pa1983

That makes no sense at all. You want to have less ram use because KSP is 32bit and limited to 4Gb or so. On 32bit windows systems about 3.5Gb or less is available to the game because some addresses is reserved for hardware , for example some is reserved for the graphic card and that can depend on the amount of memory on the card. Do you know the cost of ram? Its DIRT cheap compared to most other hardware these days. Even a 500 buck laptop has 6-8Gb of ram. The ram cost for a mid to high end system is very low compared to CPU, GFX and other components, 10% or so. Last system I built the ram was about 10% of the cost and that was 16Gb 1600Mhz DDR3 so I would have saved about 5% total cost going with 8Gb ram and that system relies on the HD4000 graphics in the i7 3770 processor. Its about time people realized 64bit systems have been around for at least 40 years and on the x86 PC side its been around for 10+ years now and non x86 PC's even longer like Alpha and yes windows existed for Alpha and alpha was always 64bit so not like Microsoft did not have experience. I got on collecting dust. Can run Windows 2000 Beta and thats 64bit. Makes no sense to minimize ram use if it can be used to speed up the system. Only time ram is wasted is if its not used or the program is poorly written. Every modern OS will use what ever free ram there is to speed up the system by caching pages in ram from secondary memory (disk, networks etc). And if your running in to memory issues there is a working KSP 64bit linux version that do not have this problem. Also the game industry have finally started to go 64bit all out and 2014 years AAA titles will in some cases be 64bit exclusive to get around limits imposed by 32bit and 6Gb of ram is Minimum for many games even 8Gb or more is recommended and a quad core processor for 2014 big game titles. So yea 8Gb is MINIMUM this days, 16Gb is recommended if your planing to play future games, do other demanding stuff and have the system around for 2-3 years. Even 5-6 year old systems have 4Gb ram this days so assuming there still running and not broken the ram minimum for KSP is not a problem. There have been 32bit games for PC since 2004 at least. I know I got the Honor of trying out Unreal Tournament 2004 Demo for 64bit linux before the demo was officially out when Icculus (Dev) let a few of use try it out on our systems. The advantages back then could be discussed sense 1-2Gb ram was much then and I had 2Gb but even since most other gaming company have made very little progress in the 64bit game market. The 4Gb limit includes the VRAM on the GFX so There is no logic in fixing the symptoms rather then fixing the actual problem and that is that 4Gb of ram is to little for progress to continue. I have never run in to memory issues in KSP. But i usually have 1-2 mods at once only.

I have not timed it. But I run a relatively fast system, plenty of ram, ssd etc and under linux tough in wine but I cant stand windows. But I do not care as much for bad menus and load times as long as that dont effect the game once you are playing. But maybe people are to use to a steady but low frame rate or real time rendering being less then 1.0x aka in the yellow. I take slow load times any day if the game play actually is fast and responsive. Its not like you jump back to KSC very often in the middle of a mission. Only time I find load times annoying is spawning a big craft from SPH/VAB. But sure improvements there would be nice in the future to, most definitely.

I honestly dont remember. load times between scene is not a big problem for me. 3-4 sec there is not as frustrating as constant frame rate issues and lag. If that was improved the question then becomes, was that a part of the overall speedup he had done to the code or a separate speedup? I dont know. 0.22 will be out soon I suppose so we will be able to test how it actually performance then. But judging from previous "small performance enhancements" if your system realy dont suck and just barely gets by on the minimum recommended specs I would not hope for a noticeable performance boost. If we get one I would be happy but I dont think its realistic to hope for it. Just my opinion. Will see what the lay of the land is very soon I think.

On the other hand 0.18 had a total of 4 sub versions to and if I recall some value other then bug fixes where added in those. I still prefer if they fix real issues rater then adding more stuff that will only pay off in the final version of the game. Might rather save that and release career when its done. The fact that you have unlimited budget and no real missions in career only makes it a limited sandbox mode. Not that interesting even if I wish it was. But I suppose will see how long people find it entertaining.

