop missile. In Rocksim I could not finish without a crash. Both on the Mac and Windows 10. Having coded for over 35 years, from 8080 assembler, C, C++, Java, and now Kotlin, I simply don't understand how Apogee can live with themselves. ROCKSIM is so easy to crash. Just do a few cut/pastes or deletes. Hell I can not get EITHER version (Mac or PC) to create a short 2.6 inch elliptical nose cone without crashing.

Openrock does not crash as in "go bye bye"! Yea! It just is still the totally counterintuitive simulation UI as well as more bugs than a Louisiana Bayou (especially rendering)! Wow. What a sad sad state of affairs. Well I guess the community is small. So there is no real business proposition to make it better. I just though by v10 of RS, maybe stability would have been baked in.

I expect OR to be a fail since its free and coded by a 100 people or more. To be clear, it is actually its more stable than RS. But its simulation UI is still, years later, crap. And as I said, it's rendering has more bugs than a subtropical rain forest. Just ranting... I thought things would be better.

CHEERS!
 
The beauty of open source software is that you can take your 35 years of software development experience and use it to make Open Rocket better. It is written in Java, so it should be right up your alley!

I haven't personally encountered any rendering bugs in OR, but that might be specific to some OSs/graphics hardware. I agree that the plotting interface isn't super intuitive, but again, if this is a program that you get value from and have the skills to contribute to, what's stopping you from making the changes you want? I'm sure a lot of people would love it if the plotting interface was more intuitive.
 
I expect OR to be a fail since its free and coded by a 100 people or more. To be clear, it is actually its more stable than RS. But its simulation UI is still, years later, crap. And as I said, it's rendering has more bugs than a subtropical rain forest. Just ranting... I thought things would be better.
1) If by rendering you're referring to 3D rendering, then yes we know about this. We are in the process of switching 3D libraries to remedy this but it's going to take a little time.
2) If you have specific complaints and/or recommendations regarding the simulation UI, then please by all means share them with us. Just saying "it's crap" doesn't really give us much to work with.
3) If there are indeed actual crashes (as you stated in your first sentence) then please report them. We have eliminated a lot of them over the course of the beta period.
 
Also: if you aren't already, use the latest beta version of OpenRocket, it has a better UI for the simulations (not the barf-inducing grey background) and (hopefully) less crashes and bugs.

But as Neil stated: an open-source program can only become better if you report bugs and possible improvements.

Also: your estimate of how many people actively develop OR is overshot by a factor of about 10.
 
I can crash them both within a few minutes. Especially ROCKSIM which is, IMO, unforgivable given I have paid for it since v5. I guess I just love misery. Literally I was designing a 2.6 inch based RedTop missile. In Rocksim I could not finish without a crash. Both on the Mac and Windows 10. Having coded for over 35 years, from 8080 assembler, C, C++, Java, and now Kotlin, I simply don't understand how Apogee can live with themselves. ROCKSIM is so easy to crash. Just do a few cut/pastes or deletes. Hell I can not get EITHER version (Mac or PC) to create a short 2.6 inch elliptical nose cone without crashing.

Openrock does not crash as in "go bye bye"! Yea! It just is still the totally counterintuitive simulation UI as well as more bugs than a Louisiana Bayou (especially rendering)! Wow. What a sad sad state of affairs. Well I guess the community is small. So there is no real business proposition to make it better. I just though by v10 of RS, maybe stability would have been baked in.

I expect OR to be a fail since its free and coded by a 100 people or more. To be clear, it is actually its more stable than RS. But its simulation UI is still, years later, crap. And as I said, it's rendering has more bugs than a subtropical rain forest. Just ranting... I thought things would be better.

CHEERS!

Opinions Vary. I know nothing about Rocksim, but Open Rocket works great on my ideapad Laptop.

Few things in life are free.
And even fewer free things are as great as Open Rocket.
And even fewer free things have folks available, literally right at your fingertips, to help with any issues that may arise.

Time to count your blessings.

"You can please some of the people all of the time, you can please all of the people some of the time, but you can't please all of the people all of the time"." ― John Lydgate
 
I can crash them both within a few minutes. Especially ROCKSIM which is, IMO, unforgivable given I have paid for it since v5. I guess I just love misery. Literally I was designing a 2.6 inch based RedTop missile. In Rocksim I could not finish without a crash. Both on the Mac and Windows 10. Having coded for over 35 years, from 8080 assembler, C, C++, Java, and now Kotlin, I simply don't understand how Apogee can live with themselves. ROCKSIM is so easy to crash. Just do a few cut/pastes or deletes. Hell I can not get EITHER version (Mac or PC) to create a short 2.6 inch elliptical nose cone without crashing.

