I wonder when the decision was made to integrate FinePrint? Because that was a lot of people's guess about what was being integrated for 0.25, and Arsonide confirmed then it wasn't his mod.
I haven't really played in a while, but 0.26 sounds great. It seems like they're getting really close to feature complete.
Also I kind of hope the upgradable buildings start with their current sizes/shapes, and upgrade to huge/ridonkulous things. I imagine that opens the gates for modders to give us different building models or themes
Fine Print is a great choice. Other mods might be more mechanically useful/interesting but I think it's the one I've used that straight-up adds the most fun to the game.
Fine Print expands the already existing system and deepens it in a logical way that doesn't break from the stock experience. Definitely makes sense as a choice! Congrats, Arsonide!
0.26 looks great already. More biomes, better editing tools, and upgradable buildings, on top of integrated Fine Print. Sign me the fuck up.
Oh god yes, they're going to revamp aerodynamics (Eventually). I'm hesitant to use mods to change underlying game mechanics that much just because I like to stay within the intent of the game. But FAR has been a long standing temptation just because the aerodynamic model is great for rockets, and bad for anything else. :P
I'm kind of disheartened by no mention of either 64-bit or memory/texture issues.
I mean, I'm sure they know what they're doing, but I think they've just given up on that entirely.
0.25's 64-bit flat-out refuses to read saves from 0.24 (first time I've ever encountered that issue--weird!), and OpenGL, while solving the memory problem, really chops away at frames and has a few graphical glitches.
I'm not sure what this means, but I think I'm kind of dreading the release of 0.90. I think we're going to get a bunch of great new parts to use, but none of the existing issues are going to be fixed. Especially the one I was most excited about in 0.24--64-bit Unity.
I guess it's beyond their control. :?
+1
Options
Nova_CI have the needThe need for speedRegistered Userregular
Unity is very much beyond their control and they've already stated that the big 64 bit bugs are Unity bugs, not KSP bugs. So we're waiting on Unity's developer to sort that out. It'll get sorted, I'm sure, so I'm not too worried there.
Actually, going with my station thing, optimization needs to be high on their list. Since KSP is single threaded and will remain so, they need to cheat the physics somehow. That welding mod, the one that reduced the number of individual objects, is a good idea although it does decrease the accuracy of the simulation. Something like that maybe? I dunno. But the joy of KSP is adding complexity as you become better at design and piloting.
That Mun base that I built a couple threads ago? KSP didn't have the greatest framerate when it was complete. Still usable, but jerky. The old Valkyrie project I did a while ago, where I assembled a large inter-planetary ship at a space station also had major framerate issues. I had an i7-2600k at the time. Just installed an i7-4790k. Should be better, but as far as I can tell from various performance tests from Tom's Hardware, not by much.
I know the stuff Scott Manley does is about 10x as complex as anything I've done. Obviously there are going to be limits, but the system becomes bogged down really fast.
+4
Options
Nova_CI have the needThe need for speedRegistered Userregular
Like, they could do a version of that welding mod where parts like RCS, docking ports, batteries, etc, are 'welded' to the part they're attached to. Major joints like between engines and fuel tanks and separators remain, but parts that aren't supposed to disconnect, or don't need flex calculations on their own, should be combined with the structural object they're attached to.
Of course, this means that an impact would destroy everything instead of the parts breaking apart (Unless they got fancy and deconstructed it once the structural part was destroyed). I think that is an acceptable downside to reducing the performance impact of large rockets and stations.
I just wish they'd devise a better way to load up only the textures they need.
Or barring that, manage them better. Switching to OpenGL cut total memory use by around a 1/3 for most people--unfortunately, OpenGL is a buggy mess for most people in KSP.
Like, they could do a version of that welding mod where parts like RCS, docking ports, batteries, etc, are 'welded' to the part they're attached to. Major joints like between engines and fuel tanks and separators remain, but parts that aren't supposed to disconnect, or don't need flex calculations on their own, should be combined with the structural object they're attached to.
Of course, this means that an impact would destroy everything instead of the parts breaking apart (Unless they got fancy and deconstructed it once the structural part was destroyed). I think that is an acceptable downside to reducing the performance impact of large rockets and stations.
Yeah, if we can't get adequate performance with zillions of discrete parts, this is what I'd like to see. Automatic welding of certain parts, and maybe allowing manual welding of small subassemblies. It'd be a decent compromise.
