Kerbal Space Program 2: Master Post

[Danilo Coelho]

https://www.gamespot.com/articles/kerbal-space-program-2-will-feature-a-bigger-more-/1100-6469648/

[Redneck]

  On 9/6/2019 at 12:07 AM, LekkoBot said:

I very much doubt this and I hope this isn't the case, there is no need for there to be voxel terrain to make a good game.

Expand  

so you wouldn't be interested in mining operations or shielding you base from solar radiation or cosmic rays? How do you make or refine elements for parts having never left some kind of hole or disturbing the ground in any way? It just happens "magically"? Wheres the realism in that? So you are saying if they added voxel terrain it would make it a bad game? Why? Why would that be bad? That to me is like saying there is no need for fire coming out the exhaust end of a rocket motor we have a accelerator, we know we are moving. Why not do it?

[PunkRockZoologist]

  On 9/6/2019 at 3:39 AM, Redneck said:

so you wouldn't be interested in mining operations or shielding you base from solar radiation or cosmic rays? How do you make or refine elements for parts having never left some kind of hole or disturbing the ground in any way? It just happens "magically"? Wheres the realism in that? So you are saying if they added voxel terrain it would make it a bad game? Why? Why would that be bad? That to me is like saying there is no need for fire coming out the exhaust end of a rocket motor we have a accelerator, we know we are moving. Why not do it?

Expand  

Voxel/editable terrain would be pretty neat, but it would definitely be beyond the scope of what they're planning of KSP2. And they've stated outright that there will be "no terraforming" or other kinds of terrain editing. So yeah, mining will just extract resources from the solid ground. Not necessarily realistic, but the realism of Kerbal has always been to do with the space flight more than the infrastructure supporting it.

[Redneck]

  On 9/6/2019 at 4:17 AM, PunkRockZoologist said:

Voxel/editable terrain would be pretty neat, but it would definitely be beyond the scope of what they're planning of KSP2. And they've stated outright that there will be "no terraforming" or other kinds of terrain editing. So yeah, mining will just extract resources from the solid ground. Not necessarily realistic, but the realism of Kerbal has always been to do with the space flight more than the infrastructure supporting it.

Expand  

i hear ya. Was just thinking though a ice covered moon like Europa with a ocean under the ice. How you going to get to it without drilling?

[PunkRockZoologist]

  On 9/6/2019 at 4:18 AM, Redneck said:

i hear ya. Was just thinking though a ice covered moon like Europa with a ocean under the ice. How you going to get to it without drilling?

Expand  

Now *that* would be awesome. They did have the ice moon in the trailer, and they've said some cryptic things about making the other planets more interesting, so maybe there will be ice over oceans.

But I wouldn't count on it, just because they've also said no caves/negative space on planets. So I guess we'll just have to wait and see.

[Balrog]

Will Asteroid impacts on celestial body surfaces be a bit more realistic? I mean, seriously... an asteroid larger than a U-Haul is not going to slow down and reach a slow terminal velocity, and then bounce off Kerbin's surface and roll to a stop.

[HebaruSan]

We have an announcement for KSP 1.8; presumably after that there will be KSP 1.9.

@UomoCapra, can you comment on plans for the numbering after that point? I hope the next version would be 1.10 rather than 2.0, for obvious reasons.

[Danilo Coelho]

 

 

[pandaman]

  On 9/6/2019 at 2:14 PM, HebaruSan said:

We have an announcement for KSP 1.8; presumably after that there will be KSP 1.9.

@UomoCapra, can you comment on plans for the numbering after that point? I hope the next version would be 1.10 rather than 2.0, for obvious reasons.

Expand  

If I understand correctly that is not how it works.

The decimal points are not actually 'decimal points', but just separators between up-counting numbers to designate different 'builds' (or sub-versions).  They could in theory use any symbol (commas, dashes, plus signs, or whatever), but convention uses the decimal point. 

After 1.9.x we will go to 1.10.x, 1.11.x potentially up to 1.9999.x  and beyond.

So KSP 1 will always start with 1.x  

The second mumber is the 'main version' currently 7 (1.7.x) 

The third is the specific build number of that version (eg  1.7.1234) 

So the second number will increase, (usually by 1) each release, and the third (and 4th if needed) will be as big as it needs to be, depending on how quickly they get a build they like.

Also the first number can be anything, it is more like a name label.  A lot of commercial software will use the year of release (eg 2019.7.1234).

At least that is my understanding.  If I got this wrong please correct me and send someone round to give me a slap.

Edited September 6, 2019 by pandaman

[HebaruSan]

  7 minutes ago, pandaman said:

So KSP 1 will always start with 1.x  

Expand  

Well, that's your speculation. In reality it's a decision made by those assigning the numbers, and I'm asking in the hopes of getting an authoritative answer to inform the CKAN team's plans of how to support KSP2.

[pandaman]

  On 9/6/2019 at 4:24 PM, HebaruSan said:

Well, that's your speculation. In reality it's a decision made by those assigning the numbers, and I'm asking in the hopes of getting an authoritative answer to inform the CKAN team's plans of how to support KSP2.

Expand  

True, they are not 'bound' into continuing using 1.x if they chose not to, but 2.x would be a little cruel.

They could really confuse everyone and use 3.x.x 

I do see your point in wanting clarification to avoid future problems.

[razark]

  On 9/6/2019 at 4:24 PM, HebaruSan said:

In reality it's a decision made by those assigning the numbers...