We are talking about the very specific on he was working at that increased performance in the order of 30% on an i7 or more on slower processors according to him in KSP TV interview with Mu a few weeks ago. You know 30% is a LOT. Any other performance optimization was never mentioned. People disk ask Mu during the interview if the Ocean bug that causes frame drops when your looking at the horizon was addressed and Mu sad it was not. He had looked at it but still had not done a fix for it and this was at the time his 30% performance optimization was submitted so yea we can count that bug out as fixed to. So I cant realy see what they could have done that would beat fixing either of those in 0.22. I would be very sceptical if there is the slightest improvement in performance. 30% people would have noticed, few percent on its own would be visible only in benchmarks. Watch the last few KSP TV episodes and the interview is in there. Just sad it did not make it. A lot of people looked forward to get some much needed performance boost.

Performance Optimization was NEVER put in QA even if it was submitted, was to late. I never sad 0.22 did not go trough QA. BIG difference! If people could learn to read ever.

I was to until I was told this. http://pastebin.com/Z9dx6ZLa Basically Mu's optimization wont be in 0.22, was never put in QA and why did that never mention this officially then? The howl dev intervuew in KSP TV gave the impression it was being tested for 0.22 release. So from what it seems we are getting "new" landing gears (yea like that realy makes much of a difference) and then we get RnD and I realy dont know what its good for. Useless in sandbox and career mode with out proper missions and no currency system how fun is that? Its like sandbox but limited tech... Worst case scenario all this added features will just make 0.22 the slowest version yet. Yea I was locking forward to the 30% increase on an i7 like Mu had on his machine and more for those with slower processors. But "RnD" and new less then useful features seems to attract new players rater then keeping the interest up with the old once seems to be what counts. To me every new version take longer and longer time to arrive but the added value for old player is declining. I might even wait for 0.23 or what ever version actually will speed the game up and offer some new real game dynamics that are useful because right now I feel there is little to do that I have not done all ready. Excitement is pretty much gone. I would rater see them implementing Resources for career mode then as a compromise because that would work well in sandbox to. Also I would like to see some new parts people have been asking for that dont have an equivalent in the game atm and that would allow for all kinds of new creations and give old players and new an incentive to reinvent stuff. No my interest for 0.22 went from exiting to Zero. I have not played much last month, just tried out a few ideas and holding of for 0.22 because at least the performance improvement would have made some stuff more doable then before. Now I might even hold off playing for 3-4 months assuming 0.23 will have the performance tuning. If not I probably give the game up because its to easy atm and it feels like its just going mainstream. And Im no pessimist, I was optimistic this time to that something squade was developing, showing of would actually be in the game sense it was passed on for QA but I feel tricked again to believe that would be the case. Now I dont believe anything they say in KSPTV, there dev blogs etc any more and will assume the opposite until I have the final version of the game.

Well they could simply offer difficulty settings just like Open Transport Tycoon will. If you want IDKFA setting then fine as long as its optional and we who like the challenge to play a game and not know that the game cant be lost because its so scripted and fixed just to sell it because everyone needs to feel like a winner to buy it. I find it more fun to play a game over with everything reseted. That way its easier to try new approaches from previous experience and see if that works better.

Yea it dose qualify. It was on the old list for 0.18 I think, 0.19 is possible I bought the game on the release day of 0.18. Its long because I was dumb not to speed up the video to emulate real time. I was not that familiar with the real time rendering concept in KSP at the time so did not pay much attention to that. Video is when I built my second space station I think. But I think the original K-prize submission featured a 36T jumbo tank to 350 or 400Km LKO.