Like, they could do a version of that welding mod where parts like RCS, docking ports, batteries, etc, are 'welded' to the part they're attached to. Major joints like between engines and fuel tanks and separators remain, but parts that aren't supposed to disconnect, or don't need flex calculations on their own, should be combined with the structural object they're attached to.
Of course, this means that an impact would destroy everything instead of the parts breaking apart (Unless they got fancy and deconstructed it once the structural part was destroyed). I think that is an acceptable downside to reducing the performance impact of large rockets and stations.
Yeah, if we can't get adequate performance with zillions of discrete parts, this is what I'd like to see. Automatic welding of certain parts, and maybe allowing manual welding of small subassemblies. It'd be a decent compromise.
I wonder if unwelding parts when a collision occurs would be a workable solution.
0
Options
Nova_CI have the needThe need for speedRegistered Userregular
Like, they could do a version of that welding mod where parts like RCS, docking ports, batteries, etc, are 'welded' to the part they're attached to. Major joints like between engines and fuel tanks and separators remain, but parts that aren't supposed to disconnect, or don't need flex calculations on their own, should be combined with the structural object they're attached to.
Of course, this means that an impact would destroy everything instead of the parts breaking apart (Unless they got fancy and deconstructed it once the structural part was destroyed). I think that is an acceptable downside to reducing the performance impact of large rockets and stations.
Yeah, if we can't get adequate performance with zillions of discrete parts, this is what I'd like to see. Automatic welding of certain parts, and maybe allowing manual welding of small subassemblies. It'd be a decent compromise.
I wonder if unwelding parts when a collision occurs would be a workable solution.
To some extent, maybe, since they're working with parts that are always the same, but I wonder if it's worth the extra load or programming complexity. Like, if a fuel tank explodes, the batteries attached to that fuel tank aren't going to survive the explosion, right?
I'm sure they're working on ideas for optimization and these guys are clearly smarter than me, so I just hope they do go ahead with an implementation of some kind.
Welding all the parts together by default would also tragically rob us of floppy rocket syndrome.
+1
Options
Nova_CI have the needThe need for speedRegistered Userregular
Well, I didn't say all the parts. Batteries and RCS thrusters and such don't flex. I dunno how the game handles it now, though, so maybe those objects are already skipped in the simulation when it comes to the way forces are applied.
FAR can rip off batteries and other small stick-on parts, though as far as I know most of those types of items are actually massless/dragless for stock vehicle physics. That doesn't necessarily mean the simulation completely ignores them, though.
I've been out of the game for a while, got back in and immediately made a big ass rocket, an attempt at interplanetary fun. It is my first attempt so not serious, but I wanted to see how far I'd get. Maybe not Duna surface, but a good distance.
Holy shit did they rebuild the launch pad out of paper mache and rocket fuel? I can't even launch a rocket without the damn pad exploding as soon as physics kick in. Pad: obliterated. Rocket: undamaged. :rotate:
First they came for the Muslims, and we said NOT TODAY, MOTHERFUCKERS
I accidentally did that by forgetting to put launch stability enhancers. The rocket falls an inch and destroys the launch pad. This was with a heavy lifter though.
Do you guys ever get beginners luck with your creations?
I had that to a huge degree. It wasn't all smooth sailing, as you can see from the smoking crater in my post above. That and my first stage taking out fuel tanks- despite not actually touching them, oddly. But once there were fixed, I was up, in orbit, and made it all the way to Duna surface in one go, having never done an interplanetary rendezvous before. Easy, I even made it back into orbit (sandbox, so I didn't care if anyone made it home). Seeing the thin, red atmosphere of Mars Duna was really very gratifying. Pic (nothing special):
[img][/img]
It was such a success, that I decided to take the same simple rocket, add a couple struts and that ladder I forgot, and do it again, this time to Eve. Disaster. The rocket is completely ungainly, it literally wobbles itself to pieces before LEO. Even above the atmosphere, the fairings on my tightly packed atomic motors get caught on each other and throw everything out of wack. So I must have had the luckiest lift-off ever, followed by an easy encounter with Duna, which is also not always so easy.
It is a peculiar game this one, though it teaches well the valuable rocket-science lesson that just because it works once, doesn't mean it works.
I had an experience like that one. I built a space shuttle analogue (same setup of SRBs and liquid engine) and launched it into stable kerbin orbit on my very first try (adjusting the throttle manually to ensure that the thing didn't spin horribly out of control post-SRB sep). I then managed to return it to Kerbin and land safely on the ground. I could never pull it off again.
