SSU Development Thread (2.0 to 3.0)

[STS]

I understand your wanting detailed systems, and that's fine, but most users aren't interested. They want to simulate what I outlined, and the SF proves my argument. People just want to enjoy it, not have to take an aerospace course.

I get PMs all the time from people thanking me for my addons. No brag, just a fact. The majority of them use the Fleet, because it is easy to use and easy to get. They would love to be able to use SSU, but most can't figure it out, thus their frustration. I just want people to enjoy it, that's all. You all can do what you want, but I'm tired of butting heads. It just isn't worth it.

Well, I didn´t take a aerospace course for completing a full SSU mission (I promise you) with SSU 2.0 and with that November 2011 version (1357? Or earlier? Well, I posted a video those days.) But yes, I did need help from math-people for getting the DV´s for the burns (simgroup mission-control-simulation requirement [we simulate FDO])... But you can avoid that with MFD´s. And Blixel shared his excellent orbiter calculators spreadsheets.

Wasn´t there a proposal for SSU-Lite?

 

[Donamy]

Well that's more than I've accomplished with it (don't have the time), Kudos to you. It is my belief, that you are in the extreme minority, from what I've heard.

SSU-lite was talked about, but I think it died.

 

[DaveS]

I think what scares most off is the installation procedure which isn't hard at all. Currently SSU is primarily obtained through the use of a version control system which requires a dedicated client to be installed and set up.

I think that once you have gotten to know the most commonly used client, TortoiseSVN, you'll find it is very easy to use. It doesn't have any complicated features or anything.

It allows you to easily keep your SSU installation updated with two simple clicks, first a right-click to open the context menu and a left-click to select SVN Update. From that point on, TortoiseSVN will automatically retrieve any files that have been checked in since the last revision.

Then it is just a matter of rebuilding the SSU source code, if that is what had been updated. This can been done with with either VC++2010 Express or 2013 Express. Or if you have it, the full VS versions.

 

[Urwumpe]

Well that's more than I've accomplished with it (don't have the time), Kudos to you. It is my belief, that you are in the extreme minority, from what I've heard.

Not sure if we have such a big difference in difficulty compared to SF.

The real problem is the quality. Many scenarios simply don't work anymore, others have dependencies that require proper installation. An SSU without a configuration and installation tool is much harder to be used than the simpler and more reliable SF. Maybe such a tool should get some priority.

For that we need a better quality assurance and stricter standards about our internal and external interfaces, not less strict standards.

 

[SiameseCat]

I don't think the installation procedure is a significant issue; we do have a public release, which be downloaded and extracted.

I think the biggest problem might be the lack of documentation. If you just start one of the scenarios without the checklists in front of you, you'll have a hard time getting anything accomplished. The documentation included with SSU is not very helpful.

 

[GLS]

I think SSU has its place in providing a as-accurate-as-possible-within-the-devs-knowledge simulation of the shuttle. I don't see the point of going after something that somebody else has already made. The shuttle was complex, so SSU is going to be complex as well. But, in terms of usage, I don't think following a checklist is so hard the average user wouldn't be able to it. Sure, if you want to fly an RTLS with everything in MAN it's going to be tricky, but the "normal" ascent and reentry are automated, so that's as easy as it gets. The on-orbit maneuvering... to be honest I haven't done that (apart from the odd OMS burn), so that might need some work to automate the calculation of the burn targets. The main problem I see is the mission setup, i.e. creating the scenario and mission file. That's another reason why I've been hitting the drums hard on the mission planner "app". Half a dozen clicks and any mission can be simulated -> that's the dream. :lol:
On the users getting a new version... an official version goes out when the team feels confident there are no major bugs. If it takes 2 months it's 2 months, if it takes 20 months it's 20 months, no point in going the iphone way and releasing a new version every 37 seconds. If anybody wants the "beta" version, it's all on sourceforge. Yes, we could simplify the access to the "betas" by having a compile of the latest sources (or something that works) be placed in a zip/rar and published every week/2weeks/1month/whatever.
Other than that, my main problem with SSU is that I'm not rich, so I can't work on this 24-7.

 

[Donamy]

A sample pm from Wolf, with his permission. There are more.

