[1.6.X] Surface Experiment Pack - Deployable science for KIS/KAS! (v2.7 - 13/Jan/19)

[pp3d]

4 hours ago, DMagic said:

I'll need log files to be able to help.

Thank you, i appreciate it. I am going relaunch the game and scene to capture the SEP deployment events. I did check the logs a few seconds ago and I got messages like:

[WRN 00:23:19.992] [R&D]: No Experiment definition found with id SEP_Retroreflector_Basic
[WRN 00:23:19.992] [R&D]: No Experiment definition found with id SEP_Retroreflector_Detailed
[WRN 00:23:19.992] [R&D]: No Experiment definition found with id SEP_Retroreflector_Exhaustive

but I'll post all the time stamps for any clues.

6 minutes ago, Krakatoa said:

Ah, gotcha.  Dunno if it's Pathfinder, I use it as well and don't have any issues, even have my SEP station plugged up to a Pathfinder base right now.  Only real difference I see between your setup and mine that might have any effect is that I make sure to include an antenna on my base and solar panels and one of the batteries before I get as far as turning things on.  Maybe try adding an extended communication unit and power and see if that fixes anything before you start uninstalling stuff?

I was trying to be a minimalist for the sake of the video, but I have attached to the SEP power unit the girder with the SEP solar panels or a RTG -- or connected the SEP power unit to the a base (including a pathfinder one) to provide external power ( I can post some pics I guess).The SEP seems that it's powered alright... it's just that the experiments connected to it are not. 

[Krakatoa]

Well avast.  Good luck to you! I'm curious to see what's causing the problem now.

[DMagic]

Missing experiment definitions are usually a sign of duplicate experiments somewhere in the GameData folder.

[pp3d]

1 hour ago, DMagic said:

Missing experiment definitions are usually a sign of duplicate experiments somewhere in the GameData folder.

good to know and i will look for them.. but could that explain what i am seeing? 

In regards to the log files, i am not sure what the due process is here... Do you require the entire log - from launching the game to engaging the UI for the sep experiments or just the time stamps that pertain to the latter?

 

 

[DMagic]

Entire log. 

[pp3d]

3 minutes ago, DMagic said:

Entire log. 

ok.. thanks.. i'll post it later tonight. i'll have to remove the user name that shows up in some directory searches when the game loads. 

[DStaal]

The other thing that can cause errors like that one is malformed experiment results - in any experiment that loads before the SEP experiments.  (KSP essentially concatenates all the experiment results into one file, and then loads it.  If there are any errors, it aborts loading at that point - leaving following results unloaded.)

[pp3d]

Not sure if there a prefered way for sharing logs here. If I just need to cut and paste here, I will do so... otherwise the KSP.log can be downloaded from here:http://www67.zippyshare.com/v/mY7zOKbH/file.html

Log captures a new game startup and ends in this scene here:

 

[DMagic]

It looks like a duplicate experiment error, though I'm not sure where it's coming from, I didn't see any obvious duplicates during the loading process (look for Config(EXPERIMENT_DEFINITION)... during the loading stage of the log file). 

This one showed up at the end of loading though, it's actually from Orbital Science, but it's triggered by the presence of a duplicate experiment definition. This error looks to be fixed for KSP 1.3, but for now this breaks all subsequent experiments from being loaded.

[ERR 00:39:01.711] [DM] Whoops. Something really wrong happened here, a duplicate science experiment definition may be present somewhere; stopping this contract experiment from loading...System.ArgumentException: An element with the same key already exists in the dictionary.
  at System.Collections.Generic.Dictionary`2[System.String,ScienceExperiment].Add (System.String key, .ScienceExperiment value) [0x00000] in <filename unknown>:0 
  at ResearchAndDevelopment.loadExperiments () [0x00000] in <filename unknown>:0 
  at ResearchAndDevelopment.GetExperiment (System.String experimentID) [0x00000] in <filename unknown>:0 
  at DMagic.DMConfigLoader.configLoad () [0x00000] in <filename unknown>:0 

 

[pp3d]

5 hours ago, DMagic said:

It looks like a duplicate experiment error, though I'm not sure where it's coming from, I didn't see any obvious duplicates during the loading process (look for Config(EXPERIMENT_DEFINITION)... during the loading stage of the log file). 

This one showed up at the end of loading though, it's actually from Orbital Science, but it's triggered by the presence of a duplicate experiment definition. This error looks to be fixed for KSP 1.3, but for now this breaks all subsequent experiments from being loaded.

