Squadcast Summary (2016/01/28) - What's not in 1.1 edition

[MiniMatt]

Scrappy summary as this week I didn't get to watch it live but relied upon the archived stream at http://www.twitch.tv/ksptv/profile/past_broadcasts to cobble together a summary - as this doesn't include the chat history some nuance, and indeed some salient points, have likely been missed.

This week saw Dr Turkey, Kasper & NathanKell being rather coy: lots of "not quite ready to talk about that" and "we've got lots on this but can't quite say anything about it yet". As such perhaps the only definite points which could be gleaned were things that won't be in 1.1 (tl;dr: weather, sonic booms, part malfunctions, 9th planet analog, wide ranging graphics update) but that sounds too negative a summary of what was an enjoyable hour's stream. Genuinely interesting diversions into NASA & the USAF's design choices for the shuttle, the stumbling blocks when developing for consoles, and much else besides. If you've an hour free I'd urge everyone to check it out - especially if you've ever struggled with designing a classic space shuttle.

Without further ado, my notes as I watched - again I stress I didn't see the chat stream so have likely missed bits, misinterpreted bits, and over or under emphasised bits.

(5 mins) launching a simple comm-sat into a geostationary kerbin orbit - looks a little low on dv

(6 mins) this squadcast won't be played on an inhouse 1.1 but on production 1.0.5

(7 mins) comm-sat launch is looking very smooth, and - what do I know - has plenty of dv

(9 mins) geostationary orbit changed slightly in 1.0.5. In 1.0 the solar day was a couple minutes too short, fixing this in 1.0.5 has necessarily changed geostationary orbit altitude (very slightly)

(11 mins) Ion powered comm-sat delivered, the 2nd stage is returning to a kerbin burn-up orbit - very tidy! 

(13 mins) secret footage on it's way - the tab title is "SecretEnchantedGermanShorthairedPointer" - might it be SQUAD DOGGIE??

(14 mins) secret footage is not squad doggie  , it's a unity themed meme-ified gif of a recent news story

(16 mins) KSP might not be coming to XBOX360 any time soon

(17 mins) hints at revisiting easter eggs, possibly a story chain

(18 mins) now building a shuttle

(18 mins) 1.1 will not have weather

(18 mins) ksp is currently multithreaded but the physics is all singlethread, v1.1 is allowing some multithreading of physics

(21 mins) "release date in 2016 for sure"

(23 mins) 9th planet analog not in design plans yet

(24 mins) definitely no plans for random part malfunction

(26 mins) graphical update is in the plans (feeling this is 1.2+)

(27 mins) shuttle build is hurting for lack of dv/twr mod, and missing a wing wheel

(28 mins) collision boxes around scatter objects are being looked at (didn't get the feels this would be a 1.1 feature)

(32 mins) - shuttle launch is.... unwieldy

(34 mins) v1.2 is in internal discussions, graphical updates are mooted as being seriously considered

(36 mins) shuttle seperation - during a spin - smashed off the rear engines and rudder

(37 mins) going to try to bring it down unpowered, with two wheels & no rudder

(39 mins) shuttle overheated and exploded, no survivors

(42 mins) porkjet has been working on some engine models (regex, of this parish, was asking about titan models, so a possible hint, think it implied reskins of existing engines rather than additions see clarification posts by regex & nathankell below)

(43 mins) relaunched shuttle is looking a teensy bit better

(48 mins) shuttle orbit almost achieved but unwieldy in orbit, revert to redesign, adding more vernor engine RCS to brute force control issues

(50 mins) stock delta-v indicator hasn't been shelved, the work previously done on it has not been scrapped

(57 mins) shuttle orbit acheived

(1h ish) something here I didn't pay enough attention to - NathanKell (I think) talking a bit about re-writing the thermal system for a bit more speed, warning that the thermal system wasn't that significant an overhead to begin with, and I think noting that multi-threading isn't in every case a simple panacea, here be dragons etc.

