[1.8.x] Oh Scrap!- A ScrapYard based Part Failure and Reliability Mod 2.0.1 (07/12/2019)

[severedsolo]

On 06/04/2018 at 6:02 PM, Ricktoberfest said:

Does running a simulation in KRASH count towards part reliablitity?  If not could it be implemented?  I would think a factor of 50-75% of an actual flight would be appropriate if able.  

Sorry, I missed this the first time. As I said on the KRASH thread, I really like this idea and will probably implement it. I probably won't make it that big a buff by default, but I may make it configurable.

[Beetlecat]

Just a quick chime-in to voice my appreciation for this mod!

I've tinkered with part failure mods (Dang it! particularly) as well as Kerbal Krash System (I'd love to see that get integrated down the line)before, but never in a serious, deliberate career game where each part failure can completely ruin a mission. I've just fired up a new career with the USI sounding rockets (and a contract pack) and KCT, and it's taken many, many attempts to get a simple rocket to launch to complete my first contract:

• First attempt, main stage does not fire, launched via second stage only -- probe and science bits don't make it very high. Status: whoops! This is gonna be fun!
• meanwhile, I'm test-flying a rudimentary airplane--roughly following along the Kerbal Space Agency's game plan...  thankful to have those SXT parts in the first tech node...
• Second attempt: rockets stage and fire properly, but parachute fails instantly upon deploy. Status: crunch.
• Third attempt with same rocket design:  rockets fire properly, instruments checked at ~5000m, 'chute deployed---but fails a few 100 meters later, status: argh.
• Forth attempt: flight profile normal -- slight hitch in staging, but still works. Trajectory angled for water descent -- 'chute deploys and  water in 3, 2, 1 -- instrument assembly sinks like a stone. status: wah-wahhhh.

Did I mention Kerbal Construction Time? For this one contract alone, I queued up 3 more of the same rocket just to deal with the now-expected part failures. An entirely new aspect of planning and preparation is needed due to waiting for craft assembly and tech node unlocking, not to mention now feeling utterly paranoid about sending *ACTUAL* Kerbals into the air.

That ABORT button is about to get more use now than it has in the many years KSP has been on my hard drive.

I can't wait!

Edited April 11, 2018 by Beetlecat

[severedsolo]

Minor bugfix release:

Oh Scrap! 1.3.1 Released

• Fixed NRE spam when no part is selected in the editor.
• Fixed NRE spam when KRASH fails to load properly.
• Stop unnecessary failure checks when we know the failure is going to be suppressed anyway (KRASH simulations, Missions, Editor)
• Fixed failed part highlight being suppressed when the part also happens to be the worst part on the ship.

 

[Sokol_323]

Hello @severedsolo! I think I'm becoming a fan of your mods. I put this mod in version 1.3.1. Now I moved my career to 1.4.2 and again put this mod. And there's no way I can get to Mun. It's some kind of "Groundhog day". Three starts, and three Apollo-13! Frankly speaking, when once again there was a fuel leak, which deprived poor Jeb and his team of chances not only to sit on the Mun, but also to return home, I wanted to tear my computer apart!

I know how difficult the path to space was and how many failures and problems there were on this road. This mod makes the game stunningly realistic.

P.S. If we have so many problems on a simple flight to moon, what will happen when we fly to Dune???

Edited April 12, 2018 by Sokol_323

[severedsolo]

3 minutes ago, Sokol_323 said:

P.S. If we have so many problems on a simple flight to moon, what will happen when we fly to Dune???

Heh, I'm currently at the point in my game where my Mun landers never fail, so I got a little arrogant and threw together an Eve Orbiter without bothering to do a pad test.

Poor jeb was diving out of the pod to apply duct tape every few minutes to two of the life support containers, until they finally gave up and couldn't be repaired any longer. Made it home with 2 days of Oxygen and 10 days of Water left. TAC was screaming at me that I was running out of resources.

The good news is that your engineers are learning from those failures and every one will reduce the chances of that part failing again in the future.

Edited April 12, 2018 by severedsolo

[Sokol_323]

3 minutes ago, severedsolo said:

The good news is that your engineers are learning from those failures and every one will reduce the chances of that part failing again in the future.

This of course so. The old rocket I fly as well on small problems have long ceased to pay attention. However, flying to the moon is very expensive. This is probably comparable in cost to the launches of real Saturn-V. I'm afraid that a couple more failures and the government will simply stop giving us money and will have to collapse all projects, as happened with the real Apollo.

In fact, I really like this game.

[Gordon Dry]

It seems that "Highlight worst part" only works when the vessel was launched in the active session / scene.

When I load a vessel from the tracking station this feature doesn't work anymore - and does not write anything suspicious to the log, no NRE all the place.