Hi Donamy. I'm writing to you just to let you know I am one of those (many) orbinauts who totally agree with your point of you about what the SSU "should be".
I have zero knowledge about IT stuff hence I am not in a position to criticize anything or anyone in the SSU team, but as an avarage Orbiter user and space shuttle enthusiast I can confirm it is quite frustrating to deal with SSU (in terms of installing as well as running scenarios). I really don't know why but the SSU is by far the most "difficult" addon to "play" with in Orbiter (and that has nothing to do with the fact that is intetnded a very complex and realistic machine). I totally agree with you when you say SSU should be a nice user friendly Space Shuttle that most of the orbinauts would enjoy (and not, as you said, a single developer property). As a user I have the feeling this project has nothing to do with an addon for Orbiter but it seems to be more an ambitious and ultimate developers challange. I think you guys "launched" the SSU project 4 or 5 years ago: according to the actual developement status how long will it take to get a final full release?.... Maybe forever, if the team wants to replicate the Shuttle in all details, drawing every single bolt the way the real shuttle was... It is a real shame cause I do really think you are right when you say we could have a nice realistic space shuttle that does all the things needed to accomplish a full STS mission, still not being the exact replica of the real thing. After all (to mention you one more time) we don't work for NASA and we are not going into space. We (the orbinauts) are just waiting for a new Space Shuttle that is able to do what the SF is currently not able. I think very few of us would complain if the total lenght of the ship is 20 cm longer or shorter than the real Shuttle..

 

[Urwumpe]

A sample pm from Wolf, with his permission. There are more.

But this then really begs the "stupid" question: Sure you both are not mistaking SSU for a second Shuttle Fleet?

Shuttle Fleet is a nice and userfriendly Space Shuttle add-on. It is available. It is also much more reliable.

Maybe, one day, it will not be it. But this one day might be in the far future and it does not mean that SSU might not form one day the base for a second S-F add-on.

But then: What about now? Should we really do something, that is already there, exactly the same way, as it already is?

Somebody should flame me, if I am now quoting things wrong but pretty often in the past years, we have all stated the following goal:

SSU should imitate the real procedures of the Space Shuttle, make use of a 3D VC, and maximally assist the player by adding AI crew members and a AI mission control.

If this is now not the goal of everyone ... we have a problem.

This is no reason to not use the source code of SSU to create a SSU lite. We are open-source, you may create winged elefants, as long as you are not calling it an elefant (ultra).

 

[Wolf]

Just a side note, since I've been "brought into this thread": speakin for myself I never expected the SSU to be just another SF. I love the SF but I'd also love to be able one day to perform an STS mission "by the books". Does this mean SSU must necessairly be the exact replica of the real Shuttle?... You guys know much better then me of course.
I know it is not appropriate to make comparison with others sims and addons but what about something like the PMDG B737NGX product?.. It is a very realistic ship with most (not all) the original 737 NG features (as far as performance, systems logic, failures, aerodynamics, etc.) and for proper use you do need the official manuals and QRH, but still it is far from the real airplane (anybody can fly it with some study without getting a full type rating at Seattle). I am just referring to the philosophy behind this product; I know I am talking about two different universes and I don't know anyhting about coding, hence here is a real "stupid question": would it be possible to consider something similar for SSU development/release?

 

[Donamy]

What I wanted, not that it holds much water or not, was a SF on steroids, and it was more or less to a certain extent. Then one member (not saying who) took it over and and broke everything that worked, no more launch scenarios worked and all animations were screwed. We had a nice SSU-lite and it got killed.

 

[STS]

SSU should imitate the real procedures of the Space Shuttle, make use of a 3D VC, and maximally assist the player by adding AI crew members and a AI mission control.

Well, I would also like to have that but it should be flexible, and able to turn on and off.

I'd also love to be able one day to perform an STS mission "by the books".

It is currently possible, as I already explained.

And for Donamy: Could you please explain to the world what happened with SSU-Lite? Those "Certain user" comments are not very ethic, I think.

 

[Urwumpe]

Well, I would also like to have that but it should be flexible, and able to turn on and off.

There is a problem: We can't turn the subsystem simulation off. Thats part of the architecture. We can only make things easier by adding more assistance or manipulating tolerances of the subsystem simulation or making things less error prone (error as in: measurement errors).

For example, we could add a menu to SSU, that does certain procedures automatically, like preparing for launch. It just requires some design work and some lot of coding - it won't be implemented in a minute.

For example, from what I learned in the past years, it shouldn't be done by hardcoding things in C++. We should provide an external domain specific language ("a simpler programming language for a very special task"), that allows adding new procedures without requiring to know C++. Also, such a language should be like "As long as the following conditions are existing, perform the following sequence." or "When these conditions exist, do the following action."

