A question: Since gravity can and does alter the paths of photons, why couldn't it alter the path of the particles just as much?
It does. But gravity doesn't alter the path of photons more than 1 degree (hardly! it's more like tiny fractions of a milliarcsecond!), which is all any of the cosmic ray detectors can resolve to. This should give you an idea of what we're talking about - even only given a pointing resolution of 1 degree, we still can't find anything in that direction which could generate a particle of that energy.
Another: Could a pair of very massive, closely orbiting black holes provide an acceleration mechanism (not to those energies, but seems to me that they could kick particles out at very high fract-c)
I thought I mentioned this in a different post.:) You can't accelerate efficiently with gravity, since it's so weak. Yes, it's strong near a black hole, but the region where it's strong near a black hole isn't large enough to allow particles to remain for very long. There are three things you need to accelerate particles, as far as we know ("stochastic acceleration") - 1) something to steal energy from (i.e. a shock wave), 2) something to prevent a particle from escaping (like a magnetic field), and 3) a large area over which to accelerate (i.e. a few lightyears in size). Faster shock wave, higher energy. Tighter confinement, higher energy. Larger area, higher energy. Unfortunately gravity's "acceleration region" and its "confinement region" are coupled, so you can't get to high energies, because when you increase the area, you reduce the confinement, so particles don't stick around long enough to accelerate to such energies.
Black holes do in fact accelerate particles (in jets). Active galactic nuclei are thought to be caused by black holes whose jets are pointed close to us. Blazars ("extreme" AGNs) are thought to be caused by black holes which are nearly pointed straight at us. Seyfert galaxies, radio galaxies also are other classes of "angles" of galaxies with big black holes in the center. Closely orbiting black holes would likely produce very high energy jets, but I'm not even sure they would exceed the energy from the jets produced by the merged black hole.
Active galactic nuclei probably do produce a lot of the 10^14-10^19 eV particles, but they probably do so through shock waves propagating through the galactic magnetic field caused by the jets. The problem is that a larger black hole wouldn't produce that much of a stronger shock wave, and the magnetic field and accelerating region are set by the galaxy, not by the black hole.
Is it possible that these particles are related to GRBs? Might explain a few of the oddities away.
One of my closest friends has a keen interest in looking to see if there's an enhancement of cosmic rays in any way correlated with GRBs. We've got 10 ns timing on the arrival of the particles, so we could definitely correlate them with GRBs. It's worth noting that there would probably be a time delay since the photons aren't affected by magnetic fields, and the particles are (however slight), but hopefully that systematic would be determinable after enough data.
But we *have*; and given the repeated observations...regardless of the curve, there will *still* be some particles at the extreme ends. Given that our detectors are still fairly primitive, it's possible that the high-e events are statistically more likely to be detected, is it not?
(Yes, I know we've seen them. Otherwise I'd be in a different field right now, and the waiters in Malargue, Argentina wouldn't all know me by name.:) But the fact that we've seen them probably implies that we don't know the source, not that we don't understand the propagation.)
I think you're missing what I'm saying - the only way we could've seen any of these particles is if they came from less than 50 Mpc. The GZK effect gets much, much stronger as you go to higher energies. After 50 Mpc, a particle that starts off at 1E21 is below 6E19. Same with particles of higher energy. Astrophysically, that's right in our backyard. There's nothing we know of that could accelerate particles like that. There's an additional problem, which is the fact that there's a spectral change in the 10^19 range that we can't explain, either. Spectral changes occur when acceleration mechanisms change. Supernovae fall apart in the 10^14-10^15 range, and there's a spectral change there, too. The fact that the spectrum continues after 10^19 (in fact, it flattens) implies that there's a new source that's "turning on" in that energy range.
As far as I understand it, the mw background is the peak energy distribution, not the total one. It's what we built our detectors to observe because it's easiest to observe - sorting random noise from the detector noise in gamma/XR is damned difficult.
It's a pure blackbody distribution, with a characteristic temperature at 2.7K. In order to have a particle out of that spectrum at 10^20, it would need to have about 10^23 times the most probable value. I haven't done the math, I'll admit - but something like e^-(10^23) probably times even the size of the Universe probably doesn't even equal *1*. (Oddly enough e^(-(10^23)) times the *number of bits* in the Universe through the end of Time wouldn't even be 1)
We don't know that, because we don't have a source for those protons. Sure, there's nothing on that vector, but even ultra-high-c factor particle paths can be altered by gravitational fields - which were not mentioned in the OMG webpage)
Yes, we do. The GZK cutoff is not "experimental physics". It's the delta++ resonance. This is stuff that they did in the 1950s, and has been extremely well studied since then. It's just proton-photon interactions. Unless Lorentz invariance is wrong (which is possible! but you should read the paper on that suggestion, and it's very, very bad), we know that GZK will slow a particle travelling faster than 6E19 couldn't've travelled more than 50 Mpc.
And gravity doesn't alter particle's paths anywhere near as much as magnetic fields do, and those particles have such high speed that neither gravity, magnetic fields, or anything else could possibly alter their path more than a tenth of a degree.
The "OMG particle"'s webpage (which, by the way, I've never heard it called - which is... odd) is a little sparse. Try the Auger homepage, this UNM site, or this LSU site. If you've got access to Science magazine, also here.
There are actually many, many more interesting things going on in the UHECR field which I haven't even mentioned.
Well, clearly you are wrong, else these would not have been detected -- something did accelerate them this much, and I doubt it's not, as you say, "in the Universe".
You know, that statement seems so obvious, but... it's wrong.:)
The particles could've been formed by the decay of massive, ultra-high energy structures or particles. This is known as a "top-down" approach.
However, I should've said "nothing that we know of", not nothing. But the top-down approach could be real.
My understanding is that smaller black holes radiate energy faster than larger ones because of the different ratio of energy per "surface area" calc thru the schwarzchild radius. Higher energy density, in effect, although that's not quite accurate either. Larger ones radiate less - and therefore decay slower - function of energy density radiated thru a ^4 surface area. (argh;)
Yes, you're correct, and that's what I'm trying to say. What you hve to remember is that large black holes become small black holes, and the only time that you actually see black holes emit particles with any frequency is when they finally evaporate, and that happens at the same mass for every black hole. Black holes that start big evaporate more energy (to conserve mass, of course) than black holes that start small, but they evaporate it over massively increased timescales, and the energy that's radiated is so pathetic it's dozens of orders of magnitude away. Small black holes might give off enough energy, but in order to compensate for the observed flux, they'd be happening in our atmosphere (which... is possible, but really, really not likely, and not very consistent with other things we see).
The final black hole evaporation is the same for every black hole.
What about mergers of black holes? Can they produce such energies (there are some who think that there are two multi-M-stellar mass black holes at the center of our galaxy - are there potential merge scenarios that can produce those energies?)
The merger of the galaxy would produce more energy, because the scale is larger even though the magnetic field is smaller. It's still too weak to accelerate particles to those scales, by several orders of magnitude. The black hole merger itself won't produce much energy. How would it? Gravity can't really accelerate particles very well, because it can't repel, only attract. Fermi acceleration is one of the most efficient acceleration processes known, and it basically goes like this - imagine a big wall, moving towards you, and you throw a tennis ball towards it. The tennis ball moves back at you faster, because the wall's moving. Now imagine there's "something" which turns the tennis ball around and runs into the wall again (magnetic fields, typically), and repeat the process. Gravity simply couldn't do it because once you exceed the escape velocity, you're gone, and the only places that you have a high escape velocity are too small to keep a shock wave in the region for very long.