Openrock does not crash as in "go bye bye"! Yea! It just is still the totally counterintuitive simulation UI as well as more bugs than a Louisiana Bayou (especially rendering)! Wow. What a sad sad state of affairs. Well I guess the community is small. So there is no real business proposition to make it better. I just though by v10 of RS, maybe stability would have been baked in.

I expect OR to be a fail since its free and coded by a 100 people or more. To be clear, it is actually its more stable than RS. But its simulation UI is still, years later, crap. And as I said, it's rendering has more bugs than a subtropical rain forest. Just ranting... I thought things would be better.

CHEERS!
You quote about OR being counterintuitive is spot on
 
You quote about OR being counterintuitive is spot on
Its totally ridiculous you have to create a configuration before you can 'launch it'. It's been like that from almost day one. WTH are they thinking? I just what to load an engine, set the delay (or any recovery event(s) ) and launch. WTH would I want this to be a two step process? Half the time the engine UI does not update leaving me with a 'no-engine' configuration. OR is now on what? Its 10th, 20th, release? This is your QA? I already know RS has zero QA.

As for open source and writing myself, if I HAD time for that, I'd have done it. That does not excuse rocksim which I can crash on any PC or Mac, and have for years, in only a few minutes. All you have to do is a few copy pastes, abort a couple of part additions, delete one or two and it goes bye bye. But OR? I'd have expected the community to take more pride.

I knew writing this would get me hate. But it needs to be said for those that might give Rocksim authors money. Don't bother. Open Rocket is a terrible UI but you can actually get through a scratch build without crashing. (BUT SAVE AFTER EVERY STEP).
 
Its totally ridiculous you have to create a configuration before you can 'launch it'. It's been like that from almost day one. WTH are they thinking? I just what to load an engine, set the delay (or any recovery event(s) ) and launch. WTH would I want this to be a two step process? Half the time the engine UI does not update leaving me with a 'no-engine' configuration. OR is now on what? Its 10th, 20th, release? This is your QA? I already know RS has zero QA.

As for open source and writing myself, if I HAD time for that, I'd have done it. That does not excuse rocksim which I can crash on any PC or Mac, and have for years, in only a few minutes. All you have to do is a few copy pastes, abort a couple of part additions, delete one or two and it goes bye bye. But OR? I'd have expected the community to take more pride.

I knew writing this would get me hate. But it needs to be said for those that might give Rocksim authors money. Don't bother. Open Rocket is a terrible UI but you can actually get through a scratch build without crashing. (BUT SAVE AFTER EVERY STEP).
First of all OR release numbers are THE YEAR that version was released not consecutive iterations so the current v22 (2022)is not 7 versions after v15 (2015) its the 2nd one as there were 7 years between releases, early versions of OR did use a traditional method of tracking iterations. The number of people working on OR is not a 100 (though over time it maybe), but probably more like 10-15. There have been less than 10 full versions but a total of 37 releases (mostly patches) versions of OR and each one has added features and improved dramatically. It works, sometimes its buggy but, it works.
 
Not to be argumentative, but I've read through your posts and the closest you get to actually describing an issue is this:

Its totally ridiculous you have to create a configuration before you can 'launch it'. It's been like that from almost day one. WTH are they thinking? I just what to load an engine, set the delay (or any recovery event(s) ) and launch. WTH would I want this to be a two step process? Half the time the engine UI does not update leaving me with a 'no-engine' configuration. OR is now on what? Its 10th, 20th, release? This is your QA? I already know RS has zero QA.

The current version of OpenRocket is just that simple. After your rocket design is complete, go to the Motors & Configurations tab and pick your motor, verify the delay, and get the results.

Attack.Rev_01.png

If you believe that this can be done in fewer steps, I'd be very interested in knowing how you would approach doing so.

Complaints without suggestions for improvements are not helpful to anyone.
 
Great, write a couple of complete sentences and put them in a bug report!

I only build supercomputers for a living, so it's possible I'm just used to things being unnecessarily cantankerous.

In all fairness, he didn't assert you weren't competent based on your professional experience, it is that he is more competent as he has written code for more computers than the number of cars you and your parents have owned.

You and I have never met, so I'm not sure of the history of you and your family's automotive procurement frequency and/or history. Based on my and only my history, I believe the answer would be approximately 20-25, but in all fairness, I was born at a very young age and don't necessarily remember the first few cars my parents had. Thankfully we always purchased and never leased, so it keeps that calculation simple and its VERY good that siblings weren't included, as my brother has probably bought and sold more than a hundred cars and I have no idea if he could even figure out how many over the years. This year is more than 10 so far, so the more than a hundred number is likely low.