I dont see why you could not save and load ships from both a shared folder and a folder connected to the particular save. Shared crafts already exists but you cant save to the shared folder from what I know atm. And yes a Failure mode should exist. Go bankrupt and its over. I dont see why you could not play for ever tough but that would basically bee sandbox but with the cost if running the crafts, fuel etc. But some might like that. To fund your own projects you have to doe some contracts to even if those contracts probably are similar to the once you already made before. If your a new player and need som practise there is sandbox mode. Tough I dont see why people cant learn to play carer and not necessarily fail because there new.

I have not sad anything about reseting a tech tree. I only sad it fits well with the tech tree. A new player will ahve basic contracts, sub orbital, lko missions etc requiring basic parts. A player that have progressed further and are doing more advanced mission will also have done more science and unlocked more advanced parts making those harder contracts possible. And whats wrong with sandbox mode to actually learn to play the game? OpenTTD is based of a game almost 20 years old and its still being developed and played adn you can fail and have to start over but you can also play on for ever and apparently people just love that concept. I doubt KSP will be developed 15+ years from now still. How hard do you think it would be for squad to write some code that takes planets position in to account so it wont offer you impossible missions and timeframes? Also time is realy a none issue. To do a specific mission the time for transfer windows are more or less known already so they wont change much. If you do a stupid burn well thats part of the failing at the game part and if your lucky you only loose some cache by completing the missions a few days late but still make a profit. So as I sad not every one can be a winer all the time thats just plain boring but I suppose thats what there teaching kids and what gamers want this days. No challenge. And the way the game works now, if you abort a mission it will go back to the time when you launched it so not time is lost. So if we assume you can abort time dont realy mater much. Time spent building is more ore less irrelevant because as fare as I know it do not count now so probably wont then either. No reason why it would because you could just import a ship you made before to save time... exploit any one? So theres no reason for time to count in VAB/SPH. So time limit wont be a factor unless you miss your transfer window due to pilot error or badly designed craft, even then you could just abort and try again. Time building wont mater either because it wont count. So in the end time realy only mater if you take more then one contract, you complete one contract while you have not done anything on the other contract during that time period then you have lost time. From a game perspective the only good reason to take more then one contract is if it allows more then one contract to be completed with the same lunch craft like satellites or probes and that would save you money and be a good strategy. In the end, for those beginners that dont like to learn from there mistakes the hard way they can just "abort" and try it all over again and thats fine with me I do it to if I dont feel like trying again. For more hardcore player that like to except there mistakes that might cost them a perfect score or make a mission go in to the red they can still play on and try to make up for it in the next mission by making a bigger profit there and from my experience people will do both approaches from time to time. Its not like where talking super mario bros where you cant save and will hit game over in 15 minutes or less if you never played the game. Most of the "I dont like this idea because this and this will happen" is just ridiculous because its based on assumptions that are not even realistic. Why would squad write a program that gives you contracts you cant possibly do? Why do you assume contracts will be awarded with a time limit thats not doable? Its more likely contracts will come and go as "transfer windows" are available. Some missions dont realy have windows others do and those that have are harder missions so fits with the added difficulty of harder missions any way with better payout. But all this assumes squad wants a system where you are paid in advance to fulfil a contract. Anything can be done. Some times I think people forget this is a computer program, you can write ANYTHING with the right skill set. Computer is a flexible tool not a fixed function machine.