(1h 4 mins) 1.1 will not have sonic booms, something to consider for future versions

(1h 5 mins) Fin

Edited February 1, 2016 by MiniMatt
42 minute clarification

[Mitchz95]

34 minutes ago, MiniMatt said:

(34 mins) v1.2 is in internal discussions, graphical updates are mooted as being seriously considered

A bit confused; does that mean it is or isn't being seriously considered?

Thanks for the summary.

[MiniMatt]

26 minutes ago, Mitchz95 said:

A bit confused; does that mean it is or isn't being seriously considered?

Thanks for the summary.

Sorry, confusion is entirely down to my scrappy transcribing and affection for tautologies.

Gist I got was that "the scope of 1.2 is already being discussed internally, and wide ranging graphical updates are at the forefront of those discussions" - but couch that with enough wiggle room to back out if unforseen problems or exciting new things to focus on emerge - ie. they're not yet ready to categorically state "these are our plans", at least not in a world where dev statements are taken as gospel and people throw a hissy if "promises" aren't kept.

[Majorjim!]

48 minutes ago, MiniMatt said:

(1h ish) something here I didn't pay enough attention to - NathanKell (I think) talking a bit about re-writing the thermal system for a bit more speed, warning that the thermal system wasn't that significant an overhead to begin with, and I think noting that multi-threading isn't in every case a simple panacea, here be dragons etc.

 

The thermal system is not good at present and I am very happy to see it being toned down for performance gains. Slightly concerned by his comment though.. What exactly do you/he mean by "here be dragons"? I am aware of its usage, I am curious as to why it was said.

Edited January 31, 2016 by Majorjim

[Norpo]

4 minutes ago, Majorjim said:

The thermal system is not good at present and I am very happy to see it being toned down for performance gains. Slightly concerned by his comment though. What exactly do you/he mean by "here be dragons"? I am aware of its usage, I am curious as to why it was said.

Based on the context I'm going to guess that he means that in regards to multithreading; Multithread programming isn't exactly easy; and you're lucky if it goes near-perfectly without bugs. This image sums it up quite nicely:

[Majorjim!]

Just now, Norpo said:

Based on the context I'm going to guess that he means that in regards to multithreading; Multithread programming isn't exactly easy; and you're lucky if it goes near-perfectly without bugs. This image sums it up quite nicely:

Riight ok, so the 1.1 update will break the game. ok cool got it. My soul is prepared.

Edited January 31, 2016 by Majorjim

[Thiagobs]

8 minutes ago, Majorjim said:

The thermal system is not good at present and I am very happy to see it being toned down for performance gains. Slightly concerned by his comment though.. What exactly do you/he mean by "here be dragons"? I am aware of its usage, I am curious as to why it was said.

" that multi-threading isn't in every case a simple panacea, here be dragons etc."

I guess it should be read as "Multi-threading isn't a solution for every problem (panacea) and may even cause unforeseen problems (here be dragons)"

[Majorjim!]

Just now, Thiagobs said:

" that multi-threading isn't in every case a simple panacea, here be dragons etc."

I guess it should be read as "Multi-threading isn't a solution for every problem (panacea) and may even cause unforeseen problems (here be dragons)"

Those are the conclusions I came to as well. I was just hoping for a little more clarity from the op.

[katateochi]

Thanks for the summary! 

1 hour ago, MiniMatt said:

(tl;dr: weather, sonic booms, part malfunctions, 9th planet analog, wide ranging graphics update)

well, can't say I was expecting any of those in 1.1, so that's good, no change there! 

But had part malfunctions been brought back onto the possible list? I thought that was something that they'd said they never wanted to add.  imo things explode quite enough without adding a dice roll of doom.
By weather does that just mean visual clouds or was that talking about having actual wind that could knock you off course.  the stock game really needs clouds, but only if they can be done better than the current mods do (which are pretty sweet these days). Not sure how I'd feel about wind.  Would be great for the more aircraft oriented folk, less good for those who want repeatability in their rocket launches. Well if they do add wind, I just hope the update is called the "I'm a leaf" update.