Either way, it appears that he has written code for 20-25 computers if my history is similar to yours. A super quick, non-verified google search indicates that Americans buy between 9.4 and 13 vehicles over their lifetime on average, so that would be 28-39-ish for you and your parents, therefore code has been written for 28-39 computers based on my interpretation of his statement. (In all fairness, I assumed he was talking about the cars that you and your parents have owned. . . it is possible that he meant cars you've owned + the number of parents you have or he has written code for your parents, possibly).

Hopefully some of the above is not taken seriously and is at least mildly humorous. I've had zero experience with RockSIM, but used wRASP back in the day and think OpenRocket is awesome for what you pay and how responsive customer support is. I don't use it for design (sketch to CAD to swing test most of the time, but I do use it to help pick delays and estimate altitudes at times if I'm being serious or trying to achieve a particular goal).

Sandy.

[edit 12:26am 06282022: Added more so there was at least a little 'content' even if not useful.]
 
Last edited:
I too have been a Rocksim user forever. The current release is reasonably stable. The immediately preceding release was unstable. It still crashes occasionally when I try to use some of the more complex features, but it's manageable. I'm old school - save often. I suspect most of the issues are memory related. It's old code, so it is what it is.

Now, EngEdit crashes a lot, definitely memory related. And updating the motor files within Rocksim itself is archaic as all get out. Without Thrustcurve, managing motor files would be a nightmare.
 
You have know idea What you are talking about. Thanks for the waste of bandwidth. I know better. I've written code for more computers than cars you've owned. And your parents.

Oh we're all very impressed.

You've admitted that simple applications crash on your computer.

You have an OS hygiene problem.
 
Last edited:
You have know idea What you are talking about. Thanks for the waste of bandwidth. I know better. I've written code for more computers than cars you've owned. And your parents.
That’s not likely to be a very high bar. Code can be transferred widely between machines for minimal difficulty and cost. Purchasing a vehicle is a significant financial decision in people’s live, and often a milestone.

For what it’s worth, I had stability issues with the trial version of RockSim but the legacy versions of OR operate just fine on my machine. It’s nothing fancy, I bought it two years ago for like $800, if that gives you any idea how capable it is. If a simple text-based message on a standard forum website is enough of a waste of bandwidth to complain about that loudly, maybe consider an upgrade to your machine or your network?

I haven’t had a lot of issues with the UI either. I will agree that it’s not very easy to learn new functions, but then again, I haven’t actively sought out tutorials or looked for accompanying documentation, sooo….maybe either of those might be a good place to start? I can get a stability margin, an altitude prediction, an optimum delay, a ground hit speed, a deployment speed, an estimated mass, and a bunch of other figures that look right, so that’s enough for me.

I’ll reiterate the messages from the others and encourage you to file bug reports. Software doesn’t improve by conducting personal attacks, especially since some of the developers are active here. Please don’t discourage them from conducting the work they do. It is valuable, and in the case of OpenRocket, we owe them our patience and gratitude. Apogee I could understand raging at, given that they’re selling the software for a not insubstantial sum, but is this really an avenue and attitude that will solve anything? A simple question would have accomplished much more.

(Minor edits)
 
Last edited:
If you believe that this can be done in fewer steps, I'd be very interested in knowing how you would approach doing so.
In a potential futile attempt to extract something useful out of this thread, let me take a guess where the complaint is. While the latest betas have definitely eliminated some mouse clicks and smoothed the overall process of creating simulations, there is still an aspect that some might find bothersome/cumbersome/unintuitive.

Right now, a flight simulation consists of two separate entities: the motor configuration and the simulation, and this (I think) is the crux of the complaint. There is a one->many relationship between the two. Although the mouse clicks have been streamlined, there is still a two-step process to first create a motor configuration, then simulate using that configuration. Motor configurations include deployment options. The simulations have a separate set of editable parameters.

This is a very flexible arrangement, but splits all the information for a particular flight sim in two separate tabs. If I want to change a particular sim, I may have to bounce to a different tab to change the motor config.

An alternative, more direct approach would be to include motor configuration and deployment information as part of each sim. This is probably analogous to the way Rocksim does it, if I recall correctly. Create a new sim, select motors, set sim parameters, and go. Editing any parameter of a sim would all be in one place. This approach might sacrifice a bit of flexibility, and/or make it a little harder to share motor configs among multiple sims (I would need to think about a good way to do it). But it would probably be more straightforward. It would also lend itself to something more like a wizard UI to set up a sim.

I hope folks will let me know if I'm hitting (or missing) the mark here.
 
In a potential futile attempt to extract something useful out of this thread, let me take a guess where the complaint is. While the latest betas have definitely eliminated some mouse clicks and smoothed the overall process of creating simulations, there is still an aspect that some might find bothersome/cumbersome/unintuitive.

There's definitely significant potential for confusion in the current workflow. Clearly it's confusing the OP. I have a couple thoughts but will put them in another thread, this one's gross.