There's a famous figure that everyone shows during GZK cutoff and UHECR (ultra-high energy cosmic ray) talks which shows the cutoff line for particle acceleration to 10^20 eV, and all the high energy processes in the Universe below it.
AFAIK even the decay of a very large (hundreds of stellar masses++) black hole can't produce protons with that energy. I'm not sure about mergers of galactic center black holes, tho - but I'm sure they've taken those into account.
Black hole decays don't produce that much energy per unit time - Hawking radiation happens in exactly the same way for every black hole, so every black hole ends in a burst of exactly the same amount of energy. Large black holes simply take far longer to reach the "fizzle-pop" stage, but in the end, the pop is exactly the same for each one.
This is because the amount of Hawking radiation a black hole produces is dependent solely on its mass. Big black hole, very little Hawking radiation. Small black hole, lots of Hawking radiation. The reason you have a "pop" is that as the black holes radiate, they get smaller, and so there's a runaway effect. What I'm saying is that a big black hole's decay is identical to a small black hole's decay, just separated by several lifetimes of the Universe.
The problem is that you have to consider how the particle gets accelerated. There's really only one way to accelerate them, and that's via magnetic fields. It's the only thing that can convey enough kick to the particle. We currently believe, for instance, that supernovae are responsible for almost all of the cosmic rays we see - when the supernova goes off, the shock wave accelerates particles via something called Fermi acceleration, and you kick particles up as high as 10^14 or 10^15 eV. Higher energies are formed by even bigger accelerating regions, but they're all shock acceleration. But nothing in the Universe is big enough and with enough magnetic field strength to accelerate particles to 10^20 eV. Not neutron stars, not active galactic nuclei, nothing.
I wouldn't think there would be an absolute cutoff there, but a (granted very steep) curve; which means that some very tiny fraction of particles *could* make it here with those energies (and who's to say what original energy level that particle started off with?)
No, of course it's not an absolute cutoff - however, the slope is somewhere in the neighborhood of E^-10 or so, which may as well be an absolute cutoff. No matter how hard you try, you basically can't get much above 6 x 10^19 for more than about 50 megaparsecs. If the GZK cutoff really does exist (which... well, it better, it's very basic physics) then in the absence of sources we don't understand (which is what we think we have), we never should've seen these particles. The "normal" processes which generate particles less than 6E-19, convolved with the GZK effect, would've produced a flux so freaking low we never would've seen it.
what are the odds that the particle in question could have resulted from the Big Bang energies once protons and neutrons started to form from the 'soup'? I realize it would have been traveling for quite a while and the odds would be infinitely small, but still, the mw background is just an average temperature, is it not?)
Actually stuff that's formed from recombination era would be microwave background energies - because, well, that's what the microwave background is.:)
But anyway, it's not just that we saw one particle, because the thing is, the detectors didn't run for that long, and they weren't that large (i.e. their acceptance was quite low). They would've had to have gotten astro-freaking-phenomenally lucky in order to see one that far away from the expected. It gets even worse when you have other detectors come online that also see those energy events.
It's not the individual particles that interest us. It's the fact that there seems to be a real spectrum out there - there's something actually producing these energies, and either A) it's close, or B) we don't understand interactions at high energies, or C) all of the cosmic ray physics people are smoking something. Considering B) basically implies that one of the fundamental tenets of relativity is wrong - which would be bad , I'd like for it to be A, but I've got a feeling it'll turn out to be C.:)
The really interesting part is that we don't really know what process would produce such a thing.
Actually, it's worse than that. Not only do we not know what process would produce such a thing, we don't know how it would've gotten here in the first place. Above 6 x 10^19 eV, particles should interact with the microwave background, and lose energy (the "GZK cutoff"). In essence, there's a cosmic speed limit. The only way that particle could've gotten here is if it came from very close (so it didn't have time to slow down yet) - very close. Which makes the problem of "how the heck was this made?" even worse.
To be fair to our hard-working particle physicists, you would need a VERY large detector hovering high in the air if you wanted to catch these things in nature.
Hell lets stop pretending anymore. Lets create corporate representatives and get our elected ones back. All these probablems coming from the fact that corporations are seen as people in the eyes of the law. Make the House, the Senate and the Market. People elect the first two, corporations vote for the third and can't contribute to the others.
That's not a bad suggestion...
On the idea of changing government, however, there was a very good suggestion in the Mars series by Kim Stanley Robinson - make one branch of the government like jury duty - that is, a random selection from eligible voters, and required service.
This has the benefit of being basically uncorruptible - it will be very expensive and very difficult to buy someone who doesn't want to be there, and who more than likely won't be there at the next changeover. If you stagger terms of office as well, that'd ease some of the transition periods.
You'd probably want to limit the scope of that body in the constitution, but so long as it can prevent amendments, it will have sufficient power to combat significant government corruption.
Then you'd have a three body legislature: one where the people elect the representatives, one where the representatives are selected randomly, and one where the corporations select representatives. It'd be worth trying, at least.
So you want to say that a PSU that has 10W heat with 100W load, which double to 20W as the components age, is in ANY way worse than a normal PSU that wasted 30W to begin with?
Depends how the design was handled. If the components were placed to handle 30W, then no, it's strictly better. The problem is that most people don't do a full thermal analysis on a design - they simply do it, then check its temperatures while running, and adjust things if it's too bad.
This means that most (cheap) designs are marginal - that is, they don't work a significant amount outside of their specified thermal envelope. A design that's marginal at 10W that encounters 20W is likely to fare worse than a design that's marginal at 30W that encounters 40W. Most tolerances built in are percentages.
So you're right, if it's a quality design, with a lot of tolerance built in. However, I doubt that most power supplies are quality designs...
I agree that there is not much performance loss in writes, but there is still SOME performance loss.
A properly designed ATA RAID controller will have no performance loss at all. The writes will simply be cloned at the RAID controller and go across the individual busses simultaneously. From the OS perspective, it has issued only one write. From the drive's perspective, the situation is identical to a single-drive situation (as each channel is only loaded with 1 drive).
With more complicated solutions (mismatched drives, non-full RAID solutions) of course you're going to have performance loss - however, in those cases you may still get a performance loss in any RAID situation. Specifically if you've got non-full RAID solutions, RAID0 can still even slow down as compared to a single isolated drive, as you could clash with requests for the non-RAID portion whereas in the independent drive case, you never would.
However, the real world implications are few: RAID1's write performance is almost always identical to single-drive, and its read performance is about 20-30% higher than single drives.
There is a reduction in performance for writes (twice as much data must go through the bus)
That's also not true. Since the bus speed typically is much higher than the drive's actual sustainable data speed, RAID1 in hardware generally doesn't show any performance degradation from writes, unless you have extremely, extremely fast drives on a slow bus. On a dual-channel ATA RAID solution, fundamentally this doesn't have to cause any slowdown, as one write command issued to the controller simply becomes a write on each of the attached channels, at the same time.
In software RAID all it really does is up the CPU usage.
