Kerbal Space Program 1.12.3 is live!

[Lisias]

5 hours ago, bonyetty said:

Black screen and half an orange loading bar is all i can get with 1.12.3 for Mac?

Hi! Mac Masoch   User here. At least for a clean install (with both DLCs), the thing loaded fine on a MacMini 6,2 running Mojave.

Open a Thread with the full KSP.log and Player.log (I think we may need this one too) here if you have modded it, or here if it's vanilla.

Spoiler

 

 

Edited December 14, 2021 by Lisias
better phasing

[Anth]

@dok_377Heres where it be useful:

1. If I have a base that has some robotics that was made only for the trip, when I get to the destination I extend the robotics and then lock them they stay locked and if I designed the base piece/vehicle correctly the drift happens once (and slightly) and never again because the craft was only made to have its robotics used once.

2. If I have a craft that is in storage at my base and I use it once in a while but I visit the base repeatedly then that craft doesn't 'drift' repeated times. it just 'drifts' every time I actually need to use it.

I don't know what the code looks like for the robotic drift lock fix compared to the docking port fix but I did test it in its early stages for @JPLRepo The docking port fix was basically one line of code. The robotic drift lock fix was more complicated. I am betting he put his hand up to Private Division and said, 'hey guys lets do this one last thing for the KSP1 community?' with the possibility of getting his hand slapped down. But they said yes instead. Private Division didn't need to do this patch. And @JPLRepodid most of it in his own spare time.

Thank You @JPLRepofor doing this. 

Edited December 14, 2021 by Anth12

[Gotmachine]

13 hours ago, SQUAD said:

You can still  help us find bugs/issues by reporting them into our bugtracker

Do we really ?

The last time the public bugtracker saw any official activity was 4 months ago, it has become an untriaged mess, issues fixed in 1.12+ weren't even closed there.
I've posted a few issues or comments on existing issues with full details on the exact cause of some (rather major) bugs. Some of those have trivial fixes. If anybody was giving a look from time to time, those issues would have been noticed, I think...

That's fine. KSP 1 development has stopped and no resources are being dedicated to its development anymore, life goes on.
But those kind of statements are misleading, and the right course of action would be to make the bugtracker read-only,  disallowing adding new issues until someone has the time to triage it. An unmaintained bugtracker is worse than a closed one.

[kerbiloid]

Spoiler

8 hours ago, bonyetty said:

Black screen and half an orange loading bar is all i can get with 1.12.3 for Mac?

The Mac users can order a loading bar of any official color.

Spoiler

 

 

[Serenity]

Thank you so much for this important fix!

This is one of the nicest Christmas gifts someone can get!

Edited December 14, 2021 by Serenity

[Just a random person]

Thank you so much!

[panzer1b]

Im happy to see some effort put forwards especially to solve this rather gamebreaking bug.  That said, im really hoping that KSP2 isnt as broken as the last few updates of KSP1, and im pretty damn certain that there wont be any fix to the extremely infuriating memory leak/performance degradation when you spend enough time in KSP and do alot of swapping vessels/launching/relaunching.  Its playable, but high part counts (as in anything over 600) coupled with this mem leak makes it very hard to enjoy the game without consiantly restarting the game every hour or so...

 

Its been a great run all these years, and while i dont have any hopes for my personal pet peve being fixed (mem leak), at least the game was well worth the minimal cost and still remains fun to this day...

Heres to all this time spend in KSP1 ...

And a rather fitting screenshot, the bottom is my aurora from like 0.25, then a mid one in prolly 1.1, and finally my most recent model that i finished a few weeks ago...  Now if only i could find my 1st warship i ever made in KSP from like 0.22 or 0.23 when i got the game, i have like 2GB of craft files archived and i forgot where it was and what it was even named...

[AndrewsonPaul]

 good launch rocket  1.12.2

[Casper_83]

Does this mean that this is fixed?