I'm currently using 22.02.beta.01. For what it's worth, the only stability issue I had in previous versions revolved around Undo actions. I haven't had any of those problems in this version. Biggest glitch I have in this version, and it's a minor one, is that in the 3D views the display of stability and apogee calculations doesn't update as I make changes to the design. I have to switch to 2D to have them update. (this display glitch is resolved in beta.04)
 
Last edited:
Biggest glitch I have in this version, and it's a minor one, is that in the 3D views the display of stability and apogee calculations doesn't update as I make changes to the design. I have to switch to 2D to have them update.

I believe that this has been resolved in the current 22.02.beta.04.
 
I would like to take this opportunity to point out that while I don't think OR is perfect, it is an AWESOME piece of software that does 99% of what I need it to do, and does it pretty well. The fact that I can use it FOR FREE is a cause to thank the developers who have (and continue to) donate their time to the effort. Thanks!
It's perfectly fine to report bugs, shortcomings, and suggestions, but to complain about free software?
 
Even a buggy computer generated sim is way better than a grumpy old dude mindsiming!

"Back in the day all we had was a Big Chief tablet, a carpenter's pencil and slide rule. If we didn't get it right there would be hoards of godless commies running amok in the streets and on the farms." Put away your balsa nose cones and toxic sanding sealer and make the great leap into the 21st century. "Me computer and phancy phone ain't workin' too good but the rebuild on me carburetor is workin' sweet! Now let's fly that fine looking oddroc!" OH THE HORROR! RUN FOR THE HIILS! GOT MY CAPS LOCK KEY STUCK AGAIN! ;)
 
I would like to take this opportunity to point out that while I don't think OR is perfect, it is an AWESOME piece of software that does 99% of what I need it to do, and does it pretty well. The fact that I can use it FOR FREE is a cause to thank the developers who have (and continue to) donate their time to the effort. Thanks!
It's perfectly fine to report bugs, shortcomings, and suggestions, but to complain about free software?
Could not agree more! And for what it's worth, beta 04 and the current 15.03 release have been very stable and mostly bug free for me (aside from the 3D views that is). Beta 04 has been so good in fact, that I now use that by default over the current official release version. The way one adds motors into a design in OR seems fine to me.

I have nothing but gratitude and appreciation for the folks who are working hard to get the new release out!

I tried the free trial of Rocksim recently, and found it to be less intuitive to use than OpenRocket, but I suppose it depends on what you are used to using.
 
When MS Windows and Office both require periodic patches...and considering that Office is extremely expensive to buy, somewhat less so when it's leased (though still exorbitant IMHO)...a program that is incredibly useful to the rocketry community and costs nothing is hardly deserving of badmouthing.

I notice that OP said he "can crash them both in a few minutes." That sounds like one is trying to crash the program. If instead the programs crash on routine use, that might be a different story, but it does not surprise me that someone who is coding-savvy could force a program to crash. I'm anything but savvy when it comes to programming---the last software I wrote was in machine language for a Commodore 64, for data acquisition---but I can crash Windows without a whole lot of trouble (inadvertently, unfortunately).

(As for Office, I had to go through the final Word file for Experimental Composite Propellant 2e THIRTY-SIX TIMES to fix all the screwups that it did with figures and fonts, and get it so that it would properly generate an Acrobat file. Stupid thing crashed uncounted times.)

Best -- Terry
 
I notice that OP said he "can crash them both in a few minutes." That sounds like one is trying to crash the program. If instead the programs crash on routine use, that might be a different story, but it does not surprise me that someone who is coding-savvy could force a program to crash.
If the OP’s attitude were better I’d say to extend an offer to join the team. Those skills could be put to use in testing releases, but he has no people skills to go with them
 
If the OP’s attitude were better I’d say to extend an offer to join the team. Those skills could be put to use in testing releases, but he has no people skills to go with them
There is a standing offer to everyone to join the team, in whatever capacity you can. Filling a bug report is a big help; so is tearing off an issue, fixing it, and filling a PR to get the fix incorporated in the project. You can expect suggestions for improvements especially as you're getting familiar with the code base, but I think we're pretty gentle. There's no experience or education requirement; if you write good code and it works you've clearly got the skills we need.

And you'll start right out at the same pay as the rest of us!
 
I have had a couple of crashes using Open Rocket on Windows over the years and I can't think of a time I was able to reproduce the crash. For me, it has been as reliable as most other programs run on Windows.

I've used it on Linux too and cannot recall any issues. The crashes I have had have probably been more related to the OS than the OR build.
 
Maybe it's just me, but I took the OP's post as a rant/vent. So maybe it's not the most rational or logical, but he/she is frustrated and annoyed and wanted an outlet.

Just my take.
 
Back
Top