Im all for this idea. Why would it be a problem? Its not like it was a problem in games the last 20+ years? Transport Tycoon Deluxe, very popular still (Called OpenTTD now) there you had to take a Loan, loan had interests and if you could not pay your interests to the bank it was Game over and you had to start over or load a save. Sim city had similar functions. To stop exploiting a set date for the mission to be completed would be a part of the contract just like real life, every day spent past that deadline would cost YOU money that you would pay back just like real aircraft manufacturers do when some one orders a craft and its not delivered in time. In the advent you want out of a contract for any reason you would have to pay back what you where given plus a severance fee based on a percentage on that contract. Now if you try to exploit this by just taking a bunch of contracts you will soon bleed money and only way to have a positive bank balance would be taking more contract but that would only make it worse as you will bleed more and more money every day until you go in the read faster then you can make up for it. After a year in the red you as the player would loose you management position at KSC and well that would be game over. The more contracts you take and fail on the worse your reputation gets so you get fewer contracts with less payout margins so in the end if your just trying to cash in fast you would not get any contracts eventually until you actually fulfil some of those contracts and get your rep up. I realy hate the idea that every gamer should be allowed to win a game. In the old days there where not even save files so you COULD fail. Now days most games must be run on the hardest difficulty to offer a challenge. This system would award players that take contracts and completes them on time. More contracts would be given that are harder to achieve but with a bigger payout and margins then for a player that fails to deliver on set dates. So for a good player there would be lots of contract to opt from with a wide range of difficultly and payouts. For the newcomer easier contracts would be the only option until the player rep improves. Those would fit well with the Tech tree and the learning curve of new players. Old players will most likely do the easy missions pretty fast and soon get more challenging missions. This way the game is relatively balanced but you can still fail hard but you can also show what you got if your a skilled player.

There are documentarys on the subject. I think the conclusion was that it could not be made back then due to wight but to day with modern reactor technology and shielding it could probabley be done.

Think the heaviest I did that did its job was 245 or 248tons with a jumbo tank as payload. It was the old Falcon IV Cargo Transport if I recall. Looking back at that craft now its realy a flying brick Cant even glide with out power. More of a rocket with control surfaces and landing gears then a spaceplane hehe. I knew little about spaceplanes back then, where not that many doing spaceplanes either compared to now. Still do have a video of the craft. http://youtu.be/XROy7-eMVRY I think girders where reduced in wight in a later version of KSP (any one can confirm?) so it drooped to 235 or 238 tons. Any way that craft and most crafts are lighter then that. My craft was not 248t or what ever it was less because landing gears have no wight in the game except in VAB/SPH. I dont think the bug was so well known then. I sure did not know about it. One reason landing gears should be mounted last because SPH/VAB assumes it has wight so CoM will be off.

Yea the payload door hack is pretty bugged in 0.21. I have not found any other way of doing it that is more reliable tough. But there are mods for hinges that should work for those that wants to replce them. I really hope there are hinges in 0.22 but I doubt it. Its really needed. The vertical speed indicator is pretty crappy yes Not a precision instrument but its accurate enough between -100 to a 100 m/s. Realy good way of staying level and keep climbe rate optimal with some practice.

They look nice. Good work!

Your correct that flipping that craft was unstable. That do not mean others cant make perfectly stable low profile winged craft. I never sad your craft was stable just that low profile winged craft can be perfectly stable. I can dive at full power strait down and pull up hard and never let go of the S key and still get out of the dive with out flipping. If you dont believe me I can make a video of it. Its a Delta winged craft. All my latest craft are low profile based because its overall better concept if you ask me. Spreads the load better on the ground and in the air. For the first 6 months that I did spaceplanes I made them with mid profile wings because low profile wings percent more challenges then your typical sandwich mid mounted wing. But doing one test and conclude that they are bad because they might be hard is not a good idea. There perfectly doable. I can keep holding S and do loop after loop and it will never flip. Crafts flipping is caused by many factors not just one. If the craft is to rear heavy it will flip because the rear end is just waiting for an opportunity to travel first. Make the craft better balanced and a low profile wing will not be a problem. Most beginners also use control surfaces inappropriately. To many are often used because they are trying to compensate for ASAS behaviour or the simple fact that CoM is moving or out of place relative to CoL etc etc. More control surfaces are not better to solve misbehaving crafts. Theres a fine line that offers just the right amount of authority with out compromising stability or responsiveness. Placement of control surfaces are extremely important and if you ask me almost always never placed optimally. In the end it do not mater if a high mounted wing in theory is more stable if one can create a perfectly stable low mounted winged craft. End result is what mater. So like I sad low profile winged craft are perfectly stable and doable if done right.