You can see the recently reported Tech Report article on SATA RAID here. Write for RAID 1 and write for single disk are pretty much identical. Read is higher for RAID1. Write for RAID 0 is, of course, much higher, as you multiplex the latencies for different stripes.
So, ESPN's website has an absolute *ton* of off-site material, so there's a ton of DNS lookup, loading, etc. Both Firefox and IE load them pretty much equally for me, Firefox was a little slower. So I saved it as a file (complete), and opened it up in both browsers.
Mozilla opened it up then very fast - virtually instant. IE crashed right after loading it (and continues to crash, each time I load it like that...)
I'm not entirely sure which versions of IE you've been using, but you've been able to use the keyboard to do anything and everything in IE since at least version 4.0 - if not earlier.
Not cleanly, and not completely. Caret browsing allows you to actually select text with the keyboard, which you can't do in IE (easily). Essentially it turns a web page into a normal document, and when you hit a link you can just hit enter and it will follow it. It's also easily turned on-off (just hit F7) - this way once you get efficient at using it, it makes things extremely easy.
The generally lower speed and reliability compared to IE on Windows?
Firefox 0.9's (well, any Firefox since 0.7, basically) rendering speed is far faster than IE on Windows. Load up any complex page (like CNN, or Slashdot, etc.) so that loading times can equalize between the two.
The startup time is a little longer, but which do you spend more time doing - opening a web browser, or browsing the web?
They care. They just don't know. This is, of course, why Netscape was part of the Microsoft antitrust trial. I couldn't convince my mother to switch to Firefox - she doesn't even really know what IE is. She just clicks on the web-browser thingy, and poof, up a page comes. I'm really tempted to just secretly install Firefox for her one of these times, but she tends to actually play the wacko online games that sometimes require ActiveX (and fill the computer with gigantic amounts of spyware and worms).
My mother hates that she needs to deal with the antivirus stuff, and when something actually comes along and does some damage, she's very frustrated by it. She just doesn't make the connection back to IE, and she doesn't even *know* there's another option. Anyone who thinks that IE dominates because of its technical brilliance is a fool. IE is not technically better than Firefox - it's slower, renders pages far worse, and is missing a lot of the standard features so much so that third-party software has to fill the holes. IE dominates because it's bundled with Windows, and how Microsoft was allowed to extend its monopoly, I will never understand. Money should never outweigh doing the right thing. Sigh.
I have used tabbed browsing. I don't find it an attractive feature. I don't know why everyone seems to believe it to be the best thing out there but I just don't need it. Another instance of IE is fine for what I do.
Tabbed browsing doesn't hurt you, you know. It's just a new feature, and it's one that doesn't even change the way the web browser works. It just helps those of us that have learned to use it well.
There are two main benefits of tabbed browsing though - some people say it allows for a "spatial metaphor" of web pages, so when you're browsing a site you can keep a "bookmark" of a previous page ready for easy access. IMHO, these people are needlessly looking way too deep into this spatial metaphor crap.
The second benefit, and the reason I use it, is that for me, the web browser is different than any other application - it's usual that I'll need several browser windows open, and I don't want those cluttering the taskbar. XP is better than previous versions - it at least collapses them all to a single tab - but you can't "Alt-Tab" through windows in the same program, and in order to see all of the browser windows you have open, you'd have to use the mouse to go down to the taskbar. Especially for someone who types very fast (like me), any time you're required to go to the mouse, it's a pain. Tabs create a "browser taskbar" for me, which is a big help.
Pop-up control is a bit of a hassle but Google's toolbar stopped that stuff for me and added Google to my toolbar. I understand that a browser would be better with it but it certainly not necessary.
This is an advantage of Google, not of IE. You still need to download the Google toolbar, and figure out how to use it. Yes, that's pretty simple - but so is configuring Firefox, especially under Windows. You can't tout IE's out-of-the-box features, and then compare a modified version of IE to Firefox.
I don't know what caret browsing or form management is. I can't imagine that Joe Blow would need it if I (as a regular Slashdot whore) don't know what it is.
Caret browsing allows using the keyboard to navigate, a la Lynx. Again, for those of us who type fast, this is a massive improvement.
I don't need "block images from..." I guess ads are a pain in the ass. Most users just deal with them. It's certainly not something I would require in a browser to make it functional.
You would if it was over a slow link. Then the difference between loading ads and not loading ads becomes immensely significant.
I guess we differ on our opinions of what is necessary. I think easy configuration, BEST configuration out of the box, and 100% perfect rendering on ALL pages being most important. Firefox/Mozilla don't offer those things to me.
Wait, for a second at the end, I thought you were going to say "IE", not Firefox/Mozilla. Firefox renders pages far better and far more standards-compliant than IE does. Not only that, but it does it far faster, too. Startup takes a little longer (I guess it helps when your render DLL is already loaded...) but page loads are far faster under Firefox.
Claiming that IE has 100% perfect rendering on all pages is a joke. Just google for "IE rendering problems", and enjoy the flood. The simplest, common problem: IE doesn't alpha-blend with PNGs. Considering people have been moving away from GIF slowly but surely, and PNGs have become far more common, this is really unacceptable.
If you want a browser that has the best rendering on most pages, you want Firefox, or Opera, or even Safari, but not IE.
If you're only going to be able to move 300 meters/day then you're going to need a lot of days to get anywhere beyond your landing zone. To get anywhere, you'll need a power source that doesn't crap out after 3-6 months.
Absolutely. I completely never disagreed with you there at all. For long-range rovers, you have to have a nuclear power source. But "long range" means tens of miles over probably a year, and it does not mean driving around at speeds much faster than a human walking. Those limitations simply come from limitations of robotic teleoperation.
The problem with nuclear is that you then need to justify (and pay for!) multiple years of operating funds, which will be very difficult to do.
You can't simply ramp up and down your staff as things get more and less interesting. People don't like to work like that.
well gee, it can spend the time it's waiting taking a gander at its new environment. And if you have scads of power, then you can crank up your transmitter's power and send more data in a shorter amount of time.
In its new environment, it will have no idea what's interesting or not, and the ground controllers can't really plan ahead, either, as they don't know what the area looks like either. The only thing it could do is take panoramic images, which would take any rover a few minutes, if not seconds.
You can't "crank up the power and transmit more data" - data rate is frequency-limited, not power limited. A nuclear powered rover would still transmit data at the same rate as the current rovers. The only way you get a higher data rate is with a higher frequency or a better encoding scheme.
We appear to be talking past each other. You're right that it's a bumpy place so you have to move carefully. My main point is you can do better than 100 feet a day. Heck, even one mile a day is 50 times better and you're still crawling.
(Point of note - they're more in the range of 100 meters a day, not 100 feet a day)
Moving quickly brings its own problems. You need to make decisions a lot quicker, and analyze problems a lot faster. They're still not that comfortable with the onboard route-determination algorithms. The capability of moving faster doesn't mean they'd do it. Look at Spirit, which has seriously booked it in the past few weeks compared to what it had been doing, and several of those days, the trip was cut short because of software issues (it didn't think the route was safe). The capability of moving faster may not have actually led to faster net movement, just faster "go, then stop" - remember, JPL has to wait for telemetry uplinks to find out how Spirit did on a previous command pass, which means that if Spirit had had the capability of moving faster, all it would have done was waited around longer.
