EndlessWaves
-
Posts
1,359 -
Joined
-
Last visited
Content Type
Profiles
Forums
Developer Articles
KSP2 Release Notes
Posts posted by EndlessWaves
-
-
I'd have thought maximum rigidity would be gained by splitting the lower two stacks in half and mounting one half to each tricoupler, you can then join them together in the middle using just two struts if needed.
-
[spoiler="something like this?"][img]http://i.imgur.com/TNYw7.jpg
The img tag above won't work if copy and pasted by the way, I had to add an invisible space to get the code to display alongside the result as this forum doesn't support the usual copy and paste friendly way of faking it.
-
They'll attach to the three fuel tanks when those fuel tanks themselves were placed using tri-symmetry (and so on up the chain). If you placed them individually I don't know if you can switch.
-
Alternatively, build a more stable lander with a lower centre of gravity and/or more widely spaced landing legs.
-
What you want above all else is a high clock speed with good architecture, so a 2.2gHz i7 quad-core won't be as good as a 2.8gHz i5 dual core or something. Assuming of course they're the same generation and whatnot. A decent GPU certainly doesn't hurt either.
Core speed is only comparable within a range using the same design. A 3Ghz pentium 4 from nine years ago is going to be far slower than your 2.2Ghz i7. Even different models of current processors have big differences in how well they perform at the same clock speed.
A further note. I believe it has been said by the devs that the physics engine, Unity, will be incorporating Nvidia GPU PhysX at some point in a future release. Hence, if you are considering a computers specs primarily from the vantage of running KSP on it a Nvidia graphics card would be a better choice then an ATI.
Have they specifically said they'll be using GPU-accelerated effects? Most games using the PhysX engine don't include them and run equally well on AMD and nVidia cards (because it's all done by the CPU).
-
KSP is in beta, so future patches will add additional options and make the existing code more efficient. The Mystic Kerbal of Minmus might know how the game will turn out but no human does (the devs will have a target they're aiming for but stuff can end up better or worse than expected and most are reluctant to share that sort of information in case it's taken as a promise).
All you can do now is build a reasonable general machine, perhaps a little more CPU heavy than some gaming machines, and wait and see how it turns out. It's very unlikely that KSP won't run well on a machine bought new now - unless it keeps selling and development goes on for a couple of years longer than expected.
If you want specific suggestions then you'll need to specify which currency you're using. Several English-speaking countries use the dollar sign for their currencies and I'm sure we've also got the odd foreigner on the forum from somewhere else using dollars, such as Hong Kong.
EDIT:
As for quad core vs. dual core. Generally LESS is better, not more, given otherwise identical performance at full speed. A dual core i3 may be just as fast as a quad core A6 but will retain that performance is more circumstances as loads that are split into three parts or two parts can run at full speed rather than leaving a core or two idle and losing performance. Loads that aren't set up for multi-core processors at all will only run at half performance rather than a quarter.
I don't know how many cores KSP can currently use and what any future plans may be but actual performance is what matters, not the number of cores.
-
No editing until you get to 5 posts I believe.
-
I disagree with EndlessWaves on the power supply deal. As long as it is good quality, you can't go wrong with having too much power. The extra 40 or 50$ the 700W will cost you over the 400W, It is well worth it. I have had too many issue sin the past with computers crapping out from an underpowered PSU to skimp on that.
700W is the sort of power supply you buy for a system with twin 150W cards, not a single 40W/70W card (depending on whether it's kepler-based). I guess you're somewhere in the US if you mention dollars so prices are probably similar to here in the UK. Would you really spend the money on the extra spare capacity for the power supply instead of getting a 7770 or 560Ti? It'll depend on his local prices as to what's cost effective in this case though so there's no harm in looking into it.
the video card???? not at all balanced with the rest of the system video card i think you ought to bump it up to a 660 or better it should be out now very recently meet its released dateThere's no such thing as a 'regular' 660. The 660 Ti is the only model. It may also be well above his budget, it'll depend on local prices again but it could well push up the total price by an extra quarter. Assuming it's even available there of course, it's debut release was only last week.
