KSP Weekly: Space Junk

[passinglurker]

4 minutes ago, basic.syntax said:

This is one of few things we can know about their style guide: hand-painted AO.

 

yet another infamous RD quote that didn't age well at all considering just this week they revamped a recent MH part because it wasn't up to standards.

I think its safe to say that anything RD said during MH development just doesn't hold true any more.

[basic.syntax]

I'm only referring to hand-painted AO. If that's still true, then your blanket statement "anything RD said" should be qualified. Also, the MH parts were up to the standards Squad wanted, when MH was released. Could the work have been better? Always. But its not quite fair to bash the MH panels with respect to new standards in v1.5.

Separately, I cast another vote for panels-into-stock. I tried to cast a bunch of them by holding a poll, but I was too clever for my own good, in wording the poll questions. 

 

[passinglurker]

2 minutes ago, basic.syntax said:

I'm only referring to hand-painted AO. If that's still true, then your blanket statement "anything RD said" should be qualified. Also, the MH parts were up to the standards Squad wanted, when MH was released.

RD came after Porkjet established baking as the standard through his space plane parts. You can't exactly chalk up the simulation of light to "space planes and rockets are different" so purely airbrushing the AO like RD's done on stream should never have been part of thier style guide for making history (ergo the infamy of this quote), and even with this "style guide" MH's art assets still came out inconsistent with each other that's a pretty big clue whatever they were trying to do in that regard fell apart at some point if they were ever serious about it at all and it wasn't just RD making excuses.

You're really not doing Squad any favors digging up these quotes it just highlights various times and ways player trust was eroded because they made excuses and failed to even implement thier excuses well within thier own little DLC microcosm. Something has clearly changed for the better though Squad's now actually listening to and responding to feedback in ways we haven't seen since 1.2 instead of pointing to the loudest fans and saying "they're fine with it we don't gotta fix nothing" like RD had essentially done in this quote, but either way we have no basis to assume this quote holds any clue to the contents of thier internal style guide that still holds true.

[Poodmund]

2 hours ago, basic.syntax said:

This is one of few things we can know about their style guide: hand-painted AO.

 

I remember that post and it's very insightful but is also around a year old so I there could be process changes or staff changes that weren't around during that time of discussion.

It's a shame, in my personal opinion, that RoverDude hand paints his AO on his parts (based off seeing his art streams in the past) as I feel that using a renderer to bake it into a shading layer gives a much more natural, precise and crisp look and I feel that it would really elevate his parts to the next level but it is obviously the artists and Art Director's prerogative as to how they do their work. Looking specifically at the the probe core parts that have been shown it doesn't look like this technique has been used... however, there is almost no mesh geometry in the parts that would cause much occlusion from an ambient light (a point I brought up in my feedback).

I wonder if any of the art team would be willing to share their views on AO Baking Vs Hand Painted Lighting but I would totally get it if they would rather stay silent as whatever they write then seems to be taken as gospel and then pulled apart a year down the line.

Edited September 23, 2018 by Poodmund
Spelling

[klgraham1013]

5 hours ago, basic.syntax said:

This is one of few things we can know about their style guide: hand-painted AO.

 

I mean, that post says there are guide lines on which black and whites to use, and we've all seen how that's turned out.

[scimas]

@passinglurker just a humble suggestion. Please proof read your posts. For all the good insight in your contents, you miss a lot of punctuation and capitalization. It becomes difficult to figure out where one sentence ends and another starts. 

[panzer1b]

11 hours ago, basic.syntax said:

Separately, I cast another vote for panels-into-stock.

Ive been behind this since the day they announced those panels.  There is absolutely nothing of "historic" significance about contruction panels, and those are some parts that could really be useful for alot of people that dont quite like the whole everything cylindrical approach squad seems to be taking to KSP.  With the exception of MK2/3 parts and a few notable things everything is bloody circular, what sci-fi movie (thats even worth watching) uses entirely cylindrical ships in space battles, none that im aware of?

Now i know its very unlikely to ever happen and would probably annoy some people that own the expansion (i own it, but i still support having those panels put into stock), but those things should really have been given to everyone rather then put behind a expansion that is focused on historical missions and not making square vessels that would never fly IRL.  Still, what id really want stock are panels of that sort but like 4 times the current size so i can actually make low part count starbases (and perhaps even make a cruiser or drednought class vessel) which is currently impossible in stock with the part count they would require.

 

So yeah, for all its worth, if its even worth anything, i also vote for panels in stock.

Edited September 23, 2018 by panzer1b

[RealKerbal3x]

1 hour ago, panzer1b said:

With the exception of MK2/3 parts and a few notable things everything is bloody circular, what sci-fi movie (thats even worth watching) uses entirely cylindrical ships in space battles, none that im aware of?

KSP is a game based on real space exploration, rather than Star Wars. Generally spacecraft IRL have a cylindrical cross section, so that’s the reason behind the cylindrical trend KSP’s parts take. It wasn’t really made with building space cruisers in mind.

Though I agree with you on the panel thing. They aren’t really ‘historical’ parts, so the base game would be a better place for them.

