Bizz Keryear
-
Posts
333 -
Joined
-
Last visited
Content Type
Profiles
Forums
Developer Articles
KSP2 Release Notes
Bug Reports
Posts posted by Bizz Keryear
-
-
Sorry that was a typo.. since I was testing it in ksp directly (not using archive) there was a then in the actual script!Bizz have you tried it with math? Space computer uses eta:apoapsis in a math statement.
when eta:apoapsis < 120 + tTime then { do something. }.Maybe try.
when eta:apoapsis < (120 + tTime) then { do something. }.That might fix it.
I also just noticed that I am using THEN I am not sure if that is still required in when statements. Honestly I am not sure why it ever was.
As long as you are using a kOS version that is capable of using thrust and steering (my experience was: 0.82 steering working but throttle broken, 0.83 throttle working but steering broken)If doesn't use it. Seems like it isn't required but I haven't tested it. I have enough trouble already without sabotaging my self.
Good news.
I wish to announce that Space Computer can now officially launch to orbit and then preform a transfer and then preform a powered landing on Kerbin... All from the menu!
Put it on the wiki, I want to have it there anyway. I give you an example what it will be like:The recursive ON statements where what was breaking the whole lot. Once I fixed that, everything worked as it should have.
I will probably switch to hosting the scripts on pastebin to clean up the front page. There will be a lot more coming.
- Site with anything in it (download and included scripts): http://kos.wikia.com/wiki/Example_-_Descend_Lander/Skycrane
- Example script displayed as it will included/downloaded: http://kos.wikia.com/wiki/Example_-_Descend_Lander/Skycrane/bodystats.txt?action=raw
- actual wiki page of the script (for editing): http://kos.wikia.com/wiki/Example_-_Descend_Lander/Skycrane/bodystats.txt
Need rights? No problem I will give them t you. (wow thats sound desperate)
I don't mean that you should drop this thread here. Its better for communication anyway.
I am not so sure that I like the menu system now that it is working. I am looking at being able to setup mission packs that will execute based on settings in the mission pack file.
You tell it to "run Mymission" and it will carry out your mission as described in the file. Probably a thousand different ways to do that. I shall have to think about it and try some things.
I have still yet to remove a ton of constants to be replaced with variables based on SOI, set parameters, Etc.
I see the mods for the mods have popped up to allow you to poll sensor information. I probably won't bother with that too much. I expect Kevin will add that functionality directly to kOS.
I also think so, but for now (and also cause this mod exists) he will concentrate on other things.
Hopefully bug fixing.
Edit: For some reason I still don't understand eta:apoapsis is still broken in the (old) space computer orbit script. while it was working in my test program.
clearscreen.
set ttime to 10.
lock steering to up + R(0,0,180).
lock throttle to 1.
stage.
when eta:apoapsis > 20 then {print "simple eta works".}.
when eta:apoapsis > 20 + ttime then {print "plus time works".}.
when eta:apoapsis > ( 20 + ttime + ttime ) then {print "brackets work".}.
until <liquidfuel> < 200 {
if stage:liquidfuel = 0 and stage:solidfuel = 0 {
stage.
print "stage compair work".
}.
}.
print "full ressource compair work".in this program everything is executed.
That also means throttle and steering are working again.
- Site with anything in it (download and included scripts): http://kos.wikia.com/wiki/Example_-_Descend_Lander/Skycrane
-
Quick question.
Can one write functions in kos, or do you have to use variables (which are all global) and call scripts that use the same variable names?
No functions (yet?), but we can call programs with parameters. And lately (dunno if that is a bug or a feature) all variables are global.
edit: reading is an art I am failing sometimes.
If you use programs with parameters the name of the variables can change.
like "run a(x)."
and a is:
declare parameter xina. //read as x_in_a
set y to xina^2.
print "x in a is: "+xina+" , and y is: "+y.And now I am sure that the globaliness of variables is a bug.
-
Sorry for the double post.
But I think the topic is so different that it needs a post of its own. got sidetracked before posting
Umm is here anyone who can come up with a vessel and a Kerboscript that is capable of testing every command?
Something we can use after each update so we can run a quick test on what is broken?
It should be a collab project.
http://forum.kerbalspaceprogram.com/threads/52004
anyone who is willing to help?
-
I use this forum cause I think its better than the wikia to communicate.
Ok, that will be a collab project in the kos wikia.
Umm is here anyone who can come up with a vessel and a Kerboscript that is capable of testing every command?Something we can use after each update so we can run a quick test on what is broken?
We are in need of a vessel and a kerboscript.
The vessel should be able to keep up with the script but should kept as simple as possible.
The script also should kept as simple possible but it should not only contain all commands but as much alternations as possible.
It also should give a reliable feedback. hopefully print, log, and switch to will not break.