Opportunity, on the other hand, in Meridiani, could've benefited from a faster movement capability. But again the other problem is that wheel slippage is also a problem - the dust is very fine, and moving quickly is a serious concern. And Meridiani is basically the only place on the planet that's that flat and easy to move in.
Now, of course, after the route-determination algorithms have been exercised, they might be more comfortable at moving faster. But it still would be rather nerve-wracking. After all, how would you know where you're going? On Mars, for a robot half the height of a human, the horizon is ~1.4 miles away! The people at JPL would have no idea what lies ahead, unless the rover stopped, and transmitted pictures - which it can't even do all the time. Which means that there's basically a maximum limit to how fast a rover can move during the day, and it's probably significantly less than a mile.
In other words, I doubt even with a nuclear powered rover that the rover would move very fast. You're teleoperating a robot from another planet, and you're doing it blindly. It's just not something that can be done quickly.
Did you miss the part about shrinking it down to modern geomerty, meaning it would run faster on less power (read less heat) than the original? Sure a 90nm i486 isn't going to run at 3.6GHz like a P4, however I expect it would run a good amount faster than a 486DX2-66 once did.
Unfortunately, nothing will beat the architectural gains which have advanced since the 486 era, and the "worst case" pipeline waits will keep your clockspeed at an insanely low level.
Let me try to explain. The 486 had a 5 stage pipeline - fetch, decode, dispatch, execute, and writeback. Now, each of those pipeline stages isn't going to take the same "minimum" amount of time - some of them are fixed by things other than switching latencies. So, say your execute stage is fine taking only 1 clock cycle up to, say, 2 GHz (a minimum latency of 500 ps), but your decode stage, simply from physical concerns, is going to take at least 5 ns to complete. This means that the maximum you can ramp the clock speed up to is 200 MHz, because each stage in the pipeline has to take 1 clock cycle, so if 5 ns is your minimum, you'll have a max clock speed of 200 MHz.
The solution, though, is obvious - break that "5 ns" decode step into multiple pipeline steps - say, 5 of them, each taking 1 ns each. Now your maximum clock frequency is 1 GHz. The problem is that your pipeline is now 9 stages long, and you have a new architecture - which is precisely what Intel did several times over to allow the clock speed to ramp.
And that's just the pipelining limitation. There are other architectural problems with "ancient cores" as well. One basic problem is that the x87 floating-point architecture is crap. It's stack-based, which means you can only do math with the "stack head". So in order to store things in the registers, you need to use the FXCH instruction to switch the stack head and one of the registers. Well, modern CPUs (the P3 and the Athlon) got around this by saying "we'll make FXCH be a zero-cycle execute when paired with an arithmetic instruction (and after the Pentium, screw it, they're free totally)". Since the modern CPUs can decode more than one instruction per cycle (3 for an Athlon), and the FXCH instruction only lives up to the decode stage, you're really not hurt, as the FXCH fills a pipeline stage that probably would've been left empty anyway. Now consider the P4, which was designed to try to encourage people to move away from x87: it does not have a zero-cycle FXCH, and its x87 performance is abysmal. (The 486 does not have a pipelined FPU, nor a free FXCH instruction. It would be even worse.)
And I haven't even mentioned register renaming yet, which works around the register limitations of the x86 ISA by creating registers that the software doesn't know about, but which the hardware can "cheat" and recognize certain compiler patterns which work around the register limitation.
In short - many core 486 CPUs would suck. Even many core Pentiums would suck. Architecturally, they're old, dead ends. The best designs for multicore processors would be the P6 design (PPro/PII/PIII/PM) and the Athlon design (K7/K8 - while the K8 is "new", it's about as new as the PM is to the P6 design). Curiously enough, Intel is likely to go with a multicore PM, and AMD is likely to go with a multicore K8.
It should also be noted that a 486DX had a transistor count of 1.2M transistors. A P3 had a transistor count of 9.5M transistors. That's an increase of about 8X - however, the P3 also has twice the data width (64-bit rather than 32-bit), 4X the L1 cache (32KB rather than 8KB), and had two instruction set enhancements tacked onto it, as well as massive architectural improvements, including, essentially, multiple versions of the 486 execute engines inside it. An 8X increase in size for those enhancements is not crazy at all.
Hell, the pioneers only covered 15 miles a day and they crossed a continent in a few months.
Lewis and Clark nearly died, and they had significant help from indigenous people already, and it still took two years. And as much as people want to believe rovers are more adept than humans are, they're not. We don't have information on the surface with resolution to the meter level, so the rover would have to rely on the information it got itself, which means it would be travelling slowly.
Pioneers were travelling on roads.
The machine is maxed out before it hits relatively smooth terrain where you can move quickly without fear.
What smooth terrain? This is Mars - massively cratered surface features, rocks everywhere. There's no place where you could travel reasonably fast with a wheeled vehicle, other than maybe some stretches in Meridiani.
Even ignoring that, do you even realize what Mars contains? Mars has an escarpment which is over a mile high. It has a canyon system which is the size of the entire United States. It has a canyon which is so huge that the walls of the canyon are under the horizon. Mars isn't an easy playground for rovers. Trying to traverse the entire planet would take years for a rover, if not decades.
does anyone here know why the would have been better able to survive the extinction event 65m years ago?
Well, they probably didn't survive. They likely died in massive numers, just like everything else, because there was nothing for them to feed on.
The difference is that smaller dinosaurs wouldn't've gone extinct - there were so many of them that even with their population going down by a few orders of magnitude, they still existed. Imagine trying to kill off all lions in the world. It never would have been hard - they're too high on the food chain, and there's far too few of them. Let's say there are only 1000 lions alive. Now, a virus that kills 99% of all lions might cause the species to go extinct (0.01*1000 = 10). A virus that kills 99% of humans, however, won't kill us off (0.01*6000000000 = 6000000)- it would just suck.
Scientists were never talking about beings like the Tyrannosaurus Rex evolving into a bird. The T. Rex went extinct, rather abruptly. Smaller dinosaurs may have survived. Flight capable dinosaurs may actually have been selected for, because flight increases the range over which they can feed, and in a food-starved ecosystem, this would be paramount.
The evidence suggests that all life would have died off not just the big reptiles.
Woah, it wasn't that big of an impact! It's only a 10 km impact on an object that's 40,000 km around.
Read the link I posted - the impact itself would've caused large tsunamis, and may have killed off an entire generation of plant life, but their seeds would likely have survived (inert) and begun to grow when sunlight fell on them again.
The meteor theory gives no compelling explaination as to why some things survived and others didn't, and why those particular survivors?
Huh? That answer's obvious - large things went extinct. Small things didn't. The population of both dropped significantly, but things that require huge amounts of food have less granularity available - a massive drop in food availability will simply kill off the species, rather than dramatically lowering the population numbers. If you reduce a population of 100 by a factor of 100, they're now a population of 1, and they will go extinct. If you reduce a population of 10,000 by a factor of 100, they're now a population of 100, and they can survive.
So the small creatures didn't survive the catastrophe - they died off just like everything else when the global energy input dropped by orders of magnitude. But they didn't go extinct, which is the difference.
Slashdot Mirror
A question: Since gravity can and does alter the paths of photons, why couldn't it alter the path of the particles just as much?