 

 

[Knokke]

The bug with docking ports doesn't seems to be fixed at all. Worse, it seems to have added more bugs instead of fixing anything. Now it is impossible to deploy an experiment from the inventory of a Kerbal to the surface (bug happened while trying to have an engineer deploying a control station in the tundra biome next to the Glacier Lake launchpad).  The experiment disappears from the Kerbal's inventory when deployed, but doesn't deploy on the ground. The icon stays floating in the air, following the Kerbal as if the inventory menu was still opened, showing only the UI icon of the control station. It was possible to deply an experiment without issue before 1.12.2. I don't know if this bug was introduced with 1.12.2 or 1.12.3

It's not an issue with any mods: this is from a fresh vanilla install.

[KerikBalm]

On 12/14/2021 at 7:51 AM, Anth12 said:

@dok_377Heres where it be useful:

1. If I have a base that has some robotics that was made only for the trip, when I get to the destination I extend the robotics and then lock them they stay locked and if I designed the base piece/vehicle correctly the drift happens once (and slightly) and never again because the craft was only made to have its robotics used once.

2. If I have a craft that is in storage at my base and I use it once in a while but I visit the base repeatedly then that craft doesn't 'drift' repeated times. it just 'drifts' every time I actually need to use it.

I don't know what the code looks like for the robotic drift lock fix compared to the docking port fix but I did test it in its early stages for @JPLRepo The docking port fix was basically one line of code. The robotic drift lock fix was more complicated. I am betting he put his hand up to Private Division and said, 'hey guys lets do this one last thing for the KSP1 community?' with the possibility of getting his hand slapped down. But they said yes instead. Private Division didn't need to do this patch. And @JPLRepodid most of it in his own spare time.

Thank You @JPLRepofor doing this. 

So if I am correct, this patch means that no additional drift occurs when parts are locked.

As it was, no relevant drift occurs in orbit when not thrusting and not rotating... ie no forces on the robotic parts. Perhaps tiny drift occurs due to floating point errors?

It seems to me that the issue is just using robotics on the surface of a body... but low gravity and/or low mass loading should be fine.

I never noticed significant drift with rotors that only had some props on them.

Still, I think I will try to have everything get back to 0 G before saving

[Anth]

1 hour ago, KerikBalm said:

As it was, no relevant drift occurs in orbit when not thrusting and not rotating... ie no forces on the robotic parts. Perhaps tiny drift occurs due to floating point errors?

I have never found any proof that floating point errors affect robotic parts and I have done months of testing of robotic drift to try to report it on the bug tracker in the right way. I even made a mod that reports parts coordinates to understand how it worked. Nothing pointed to floating point errors.

I have have heard from multiple sources that floating point errors cause robotic drift. I have found no evidence of that.

 

 

[KerikBalm]

8 hours ago, Anth12 said:

I have never found any proof that floating point errors affect robotic parts and I have done months of testing of robotic drift to try to report it on the bug tracker in the right way. I even made a mod that reports parts coordinates to understand how it worked. Nothing pointed to floating point errors.

I have have heard from multiple sources that floating point errors cause robotic drift. I have found no evidence of that.

 

 

So then, as long as one locks parts when there is no acceleration (ie, not thrusting, not rotating, not in an atmosphere nor on a surface), then there won't be any robotic drift?

Edited January 8, 2022 by KerikBalm

[DerkyJerkyWreaksHavoc]

On 12/13/2021 at 1:25 PM, KSPStar said:

Hello Kerbonauts!

Kerbal Space Program 1.12.3 is live!  This small patch addressed drifting issues with robotic parts and rotating docking ports.

Changelog:

=================================== v1.12.3 ============================================================
+++ Bugfixes
* Fix AOORE in Assembly Loading of Duplicate Mod DLLs causing issues with loading of some mods.
* Improved robotic parts drift when robotic parts are left locked. Players can now quicksave/reload or timewarp to reset part placement. Note: The drift issue can only be corrected when the robotic parts are set to locked mode.
* Improved rotating docking ports drift when rotating docking ports are left locked. Players can now quicksave/reload or timewarp to reset part placement. Note: The drift issue can only be corrected when the rotating docking ports are set to locked mode.