 

[Donamy]

Well, I would also like to have that but it should be flexible, and able to turn on and off.



It is currently possible, as I already explained.

And for Donamy: Could you please explain to the world what happened with SSU-Lite? Those "Certain user" comments are not very ethic, I think.

I will try to be more clear. There was an earlier release of SSU that actually worked. It was IMO "SSU-lite" then it was changed, and all launch scenarios were broken, and animations didn't work, not to mention ruined texture mapping. When I asked why, I was told I had to compile a new version, it's easy. Maybe for all team members here, but not myself or alot of other users. Is it too difficult, to have an out of the box download, from OH? If it is, most average Orbinauts won't use it. And if that isn't a concern, then I think it's a shame and really a waste of our time. I do get alot of enjoyment doing it, but not if others can't.

 

[SiameseCat]

What's wrong with the version on Orbithangar (http://orbithangar.com/searchid.php?ID=6389)? Anyone who wants the latest, potentially buggy version, can checkout the source from SVN and compile it. Otherwise, just download the formal release and use that.

---------- Post added at 06:00 PM ---------- Previous post was at 05:58 PM ----------

What I wanted, not that it holds much water or not, was a SF on steroids, and it was more or less to a certain extent. Then one member (not saying who) took it over and and broke everything that worked, no more launch scenarios worked and all animations were screwed. We had a nice SSU-lite and it got killed.

"SF on steroids" was never my intention. From the start, I've always wanted SSU to be a complete and realistic simulation of the space shuttle.

 

[DaveS]

I will try to be more clear. There was an earlier release of SSU that actually worked. It was IMO "SSU-lite" then it was changed, and all launch scenarios were broken, and animations didn't work, not to mention ruined texture mapping. When I asked why, I was told I had to compile a new version, it's easy. Maybe for all team members here, but not myself or alot of other users. Is it too difficult, to have an out of the box download, from OH? If it is, most average Orbinauts won't use it. And if that isn't a concern, then I think it's a shame and really a waste of our time. I do get alot of enjoyment doing it, but not if others can't.

Are you perhaps talking about Space Shuttle Deluxe, upon which SSU is based? Or do you mean SSU V1.25? If you do mean the latter, then that was just a very early release.

The plan always has been a detailed simulation of the Space Shuttle. I think somewhere, sometime you misunderstood the goal of SSU.

 

[Donamy]

What's wrong with the version on Orbithangar (http://orbithangar.com/searchid.php?ID=6389)? Anyone who wants the latest, potentially buggy version, can checkout the source from SVN and compile it. Otherwise, just download the formal release and use that.

---------- Post added at 06:00 PM ---------- Previous post was at 05:58 PM ----------


"SF on steroids" was never my intention. From the start, I've always wanted SSU to be a complete and realistic simulation of the space shuttle.

The version on OH is virtually the Fleet. No RMS VC, no switches working really. There was a release version that worked, but that is gone.

I understand you want a complete and realistic version of the Shuttle , but can we at least, in the mean time have a working public release ?

 

[GLS]

The version SC linked (2.0) is about 1 year old... there were some improvements in the meantime, but I think that version already does a lot. As far as I remember there was 1.25 before that and that's the extent of my memory.
I'd just like to say something to the users (and devs) that think there's unnecessary work going on: my goal is to flip some switch and have something happen that closely resembles what would happen in the real thing, and that implies some logic "behind the curtains". Sometimes we can just "hack" some half-dozen lines of code and it's good enough. But most times it takes many hours reading (and searching) documents to know what to do, and then some architecture has to be developed to have some simplistic simulation of some system in order to have the wanted behavior. I don't have any interest in having SSU simulating every nut and bolt, but this was a complex vehicle, so the development is going to be somewhat complex.
I always looked at the SSU team as a group of people with different skills and goals, and thought that was our strength... it's sad that it appears to be our weakness

 

[Donamy]

And it all happened with one screw up.

 

[SiameseCat]

The version on OH is virtually the Fleet. No RMS VC, no switches working really. There was a release version that worked, but that is gone.

I understand you want a complete and realistic version of the Shuttle , but can we at least, in the mean time have a working public release ?

Have you tried the version I just linked to? The RMS VC is working. The Entry & TAEM DAP is implemented.

 

[Donamy]

the SSUversion 2.0 on Orbithanger?