:) You can't accelerate efficiently with gravity, since it's so weak. Yes, it's strong near a black hole, but the region where it's strong near a black hole isn't large enough to allow particles to remain for very long. There are three things you need to accelerate particles, as far as we know ("stochastic acceleration") - 1) something to steal energy from (i.e. a shock wave), 2) something to prevent a particle from escaping (like a magnetic field), and 3) a large area over which to accelerate (i.e. a few lightyears in size). Faster shock wave, higher energy. Tighter confinement, higher energy. Larger area, higher energy. Unfortunately gravity's "acceleration region" and its "confinement region" are coupled, so you can't get to high energies, because when you increase the area, you reduce the confinement, so particles don't stick around long enough to accelerate to such energies.
It does. But gravity doesn't alter the path of photons more than 1 degree (hardly! it's more like tiny fractions of a milliarcsecond!), which is all any of the cosmic ray detectors can resolve to. This should give you an idea of what we're talking about - even only given a pointing resolution of 1 degree, we still can't find anything in that direction which could generate a particle of that energy.
Another: Could a pair of very massive, closely orbiting black holes provide an acceleration mechanism (not to those energies, but seems to me that they could kick particles out at very high fract-c)
I thought I mentioned this in a different post.
Black holes do in fact accelerate particles (in jets). Active galactic nuclei are thought to be caused by black holes whose jets are pointed close to us. Blazars ("extreme" AGNs) are thought to be caused by black holes which are nearly pointed straight at us. Seyfert galaxies, radio galaxies also are other classes of "angles" of galaxies with big black holes in the center. Closely orbiting black holes would likely produce very high energy jets, but I'm not even sure they would exceed the energy from the jets produced by the merged black hole.
Active galactic nuclei probably do produce a lot of the 10^14-10^19 eV particles, but they probably do so through shock waves propagating through the galactic magnetic field caused by the jets. The problem is that a larger black hole wouldn't produce that much of a stronger shock wave, and the magnetic field and accelerating region are set by the galaxy, not by the black hole.
Is it possible that these particles are related to GRBs? Might explain a few of the oddities away.
One of my closest friends has a keen interest in looking to see if there's an enhancement of cosmic rays in any way correlated with GRBs. We've got 10 ns timing on the arrival of the particles, so we could definitely correlate them with GRBs. It's worth noting that there would probably be a time delay since the photons aren't affected by magnetic fields, and the particles are (however slight), but hopefully that systematic would be determinable after enough data.
Sufficient to say, they have no evidence so far.
But we *have*; and given the repeated observations...regardless of the curve, there will *still* be some particles at the extreme ends. Given that our detectors are still fairly primitive, it's possible that the high-e events are statistically more likely to be detected, is it not?
:) But the fact that we've seen them probably implies that we don't know the source, not that we don't understand the propagation.)
(Yes, I know we've seen them. Otherwise I'd be in a different field right now, and the waiters in Malargue, Argentina wouldn't all know me by name.
I think you're missing what I'm saying - the only way we could've seen any of these particles is if they came from less than 50 Mpc. The GZK effect gets much, much stronger as you go to higher energies. After 50 Mpc, a particle that starts off at 1E21 is below 6E19. Same with particles of higher energy. Astrophysically, that's right in our backyard. There's nothing we know of that could accelerate particles like that. There's an additional problem, which is the fact that there's a spectral change in the 10^19 range that we can't explain, either. Spectral changes occur when acceleration mechanisms change. Supernovae fall apart in the 10^14-10^15 range, and there's a spectral change there, too. The fact that the spectrum continues after 10^19 (in fact, it flattens) implies that there's a new source that's "turning on" in that energy range.
As far as I understand it, the mw background is the peak energy distribution, not the total one. It's what we built our detectors to observe because it's easiest to observe - sorting random noise from the detector noise in gamma/XR is damned difficult.
It's a pure blackbody distribution, with a characteristic temperature at 2.7K. In order to have a particle out of that spectrum at 10^20, it would need to have about 10^23 times the most probable value. I haven't done the math, I'll admit - but something like e^-(10^23) probably times even the size of the Universe probably doesn't even equal *1*. (Oddly enough e^(-(10^23)) times the *number of bits* in the Universe through the end of Time wouldn't even be 1)
We don't know that, because we don't have a source for those protons. Sure, there's nothing on that vector, but even ultra-high-c factor particle paths can be altered by gravitational fields - which were not mentioned in the OMG webpage)
Yes, we do. The GZK cutoff is not "experimental physics". It's the delta++ resonance. This is stuff that they did in the 1950s, and has been extremely well studied since then. It's just proton-photon interactions. Unless Lorentz invariance is wrong (which is possible! but you should read the paper on that suggestion, and it's very, very bad), we know that GZK will slow a particle travelling faster than 6E19 couldn't've travelled more than 50 Mpc.
And gravity doesn't alter particle's paths anywhere near as much as magnetic fields do, and those particles have such high speed that neither gravity, magnetic fields, or anything else could possibly alter their path more than a tenth of a degree.
The "OMG particle"'s webpage (which, by the way, I've never heard it called - which is... odd) is a little sparse. Try the Auger homepage, this UNM site, or this LSU site. If you've got access to Science magazine, also here.
There are actually many, many more interesting things going on in the UHECR field which I haven't even mentioned.
Well, clearly you are wrong, else these would not have been detected -- something did accelerate them this much, and I doubt it's not, as you say, "in the Universe".
:)
You know, that statement seems so obvious, but... it's wrong.
The particles could've been formed by the decay of massive, ultra-high energy structures or particles. This is known as a "top-down" approach.
However, I should've said "nothing that we know of", not nothing. But the top-down approach could be real.
My understanding is that smaller black holes radiate energy faster than larger ones because of the different ratio of energy per "surface area" calc thru the schwarzchild radius. Higher energy density, in effect, although that's not quite accurate either. Larger ones radiate less - and therefore decay slower - function of energy density radiated thru a ^4 surface area. (argh ;)
Yes, you're correct, and that's what I'm trying to say. What you hve to remember is that large black holes become small black holes, and the only time that you actually see black holes emit particles with any frequency is when they finally evaporate, and that happens at the same mass for every black hole. Black holes that start big evaporate more energy (to conserve mass, of course) than black holes that start small, but they evaporate it over massively increased timescales, and the energy that's radiated is so pathetic it's dozens of orders of magnitude away. Small black holes might give off enough energy, but in order to compensate for the observed flux, they'd be happening in our atmosphere (which... is possible, but really, really not likely, and not very consistent with other things we see).
The final black hole evaporation is the same for every black hole.
What about mergers of black holes? Can they produce such energies (there are some who think that there are two multi-M-stellar mass black holes at the center of our galaxy - are there potential merge scenarios that can produce those energies?)
The merger of the galaxy would produce more energy, because the scale is larger even though the magnetic field is smaller. It's still too weak to accelerate particles to those scales, by several orders of magnitude. The black hole merger itself won't produce much energy. How would it? Gravity can't really accelerate particles very well, because it can't repel, only attract. Fermi acceleration is one of the most efficient acceleration processes known, and it basically goes like this - imagine a big wall, moving towards you, and you throw a tennis ball towards it. The tennis ball moves back at you faster, because the wall's moving. Now imagine there's "something" which turns the tennis ball around and runs into the wall again (magnetic fields, typically), and repeat the process. Gravity simply couldn't do it because once you exceed the escape velocity, you're gone, and the only places that you have a high escape velocity are too small to keep a shock wave in the region for very long.