[LoSBoL]

4 hours ago, panzer1b said:

So yeah, for all its worth, if its even worth anything, i also vote for panels in stock.

I can't blame SQUAD one bit for making the MH expansion contain some parts that could be considered 'stock'. freeloading doesn't pay any bills, and there is only a certain amount of travel that a 'one time investment' can carry.  

[LegendaryAce]

Wow... Even after the lively discussion in last week's Weekly, we still don't have anything for Enhanced Edition. Squad really doesn't notice us asking for news or really care, do they. I'm calling it now. We're done getting updates. Thanks guys. You've made me hate you once again.

If anyone cares (let's be fair, they don't), I'll be on Elite Dangerous waiting for Ace Combat 7 if you need me.

[Terwin]

On 9/22/2018 at 10:03 PM, steve_v said:

Because this needs to be fixed in the core game, by those who introduced it.
Because irony in having a custom controller in the launch video while breaking controller support.
Because I am utterly tired of every single major release shipping with BS regressions and game-engine bugs.
Because, as much as I would otherwise like to, I will not buy the DLC if Squad keeps up the broken releases.

While we're on questions and AFBW, if nlight and yourself are capable of implementing proper input support, why isnt't Squad?

I agree, as joystick support for Linux was broken by moving to a version of unity where it is broken, the only real fix is for Unit to release new versions where it works.

Has Unity already done that?  If so, then what you should really be asking for is for another engine upgrade, if not, then you are barking up the wrong tree.

Regressions are a fact of life for software upgrades.  You have limited testing resources and those are best spent on platforms and versions where the majority of your clients operate.  If you want to use unpopular peripherals on an unpopular platform then you need to either A) be a big enough client spending enough on-going funds to warrant special attention(as in you specifically represent multiple percent of their yearly income for the project), or B) know that your stuff will break with great frequency and get very few resources for either regression testing or repairs.  That is just how the economics of the situation work out.

Who says Squad can't?  I am sure they could, but only if the economics of the situation warrant it.   

Edited September 24, 2018 by Terwin
testing, not resting

[nestor]

On 9/22/2018 at 2:40 PM, passinglurker said:

So before I get into artistic feedback I'm a little confused here @nestor in previous weeks it was said that the revamp was first focusing on early game parts but at the same time we've been getting a lot of probe related parts does this mean probes are being moved to earlier in the tech tree? It is certainly a common request I just hope it would be properly planned out, balanced, and play tested if it were the case.

There are a bunch of parts we want to improve. It's hard to pick which ones we should do first  (everyone has an opinion as we can see in the forum). The date the parts were made is not the only variable we are looking into. This time we picked some parts from the beginning of the tech tree and the probe cores, next time we will pick something else. Share your suggestions for the next ones.

[Redneck]

3 minutes ago, nestor said:

There are a bunch of parts we want to improve. It's hard to pick which ones we should do first  (everyone has an opinion as we can see in the forum). The date the parts were made is not the only variable we are looking into. This time we picked some parts from the beginning of the tech tree and the probe cores, next time we will pick something else. Share your suggestions for the next ones.

keep up the good work man and thanks for listening 

[qzgy]

5 hours ago, nestor said:

Share your suggestions for the next ones.

Engines perhaps? The smaller sizes, such as the 1.25 and the 2.5. While there's nothing quite wrong with them, given fow the art update is looking I think those would begin to clash at least a little with the new parts. 

[SpacePigeon]

The QBE...

It looks so minimalistic! 

I love it.

[steve_v]

13 hours ago, Terwin said:

the only real fix is for Unit to release new versions where it works.

Not necessarily. As I mentioned, AFBW does it without help from Unity. Squad could too if they could be bothered - simply ignore Unity's perpetually excrementse input handling and use SDL directly, there are C# bindings that work fine.
If the community can fix this, I see no reason a professional game dev house can't. All I do see is a bunch of poor excuses.

 

13 hours ago, Terwin said:

what you should really be asking for is for another engine upgrade

If it actually fixes the problem, doesn't come with a bunch of new regressions, and happens in a timely manner, sure. I don't really care how it gets fixed anyway.
 

13 hours ago, Terwin said:

Regressions are a fact of life for software upgrades.  You have limited testing resources...

Regressions are a fact of life when you don't bother to test your excrements, and if every upgrade breaks as much as it fixes we will never get anywhere. Accepting major regressions as inevitable is also lousy software engineering.
If Squad can't test or can't be bothered testing something as basic as joystick detection on a "supported" platform, then I have to question both their motivation and their competency.
Considering that they managed to ship a copy of the liblingoona shared object incompatible with the Ubuntu release  recommended on their own site, I strongly suspect that they don't bother to do any testing on anything but Windows.
Considering that they left the 1.1 Linux build in a state where it crashed regularly with memory management errors for an entire release cycle, I suspect they don't bother fixing anything but trivial bugs on anything but Windows either.
 

13 hours ago, Terwin said:

That is just how the economics of the situation work out.

The Linux build is the same price as the Windows and MacOS builds, so I expect comparable functionality on all these platforms. How Squad achieves this is not my business.
If you advertise your product as working on a particular platform, you test it to make sure it really does work properly before you ship it.