Update: I created a Wikipage.
-
bugs in ksp / kOS.
well if I am serious its not just one.
Its Global warming/Climate change, Overpopulation and Sustainable Energy.
Also critical but not as much as above Poverty/Distribution of wealth and Education.
When fossil energy runs out and we haven't fixed our energy consumption / sources then Overpopulation will be fixed since Climate change and diseases will kill most poor and and or uneducated. (well uneducated means in most cases poor)
Well, War ever was and ever will be there as long as we are humans. I think (its sad but true) that is part of our nature.
If fossil really is running out ... then there will be war, and the war for the final resources will make look the WWII like a quarrel of kindergartners in a sandbox.
-
I just want to make sure everyone is aware that while I may have created the kOS Wiki and added some initial content, Bizz Keryear has spent far more time than I keeping it up to date. I've essentially just stayed out of his way because he's doing such a great job in my opinion. Thanks, Bizz. I have definitely not regretted making you an admin as you have done far more than I would ever have time to get done.
*blushes*
Yes, true.Not to mention that Bizz is not a native english speaker/writer, as far as I understand.
Please feel free to correct any bad grammar / typo / spelling you come across.
And if its not understandable what I am try to tell or if I use terms or proverbs wrong.
You can also edit it, but if you not understand it yourself just add comment or open a thread in the wiki forum.
-
Yeah, could be a random problem. I know with 0.82 it was mostly random it worked sometimes but after it worked once it didn't seem to work after that, for me anyway. The randomness could be due to different mods installed. I don't use MJ but I have module manager still in the game folder (from when I was testing RT2). Also got Kethane, ISA MapSat, Quantum struts, procedural fairings and some others. Though they're not being used in my current dealings with kOS, but they could still be interfering somehow I guess.
It was just a basic lock steering, after initial staging to launch (initPitch was set to 0)
stage.
lock steering to up + R(0,initPitch,180).I believe I also tried "lock steering to prograde" and "lock steering to heading 90 by 45" manually and it did nothing. I can try it again later just to make sure. This is in 0.83 just so we're on the same page.
I switched back to 0.7 for now trying to fine tune my kOS challenge code to get a slightly higher PE altitude.
On an unrelated note, not a big deal, but I thought the log function saved to archive. Seems that it saves to where ever the program is run from, which I guess works, being as if you were out of range it couldn't save to the archive...soo I guess that was just my misunderstanding lol DOH!
Confirming in .83, lock throttle works fine - lock steering not so much.OK, I have found it.
Lock steering works in general, but it seems that using variables in lock steering is broken!
So while lock "lock steering to heading 90 by 45." works
set pitch to 45.
lock steering to heading 90 by pitch.does not.
edit : umm ... in 0.82 it still has worked and I belive I got it to in a test program once...
but now it don't.
edit2: So I mean I also can confirm this for 0.83.
edit3: wheelsteering is also broken.
-
Yep, and lock steering is also broken(still? dont remember if it was broken in 0.82Right...I was having problems with in in 0.82 lol how quickly I forget). But hey, throttle works now. LOL
Could you post the code? because in my case lock steering still works.
I think it may the same is as with eta: in some cases it works in some not.
-
Nice!
For me, in 0.82 pretty much all locks were broken, although it seemed like lock steering would work sometimes, but not all the time, like after a certain time/altitude (when controlling throttle manually).
Also, I'd like to add, it might be a good idea for anyone reading this, to save the previous version zip files you download, so incase you do run into bugs like in that last few updates, you can go back to a previous version that was working. I've got 7 versions from 0.46 to 0.82 (missed 0.8 and 0.81 though) and it takes up less than 3mb. Though you could keep just one previously known working version and delete the others I guess.
Lol from Kevins Dev Blog: "Wheel throttle could be set to 100 with hilarious results" Makes me wish I was testing rovers now.
Ok, so does anyone know where these "flagrant errors" are logged to?
sorry, but that got fixed... maybe because it wasn't game breaking.
I have some Versions from 0.5 on. I also have 0.8 but not 0.81. 0.81 was too fast replaced with 0.82.
They are all on my google drive. See the issues page on the wiki.
Edit: I think the code needs a complete overhaul. The eta thing is too wired.
But I am too stupid to find out when it works and when not.
-
0.83 space computer STILL broken same error on eta going to go make a stink bout it in main thread in hopes its finally resolved. So for the time being guys stick with 0.71.
Well the funny thing is: I wrote a program with only three commands, and 4 lines.
if eta:apopapsis > 10 {print "if ok".}.
wait 1.
when eta:apopapsis > 10 {print "when ok".}.
wait 50.When sitting on ground and entered on the command line both if and when are fired.
When sitting on ground and in program only when is fired.
When flying both are fired.
sorry no screeny of the flying version.