Is it just me or do kerbals now fling themselves spinning out of command pods when they go EVA?
I find that happens most often when there's something over the hatch - the Kerbal would climb out the hatch, but their helmet would clip on the Science Jr. I put on top of the capsule sending them spinning off. Giving them more clearance seems to solve the problem for me.
So Squad says they want Kerbals to get more interesting the more reputation each one gains. Maybe some sort of perk system? There's a lot of cool ideas they could run with and I've been having fun thinking of some:
Scientist tree, scales inversely with Stupidity score:
- "Keen eye", X% bonus science from EVA reports
- "Geology Specialist", X% bonus science from surface sampling
- "Laboratory experience", X% bonus science while this kerbal is stationed in a science lab
- "Diligent experimenter", X% bonus science to samples pulled out of the materials lab or goo canister.
- "Electrical engineering", can repair one solar panel per mission, though the repaired model is covered in duct tape and less effective.
- Possibly a unique EVA experiment as the "top tier" perk?
Piloting tree, good for any kerbal:
- "Steady hand", improved SAS
- "Flight Experience", crafts stall at a slightly higher angle of attack, once they put better aerodynamics in
- "Hard Landings", landing gear are X% more durable
- "Push the envelope", control surfaces and thrust vectoring can exceed 100% of normal by a small amount.
- "Maneuver training", top level skill that basically gives you MechJeb's SmartASS as long as this kerbal is in a command pod.
BAD/A55 special training program, requires Jeb or higher level of courage, imbues the kerbal with the "BADA55" trait:
- "Stunt driver", rovers go X% faster and the tires are more durable.
- "Iron grip", X% more grip strength on ladders, can "grip" any part of a ship at regular ladder strength.
- "Skydiver", this kerbal gains a one-use EVA parachute when exiting a MK1, 2 or 3 plane cockpit.
- "Tough as nails", this kerbal has a chance to survive the destruction of his or her command pod due to collision, though the chance would diminish based on impact speed.
- "Prototype jetpack", holding space while operating the jetpack will consume 4x as much fuel but double the pack strength.
Skydiver and Tough as Nails would be especially good for test pilots working the kinks out of planes.
I actually don't much care for the idea of Kerbal experience/perks. It feels like needless RPG complexity in a game that's already super-complicated. I don't want to have to worry that a given craft design will only work if it's being piloted by a Kerbal with a specific bonus, juggle crew assignments to minmax science output, etc. It sounds tedious.
The way I would see it being less tedious would be if the skills within a category aren't mutually exclusive, and you'd get training points based on the reputation gained on the mission. That way you'd want to focus on one scientist, one pilot, and maybe a test pilot. You'd want to invest in a small number of kerbals and take great care of them, rather than spread out skills and micromanage. Because once you unlock all the experiments, you almost never send a kerbal somewhere without all of them, thus you'd want your one kerbal to have as many perks as you have unlocked. Juggling would be tedious but there are ways to make it not worth it or even counter-productive to do.
And on ships being only flyable by an expert pilot, yeah maybe the stall-angle perk would be a problem there, but the rest are just sorta quality of life bonuses. No X% more lift or less drag or that sort of thing. I really like the idea that a talented kerbal pilot does something. Also the idea that a young rookie could learn valuable skills from doing routine stuff like hauling fuel up to a space station or doing aerial surveys or putting satellites into orbit, while the well-trained senior crew are off exploring planets. And incentivizing the use of bigger command pods / larger crews would be neat. Right now the only reason to send 3 kerbals to Duna is if sending just one is too easy.
I actually don't much care for the idea of Kerbal experience/perks. It feels like needless RPG complexity in a game that's already super-complicated. I don't want to have to worry that a given craft design will only work if it's being piloted by a Kerbal with a specific bonus, juggle crew assignments to minmax science output, etc. It sounds tedious.
Then don't do it. Nothing says you ever have to minmax anything, and this is doubly true for KSP. So what if you didn't get all the possible science you could have, that just gives you a reason to build another rocket and do it again, or do something else for more science. If a design requires a specific Kerbal and/or perks to make it work, then the design is flawed and needs to be reworked.
The exact complaint you have here was used against the tech tree, the economy, and just about every addition to the game. Including the Mun when that was first added.