+++ Modding
* Add UpgradePipeline support for PartModule renaming.

 

all the more reasons to learn coding. Nice Job!

 

[Anth]

5 hours ago, KerikBalm said:

So then, as long as one locks parts when there is no acceleration (ie, not thrusting, not rotating, not in an atmosphere nor on a surface), then there won't be any robotic drift?

This video shows orgPos and orgRot in the action menu. This what my mod does. 

• orgPos is the distance away from the root part x,y,z
• orgRot is the rotation x,y,z,w

 In the video they are changing every time I press F5, that exact number is recorded into the persistent file and the quicksave each time I do it in the video. This is what causes robotic drift.

That deviation in space looks to be around 0.000001 metres to 0.000003 metres as long as the craft is at full rest and isn't flexing (recovering from a wobble for example) if its unlocked.

Its extremely slight, but its not eliminated...

 

 

Edited January 9, 2022 by Anth12

[KerikBalm]

8 hours ago, Anth12 said:

That deviation in space looks to be around 0.000001 metres to 0.000003 metres as long as the craft is at full rest and isn't flexing (recovering from a wobble for example) if its unlocked.

Its extremely slight, but its not eliminated...

Ok, well then, looks like I will do what I was doing, just not as much ... editing of the save file to reset the part positions/ undo drift.

So, for a robotic craft, I will save before unlocking anything, then go copy the save file text corresponding to that shift to a new text file. When the craft is noticably affected by robotic drift, I will again copy the relevant ship text from the save file, us MS words compare function to compare the drifted state to the original, and "revert" those changes, while keeping the other changes such as bodies visited, science stored, fuel, orbit, etc

[Anth]

0.000001 metres to 0.000003 metres isn't too bad

[Krazy1]

On 12/13/2021 at 10:25 AM, KSPStar said:

You can still  help us find bugs/issues by reporting them into our bugtracker,

What about the repeatable crash to desktop I reported that wasn't even confirmed? You're not going to fix that? It seems like bug reporting is a waste of time.

I want to be excited about KSP 2 but I can't be if KSP 1 isn't left in good shape at the end. 

[Anth]

@Krazy1Does the mentioned counter in that bug report stop the crash to desktop? 

I have over 7000 hours in KSP at a guess. The more complicated the craft or mission I am trying to do the more bugs I have to counter.

There's just too many issues with the game to take it to the point of being in 'good shape' without spending too much money on it with no real return on investment.

The following is a theory that I have:

• T2 bought KSP to make KSP2.
• T2 got Squad to release 2 DLC's to see how interested the current fanbase was still interested in the game to see how much money to throw at KSP2
• T2 could have helped Squad with more support as to release more quality updates but didn't.
• KSP1 was being supported to keep people interested for KSP2

Even the contrast of going from a buggy KSP1 to a stable smooth as silk (hopefully) KSP2 will increase sales.

The Outer Worlds is a game released under the Private Division banner. Its sales were so successful they are making a sequel. 

T2 don't care about KSP1. They care about KSP2 because its going to make more money than The Outer Worlds will. A lot more.

[AtomicTech]

7 hours ago, Anth12 said:

@Krazy1Does the mentioned counter in that bug report stop the crash to desktop? 

I have over 7000 hours in KSP at a guess. The more complicated the craft or mission I am trying to do the more bugs I have to counter.

There's just too many issues with the game to take it to the point of being in 'good shape' without spending too much money on it with no real return on investment.

The following is a theory that I have:

• T2 bought KSP to make KSP2.
• T2 got Squad to release 2 DLC's to see how interested the current fanbase was still interested in the game to see how much money to throw at KSP2
• T2 could have helped Squad with more support as to release more quality updates but didn't.
• KSP1 was being supported to keep people interested for KSP2

Even the contrast of going from a buggy KSP1 to a stable smooth as silk (hopefully) KSP2 will increase sales.