But if I use orbit.txt (from space computer) which btw worked perfectly fine 0.7 then I get the unrecognized eta: term.
I checked the code over and over again. I can't find an error.
On the other hand I just proved that eta:apoapsis works .. on command line or programs...
-
I believe the "LOCK TO" code breaks a bit further with every new version. In 7.0, every LOCK TO that isn't locked to altitude seems to stop working as soon as any loop or wait is started. And it sounds as if the problem got worse with the 0.8x series.
Also locked variables also get only updated outside from loops when called.
When not they stay at the last value.
Edit: There is something really wired going on.
While a in a program that I wrote eta:apoapsis works perfectly fine.
It doesn't work in space computer anymore... in 0.7 it has!
A moment I thought it has to do something with single and multiple lines, but that's also not true.
stage.
wait 10.
print "after wait.".
if eta:apoapsis > 10 {print "if 1 line ok". }.
when eta:apoapsis > 10 then {print "when 1 line ok". }.
wait 10.
if eta:apoapsis > 10 {
print "if multi line ok".
}.
when eta:apoapsis > 10 then {
print "when multi line ok".
}.
wait 50.All lines were executed.
-
Ever since .8, steering doesn't work. Mechjeb steering works fine, so it's not the rocket, and I'm using the same program I used in .7, so it's not that either.
can't confirm that. Made a program to test the eta: problem in 0.82 and used manual throttle (cause I had known that that was broken)
and it had 3 lock steering commands.
- lock steering to up + R(0,0,180).
- lock steering to heading 90 by 90.
- lock steering to heading 90 by 45.
That worked like a charm... So not broken for me. Maybe only with mechjeb?
edit:
-=[ Does Your System Have Bugs? ]=- 4/97
.--. .--.
_ ` \ / ` _
`\.===. \.^./ .===./`
\/`"`\/
, | | ,
/ `\|`-.-'|/` \
/ | \ | \ .--. .--.
.-' ,-'`| ; |`'-, '-. _ ` \ / ` _
| | \| | `\.===. \.^./ .===./`
| | ;| | \/`"`\/
| \ // | , | | ,
| `._//' | / `\|`-.-'|/` \
jgs .' `. / | \ | \
_,' `,_ .-' ,-'`| ; |`'-, '-.
` ` | | \| |
| | ;| |
.--. .--. | \ // |
_ ` \ / ` _ | `._//' |
`\.===. \.^./ .===./` .' `.
\/`"`\/ _,' `,_
, | | , ` `
/ `\|`-.-'|/` \
/ | \ | \
.-' ,-'`| ; |`'-, '-.
| | \| |
| | ;| | .--. .--.
| \ // | _ ` \ / ` _
| `._//' | `\.===. \.^./ .===./`
.' `. \/`"`\/
_,' `,_ , | | ,
` ` / `\|`-.-'|/` \
/ | \ | \
.-' ,-'`| ; |`'-, '-.
| | \| |
| | ;| |
| \ // |
| `._//' |
.' `.
_,' `,_
` ` - lock steering to up + R(0,0,180).
-
Oh ... I haven't seen that 0.83 is out..
edit found out throttle fixed (at least kevin says that) and eta: still broken.
-
[...]
Meh ... I wish the people would read all sources.
See github. eta: is broken as well (in programs at least)
-
I'd like to see this as well.
Something of similar form factor to an AR202 style mechjeb, where it can be easily attached to an existing vessel or used on an ultralight probe assembly without having to make structural changes.
I was playing with it last night and managed to automate my kethane scanner such that the sensor automatically turns off as the battery runs down, and turns back on as the battery recharges. KOS is apparently able to fire action groups while a vessel is warping, making it quite the solution for my 1000x kethane scanner.
I'll probably add other power management tasks to it tonight, and possibly even a 'phototropism' mode where it attempts to orient the vessel for maximum attainable sun exposure.
In the wiki http://kos.wikia.com/wiki/Mods I described a way to make all stock drone cores a kOS computer.
That would work for radial parts like flight engineer, too
Is there an older version available anywhere until there's an update? .82 doesn't seem to work (i.e. "lock throttle to 1.")Well, since the problems the current version has I have uploaded every version I had to google drive. Link in the wiki http://kos.wikia.com/wiki/Issues
-
No its not particularly realistic. Usually rocket engines fall into three broad categories.
1) engines that are designed to ignite only once, with a specific single use igniter. These are usually first and second stage engines.
2) non-hypergolic engines designed for multiple restarts. These have different igniters and do have a design life, but aren't like shotgun shells: you don't have a specific set number of uses. Exmaple of such an engine is the new Merlin-1D vacuum engine (with dual igniters for redundancy) or the Centaur. This is not the most reliable of methods, it can fail, which leads to #3
3) hypergolic engines. Put oxidizer in contact with fuel and it spontaneously ignites. Example would be the AJ-10 from the Apollo CSM. Suffers from generally a relatively poor ISP, but if you need it to start every time, on time, this is the most reliable method.
