Devnote Tuesday: Wednesday Edition IV

[SQUAD]

Hello everyone!

Work continues at a solid pace here: both patch 1.1.3 and the plans for update 1.2 are taking shape. We’re sure you’re curious about what we’ve been doing, so let’s get these devnotes started (say heeeee-hey, whoooo-oh!)

To start with the best news we can bring: as far as we can tell we’ve hunted down the crashes that were causing issues with version 1.1.2 and we’re very eager to release the patch. Mike (Mu), Nathan (Claw) and Jim (Romfarer) also spearheaded an effort to plug a number of memory leaks and reduce memory footprint overall, which should increase the stability of the game even further. Currently the patch is in “feature lock”, meaning we’re focusing solely on fixing known issues, and we’re preparing for experimental testing before we release. It’s taking a bit longer than usual to develop this patch, but we’re sure the quality will be worth the wait.

Just before we went into feature lock, Jim changed the game to pause when you’re in career mode and enter a KSC facility. This should prevent contracts expiring when you’re looking at the administration center, or hiring new astronauts, or selecting new contracts, and you haven’t killed timewarp before entering the building. You can now relax and take all the time you need to pick the contracts you want.

Progress has also been made with regards to the decaying orbits. Bill (Taniwha) and Nathanael (NathanKell) receded into their offices, crunched a lot of numbers, tested hundreds of cases and came back with a number of code changes that have been described as magical. Spinning references frames and sharpened swizzles had to be slain but orbits are now rock solid. Nathanael also fixed a number of miscellaneous issues: asteroid mass indicators giving faulty information, and an issue where a faulty PartModule could kill all parts if it failed to load or start properly. A good old try/catch method now limits the damage in these cases.

Brian (Arsonide) tackled an issue where  EVA’d kerbals interacted violently with wheel suspension raycasts. The Kerbals are now moved onto a special “EVA” layer in the game which ignores the suspension colliders. A side-effect was that this new layer was not being lit by any light sources in the game, which required Brian to look up all light sources in the game and adding the layer to the culling masks. A large task, but we can’t make it too easy for Danny2462 to break the game!

The patch is currently in QA testing and serves as a first test for the new QA and experimental test procedures that Ted and Dave (TriggerAu) have redesigned based on the feedback from the 1.1 testing periods. New trackers and workflows should help to streamline the testing process and communication.

On to the KSP 1.2 news then. We’ve decided to (again) overhaul large parts of the wheels coding for 1.2. Vehicle Physics Pro, the middleware used for the wheels, has received many updates since we last updated it, and between those updates and the large amount of patching that we implemented for both Unity issues that are/will be fixed by an update to version 5.3 or 5.4 it’s easier to start with a blank page.

Career mode will receive some changes as well: in addition to the tweaks to contracts we talked about last week, Brian will look into ways to make reputation more valuable and useful in the game, and he’s also looking to give players more control over the contract weighting.

The telemetry and antenna updates are being dusted off, and have been made to work with the latest Unity versions. Bob (RoverDude) is finishing up the balance of the starting antennas and is working on adding a new, non-deployable version of the Communotron 16 that is suitable for atmospheric use. This will come in handy for communication with unmanned aircraft and rockets alike.

On the community front Andrea (Badie) and Kasper (KasperVld) executed a forum update last week which, despite a few hiccoughs along the way, succeeded. A minor update is coming up later this week to deal with the username mentions generating broken links, but that should not cause much problems at all. In case you missed it, Kasper also hosted a quick AMA on the Kerbal Subreddit, and time permitting is looking to host one on the forums as well.

Testing the console versions of the game is progressing at a good pace as well, and the game will re-enter certification once a few final issues have been worked out. Pablo (Paul Amsterdam) has been helping out with the testing and created a short console trailer.

Mathew (sal_vager) lifts our collective spirits once more with an inspirational poem:

The red, the blue, the green.

We toil in these redmine fields
Our hands calloused in pain
Faces shine by dim screen light
While keys compress membrane

Screen refreshed we see the blue
Our hopes and spirits rise
We search out the foe but they’re not found
Joy becomes our eyes

Fields change from blue to green
No sight is more yearned for
Souls soaring, job done well
Our foe defeat, bugs no more

That wraps up the devnotes for this week! As always you can follow us and ask questions on our forums, on Facebook and Twitter, and on Reddit.

[Galileo]

Can't wait!!! Love the poem btw!  