The Outer Worlds is a game released under the Private Division banner. Its sales were so successful they are making a sequel. 

T2 don't care about KSP1. They care about KSP2 because its going to make more money than The Outer Worlds will. A lot more.

That actually makes a lot of sense!

[Lisias]

12 hours ago, Anth12 said:

There's just too many issues with the game to take it to the point of being in 'good shape' without spending too much money on it with no real return on investment.

This is not exactly the most desirable of explanations, as this means that Squad/TT2 were deceiving us all this time. What may make them liable to a class action, more or less the same as happened to CD Projekt RED due Cyberpunk: they are intentionally selling us a flawed product.

 

12 hours ago, Anth12 said:

T2 don't care about KSP1. They care about KSP2 because its going to make more money than The Outer Worlds will. A lot more.

Perhaps. But this doesn't changes my previous statement - if KSP1 is flawed beyound fix, keeping them on sales are not exactly the best of the moves - at least under the legislation of some USA states.

I bought KSP on the 1.3.1 era, but only started to play with it about 1.4.1 - and until 1.4.3, I was pretty happy with the product (being the reason I'm playing on this version nowadays). So I'm one customer that, well, is not really affected by current KSP problems. But anyone that bought the game recently may not think the same.

IMHO, Private Division should spend a little more money with at least one good developer to tackle down the worst problems on the bug track - taking advantage that some skilled 3rd parties had even posted the solution for some of the problems there, and issue a KSP 1.12.3 release the fast as it's possible. This can be cheaper than trying to clean up the stain later, if by any reason KSP2 is launched with some bugs (and it's almost surely it will - the tracked record of the development doesn't inspire too much optimism).

Actions have consequences. Lack of actions too.

[MechBFP]

3 minutes ago, Lisias said:

This is not exactly the most desirable of explanations, as this means that Squad/TT2 were deceiving us all this time. What may make them liable to a class action, more or less the same as happened to CD Projekt RED due Cyberpunk: they are intentionally selling us a flawed product.

 

Perhaps. But this doesn't changes my previous statement - if KSP1 is flawed beyound fix, keeping them on sales are not exactly the best of the moves - at least under the legislation of some USA states.

I bought KSP on the 1.3.1 era, but only started to play with it about 1.4.1 - and until 1.4.3, I was pretty happy with the product (being the reason I'm playing on this version nowadays). So I'm one customer that, well, is not really affected by current KSP problems. But anyone that bought the game recently may not think the same.

IMHO, Private Division should spend a little more money with at least one good developer to tackle down the worst problems on the bug track - taking advantage that some skilled 3rd parties had even posted the solution for some of the problems there, and issue a KSP 1.12.3 release the fast as it's possible. This can be cheaper than trying to clean up the stain later, if by any reason KSP2 is launched with some bugs (and it's almost surely it will - the tracked record of the development doesn't inspire too much optimism).

Actions have consequences. Lack of actions too.

I highly recommend you take some time to better understand what constitutes a defective product.

[Lisias]

9 minutes ago, MechBFP said:

I highly recommend you take some time to better understand what constitutes a defective product.

I highly recommend Private Division to do so. I'm just the messenger, I'm not the one selling KSP1.

[MechBFP]

36 minutes ago, Lisias said:

I highly recommend Private Division to do so. I'm just the messenger, I'm not the one selling KSP1.

Oh, well that is good then.

[Krazy1]

On 1/9/2022 at 9:54 PM, Anth12 said:

Does the mentioned counter in that bug report stop the crash to desktop? 

I saw that. IDK I'll try it sometime. 

On 1/9/2022 at 9:54 PM, Anth12 said:

The more complicated the craft or mission I am trying to do the more bugs I have to counter.

Yeah, or the longer I play in a session. I made a new save game with 1 ship and it still crashed for that particular bug. 

On 1/10/2022 at 10:34 AM, Lisias said:

the tracked record of the development doesn't inspire too much optimism

I started playing on 1.8... not very optimistic.