phoenix_ca

NovaPunch.

That's a pretty neat camera, and I especially love the creative way it's getting around some fundamental problems (one of the major ones that is fast approaching, and the one that it mainly works-around, is that a single photocell must be at least 4micrometres wide, as anything smaller is useless...for some complex mathematical reasons I'd rather not get into; wikipedia has good explainations). Such a device would probably be used in SLR cameras with some creative re-engineering. The serious problem is with the lens though. Tbh, we aren't all that great at making perfect lenses. Many SLR cameras today have sensors that easily out-resolve many of the best lenses. This thing can clearly out-resolve even the best lenses on the market...and that really defeats the purpose. In terms of creative application, I don't see this being all that useful, since to get the image a photographer would want, they'd still need a medium- or large-format size lens to accommodate it. Obviously though, for surveillance, science, and a few other specific applications, this kind of design is pretty awesome. Though...where's the focus mechanism? We can make a sphere out of glass...that's not so bad. However, I don't know of any material off the top of my head that could deform the shape of a sphere like that, with the required optical properties. (It's possible to change the focus by moving the sensors, but I suspect that the reason such a system didn't arise out of natural evolution is the same reason we'd try to avoid it: it's really, really complicated. Adjusting the lens is easier.)

Nothing about multi-threading is as easy as that. Nothing. If it were, programmers would be doing it all the time. The amount of problems that come-up even just spitting a program into two threads is incredible. And you also lose a lot of control when doing it, without some serious programming voodoo. It's...a complex and messy problem. Multi-core processors didn't arise because it was a good idea to have multi-threaded programs. They arose because hardware engineers had essentially run out of ways to make single-core units significantly faster than their predecessors, so they slapped two on the same chip and dared software engineers to program it. Perhaps the better answer then is to overclock so long as you have adequate cooling. OC'ing the components of a laptop or iMac, say, would be a pretty bad idea, as those devices are designed with very tight limits on heat (how much thermal energy can be removed from the unit over time is quite limited, though in the case of many Apple products, the solid aluminum exterior confers an advantage, though not enough to get away with over-clocking). The biggest issue with OC'ing anything is inconsistencies that can crop-up by doing so. As an example, OC'ing a graphics card will inevitably draw more power, and produce more heat from the components. Excess heat can impact the efficiency of the VRM (Voltage Regulator Module, which produces a lot of heat), and when the VRM starts getting a bit silly, a spike in voltage to the card could occur (though it's more likely that an error is detected and the system shuts-down to prevent damage). It's why I've said that if you want to over-clock like a madman, then it's best to have a nice open platform with lots of space inside, and liquid-cool the major components. Liquid-cooling is way, way, way more effective than air cooling, and can allow for a lot of room to overclock components well beyond their rated limits.

Ahhhh, touché.

After re-reading both the OP and my post, I assure you, I wasn't misinterpreting anything. The question had already been answered, so I was posing another question. I'm wondering what would happen if you took as many Kerbins as would fit in Jool, and then stick said Kerbins into Jool, if you would get a star. Quite possibly the most awesome yet least relevant thought experiment in science to date.

How much green pigment, I wonder, could you create if you then mushed-up that zettakerbal into a fine paste and extracted the green pigments.

Considering that normal and anti-normal change as you continue your burn, 'tis the sane thing indeed. O.O

And...that'd be enough to print the side of a building at 300 dpi, if one wanted to...O.o

Another thread popped-up recently about this very thing. Slapping a MechJeb brain on the debris should do it. The downside is it won't auto-rename. Though when it comes to RC, you don't really need all that many satellites. Maybe four around Kerbin, three around the Mun and Minmus each...really, once you get the hang of putting them up there, it's easy, and once they're there, you needn't do it ever again.

Hm, maybe I should contribute some false-colour maps once they're out there. I'm reasonably handy with ze shoops.

One per frame vs. once per physics update is a clear winning tradeoff for the former. (At least for this application.) Thanks for the fix.

Ack, some physicist posted a book on here a while ago that covered orbital mechanics. Maybe if you go digging in the forums with Google search...

Dammit nhnifong, what did I tell you about roasting Kerbals?! There's not NEARLY enough BBQ sauce in that cabin to ensure proper flavour!

Who's attacking anything? I just asked a question.

Bugger the hand. Go back to the most important parts. The FREAKIN' LASER and the NUCLEAR THERMOCOUPLE OF DOOM that powers it. (Over-emphasis was entirely necessary, for those wondering.) On a more serious note...WEEEEEEE! So excited!

Try this? Seems to support layers.

Man...300MB and only a mere 1600x800. Yeepers. O.o Still, that's enough data to upsample it in photoshop easily with very little loss in fidelity. (I actually have a PS extension that can easily expand that to something insane like 10,000 pixels across...you know...if you wanted to make a whole floor out of it or something.)

The more interesting question is whether or not adding that many Kerbins to Jool would cause it to collapse and begin nuclear fusion. I suspect the answer to be "yes". Though since Kerbin already has the density of a neutron star in our universe...things might be different. Who's to say? And who are these "math peoplez"? All I see are those willing to learn, and those who are not.

Indeed. I already grabbed them. In any case it's not a big deal anymore. Context-sensitive healing brush fixes all.

I'm trying to get photos of Ike, and Jool's moons for a signature I'm working on. All I need is a decent contrasting picture of each, but the problem is the darned map lines and big ol' circle in the center of each moon when I focus on it. Soooo. Any way of tossing those temporarily? (I'm about to just fudge it and work with what I can get, editing-out the GUI-nasties later.) Edit: Hrm, well Vall, Tylo, and Bop were kind enough to have their orbital tracking line thingamajigs fade-out, but Laythe and Ike are still an issue.

Oh bugger me, you're right. E gad, it's too bloody early. >.< (No need for the link lol. I just got my brain meats all muddled. I think I need more coffee. O.o )

Yeah...um...use NovaPunch? Same bang for far fewer parts. O.o

No, 270 degrees (eugh, went to make the degree sign with Alt+0 and things asploded; stupid Windows -.- ) heading is due South, which puts you in an orbit with -90 degrees inclination. To get to a destination using the lowest delta-v, you wait for the phase angle of the destination planet to be the ideal angle given on the page you quoted. Though that blasted grey arrow clearly needs to go.

If that's the case, click all the arrows and de-warp slowly until 0.17.1.

One. 0.17. Why would you want to keep more?