[severedsolo]

2 hours ago, Gordon Dry said:

When I load a vessel from the tracking station this feature doesn't work anymore - and does not write anything suspicious to the log, no NRE all the place.

Unable to reproduce this I'm afraid, could you perhaps provide a (stock) craft file this happens on and your save? Might be a particular module that isn't working properly (although I don't see why, they all inherit from the same place).

List of silly things to check (you've probably done this, but just in case)

1) Has the part already failed? If it has the highlight won't work, because the highlighter has already been enabled.

2) Are you using version 1.3.1?

3) Has OhScrap identified the worst part. To check this just look in the UI and see if the "Worst Part" section is blank (sometimes it loses track if the worst part is staged away. Again I don't see how this would happen if you've just switched to the vessel, but you never know).

[Gordon Dry]

@severedsolo it could be 1) and it seems that this happened on loading (or let's say it happened in background)

Perhaps I was confused by this:

1. in VAB the feature highlighted all parts, but in different colors (red - yellow - green)
2. in flight mode only the one worst part is highlighted then

[severedsolo]

15 minutes ago, Gordon Dry said:

it could be 1) and it seems that this happened on loading (or let's say it happened in background)

Tbh, that's a stupid feature on my part anyway. You don't care if the worst part has already failed, you probably want to know what the next worst part is. I'll do something about that (especially as it gives you no warning at all right now.)

15 minutes ago, Gordon Dry said:

Perhaps I was confused by this:

1. in VAB the feature highlighted all parts, but in different colors (red - yellow - green)
2. in flight mode only the one worst part is highlighted then

Yeah I know about that one, I just literally have no idea why it''s happening. It's the same freaking code! in both scenes. EXACTLY the same class, loaded into both scenes. It's meant to just flag the worst part. For some reason in the VAB though it triggers the highlighter on all of them. I'm assuming this is KSP shenanigans (although actually, I kinda like the way they all light up in the VAB, so I might program that into the flight scene too.).

Edit: Ooh actually, if it's going to light all the parts up in the VAB anyway then why not make it useful. Turn them different colours based on their safety ratings.

Edited April 13, 2018 by severedsolo

[Tyko]

How often does Oh Scrap "roll" for failures on long term missions? I'm worried about it being too likely that parts will fail on multiyear missions to the outer planets.

Thanks!

[severedsolo]

2 minutes ago, Tyko said:

How often does Oh Scrap "roll" for failures on long term missions? I'm worried about it being too likely that parts will fail on multiyear missions to the outer planets.

Thanks!

See the bottom part of this post (apologies for the lazy linking, this is not my way of saying "read the thread" I just cba to type it all out again) - Note to Self: set up an FAQ.

 

 

[Tyko]

6 minutes ago, severedsolo said:

See the bottom part of this post (apologies for the lazy linking, this is not my way of saying "read the thread" I just cba to type it all out again) - Note to Self: set up an FAQ.

Thanks...it looks like the relevant info was on p11. It also sounds like I'm better off not randomly switching to long term missions because every time I switch to them i'm going to incur another re-roll. Am I reading that correctly?

<SNIP> it rolls whenever Start() or OnLaunch() fires. In practice this basically means it will fire at the following points:

• When you activate your first engine on the pad (the first Start() is ignored if OnLaunch has never fired).

• When you switch to the vessel, or it moves into physics range.

• If you stage after switching back to the vessel (thats probably a KSP bug, because OnLaunch probably shouldn't fire then, but it does).

There may be other instances where OnLaunch fires, but those are the ones I'm aware of.

Caveat: Technically it also rerolls when ScrapYard updates the tracker, but AFAIK that only ever happens on Vessel Rollout, so would be discarded as OnLaunch has never fired.

 

[severedsolo]

7 minutes ago, Tyko said:

It also sounds like I'm better off not randomly switching to long term missions because every time I switch to them i'm going to incur another re-roll. Am I reading that correctly?

Pretty much. HOWEVER - if your plan is "switch in randomly, check on it and switch away" you will probably be fine. If the failure doesn't happen before you switch away, it's like the "bad roll" never happened.

It's not necessarily a bad thing to let the failures happen though. Every time a part fails, that particular part will get a permanent buff to failure rates. (in this case by "that part" I mean "all Wolfhound engines" not "that particular Wolfhound Engine that has a specific ID from ScrapYard")

Edited April 13, 2018 by severedsolo

[severedsolo]

@Gordon Dry - I just found your problem. It's not the Worst Highlight that's not working, its the "I've failed" highlighting that's not switching back on when you switch to the vessel. It just happens that your worst part was also the failed part. I'll get this fixed up. (Don't expect anything until 1.4.3 hits though)

Edited April 14, 2018 by severedsolo

[severedsolo]

On 13/04/2018 at 4:20 PM, Gordon Dry said:

It seems that "Highlight worst part" only works when the vessel was launched in the active session / scene.