[ERR 00:39:01.711] [DM] Whoops. Something really wrong happened here, a duplicate science experiment definition may be present somewhere; stopping this contract experiment from loading...System.ArgumentException: An element with the same key already exists in the dictionary.
  at System.Collections.Generic.Dictionary`2[System.String,ScienceExperiment].Add (System.String key, .ScienceExperiment value) [0x00000] in <filename unknown>:0 
  at ResearchAndDevelopment.loadExperiments () [0x00000] in <filename unknown>:0 
  at ResearchAndDevelopment.GetExperiment (System.String experimentID) [0x00000] in <filename unknown>:0 
  at DMagic.DMConfigLoader.configLoad () [0x00000] in <filename unknown>:0 

 

I see. Can duplicate definitions be ignored rather than trigger errors? I see some duplicate definitions in tanks for example that are ignored by the mods that find them.

[pp3d]

On 4/23/2017 at 8:10 PM, DStaal said:

The other thing that can cause errors like that one is malformed experiment results - in any experiment that loads before the SEP experiments.  (KSP essentially concatenates all the experiment results into one file, and then loads it.  If there are any errors, it aborts loading at that point - leaving following results unloaded.)

do you happen to know the name of that file? and what happens if i remove it.

[DStaal]

2 hours ago, pp3d said:

do you happen to know the name of that file? and what happens if i remove it.

The file doesn't actually exist that I know of - it's just how KSP behaves.  (There's probably some memory buffer being read into or something.)

[pp3d]

3 minutes ago, DStaal said:

The file doesn't actually exist that I know of - it's just how KSP behaves.  (There's probably some memory buffer being read into or something.)

thank you all for trying to look into this. 

[cakepie]

@CobaltWolf

PR submitted for a few simple quality of life changes, e.g.:
- CCK categorization
- tags for search
- KIS stackability

Let's also look into locker balance before 1.3, what say you?

[CobaltWolf]

On 5/7/2017 at 8:08 AM, cake>pie said:

@CobaltWolf

PR submitted for a few simple quality of life changes, e.g.:
- CCK categorization
- tags for search
- KIS stackability

Let's also look into locker balance before 1.3, what say you?

Sorry for the delay! Thanks! I've pushed an update with these changes! I don't have any real feelings on locker balance so if you have any suggestions let me know!

Update is now available on Github and Spacedock

 

[cakepie]

57 minutes ago, CobaltWolf said:

I don't have any real feelings on locker balance so if you have any suggestions let me know!

Well, if nothing else, they probably shouldn't behave like magical bags of holding. =Þ

 

The issue with the lockers, I think, is probably due to KIS not computing the correct external volume -- I think it's using only the collider for its calculations.
In other words, it's measuring only the front face of the whole locker, and using that as the size of the whole thing. 
This can be solved by us overriding KIS and providing an explicit value for part volume.

I'll take measurements of the models and compare them vs KIS containers to see what makes sense. Will report back with my findings.

 

The cardboard box has a different problem. It can contain 750L and is explicitly declared with a volume of 3L. The description text suggests that it was intended to be "collapsible", i.e. the box can be hold up to 750L, but when empty it can be folded up and will take up on 3L of inventory storage space. That's a cool feature/idea that base KIS does not have. The problem with this is that the end result isn't what was intended -- the box is somehow capable of holding 750L of other stuff inside 3L of space.

This one will require plugin code to be implemented correctly. But I think it should be feasible.

[CobaltWolf]

5 minutes ago, cake>pie said:

Well, if nothing else, they probably shouldn't behave like magical bags of holding. =Þ

 

The issue with the lockers, I think, is probably due to KIS not computing the correct external volume -- I think it's using only the collider for its calculations.
In other words, it's measuring only the front face of the whole locker, and using that as the size of the whole thing. 
This can be solved by us overriding KIS and providing an explicit value for part volume.

I'll take measurements of the models and compare them vs KIS containers to see what makes sense. Will report back with my findings.

 

The cardboard box has a different problem. It can contain 750L and is explicitly declared with a volume of 3L. The description text suggests that it was intended to be "collapsible", i.e. the box can be hold up to 750L, but when empty it can be folded up and will take up on 3L of inventory storage space. That's a cool feature/idea that base KIS does not have. The problem with this is that the end result isn't what was intended -- the box is somehow capable of holding 750L of other stuff inside 3L of space.

This one will require plugin code to be implemented correctly. But I think it should be feasible.

I, uh, am not sure what you mean. The volume is defined in the part cfg. I don't believe any calculation is done at all. I'd like to find a solution for the cardboard.