I mean, I do agree with the basic premise that I'd rather have no perks system than a tedious or poorly implemented one, no doubt. But it's pretty clear that they want the game to become as much about running a space program as it is about running individual missions, at least in career mode. Maybe give sandbox mode kerbals all piloting perks?
I actually don't much care for the idea of Kerbal experience/perks. It feels like needless RPG complexity in a game that's already super-complicated. I don't want to have to worry that a given craft design will only work if it's being piloted by a Kerbal with a specific bonus, juggle crew assignments to minmax science output, etc. It sounds tedious.
Agreed 100%. Trying to level up my Kerbals sounds lame and un-fun. Having fun stats of courage and stupidity fit the tone of the game perfectly, and let us focus on the precious rocket making and flying.
Posts
http://forum.kerbalspaceprogram.com/content/312-The-Grand-0-26-Plan
I wonder when the decision was made to integrate FinePrint? Because that was a lot of people's guess about what was being integrated for 0.25, and Arsonide confirmed then it wasn't his mod.
I haven't really played in a while, but 0.26 sounds great. It seems like they're getting really close to feature complete.
Also I kind of hope the upgradable buildings start with their current sizes/shapes, and upgrade to huge/ridonkulous things. I imagine that opens the gates for modders to give us different building models or themes
It's pretty easy to mod your save game to give you 40k SCIENCE, at least.
0.26 looks great already. More biomes, better editing tools, and upgradable buildings, on top of integrated Fine Print. Sign me the fuck up.
PSN: ShogunGunshow
Origin: ShogunGunshow
Oh god yes, they're going to revamp aerodynamics (Eventually). I'm hesitant to use mods to change underlying game mechanics that much just because I like to stay within the intent of the game. But FAR has been a long standing temptation just because the aerodynamic model is great for rockets, and bad for anything else. :P
I will rub that lamp.
I will give my two other wishes away.
That is my one wish.
Thank you.
I'm kind of disheartened by no mention of either 64-bit or memory/texture issues.
I mean, I'm sure they know what they're doing, but I think they've just given up on that entirely.
0.25's 64-bit flat-out refuses to read saves from 0.24 (first time I've ever encountered that issue--weird!), and OpenGL, while solving the memory problem, really chops away at frames and has a few graphical glitches.
I'm not sure what this means, but I think I'm kind of dreading the release of 0.90. I think we're going to get a bunch of great new parts to use, but none of the existing issues are going to be fixed. Especially the one I was most excited about in 0.24--64-bit Unity.
I guess it's beyond their control. :?
Actually, going with my station thing, optimization needs to be high on their list. Since KSP is single threaded and will remain so, they need to cheat the physics somehow. That welding mod, the one that reduced the number of individual objects, is a good idea although it does decrease the accuracy of the simulation. Something like that maybe? I dunno. But the joy of KSP is adding complexity as you become better at design and piloting.
That Mun base that I built a couple threads ago? KSP didn't have the greatest framerate when it was complete. Still usable, but jerky. The old Valkyrie project I did a while ago, where I assembled a large inter-planetary ship at a space station also had major framerate issues. I had an i7-2600k at the time. Just installed an i7-4790k. Should be better, but as far as I can tell from various performance tests from Tom's Hardware, not by much.
I know the stuff Scott Manley does is about 10x as complex as anything I've done. Obviously there are going to be limits, but the system becomes bogged down really fast.
Of course, this means that an impact would destroy everything instead of the parts breaking apart (Unless they got fancy and deconstructed it once the structural part was destroyed). I think that is an acceptable downside to reducing the performance impact of large rockets and stations.
Or barring that, manage them better. Switching to OpenGL cut total memory use by around a 1/3 for most people--unfortunately, OpenGL is a buggy mess for most people in KSP.
Yeah, if we can't get adequate performance with zillions of discrete parts, this is what I'd like to see. Automatic welding of certain parts, and maybe allowing manual welding of small subassemblies. It'd be a decent compromise.
I wonder if unwelding parts when a collision occurs would be a workable solution.
To some extent, maybe, since they're working with parts that are always the same, but I wonder if it's worth the extra load or programming complexity. Like, if a fuel tank explodes, the batteries attached to that fuel tank aren't going to survive the explosion, right?
I'm sure they're working on ideas for optimization and these guys are clearly smarter than me, so I just hope they do go ahead with an implementation of some kind.