The link PrivateFlip posted above mine doesn't say what he thinks it does. Basically, it describes single shot igniters (SRBs, and first stage engines (e.g. pyrotechnic charge) and it describes multiple start systems (e.g. plasma igniter/spark plugs)
So it might realistic to have that in the big engines. Like mainsail, or skipper. But that would generally not add a challenge.
Well beside from that I am sure that a real rocket would have in that case a method to reduce the fuel consumption to a minimum without turning it off completely.
It's real: sourceWe read the same article, yet we came to different conclusions.
--------
Well at least it seems to add some new interesting aspects to the game while not really realistic.
But on other hand its not my cup of coffee.
-
Is that realistic?
I can't find source that states that liquid engines have limited ignitions.
IMHO is it a electric spark that is doing the job.
Like in any good old gasoline engine.
And that is having a few thousand ignitions ... per second. (rpm/cylinder)
So is it based on the real world or just for challenge?
--------Edit below:-------
From the German and English wikipedia I know now that are two kinds of liquid engines.
Ignition can be achieved by a number of different methods;- One can not ignite itself and once it is turned off its over. >wikipedia=> "a pyrotechnic charge can be used, a plasma torch can be used"
- The other however can be ignited (almost) unlimited
subtypes:- wikipedia=> "or electric spark plugs may be employed. "
- wikipedia=> "Some fuel/oxidizer combinations ignite on contact (hypergolic),"
- wikipedia=> "or electric spark plugs may be employed. "
wikipedia=> "and non-hypergolic fuels can be "chemically ignited" by priming the fuel lines with hypergolic propellants (popular in Russian engines)."
I've just (while writing) realized that might be the one and only that is limited cause the amount of hypergolic propellants is also fairly limited. But I am not sure if its a one time ignition method.
- One can not ignite itself and once it is turned off its over. >wikipedia=> "a pyrotechnic charge can be used, a plasma torch can be used"
-
Can I suggest a feature? I can't write C#, but I would like a command for entering text to my programs.
Something like 'INPUT TO X' like in BASIC- recieves input and sets variable X to it.
EDIT: I had another idea:
We currently lack any file reading commands, so I might suggest:
'READ "FILE" TO X' Reads the content of the file FILE into variable X
'READLINE X FROM "FILE" TO Z' Reads line number X in file FILE into variable Z
Nice, we also discussed about that, but I think all of us would prefer to have the existing features working as intended before having new features.
Having then (new features) are nice, but all for naught if the rest isn't working.
Wait.. What? There are reddit challenges for kOS?[...]I've been using that temporarily in my attempt at the kOS reddit challenge.
-
Here is a little script that extrapolates a planes vertical speed and tries to converge the altitude to some target altitude.
Testing it on a B9 I8-L Bradbury it was able to keep it stable within 2.5 meters of the target (11500m). On heavier designs like the B9 D-175 Strugatsky it struggled to keep within a couple 100m as there is just too much momentum and not enough pitch authority at that altitude.
Random question: What exactly is the semantic of locking variables ? When exactly are they updated ?
I didn't use lock variables for a while now, cause they where broken within loops.
Maybe Kevin has worked on that, but since 0.82 (? dunno hadn't tested 0.8 -loops broken- and 0.81 -was too slow or 0.82 was released too fast- so far) Lock throttle and wheelthrottle is broken.
-
Well, the PRINT AT () and Floor() functions have come in super handy...
Altitude (x) and surface speed (o) launch profile:
[...]
Altitude (x) and surface speed (o) reentry profile:
[...]
Terrain elevation (x) profile:
[...]
Neat... Feel free to make a wiki page
-
Oh, man.
It works fine as long as you use it on the command line.... but in programs its broken.... and only for when. Its reported on git
-
[...] the more likely there will be strange bugs.
Yeah, 2+ meter cockroaches...
-
umm, what else except the throttle is broken?
as I said I have just in the moment 0.82 running and it displays me very well the eta:apoapsis.
-
Somehow with k-OS I lately feel like I am here: http://www.batoto.net/comic/_/terra-formars-r4747
So many huge bugs....
And since 0.5 it only gets worse...
kOS Test lab - a collab project
in Science & Spaceflight
Posted · Edited by Bizz Keryear
If I would have ideas about that I wouldn't need help.
edit: That sounded way more snooty than I wanted to be ... actually ... I only wanted to say that I right now have no clue what and how to do.
Actually my rockets are working, but not that good. And my programming skills ... well ... they usually work (after removing all typos ... lots of typos) but they are far from genius.