I'm glad to see that a graphical updates (and several of the other things mentioned) aren't planned for 1.1.  It's going to be a big release anyway with plenty for the QA team to deal with (not to mention the modding community once it drops).  I'd also like to get a better comparison of how 1.1 performs vs 1.0.5, a big graphics update and other new features would make that comparison harder to make.  If 1.1 was "just" the port to U5 and the changes necessary to make that work, I'd be very happy. 

[MiniMatt]

Yep, again this is just confusion arising from my interpretation of a half heard off the cuff comment made at the end of an hour long stream, I'd encourage everyone to watch and listen to the stream for more clarity.

But as I understood it - and please note I really only half heard this point - it was noting that "yes the thermal system will be multi-threaded, but modestly so, it's not just a case of pressing a button and hey-presto it's multi-threaded. What's more with the overhead of the the thermal system being relatively low to begin with, writing some ornate multi-thread handling code to spread that over 16 parallel processes is risking all manner of bugs for very little gain, better to lightly breathe some multi-threaded goodness over it and keep it simple, stable, and a bit faster".

[katateochi]

28 minutes ago, Norpo said:

Based on the context I'm going to guess that he means that in regards to multithreading; Multithread programming isn't exactly easy; and you're lucky if it goes near-perfectly without bugs. This image sums it up quite nicely:

yep, those threads are sharing common resources, not thread-safe! 

26 minutes ago, Majorjim said:

Riight ok, so the 1.1 update will break the game. ok cool got it. My soul is prepared.

tbh this is my primary expectation for 1.1! So anything else is going to be a pleasant surprise! 
I'm looking forward to seeing the multitude of new ways we can break the game! I hope Danny2462 is well rested and ready for it!

Edited January 31, 2016 by katateochi

[Superfluous J]

Thanks for the summary!

4 hours ago, MiniMatt said:

(17 mins) hints at revisiting easter eggs, possibly a story chain

(24 mins) definitely no plans for random part malfunction

These both make me happy, for what the first adds and what the second assures won't be added.

[regex]

9 hours ago, MiniMatt said:

(42 mins) porkjet has been working on some engine models (regex, of this parish, was asking about titan models, so a possible hint, think it implied reskins of existing engines rather than additions)

Not reskins, no.  What I was asking about was something like the LR-87-7 (NathanKell knew what I was asking about):

This is a single engine with two combustion chambers and bells, with shared turbopumps and hardware.  I'm wrong here, two engines on a single frame.  A lot of Russian engines use this sort of layout, although they usually have fixed combustion chambers and either gimbal the entire engine or have verniers.  The LR-87-7 gimbals each combustion chamber engine on one axis for roll control, not sure if they can do pitch maneuvers.  Last time I went to the Evergreen Aviation Museum (If you live in Oregon or Washington do yourself a favor and make a trip, it's fantastic) you could only move the bells together for roll.

Interestingly, this same engine uses the fuel mixture that LFO's thermal properties are modeled from, Aerozine50/N2O4.

E: Definitely not asking about some monstrosity like the Twin Boar, which I find fairly useless because of the integrated tank.  Although if it had the tank removed and the bells tightened up, I'd be okay with that.

Edited February 1, 2016 by regex

[NathanKell]

And I said you'd be satisfied with the results of Porkjet's labors.

Being as mysterious as always.

[Wjolcz]

1 hour ago, NathanKell said:

And I said you'd be satisfied with the results of Porkjet's labors.

Being as mysterious as always.

1.1 or 1.2 labors?

[Overland]

...... I think the hype train just went from mark 3 sized parts.. To N scale

[regex]

1 hour ago, NathanKell said:

And I said you'd be satisfied with the results of Porkjet's labors.

Being as mysterious as always.

Oh, I heard you straight, was just pre-empting any posts including the Twin Boar.

[NathanKell]

1.2. 1.1 is not the part/graphics update.

[I_Killed_Jeb]

so.. 1.2 is thargoids confirmed. wait a minute..