How far I made it:
Oh KSP, how did I ever leave you.
You just made Jeb shed a single tear.
There was silence in mission control.
Except for the sound of a dropped clipboard.
You monster
Much loik DAKKA, can nevah be enuff Rokket.
I had that to a huge degree. It wasn't all smooth sailing, as you can see from the smoking crater in my post above. That and my first stage taking out fuel tanks- despite not actually touching them, oddly. But once there were fixed, I was up, in orbit, and made it all the way to Duna surface in one go, having never done an interplanetary rendezvous before. Easy, I even made it back into orbit (sandbox, so I didn't care if anyone made it home). Seeing the thin, red atmosphere of Mars Duna was really very gratifying. Pic (nothing special):
It was such a success, that I decided to take the same simple rocket, add a couple struts and that ladder I forgot, and do it again, this time to Eve. Disaster. The rocket is completely ungainly, it literally wobbles itself to pieces before LEO. Even above the atmosphere, the fairings on my tightly packed atomic motors get caught on each other and throw everything out of wack. So I must have had the luckiest lift-off ever, followed by an easy encounter with Duna, which is also not always so easy.
It is a peculiar game this one, though it teaches well the valuable rocket-science lesson that just because it works once, doesn't mean it works.
Yes I've had that problem a lot since the last update. Even grabbing back on they immediately let go.
I find that happens most often when there's something over the hatch - the Kerbal would climb out the hatch, but their helmet would clip on the Science Jr. I put on top of the capsule sending them spinning off. Giving them more clearance seems to solve the problem for me.
Scientist tree, scales inversely with Stupidity score:
- "Keen eye", X% bonus science from EVA reports
- "Geology Specialist", X% bonus science from surface sampling
- "Laboratory experience", X% bonus science while this kerbal is stationed in a science lab
- "Diligent experimenter", X% bonus science to samples pulled out of the materials lab or goo canister.
- "Electrical engineering", can repair one solar panel per mission, though the repaired model is covered in duct tape and less effective.
- Possibly a unique EVA experiment as the "top tier" perk?
Piloting tree, good for any kerbal:
- "Steady hand", improved SAS
- "Flight Experience", crafts stall at a slightly higher angle of attack, once they put better aerodynamics in
- "Hard Landings", landing gear are X% more durable
- "Push the envelope", control surfaces and thrust vectoring can exceed 100% of normal by a small amount.
- "Maneuver training", top level skill that basically gives you MechJeb's SmartASS as long as this kerbal is in a command pod.
BAD/A55 special training program, requires Jeb or higher level of courage, imbues the kerbal with the "BADA55" trait:
- "Stunt driver", rovers go X% faster and the tires are more durable.
- "Iron grip", X% more grip strength on ladders, can "grip" any part of a ship at regular ladder strength.
- "Skydiver", this kerbal gains a one-use EVA parachute when exiting a MK1, 2 or 3 plane cockpit.
- "Tough as nails", this kerbal has a chance to survive the destruction of his or her command pod due to collision, though the chance would diminish based on impact speed.
- "Prototype jetpack", holding space while operating the jetpack will consume 4x as much fuel but double the pack strength.
Skydiver and Tough as Nails would be especially good for test pilots working the kinks out of planes.
And on ships being only flyable by an expert pilot, yeah maybe the stall-angle perk would be a problem there, but the rest are just sorta quality of life bonuses. No X% more lift or less drag or that sort of thing. I really like the idea that a talented kerbal pilot does something. Also the idea that a young rookie could learn valuable skills from doing routine stuff like hauling fuel up to a space station or doing aerial surveys or putting satellites into orbit, while the well-trained senior crew are off exploring planets. And incentivizing the use of bigger command pods / larger crews would be neat. Right now the only reason to send 3 kerbals to Duna is if sending just one is too easy.
Then don't do it. Nothing says you ever have to minmax anything, and this is doubly true for KSP. So what if you didn't get all the possible science you could have, that just gives you a reason to build another rocket and do it again, or do something else for more science. If a design requires a specific Kerbal and/or perks to make it work, then the design is flawed and needs to be reworked.
The exact complaint you have here was used against the tech tree, the economy, and just about every addition to the game. Including the Mun when that was first added.
Agreed 100%. Trying to level up my Kerbals sounds lame and un-fun. Having fun stats of courage and stupidity fit the tone of the game perfectly, and let us focus on the precious rocket making and flying.