There's a famous figure that everyone shows during GZK cutoff and UHECR (ultra-high energy cosmic ray) talks which shows the cutoff line for particle acceleration to 10^20 eV, and all the high energy processes in the Universe below it.
Yah. The remaining 1% are the theories actually developed by experimentalists, not theorists. Their jobs don't depend on the theory being right.
AFAIK even the decay of a very large (hundreds of stellar masses++) black hole can't produce protons with that energy. I'm not sure about mergers of galactic center black holes, tho - but I'm sure they've taken those into account.
Black hole decays don't produce that much energy per unit time - Hawking radiation happens in exactly the same way for every black hole, so every black hole ends in a burst of exactly the same amount of energy. Large black holes simply take far longer to reach the "fizzle-pop" stage, but in the end, the pop is exactly the same for each one.
This is because the amount of Hawking radiation a black hole produces is dependent solely on its mass. Big black hole, very little Hawking radiation. Small black hole, lots of Hawking radiation. The reason you have a "pop" is that as the black holes radiate, they get smaller, and so there's a runaway effect. What I'm saying is that a big black hole's decay is identical to a small black hole's decay, just separated by several lifetimes of the Universe.
The problem is that you have to consider how the particle gets accelerated. There's really only one way to accelerate them, and that's via magnetic fields. It's the only thing that can convey enough kick to the particle. We currently believe, for instance, that supernovae are responsible for almost all of the cosmic rays we see - when the supernova goes off, the shock wave accelerates particles via something called Fermi acceleration, and you kick particles up as high as 10^14 or 10^15 eV. Higher energies are formed by even bigger accelerating regions, but they're all shock acceleration. But nothing in the Universe is big enough and with enough magnetic field strength to accelerate particles to 10^20 eV. Not neutron stars, not active galactic nuclei, nothing.
I wouldn't think there would be an absolute cutoff there, but a (granted very steep) curve; which means that some very tiny fraction of particles *could* make it here with those energies (and who's to say what original energy level that particle started off with?)
:)
:)
No, of course it's not an absolute cutoff - however, the slope is somewhere in the neighborhood of E^-10 or so, which may as well be an absolute cutoff. No matter how hard you try, you basically can't get much above 6 x 10^19 for more than about 50 megaparsecs. If the GZK cutoff really does exist (which... well, it better, it's very basic physics) then in the absence of sources we don't understand (which is what we think we have), we never should've seen these particles. The "normal" processes which generate particles less than 6E-19, convolved with the GZK effect, would've produced a flux so freaking low we never would've seen it.
what are the odds that the particle in question could have resulted from the Big Bang energies once protons and neutrons started to form from the 'soup'? I realize it would have been traveling for quite a while and the odds would be infinitely small, but still, the mw background is just an average temperature, is it not?)
Actually stuff that's formed from recombination era would be microwave background energies - because, well, that's what the microwave background is.
But anyway, it's not just that we saw one particle, because the thing is, the detectors didn't run for that long, and they weren't that large (i.e. their acceptance was quite low). They would've had to have gotten astro-freaking-phenomenally lucky in order to see one that far away from the expected. It gets even worse when you have other detectors come online that also see those energy events.
It's not the individual particles that interest us. It's the fact that there seems to be a real spectrum out there - there's something actually producing these energies, and either A) it's close, or B) we don't understand interactions at high energies, or C) all of the cosmic ray physics people are smoking something. Considering B) basically implies that one of the fundamental tenets of relativity is wrong - which would be bad , I'd like for it to be A, but I've got a feeling it'll turn out to be C.
You've just described 99% of all current theories out there.
The really interesting part is that we don't really know what process would produce such a thing.
Actually, it's worse than that. Not only do we not know what process would produce such a thing, we don't know how it would've gotten here in the first place. Above 6 x 10^19 eV, particles should interact with the microwave background, and lose energy (the "GZK cutoff"). In essence, there's a cosmic speed limit. The only way that particle could've gotten here is if it came from very close (so it didn't have time to slow down yet) - very close. Which makes the problem of "how the heck was this made?" even worse.
To be fair to our hard-working particle physicists, you would need a VERY large detector hovering high in the air if you wanted to catch these things in nature.
:) )
Or on the ground.
Say, about 3000 square kilometers or so oughtta do it.
(and for a mite bit cheaper than a particle accelerator, too
Hell lets stop pretending anymore. Lets create corporate representatives and get our elected ones back. All these probablems coming from the fact that corporations are seen as people in the eyes of the law. Make the House, the Senate and the Market. People elect the first two, corporations vote for the third and can't contribute to the others.
That's not a bad suggestion...
On the idea of changing government, however, there was a very good suggestion in the Mars series by Kim Stanley Robinson - make one branch of the government like jury duty - that is, a random selection from eligible voters, and required service.
This has the benefit of being basically uncorruptible - it will be very expensive and very difficult to buy someone who doesn't want to be there, and who more than likely won't be there at the next changeover. If you stagger terms of office as well, that'd ease some of the transition periods.
You'd probably want to limit the scope of that body in the constitution, but so long as it can prevent amendments, it will have sufficient power to combat significant government corruption.
Then you'd have a three body legislature: one where the people elect the representatives, one where the representatives are selected randomly, and one where the corporations select representatives. It'd be worth trying, at least.
So you want to say that a PSU that has 10W heat with 100W load, which double to 20W as the components age, is in ANY way worse than a normal PSU that wasted 30W to begin with?
Depends how the design was handled. If the components were placed to handle 30W, then no, it's strictly better. The problem is that most people don't do a full thermal analysis on a design - they simply do it, then check its temperatures while running, and adjust things if it's too bad.
This means that most (cheap) designs are marginal - that is, they don't work a significant amount outside of their specified thermal envelope. A design that's marginal at 10W that encounters 20W is likely to fare worse than a design that's marginal at 30W that encounters 40W. Most tolerances built in are percentages.
So you're right, if it's a quality design, with a lot of tolerance built in. However, I doubt that most power supplies are quality designs...
I agree that there is not much performance loss in writes, but there is still SOME performance loss.
A properly designed ATA RAID controller will have no performance loss at all. The writes will simply be cloned at the RAID controller and go across the individual busses simultaneously. From the OS perspective, it has issued only one write. From the drive's perspective, the situation is identical to a single-drive situation (as each channel is only loaded with 1 drive).
With more complicated solutions (mismatched drives, non-full RAID solutions) of course you're going to have performance loss - however, in those cases you may still get a performance loss in any RAID situation. Specifically if you've got non-full RAID solutions, RAID0 can still even slow down as compared to a single isolated drive, as you could clash with requests for the non-RAID portion whereas in the independent drive case, you never would.
However, the real world implications are few: RAID1's write performance is almost always identical to single-drive, and its read performance is about 20-30% higher than single drives.
There is a reduction in performance for writes (twice as much data must go through the bus)
That's also not true. Since the bus speed typically is much higher than the drive's actual sustainable data speed, RAID1 in hardware generally doesn't show any performance degradation from writes, unless you have extremely, extremely fast drives on a slow bus. On a dual-channel ATA RAID solution, fundamentally this doesn't have to cause any slowdown, as one write command issued to the controller simply becomes a write on each of the attached channels, at the same time.