-
The video card is fairly weak.
http://www.techpowerup.com/reviews/Zotac/GeForce_GT_640/6.html
Ah, I didn't realise they'd launched a kepler version. I was assuming it was the GF116-based GT640/545. Looks like they may have even improved the performance given that's a DDR3 version. I assume there's a GDDR5 version of the GK107 as well? I'd expect that to be closer to the 550Ti on that table.
-
The GT 545 (AKA the GT 640) has some poke but it's not that popular around here because the slightly more expensive cards are much more powerful. Make sure you get the GDDR5 version though, the DDR3 version is likely to be junk. The GDDR5 version is somewhere in between a 6670 and a 7750 so it's a strong entry level gaming card if the price is right.
The FX's weak point is gaming so I'd not use it in a primarily gaming system but if you want it for other uses or if it's particularly cheap where you are then it'll manage well enough.
My main concern would be the power supply. That system needs a 400-450W model so unless you've explicitly specified a 700W model for future expansion you're either getting a cheap unit that promises far more than it can deliver (and they're often poor quality besides) or you're paying a fair amount extra for something you don't need and the money could likely be used to improve everyday performance.
-
It'll depend on the weight of your lander as well as how much thrust you have. If you've got a 3 kerbal pod and a jumbo sized fuel tank you'll have to start burning long before you would with a 1 kerbal pod and mini fuel tank.
My first munar rocket required around 50,000m to slow down, the second one would do it in 15,000m.
Have a look at which height you stop at and subtract that from the height your started at to get an estimate of what you need to start at next turn. I'd guess it's probably an underestimate (gravity is stronger near the surface) so start a bit above that.
-
This is my first attempt at a space plane. I'm beginning to think the ramscoops are a bit too much for this plane, and perhaps it's a fuselage segment too long?
http://kerbalspaceprogram.com/forum/showthread.php/17692-Spaceplane-Design-Tips
-
This is my first attempt at a space plane. I'm beginning to think the ramscoops are a bit too much for this plane, and perhaps it's a fuselage segment too long?
The engine bodies and intakes don't seem to do anything currently, just stick a turbo jet directly on the back of a fuel tank (any fuel tank, rockets and jet engines both use the same fuel at the moment). Likewise the Avionics and Advanced SAS do the same job with different tunings so it's a case of using one or the other.
You haven't said what problem you have with it. You've got a very heavy back end and not much lift though so I'd expect it to want to pitch up but with the large wagglers at the front you should be able to curb most of that for level flight (although turning may be unstable). Moving the wings back as far as possible would likely help with stability, or adding a second pair to the end of or behind the current pair.
I can't see the rear wheels so if they're close together you might get it weaving on the runway a bit. Putting them at the edge of the wings would help (wasd and qe to rotate), but you might have to add a couple of braces to stop that heavy fuselage sagging if you do that.
I don't know the details of the bug but I've found with wide wheel placement veering is a very minor issue, most of my planes will get to the end of the runway without the controls being touched. Any spectacular veering is usually the result of unstable wheels, wings or just a part I forgot to make symmetrical.
-
You're wrong on that point. Wings do provide lift (although they also can deflect airstreams).
Really? Could you provide a simple demonstration model as I don't think I've ever seen wings generate thrusts at right angles when they're exactly edge on to the direction of motion (i.e. lift not through deflection). Certainly fitting them radially on a rocket doesn't cause it to rotate (which I've done a few times they make useful hard points).
I wasn't after troubleshooting a specific design - I've built a couple of dozen using lots of different layouts and engines - but rather after a discussion of the principles. What works and what doesn't in KSP's current wing physics.