Squad's approach, as far as I can tell: Ship on the release date, bugs be damned. We can patch it later if it's not to hard, otherwise just ignore the problem and up the hype for the next release. Repeat.

Edited September 25, 2018 by steve_v

[Dafni]

15 hours ago, Terwin said:

... If you want to use unpopular peripherals on an unpopular platform...

 

Are you serious here? I wont go into the discussion of Linux being an unpopular platform (anyway the game is advertised as being compatible, fwiw) but calling joysticks unpopular peripherals for a space flight sim?

I dont get it. But then again I dont follow popular trends much.

[steve_v]

8 minutes ago, Dafni said:

calling joysticks unpopular peripherals for a space flight sim?

Yeah, that kinda did my head in too.
The whole "minorities shouldn't get bugfixes / Linux is unpopular" BS really grinds my gears as well... But there's no accounting for opinions.

Edited September 25, 2018 by steve_v

[klgraham1013]

19 hours ago, Terwin said:

as joystick support for Linux was broken by moving to a version of unity where it is broken, the only real fix is for Unit to release new versions where it works.

Umm...have the programmers, I don't know, program around it?  People have done it for free.  Surely, paid professionals can do the same.

[Delay]

4 hours ago, steve_v said:

But there's no accounting for opinions.

Sorry if what I said a few posts ago sounded like ignoring your opinion. I am respecting you and your right to have an opinion. I just don't think it's a valid/reasonable one.

[passinglurker]

14 hours ago, qzgy said:

Engines perhaps? The smaller sizes, such as the 1.25 and the 2.5. While there's nothing quite wrong with them, given fow the art update is looking I think those would begin to clash at least a little with the new parts. 

Engines might be some pretty big shoes to fill.

 
I think I'd like to see them maintain good quality over multiple consecutive previews before they take a pass at competing with this. maybe other early parts or deep space parts? the service bays, the parachutes, the 2.5m pod, the 1.25m/2.5 fuel tanks, the radial decouplers, the nose cones, etc plenty to do first.

Edited September 25, 2018 by passinglurker
forgot vector as example

[Raptor9]

I didn't want to be the guy that brought them up, but since they already have been... 

20 hours ago, nestor said:

Share your suggestions for the next ones.

The LV-T30, LV-T45, LV-909, RE-M3, RE-I5, and RE-L10:  Get rid of those mandatory tank butts with their black/yellow stripes, with some bare variants available.

Or properly matching up the Mk1-3 forward node to other 1.25m parts, as the new Mk1 pod's was matched up properly to 0.625m parts in a recent Weekly preview.

And while I'm going overboard with wishes as if I was talking to a wishing well, you could change the 24-77 Twitch to be a better looking radial engine like in this picture of a Mars lander concept from Boeing, and the Mk2 Lander can to a decent looking 3-Kerbal lander module like the one in this picture (minus the RCS, exterior tanks and docking port of course; let's keep the "lego" options).

     

But if nothing else, I would settle for a better looking RE-L10 Poodle model that looks somewhat like a higher Isp vacuum engine it's rated as.

Edited September 25, 2018 by Raptor9
Mk1-3 pod comment

[klgraham1013]

Please.  Just integrate Porkjet's overhaul all ready.

[Farix]

1 hour ago, klgraham1013 said:

Please.  Just integrate Porkjet's overhaul all ready.

There are probably some legal issues with that. It really depends on if Porkjet's part overhaul falls within the "work for hire" framework or if it was a project he did on his own.

[swjr-swis]

48 minutes ago, klgraham1013 said:

Just integrate Porkjet's overhaul all ready.

I will admit I am growing more and more puzzled as to why Squad has not, and is clearly still not doing that to begin with.

I'm not going to get into a discussion about which technique or art vision is better. I'm purely thinking from a practical point of view, from the standpoints of return-of-investment and crowd management.

Here they have a whole heap of pre-made work that looks very close to being release-ready. The public at large clearly likes this stuff, as can be deducted from how much the community has been asking about future plans on integrating it, and the community efforts for 'finishing' it. Lots of brownie points could've been scored by Squad with seemingly very little effort, and if they had wanted to release/change other parts at the same time that use a different style, people would've likely not made a huge deal of it because they're still getting a lot of what they do want and like. Kids tend to take medicines they don't like with a lot less trouble when it's mixed with their favourite foods; if you so happen to be sitting on a pile of it, all the cheaper a solution.

Instead, Squad decides to completely sideline Porkjet's parts without explanation, then spends a lot of labour on other art and retouches and presents only that new stuff by itself. Not exactly surprisingly, the discontent and picketing  grows even louder. "Porkjet!" "Porkalike" "Piggie Bank!" "Bacon and Eggs!" "Birdstrike!"

It makes no sense. It took no crystal ball to predict this. Why then still risk it? There has to be some incredibly strong reason to decide to spend time, money and effort on a job that you could've known is going to spark less-than-happy reception. Doubly so when you have a pile of pacifying candy to hand out RIGHT THERE... and decide not to do so. But we're left in the dark to wonder and speculate.