In software RAID all it really does is up the CPU usage.
You can see the recently reported Tech Report article on SATA RAID here. Write for RAID 1 and write for single disk are pretty much identical. Read is higher for RAID1. Write for RAID 0 is, of course, much higher, as you multiplex the latencies for different stripes.
So, ESPN's website has an absolute *ton* of off-site material, so there's a ton of DNS lookup, loading, etc. Both Firefox and IE load them pretty much equally for me, Firefox was a little slower. So I saved it as a file (complete), and opened it up in both browsers.
Mozilla opened it up then very fast - virtually instant. IE crashed right after loading it (and continues to crash, each time I load it like that...)
Now that was funny.
I'm not entirely sure which versions of IE you've been using, but you've been able to use the keyboard to do anything and everything in IE since at least version 4.0 - if not earlier.
Not cleanly, and not completely. Caret browsing allows you to actually select text with the keyboard, which you can't do in IE (easily). Essentially it turns a web page into a normal document, and when you hit a link you can just hit enter and it will follow it. It's also easily turned on-off (just hit F7) - this way once you get efficient at using it, it makes things extremely easy.
The generally lower speed and reliability compared to IE on Windows?
Firefox 0.9's (well, any Firefox since 0.7, basically) rendering speed is far faster than IE on Windows. Load up any complex page (like CNN, or Slashdot, etc.) so that loading times can equalize between the two.
The startup time is a little longer, but which do you spend more time doing - opening a web browser, or browsing the web?
They care. They just don't know. This is, of course, why Netscape was part of the Microsoft antitrust trial. I couldn't convince my mother to switch to Firefox - she doesn't even really know what IE is. She just clicks on the web-browser thingy, and poof, up a page comes. I'm really tempted to just secretly install Firefox for her one of these times, but she tends to actually play the wacko online games that sometimes require ActiveX (and fill the computer with gigantic amounts of spyware and worms).
My mother hates that she needs to deal with the antivirus stuff, and when something actually comes along and does some damage, she's very frustrated by it. She just doesn't make the connection back to IE, and she doesn't even *know* there's another option. Anyone who thinks that IE dominates because of its technical brilliance is a fool. IE is not technically better than Firefox - it's slower, renders pages far worse, and is missing a lot of the standard features so much so that third-party software has to fill the holes. IE dominates because it's bundled with Windows, and how Microsoft was allowed to extend its monopoly, I will never understand. Money should never outweigh doing the right thing. Sigh.
I have used tabbed browsing. I don't find it an attractive feature. I don't know why everyone seems to believe it to be the best thing out there but I just don't need it. Another instance of IE is fine for what I do.
Tabbed browsing doesn't hurt you, you know. It's just a new feature, and it's one that doesn't even change the way the web browser works. It just helps those of us that have learned to use it well.
There are two main benefits of tabbed browsing though - some people say it allows for a "spatial metaphor" of web pages, so when you're browsing a site you can keep a "bookmark" of a previous page ready for easy access. IMHO, these people are needlessly looking way too deep into this spatial metaphor crap.
The second benefit, and the reason I use it, is that for me, the web browser is different than any other application - it's usual that I'll need several browser windows open, and I don't want those cluttering the taskbar. XP is better than previous versions - it at least collapses them all to a single tab - but you can't "Alt-Tab" through windows in the same program, and in order to see all of the browser windows you have open, you'd have to use the mouse to go down to the taskbar. Especially for someone who types very fast (like me), any time you're required to go to the mouse, it's a pain. Tabs create a "browser taskbar" for me, which is a big help.
Pop-up control is a bit of a hassle but Google's toolbar stopped that stuff for me and added Google to my toolbar. I understand that a browser would be better with it but it certainly not necessary.
This is an advantage of Google, not of IE. You still need to download the Google toolbar, and figure out how to use it. Yes, that's pretty simple - but so is configuring Firefox, especially under Windows. You can't tout IE's out-of-the-box features, and then compare a modified version of IE to Firefox.
I don't know what caret browsing or form management is. I can't imagine that Joe Blow would need it if I (as a regular Slashdot whore) don't know what it is.
Caret browsing allows using the keyboard to navigate, a la Lynx. Again, for those of us who type fast, this is a massive improvement.
I don't need "block images from..." I guess ads are a pain in the ass. Most users just deal with them. It's certainly not something I would require in a browser to make it functional.
You would if it was over a slow link. Then the difference between loading ads and not loading ads becomes immensely significant.
I guess we differ on our opinions of what is necessary. I think easy configuration, BEST configuration out of the box, and 100% perfect rendering on ALL pages being most important. Firefox/Mozilla don't offer those things to me.
Wait, for a second at the end, I thought you were going to say "IE", not Firefox/Mozilla. Firefox renders pages far better and far more standards-compliant than IE does. Not only that, but it does it far faster, too. Startup takes a little longer (I guess it helps when your render DLL is already loaded...) but page loads are far faster under Firefox.
Claiming that IE has 100% perfect rendering on all pages is a joke. Just google for "IE rendering problems", and enjoy the flood. The simplest, common problem: IE doesn't alpha-blend with PNGs. Considering people have been moving away from GIF slowly but surely, and PNGs have become far more common, this is really unacceptable.
If you want a browser that has the best rendering on most pages, you want Firefox, or Opera, or even Safari, but not IE.
If you're only going to be able to move 300 meters/day then you're going to need a lot of days to get anywhere beyond your landing zone. To get anywhere, you'll need a power source that doesn't crap out after 3-6 months.
Absolutely. I completely never disagreed with you there at all. For long-range rovers, you have to have a nuclear power source. But "long range" means tens of miles over probably a year, and it does not mean driving around at speeds much faster than a human walking. Those limitations simply come from limitations of robotic teleoperation.
The problem with nuclear is that you then need to justify (and pay for!) multiple years of operating funds, which will be very difficult to do.
You can't simply ramp up and down your staff as things get more and less interesting. People don't like to work like that.
well gee, it can spend the time it's waiting taking a gander at its new environment. And if you have scads of power, then you can crank up your transmitter's power and send more data in a shorter amount of time.
In its new environment, it will have no idea what's interesting or not, and the ground controllers can't really plan ahead, either, as they don't know what the area looks like either. The only thing it could do is take panoramic images, which would take any rover a few minutes, if not seconds.
You can't "crank up the power and transmit more data" - data rate is frequency-limited, not power limited. A nuclear powered rover would still transmit data at the same rate as the current rovers. The only way you get a higher data rate is with a higher frequency or a better encoding scheme.
We appear to be talking past each other. You're right that it's a bumpy place so you have to move carefully. My main point is you can do better than 100 feet a day. Heck, even one mile a day is 50 times better and you're still crawling.
(Point of note - they're more in the range of 100 meters a day, not 100 feet a day)
Moving quickly brings its own problems. You need to make decisions a lot quicker, and analyze problems a lot faster. They're still not that comfortable with the onboard route-determination algorithms. The capability of moving faster doesn't mean they'd do it. Look at Spirit, which has seriously booked it in the past few weeks compared to what it had been doing, and several of those days, the trip was cut short because of software issues (it didn't think the route was safe). The capability of moving faster may not have actually led to faster net movement, just faster "go, then stop" - remember, JPL has to wait for telemetry uplinks to find out how Spirit did on a previous command pass, which means that if Spirit had had the capability of moving faster, all it would have done was waited around longer.