When I load a vessel from the tracking station this feature doesn't work anymore - and does not write anything suspicious to the log, no NRE all the place.

I think I've fixed this but haven't had a chance to test it yet. If you get a chance can you try the latest dev build and tell me if it helps.

Edited April 15, 2018 by severedsolo

[Gordon Dry]

8 hours ago, severedsolo said:

If you get a chance can you try the latest dev build and tell me if it helps.

I will do next session.

[Beetlecat]

A quick q about the suggested use alongside KCT:

I have a plane with a fuel tank fail mid-flight (though not catastrophically -- made it home. ). I recovered the whole vessel, then hopped in to "edit" the vehicle and maybe swap out the tank. It is still actively leaking in the SPH (really cool! --is this intended?). 

Selecting it, and clicking "use new part" switches the part's menu to indicate that it's brand new, but it's still leaking and marked as the "worst part." Did I do something wrong, or have I stepped into a bug? I can totally scrap this plane, but wondered if you'd like me to hang onto it for science...

[Kwebib]

I believe I have found a bug, but I'm not sure if it's a ScrapYard thing or Oh Scrap thing. Here is the info:

KSP: 1.4.2 Windows 64-bit

Problem: Some parts do not get recovered by ScrapYard when Oh Scrap is installed.

Mods installed:

Scrapyard 1.1.0.107

KCT 1.4.0.69

Oh Scrap 1.3.1

Unmanned Before Manned

Module Manager 3.0.6

Reproduction steps:

Build a "KSC Science Rover" using parts that Unmanned Before Manned puts in Engineering 101. Launch and Recover. Without Oh Scrap, all parts show up in inventory. With Oh Scrap, the battery, antenna, and rover body do not show up in inventory. Picture in spoiler.

Spoiler

Log:

Output Log

EDIT: Might this have something to do with the fact that there is no staging on a vessel like this?

Edited April 16, 2018 by Kwebib

[magico13]

59 minutes ago, Kwebib said:

Might this have something to do with the fact that there is no staging on a vessel like this?

I'm betting it's that. It's been mentioned before that Oh Scrap! requires the vessel to be moved out of the pre-launch status in order to recover parts, which definitely happens with staging and might happen if you start driving the rover around. I believe if the MET timer moves off of zero then it should count.

[Kwebib]

1 hour ago, magico13 said:

I'm betting it's that. It's been mentioned before that Oh Scrap! requires the vessel to be moved out of the pre-launch status in order to recover parts, which definitely happens with staging and might happen if you start driving the rover around. I believe if the MET timer moves off of zero then it should count.

You're right. I just loaded up the rover and drove around and MET never starts.

Edit: I just added a launch clamp on my rover and this solves the issue.  If there isn't an easy solution to this, I could live with having to add a clamp or something else to stage.

Edited April 16, 2018 by Kwebib

[Gordon Dry]

@severedsolo No fun! I tried it with a new test career and the first 6 attempts of launching a sounding rocket failed, also when I staged through always both stages failed to ignite. Just too much of "random" failures...

[severedsolo]

4 hours ago, Beetlecat said:

Selecting it, and clicking "use new part" switches the part's menu to indicate that it's brand new, but it's still leaking and marked as the "worst part." Did I do something wrong, or have I stepped into a bug? I can totally scrap this plane, but wondered if you'd like me to hang onto it for science...

I'm going to guess you used the "Recover Active Vessel" option in KCT? - I've not tested that at all, I might have to write some special logic for that.

1 hour ago, Gordon Dry said:

No fun! I tried it with a new test career and the first 6 attempts of launching a sounding rocket failed, also when I staged through always both stages failed to ignite. Just too much of "random" failures...

That's odd, I didn't change the failure logic at all. Unless I broke something moving that code about. No probs I'll have another look.

[Beetlecat]

1 hour ago, severedsolo said:

I'm going to guess you used the "Recover Active Vessel" option in KCT? - I've not tested that at all, I might have to write some special logic for that.

Yup. That's precisely it. My solution was to re-launch with an engineer to stop the leak. Good as new!

 

[severedsolo]

Heads up - updates and support are going to be slow for the forseeable future:

A moment of silence please. My venerable old desktop has finally died. The indications are that it's a critical hardware failure (I'm suspecting motherboard, but it might be the CPU too - it's definitely something critical anyway). Sadly this machine is a second generation i5 so the chances of getting a replacement component are next to zero. Which means I need to build a new PC, but don't have the money right now (and likely won't until September). RIP old friend, you've been a very faithful companion.

For users of my mods: this means that updates are going to be a bit sparse going forward until I can scrape the money together to replace it.

I have a machine I can use as a backup, but it's my wifes main computer, which means I probably won't be able to get on it that much (she's a writer). My other alternative is a very slow netbook that definitely won't play KSP.