[cakepie]

19 minutes ago, CobaltWolf said:

I, uh, am not sure what you mean. The volume is defined in the part cfg. I don't believe any calculation is done at all. I'd like to find a solution for the cardboard.

 

My bad, I thought you were already aware of the nature of the problem since there was already github issue titled "locker balance".

Basically, the lockers have the same problem as the cardboard box (but with different underlying cause):

• AKI-L can hold 1500L, but itself takes up 320.33L when you put it into another container..
• AKI-S can hold 750L, but itself takes up 149.39L

So, you could do something like:

1. Pack an AKI-L full of whatever you like.
2. Put 4 such lockers into one AKI-L
3. Do step 2 recursively
4. ???!!!

Mass continues to be correctly accounted for, though. If you keep stuffing lockers inside lockers at some point the outermost locker becomes so massive that when you load your craft to launchpad/runway, gravity just rips the locker off because its weight exceeds what the part attachment joint can withstand. XD

I am aware that "volume" of the locker -- volumetric capacity, i.e. how much it can hold -- (1500L, 7500L) are declared in SEP part cfgs.
By "calculation" I'm talking about code in KIS, which calculates external volume -- how much space it takes up when placed in another container
KIS automatically does this for all parts, based on the model. That's where 320.33L / 149.39L comes from.

Although KIS gets reasonably "correct" figures almost all of the time, it doesn't do so well here because the locker parts are a bit different from what it usually expects.
But KIS is aware that this kind of mistake can sometimes happen, so it provides a way to override it by specifying our own value in the part cfg.

Hope that clarifies.

[CobaltWolf]

Well the lockers shouldn't be KIS storable in the first place...

[stratt76]

Been searching for a few hours now; basically my question is this: Once you've set and completed your EVA experiments, is it possible to repack the items and reuse them elsewhere? Or do you need a whole new set of equipment for each setup.

 

See, I'd like to do a rover mission doing the experiments in all the different biomes and it would be great if i could repack and reuse the stuff otherwise it's a separate mission for each biome.

Just now, stratt76 said:

Been searching for a few hours now; basically my question is this: Once you've set and completed your EVA experiments, is it possible to repack the items and reuse them elsewhere? Or do you need a whole new set of equipment for each setup.

 

See, I'd like to do a rover mission doing the experiments in all the different biomes and it would be great if i could repack and reuse the stuff otherwise it's a separate mission for each biome.

'Whether you think you can or that you can't, you're right." -Stewie Griffin

[stratt76]

Can't seem to find an answer for this anywhere, can you pack up the experiments and gear after deploying them to redeploy them again elsewhere? Or do you need a new set of equipment every time?

[cakepie]

8 hours ago, CobaltWolf said:

Well the lockers shouldn't be KIS storable in the first place...

I don't think KIS provides that option. With KIS, all parts can be assembled/disassembled in flight, and can be stored. The only requirements are:

• depending on stack/surface attachment, may require an engineer Kerbal and tools to attach/detach
• require enough Kerbals to move it based on mass of the part(s)
  • this refers to the ability to "G" grab the part and drag a hologram of it around and place/drop it elsewhere
  • not the ability to "carry" like a backpack (see below)
• cannot store an assembly of multiple parts still joined together, only single parts by themselves
• must have with sufficient space in container to store

KIS doesn't require any settings in part cfg to enable this to happen.

 

You're probably thinking in terms of "lockers shouldn't be EVA items that can be equipped" in the sense of:

• can be equipped in the hand: KIS tools, SEP screwdriver, golf club
  • part cfg has equipable = true
• can be "carried" on the back of Kerbal: KIS SC-62 Container, SEP cardboard box
  • part cfg has carriable = true

But these are just separate features that can be specified in addition to the standard KIS behavior described earlier.
 

[cakepie]

2 hours ago, stratt76 said:

Can't seem to find an answer for this anywhere, can you pack up the experiments and gear after deploying them to redeploy them again elsewhere? Or do you need a new set of equipment every time?

Yes, you can dismantle the experiments.

* the SEP solar panels don't retract, but that's not really an issue.

Edited May 14, 2017 by cake>pie

[stratt76]

Hey thanks guys for the help: here's what I did wrong: I was holding g and left clicking but not holding the click; you have to hold g and the click for the full drag action. Only took me 6 hours to figure out!  Can't believe I misunderstood it from the user manual PDF. Anyways thanks again!

[garwel]

Just another screenshot: an engineer, a scientist, a lander, and an SEP set at the Mun's highest mountain near the south pole.