Low profile wings are not a problem nor unstable. Only real problem I had was engine placment and structrual integrity but once I solved that on my crafts it wirks just fine. My lates craft can fly up side down but rolling from level fligt it usually stopps its roll at a 90 degree bank. Its perfectly posible to fly upside down but it seems to prefer rigth side up.

I use the vertical speed indicator in to the right of the altitude indicator. I just keep my eyes on vertical speed relative to my altitude so I dont go to fast, can happen with a light payload, and if I go to slow I just add a km or two to the first couple of steps so gain altitude faster. So yea you need to have an ascent profile and I now days just use what I remember from the top of my head and then you need an idea of whats a god vertical speed at etch step and then correct when you actually pitch down for the next step depending on vertical speed. So yea I kind of fly by feel this days if its a craft Im use to. So yea eyeballing it I suppose But a good ascent profile helps then its just practise.

I dont realy have an ascent profile for it but its basically the same as the kosmos or the old Jumbo. This days I fly by tracking Vertical speed and try to keep that optimal. Well seems Like I got most values wrong with 5 degrees. I rewrote it because I found the first way hard to interpret so it seems the error slipped in there early on. Its all in the video but I have updated the ascent profile. Any way you cant blindly run the numbers in the ascent profile because the way ASAS works you need to correct all the time. I suppose some on could write a kOS program for it that would make it reliable all the time. But manually you still have to keep an eye on vertical speed and pitch and well use your experience for best results.

Yea once. Some on sad it was so hard to do because of high delta V requirement even compared to Duna etc so I had to try it. I modified an existing craft used to hull payloads to LKO if I recall. Added engines and more fuel. SO it was not balanced well or perfect but that was not the point. So no the craft do not exist any more, it was just a test. But it worked and got back. But I dont see a point in landing spaceplanes on a body with out an atmosphere. The howl point of a spaceplane is to use the wings to reduce the fuel consumption. No atmosphere no spaceplane. I usually hull a 60-70 ton spaceship to LKO and send that to the body lacking atmosphere.

lower your physics delta to 0.03 that should help a lot with the performance.

I dont know what NASA would go for but I know what I would like to see some one do, NASA or other organization/company. Spaceplane! Its about time Space exploration became reusable and environmental friendly. Sure you can argue all you want that it has little or no effect but solid boosters are not the most friendly once and trashing the ocean with valuable metals is just a waste even if its just a fraction its a waste. Hydrogen Powered spaceplane would be extremely friendly to the environment and fit with what humans are trying to do on all other sectors, make our way of live having less of an environmental impact. Re-usability and a friendly fuel goes a long way. Im not saying rockets should be band just that if there was a spaceplane like the skylon it could do a big portion of the normal jobs and leave the more demanding once to rockets. It would also be nice to see what effect a reusable spaceplane like the skylon would have one the exploration of space and the space industry and the technology that would need to be invented to make the first spaceplane. Might be a failure or spawn new businesses and science. Im not much for, ooh no one has done a spaceplane so we dont know if it works jada jada. Well some one probably sad the same about going to the moon in a rocket once. Lucky for use no one listened to them. If NASA or some on else gets the funding for a spaceplane it will be built and refined until it works.

Thats why I mapped staging to backspace from day one. Spacebar is so stupid for staging. Dont know why its default in the first place.

351 intake air? that's 1755 stock intakes. Must be something wrong here because one could easily run 12 torbojets at 100% throttle at 33-34km on 12 intakes per engine and reach speeds off 2000m/s or more at 210 tons and easily climb to 40-44km altitude on jets alone. With all those intakes reaching orbital speeds and even slingshot in to space would be doable. This do not make sens, firring rockets at 31km altitude and 1800m/s with so much potential intake air don't make sens. I'm just wondering what's going on here because that intake capacity is odd.