[AbacusWizard]

6 hours ago, Veeltch said:

1.1 or 1.2 labors?

1.2 labors? Increase that by an order of magnitude and you'll match Herakles!

[Warzouz]

Thank you for the transcript.

I would very much have a graphical improvement in 1.2 AND a better sound design (more and better sounds, and more contextual musics)

[bewing]

I am concerned about the concept of clouds in the game. IRL, you have radar to tell you if there is a mountain in the center of that poofy cloud or not. If we have clouds but no radar, we're flying blind -- which is significantly worse than reality.

 

[MiniMatt]

14 hours ago, katateochi said:

...But had part malfunctions been brought back onto the possible list? I thought that was something that they'd said they never wanted to add.  imo things explode quite enough without adding a dice roll of doom....

...By weather does that just mean visual clouds or was that talking about having actual wind that could knock you off course.  the stock game really needs clouds, but only if they can be done better than the current mods do (which are pretty sweet these days)....

Re part malfunctions, the impression I got was a very firm "noooooo, no, ewww, no, definitely not". I think it's probably safe to say the idea has come nowhere near the possible list.

Re weather, I think the question implied both visual and physical effects and the answer suggested neither would be a feature in 1.1. The rest of the stream certainly suggested visual effects are on the longer term radar (ie 1.2ish and beyond), and perhaps left open either way the even longer term possibility of weather having physical effects.

12 hours ago, 5thHorseman said:

(re easter eggs)

These both make me happy...

That point for me was perhaps the most salient one of the stream. I'd encourage you to skip to that point in the video and double check the answer for yourself as to my mind it implied some serious consideration had been given already. This wasn't just an enthusiastic "oh yeah that would be cool" response but implied a "oh yeah we've got some great ideas lined up on that one but too early to say" - but as for whether that might be addressed in 1.1, 1.2, 1.3, or 15.7 is too early to say.

9 hours ago, NathanKell said:

And I said you'd be satisfied with the results of Porkjet's labors.

Being as mysterious as always.

OOOOOOH! Regex's reputation as being (adorably) grumpy*, coupled with a confident assertion that he'll be satisfied with efforts, suggests the rest of us should be ecstatic with the results of Porkjet's labours

 

* love you really regex, just teasing!

[samstarman5]

22 minutes ago, bewing said:

I am concerned about the concept of clouds in the game. IRL, you have radar to tell you if there is a mountain in the center of that poofy cloud or not. If we have clouds but no radar, we're flying blind -- which is significantly worse than reality.

So you are saying that Squad wouldn't consider that, and just add clouds on a whim?  Squad has had weather in mind for a long time. This gives me reason to believe they have had plenty of time of considering ways to deal with piloting through heavy cloud cover. And they already have a foundation for a radar system, using the framework for the mining survey scans.

 

Considering the changes forecasted for 1.1, I have to wonder how anybody could think it was not going to break current saves.  Even with KSP in release mode, I maintain a full expectation of breakage somewhere with each update.  It is always worth giving it a whirl with an old save, but I never mind starting fresh. It gives opportunity to see what comes with the start of a career or science mode.

[Curveball Anders]

17 hours ago, Norpo said:

Based on the context I'm going to guess that he means that in regards to multithreading; Multithread programming isn't exactly easy; and you're lucky if it goes near-perfectly without bugs.

Multithreaded programming is far from easy and there's a lot that can go wrong.
Luckily the switch from single to multi isn't automatic (except in very few cases).

Just moving code to a multithread capable environment doesn't require any changes or induce new bugs but neither does it give any advantages.

A multithreaded environment allows the coder to adapt the code to take advantage of the new capabilities but it takes also requires that those changes respect the rules of this new environment.
So the move to the new PhysX by itself shouldn't add new issues, nor radically better performance.

But it'll allow Squad to rethink and recode stuff using the new model.

(Been there, done that, many times :/ Often it leads to major rethink and rewrite before the effort pays off. But it the end it does.)