Edited June 8, 2016 by Berlin

[Choice // SLOTH Airlines]

Yay, devnotes just when i was going to check!

[Perry Apsis]

42 minutes ago, SQUAD said:

Mike (Mu), Nathan (Claw) and Jim (Romfarer) also spearheaded an effort to plug a number of memory leaks and reduce memory footprint overall,

YES!

42 minutes ago, SQUAD said:

Currently the patch is in “feature lock”, meaning we’re focusing solely on fixing known issues, and we’re preparing for experimental testing before we release.

YES!

42 minutes ago, SQUAD said:

It’s taking a bit longer than usual to develop this patch, but we’re sure the quality will be worth the wait.

YES!

The most reassuring message I've heard from Squad in a long time. I can't wait for the patch.

But I will.

Keep up the good work!

[Gojira1000]

You have made me glad.

[Adelaar]

Yay for the orbital decay bug to be solved. I have stopped construction on my space station(s) due to this bug, so when 1.1.3 is released I can get back to building stuff

[RW-1]

" non-deployable version of the Communotron 16 "

Wait, a non-deployable antenna? Or did you mean non-retractable?  

[Nerfclasher]

Obit decaying has been annoying but I'm glad that bug has been fixed

[cantab]

No new lead developer announcement?

[adsii1970]

Wow, excited to hear of the progress on the patch and on the memory leak issues.

[monstah]

...meanwhile, my pitchfork and torches gather dust in a forgotten corner...

[Jarin]

Crash fix yesssssssss

1 hour ago, RW-1 said:

" non-deployable version of the Communotron 16 "

Wait, a non-deployable antenna? Or did you mean non-retractable?  

"non-deployable" = "one-state-only" or always-on. Does not require deployment to function.

[MiniMatt]

Question, the answer to which I appreciate you may not be in a position to give (for entirely understandable, valid, reasons):

I, perhaps in error, gathered an impression that a major part of the focus for 1.2 would be a graphics overhaul. Is this, in any way, still on the cards and are you in a position to flesh out any detail? (eg. are we talking part models, environment textures, atmosphere/lighting effects?)

[Glaran K'erman]

Ya know, didn't even realize timewarp persisted when entering buildings, never even tried it!
@Arsonide I am really looking forward to contract weighting having more flexibility. Often find myself saying, "Oh no! But I wanted to go x and do y!" We all have our favorite body to visit and it's not always the most difficult or far away, would be nice to still have a career focus on where the user wants to go and what they want to do there.

Great poem by the way, @sal_vager!

[Tw1]

23 hours ago, SQUAD said:

 

Magnificent work. See they are getting the bugs. 

 

I remain sceptical about the rumoured new rocket parts though, Porkjet can certainly model, but design choices with the lab interior and mk1 cockpit still have me sceptical. 

I enjoy the thrown together aesthetic of many parts, -the semi improvised nature makes it makes them feel more of an achievement when you've got your junk into orbit,It's more of a challenge to produce nice looking vessels. 

To much uniformity also limits the looks we can achieve. So if the rocket parts are being overhauled, please keep the key characteristics of each one. Give it careful thought. Impress me.

Edited June 9, 2016 by Tw1

[John FX]

3 hours ago, SQUAD said:

Currently the patch is in “feature lock”, meaning we’re focusing solely on fixing known issues, and we’re preparing for experimental testing before we release.

Oh you!

You DO love us

You DO listen to our concerns

We love you too.

EDIT :

3 hours ago, SQUAD said:

in addition to the tweaks to contracts we talked about last week, Brian will look into ways to make reputation more valuable and useful in the game

 

Are you following me?

 

EDIT 2:

1 hour ago, RW-1 said:

" non-deployable version of the Communotron 16 "

Wait, a non-deployable antenna? Or did you mean non-retractable?  

 

I think they mean an antenna which will not break when in the wind but is still on and working. They mean it works without having to be deployed.

Edited June 8, 2016 by John FX

[Kerbart]

3 hours ago, SQUAD said:

(...)

The patch is currently in QA testing and serves as a first test for the new QA and experimental test procedures that Ted and Dave (TriggerAu) have redesigned based on the feedback from the 1.1 testing periods. New trackers and workflows should help to streamline the testing process and communication.

(...)

I really hope that it is not just external communication, but external as well. Most of the angst and anxiety in forum complaints comes from the feeling that bugs are not being addressed. Knowing that the bugs are being worked on, and that solving issues it not just a matter of inserting a single line MakeProblemGoAway = True; at a random spot hopefully helps in emphasizing that, yes, it's being worked on and no, it's not an overnight fix.