Opportunity, on the other hand, in Meridiani, could've benefited from a faster movement capability. But again the other problem is that wheel slippage is also a problem - the dust is very fine, and moving quickly is a serious concern. And Meridiani is basically the only place on the planet that's that flat and easy to move in.
Now, of course, after the route-determination algorithms have been exercised, they might be more comfortable at moving faster. But it still would be rather nerve-wracking. After all, how would you know where you're going? On Mars, for a robot half the height of a human, the horizon is ~1.4 miles away! The people at JPL would have no idea what lies ahead, unless the rover stopped, and transmitted pictures - which it can't even do all the time. Which means that there's basically a maximum limit to how fast a rover can move during the day, and it's probably significantly less than a mile.
In other words, I doubt even with a nuclear powered rover that the rover would move very fast. You're teleoperating a robot from another planet, and you're doing it blindly. It's just not something that can be done quickly.
Did you miss the part about shrinking it down to modern geomerty, meaning it would run faster on less power (read less heat) than the original? Sure a 90nm i486 isn't going to run at 3.6GHz like a P4, however I expect it would run a good amount faster than a 486DX2-66 once did.
Unfortunately, nothing will beat the architectural gains which have advanced since the 486 era, and the "worst case" pipeline waits will keep your clockspeed at an insanely low level.
Let me try to explain. The 486 had a 5 stage pipeline - fetch, decode, dispatch, execute, and writeback. Now, each of those pipeline stages isn't going to take the same "minimum" amount of time - some of them are fixed by things other than switching latencies. So, say your execute stage is fine taking only 1 clock cycle up to, say, 2 GHz (a minimum latency of 500 ps), but your decode stage, simply from physical concerns, is going to take at least 5 ns to complete. This means that the maximum you can ramp the clock speed up to is 200 MHz, because each stage in the pipeline has to take 1 clock cycle, so if 5 ns is your minimum, you'll have a max clock speed of 200 MHz.
The solution, though, is obvious - break that "5 ns" decode step into multiple pipeline steps - say, 5 of them, each taking 1 ns each. Now your maximum clock frequency is 1 GHz. The problem is that your pipeline is now 9 stages long, and you have a new architecture - which is precisely what Intel did several times over to allow the clock speed to ramp.
And that's just the pipelining limitation. There are other architectural problems with "ancient cores" as well. One basic problem is that the x87 floating-point architecture is crap. It's stack-based, which means you can only do math with the "stack head". So in order to store things in the registers, you need to use the FXCH instruction to switch the stack head and one of the registers. Well, modern CPUs (the P3 and the Athlon) got around this by saying "we'll make FXCH be a zero-cycle execute when paired with an arithmetic instruction (and after the Pentium, screw it, they're free totally)". Since the modern CPUs can decode more than one instruction per cycle (3 for an Athlon), and the FXCH instruction only lives up to the decode stage, you're really not hurt, as the FXCH fills a pipeline stage that probably would've been left empty anyway. Now consider the P4, which was designed to try to encourage people to move away from x87: it does not have a zero-cycle FXCH, and its x87 performance is abysmal. (The 486 does not have a pipelined FPU, nor a free FXCH instruction. It would be even worse.)
And I haven't even mentioned register renaming yet, which works around the register limitations of the x86 ISA by creating registers that the software doesn't know about, but which the hardware can "cheat" and recognize certain compiler patterns which work around the register limitation.
In short - many core 486 CPUs would suck. Even many core Pentiums would suck. Architecturally, they're old, dead ends. The best designs for multicore processors would be the P6 design (PPro/PII/PIII/PM) and the Athlon design (K7/K8 - while the K8 is "new", it's about as new as the PM is to the P6 design). Curiously enough, Intel is likely to go with a multicore PM, and AMD is likely to go with a multicore K8.
It should also be noted that a 486DX had a transistor count of 1.2M transistors. A P3 had a transistor count of 9.5M transistors. That's an increase of about 8X - however, the P3 also has twice the data width (64-bit rather than 32-bit), 4X the L1 cache (32KB rather than 8KB), and had two instruction set enhancements tacked onto it, as well as massive architectural improvements, including, essentially, multiple versions of the 486 execute engines inside it. An 8X increase in size for those enhancements is not crazy at all.
Hell, the pioneers only covered 15 miles a day and they crossed a continent in a few months.
Lewis and Clark nearly died, and they had significant help from indigenous people already, and it still took two years. And as much as people want to believe rovers are more adept than humans are, they're not. We don't have information on the surface with resolution to the meter level, so the rover would have to rely on the information it got itself, which means it would be travelling slowly.
Pioneers were travelling on roads.
The machine is maxed out before it hits relatively smooth terrain where you can move quickly without fear.
What smooth terrain? This is Mars - massively cratered surface features, rocks everywhere. There's no place where you could travel reasonably fast with a wheeled vehicle, other than maybe some stretches in Meridiani.
Even ignoring that, do you even realize what Mars contains? Mars has an escarpment which is over a mile high. It has a canyon system which is the size of the entire United States. It has a canyon which is so huge that the walls of the canyon are under the horizon. Mars isn't an easy playground for rovers. Trying to traverse the entire planet would take years for a rover, if not decades.
Assuming birds are the ancestors of dinosaurs
s/ancestors/descendants/, I'm assuming.
does anyone here know why the would have been better able to survive the extinction event 65m years ago?
Well, they probably didn't survive. They likely died in massive numers, just like everything else, because there was nothing for them to feed on.
The difference is that smaller dinosaurs wouldn't've gone extinct - there were so many of them that even with their population going down by a few orders of magnitude, they still existed. Imagine trying to kill off all lions in the world. It never would have been hard - they're too high on the food chain, and there's far too few of them. Let's say there are only 1000 lions alive. Now, a virus that kills 99% of all lions might cause the species to go extinct (0.01*1000 = 10). A virus that kills 99% of humans, however, won't kill us off (0.01*6000000000 = 6000000)- it would just suck.
Scientists were never talking about beings like the Tyrannosaurus Rex evolving into a bird. The T. Rex went extinct, rather abruptly. Smaller dinosaurs may have survived. Flight capable dinosaurs may actually have been selected for, because flight increases the range over which they can feed, and in a food-starved ecosystem, this would be paramount.
The evidence suggests that all life would have died off not just the big reptiles.
Woah, it wasn't that big of an impact! It's only a 10 km impact on an object that's 40,000 km around.
Read the link I posted - the impact itself would've caused large tsunamis, and may have killed off an entire generation of plant life, but their seeds would likely have survived (inert) and begun to grow when sunlight fell on them again.
The meteor theory gives no compelling explaination as to why some things survived and others didn't, and why those particular survivors?
Huh? That answer's obvious - large things went extinct. Small things didn't. The population of both dropped significantly, but things that require huge amounts of food have less granularity available - a massive drop in food availability will simply kill off the species, rather than dramatically lowering the population numbers. If you reduce a population of 100 by a factor of 100, they're now a population of 1, and they will go extinct. If you reduce a population of 10,000 by a factor of 100, they're now a population of 100, and they can survive.
So the small creatures didn't survive the catastrophe - they died off just like everything else when the global energy input dropped by orders of magnitude. But they didn't go extinct, which is the difference.