Expand  

https://wiki.kerbalspaceprogram.com/wiki/Version_history

Considering they went from 0.9 to 0.10, and 0.9 and 0.90 are different versions, I think we can determine that they're using version numbers correctly, and not as decimals.

A KSP 2.x.y is a conceivable version number separate from KSP2, but it's highly unlikely that a change radical enough to involve such a change would occur at this point.

[MechBFP]

  On 9/6/2019 at 4:13 PM, pandaman said:

So KSP 1 will always start with 1.x  

Expand  

Ya that’s not correct. Generally in software versioning 1.0 indicates the initial GA (General Availabilty) release of the software. A version 2.0 would indicate a major overhaul of the software in such a way that it is significantly different from the 1.x branch. 

Generally updating from 1.x to 2.x would require migration strategies, new servers, user re-training, etc. However it could even be something like changing your licensing model. It’s entirely up to the company to decide what qualifies as a “major overhaul”. 

[LekkoBot]

  On 9/6/2019 at 4:18 AM, Redneck said:

i hear ya. Was just thinking though a ice covered moon like Europa with a ocean under the ice. How you going to get to it without drilling?

Expand  

I would be fine with terrain editing but voxels are a strange way to go with smooth surfaces like planets.

[Spfnym]

On that KSP 2 Developer on Multiplayer video when they talk about performance at 05:27, Nate answers that the game won't make use of multiple CPU cores. I would ask of you to reconsider. Performance is my number one concern with the new KSP (after it's otherwise at least comparable to KSP 1).

There's only so much one can do with a single core. Real life physics get in the way of getting much more single-thread performance since one can't just infinitely scale down things.

Nate brings up that "normal hardware" should be able to handle vehicles on the scale of what's seen on trailer. He completely neglects that multicore is pretty much the norm already so the current plan would be underutilizing even the said normal hardware and holding back any that have more powerful hardware than that. I tried to look for information on this and the best I could quickly find was cpubenchmark.net, and even if it's probably biased toward more powerful CPUs, I think it says at least something that more than the top 30 of the most benchmarked CPUs in the last 90 days have at least two physical cores.

Okay, I can acknowledge that it might not be that easy to make good use of multiple CPU cores when vehicle parts are so interlinked. Still, it's something that should be given more consideration in my opinion. Here's one person's input that I managed to find at least:

  On 8/23/2019 at 3:25 PM, Kerenatus said:

Yeah that's a tough one.

i remember reading a Star Citizen development article saying that they used a "grid" technique which procedually treats certain parts of a larger vessel as a whole, calculating the interactions between those collection of parts, then calculating the parts within each collection seperately. This way the calculation of a big vessel is much more distributed.

Though theses kind of things may be too much to ask for in KSP2

Expand  

 

[MechBFP]

  On 9/7/2019 at 9:28 AM, Spfnym said:

Still, it's something that should be given more consideration in my opinion.

Expand  

You must have some insider information that the rest of us don’t know about since you are clearly aware how much consideration they have already given it. 

You got a link to that info by chance?

[Spfnym]

By that I mean "more consideration than what has been made public". That video was the first I even heard of somebody involved talking about this issue, and what is there is a very unsatisfactory couple of words for explanation.

[TheKSPBeginner]

I am excited. I’m getting a new laptop mainly for this.

[Dix2lespace]

Hi, I hope they put a true realistic clouds like that:

 

[lunardog15]

 

  On 9/8/2019 at 2:30 PM, Dix2lespace said:

Hi, I hope they put a true realistic clouds like that:

 

Expand  

i can hear my gpu screaming at that.

[Pecan]

Controverially, I think I've decided not to buy KSP 2 for a while.

I've never paid GBP £60 for a game and I certainly don't intend to start with one that is sounding more and more like a £20 DLC.  As a re-write and tidy sequel it doesn't have the excuse of 'years of AAA' development to pay for, not that that was ever valid anyway.
Still using Unity.  Still using single-core.  Bundling some new parts- and planets- mods.

Call me back when it's on sale for £30.

Edited September 9, 2019 by Pecan

[coyotesfrontier]

  On 9/9/2019 at 1:33 PM, Pecan said:

Controverially, I think I've decided not to buy KSP 2 for a while.

I've never paid GBP £60 for a game and I certainly don't intend to start with one that is sounding more and more like a £20 DLC.  As a re-write and tidy sequel it doesn't have the excuse of 'years of AAA' development to pay for, not that that was ever valid anyway.
Still using Unity.  Still using single-core.  Bundling some new parts- and planets- mods.

Call me back when it's on sale for £30.

Expand  

I wince everytime I hear that "KSP is bad because Unity". It's not, most of kerbal's bad performance and the like comes from a bad codebase, not the engine. KSP2 is not a DLC, the entire game is different from the ground up. All of the models have been created from scratch, using the old ones only as a reference.

[Kuszotke]

  On 9/9/2019 at 2:32 PM, coyotesfrontier said:

I wince everytime I hear that "KSP is bad because Unity". It's not, most of kerbal's bad performance and the like comes from a bad codebase, not the engine. KSP2 is not a DLC, the entire game is different from the ground up. All of the models have been created from scratch, using the old ones only as a reference.

Expand  

Still. The price is ridiculous. $60 isn't right.

[MechBFP]

  On 9/9/2019 at 5:34 PM, Kuszotke said:

Still. The price is ridiculous. $60 isn't right.

Expand  

You’re right, it’s too low, but they can’t set it too high otherwise they will start losing sales to the uninformed.

[KSK]

How about waiting until the finished game - or release ready game at any rate - is in front of us, before jumping to opinions on price?