(But we already do that with the bugtracker: Lesson learned from Joel Spolsky: if the user is complaining, then the developer did not enough to communicate)

[RW-1]

12 minutes ago, John FX said:

Oh you!

You DO love us

You DO listen to our concerns

We love you too.

EDIT :

 

Are you following me?

 

EDIT 2:

 

I think they mean an antenna which will not break when in the wind but is still on and working. They mean it works without having to be deployed.

I can see that in terms of a non-deployed 16, possibly having much less range if retracted and still functional, or in terms of a shorter antenna that is just on, which I can say first, it should be named "Stubby"

[Dman979]

Awesome work, @sal_vager and the rest of the crew!

Hey, Sal, wanna post something, so I can rep you directly? That was a good one.

[Claw]

29 minutes ago, Glaran K'erman said:

Ya know, didn't even realize timewarp persisted when entering buildings, never even tried it!

Entering buildings would stop time warping (that was fixed a version or two back), but the game didn't completely pause. Time moved forward at 1x rate. Now, when entering buildings such as admin or mission control, the game completely pauses. That sounds very minor, but it actually required changing a lot, since the animations and a few other things relied on specific calls that only occurred when time passed in-game.

 

47 minutes ago, MiniMatt said:

I, perhaps in error, gathered an impression that a major part of the focus for 1.2 would be a graphics overhaul.

The graphics overhaul is for rocket parts.

 

13 minutes ago, Kerbart said:

I really hope that it is not just external communication, but external as well.

We don't usually work inside the main tracker, but rather copy issues over into a development tracker. This makes it easier for the development team to see what's being focused on, and enables tester to better see what's been fixed and ready for testing. It's a smaller subset that we can better control.

[sal_vager]

4 minutes ago, Dman979 said:

Awesome work, @sal_vager and the rest of the crew!

Hey, Sal, wanna post something, so I can rep you directly? That was a good one.

Nah it's okay, I have enough rep

[juanml82]

I hope 1.2 wheels work fine.

For the time being, I'm still driving my 1.0.5 rover on Tylo and by the time I'm done getting two more biomes plus transferring the crew to the Vall rover, exploring Vall and then continue to Laythe.... well, I may just jump from 1.0.5 to 1.2

[Violent Jeb]

Good news. While i'm glad to see progress on the 1.1.x patch is going logically, its unfortunate that the focus is still ultimately split with down the road things in 1.2. work on one update at a time maybe? Why not have everyone fix bugs for a little while, so 1.1.x can come out sooner (maybe it wouldn't take longer to develop that way, with everybody working on it), then worry about the next update? I obv know nothing of this process.

At least 1.1.x will be something to look forward to from the sounds of it. esp glad about the orbits.

Good devnotes. hopefully the patch isn't too far away.

 

4 minutes ago, juanml82 said:

I hope 1.2 wheels work fine.

For the time being, I'm still driving my 1.0.5 rover on Tylo and by the time I'm done getting two more biomes plus transferring the crew to the Vall rover, exploring Vall and then continue to Laythe.... well, I may just jump from 1.0.5 to 1.2

1.1's new features made lots of new bugs. Historical evidence suggests that 1.2's new features will similarly make a lot of new bugs/instability. I expect 1.2.2/3/4 to be stable releases but have little hope for 1.2

 

Edited June 8, 2016 by Violent Jeb

[Arsonide]

Just now, Violent Jeb said:

Good news. While i'm glad to see progress on the 1.1.x patch is going logically, its unfortunate that the focus is still ultimately split with down the road things in 1.2. work on one update at a time maybe? Why not have everyone fix bugs for a little while, so 1.1.x can come out sooner (maybe it wouldn't take longer to develop that way, with everybody working on it), then worry about the next update? I obv know nothing of this process.

At least 1.1.x will be something to look forward to from the sounds of it. esp glad about the orbits.

Good devnotes. hopefully the patch isn't too far away.

This is what is happening. Everybody is focused on bug fixing, and 1.1.3 at the moment. 1.2 is just in the planning stages. The only exception to this is that there were a few features that some of us wanted to work into 1.1, but weren't quite ready for release. Those features are done for the most part, and ready for 1.2. We still enjoy talking about them in our developer logs though.

[Violent Jeb]

Well that just made my day, thanks for the clarification Arsonide. I'll be on the hype train if anybody is looking for me