But since you've all asked here is an example aeroplane:
Without wings it achieves 5008m (single fuel tank for 7 engines, deliberately limited for testing)
With wings it achieves 2350m if flown horizontally.
I could probably take one or two wings off the end and replace the silly tail with something sensible but I must be doing something else wrong as those measures won't close such a large gap.
-
There's no 'canon' for this yet.
No one would have believed, in the last years of the nineteenth century, that Kerbal affairs were being watched from the timeless worlds of space.
Another possibility would be Charizard for Charr, to borrow from another addictive game
Bop definitely seems like the tricky one. What about putting an H after it and having something like Bophran?
-
I'm struggling with space planes, I can get them to fly well enough but they all seem to perform rather poorly in comparison with stripping the wings and landing gear off and launching exactly the same machine vertically.
I'm not even sure how I should be building them. While making a better rocket is obvious - use as much power as you can keep under control - the issue of wings is mysterious to me.
I've gathered they only produce deflection and not lift so they're dead weight when pointing directly into the direction of travel and through experiments the optimal angle for lift during level flight seems to be somewhere around 35° (give or take 10°) with the effect dropping off sharply after 45°. But is this a viable approach or will current mechanics always result in this being substantially less efficient than vertical flight? The best I've achieved so far is around 50% of the height compared to vertical launch for the same machine minus the wings.
Should I instead only be looking at spaceplanes as just machines that land better than traditional rockets and still be aiming to get them going vertically as soon as possible with wings only being used as stabilisers?
-
What's wrong with Chartian? I kinda like it.
It does make it sounds like it's populated by Kerbal Chartists.
-
Well firstly, a moon is a moon; Mun is Kerbin's moon only. It's the proper name for Kerbin's moon.
This is similar to the name for Earth's moon actually being Luna. Thus, Lunar surface.
No, Luna is merely an alternate name used when it needs to be distinguished from other moons. The Moon is it's actual, full and proper name in English.
I imagine there's some method for naming the descriptive version of a proper name, eg Venusian, Martian, Murcurian, Lunar, etc.. I'm no English major though. So I couldn't really say what the proper method for deriving it for like "Eve" or "Bop" would be though.
Those words picked from other languages will often used the descriptive term associated with it there thus mars and martian or mercury and mercurial but it's often modified to suit English ears or uses. For example Venus would originally have been something like Venereal but that has another association with the goddess of love so venusian is more common now.
A few more for the list:
Charr will vary depending on whether you want to pronounce it with a long or short a (like Charred or like Charon). I would suggest Charrel and Charrl be added to the list though.
Eve could be Even, Evel (or Evil), Evemral, Eef.
Minmus could be Minute, Minus
Bop could be Boppin, Boppan, Bopapart, Boppy, Bope
Gilly of course has to be Katzenellen.
-
They're potentially heavier depending on which ones you're comparing and then there are the situations where X non-vectored engines are enough but X+1 vectored engines are required which can quite considerably add to the weight.
There's also the gyro option (SAS modules) in addition to vectored engines and winglets but they can be fairly heavy.
Does anyone use jet engines as stability controls? They have a good range of vectoring.
-
Click the box at the top of the assembly building screen that says unnamed spacecraft and it'll let you rename them.
There's a max persistent debris slider on the settings page in the main menu if you want to increase the total rubbish that can accumulate.
-
I usually brace against the direction of wobble, so if they're wobbling at the end of a stack I'd put the braces horizontally between the bottom of the stacks (on the last fuel tank). I don't actually know if it makes a difference but it looks best that way anyway.
How to make a basic plane help
in KSP1 Gameplay Questions and Tutorials
Posted
I'm still trying to work out how to make wings efficient myself. I personally prefer planes that are balanced back to front (a little more weight at the front on takeoff may be helpful as fuel tanks drain from the top of the 'stack'/most connections away from the engine). I would say build big and underpowered. You can always add more power/control surfaces but it's much harder to take it away