Assuming the things can be manufactured (big if), there wouldn't be enough fuel inside a single buckyball to make igniting it worthwhile.
It's not like you turn up the lasers to 100% and start throwing in fuel. The pulse applied to an ICF pellet is carefully shaped to achieve the maximum neutron yield. Lining everything up properly is also a pain. Plus current laser facilities are only able to do a few shots per day, at most.
I guess the problems with this are the problems with ICF in general: our technology isn't there yet.
I'm not sure when they added it; I was first there two summers ago but I can ask the other people in my group. It's called the "Center", and is across the street from the water tower and police station.
When I first went to Brookhaven National Lab (New York) I was surprised to find a sort of Bar/Grill on site. You can usually find people chatting there over a few beers. Here in California, we don't have alcohol on site, probably because Berkeley and Livermore are close to urban areas, and Brookhaven is relatively isolated.
Think further back. A few years ago, the Opteron and the Athlon64 took a big bite out of Intel's market share. That happened because Intel was arrogantly chasing higher clocks (P4) and awkward architectures (Itanium). The tick-tock strategy was adopted in response to AMD's success. If AMD hadn't embarrassed Intel so badly, I doubt we'd be seeing such rapid product cycles today.
Though, 45nm processors are currently in short supply. They're usually sold out, and are marked up considerably.
I think it's far more important that he was in charge of the Fermilab Permanent Magnet Antiproton Recycler Ring.
"The Recycler Ring incorporates a number of innovative cost-saving features including strontium ferrite hybrid permanent magnets, a low-cost high-vacuum system, and novel stochastic cooling and broadband RF systems."
Technical subjects are generally not objected to. The mathematics pages are full of some really obscure stuff, for example. The worst you can expect is a "too technical" tag, as long as you back up your contribution with a citation, or ten.
Most of the energy from DT fusion comes out in the form of fast neutrons. What's envisioned (emphasis on envisioned) is to have a lithium "blanket" surrounding the first reactor wall that will 1) be heated by neutrons 2) breed tritium for the fuel cycle. For bonus points make this a molten lithium system and run it through a heat exchanger. The rest is just a standard balance of plant: steam generator and turbine. Nothing exotic.
The main problem is dealing with all these pesky neutrons. Aneutronic fusion avoids them, but is far more difficult than DT fusion.
The truth is, we still don't fully understand how plasmas act in the real world. The article alludes to this, by mentioning turbulence and instability. Fluid models and magnetohydrodynamics just aren't detailed enough, and full-blown simulations are far too complex to be of much use on a fusion-reactor scale.
A key concept is "transport". What a fusion reactor requires is to keep heat bottled up. The ions in particular need to be kept hot so that they can fuse. What happens, though, is that heat gets dumped from the ions into the electrons (which are useless for fusion) at a rate which exceeds theoretical predictions -- one of many "anomalous transport" phenomena. (Great phrase, which you may recognize from HL.)
Bottom line: we need to do more research on fundamental plasma physics for fusion. Yet for whatever reason, fusion funding has been dropping for decades.
The guy has some publications. There's a conference paper called "Implementation of a Gigabit Per Second Millimetre Wave Transceiver on CMOS", but I don't have access to IEEE papers, unfortunately.
A spring has potential energy of 0.5*k*x^2, where k is the spring constant and x is the displacement. This clock is going to require a huge spring to provide enough energy storage, and it would be hell to wind up.
In general relativity, light follows geodesics -- straight lines in spacetime. Black holes bend spacetime, so they also bend light. (No, I didn't RTFA.)
Note that most hardware sites are geared toward gaming. As you're a graphic designer, your requirements are probably
0) color accuracy (don't know anything about this, really) 1) gobs of memory 2) gobs of storage 3) fast CPU
Then there's software limitations, which are out of the purview of most hardware sites. 32-bit Windows is limited to about 3 GB of usable RAM. Kickass hardware is useless without software that supports its features.
They're not really donating it. They are "awarding" hours of processor time on supercomputers that were paid for with taxpayer money. Energy research is the mandate of the DOE. The concept is:
1. DOE buys/builds supercomputers. 2. Scientists write proposals about running programs on supercomputers. 3. DOE chooses proposals that look cool / are scientifically interesting / don't suck 4. DOE divvies up the time pie 5. Scientists run their code and figure stuff out, write papers
Time = money, especially when supercomputers are involved.
Terabytes of scientific data don't purport to hold the answers to life, the universe, and everything. A limit on CP violation probably doesn't have anything to do with getting into heaven.
It's likely implementation dependent. In the Flash I've used, the first block has guaranteed 100K or so write cycles, so you generally stick the bad block map there. For a SSD, I'm sure things get very complex.
At Berkeley, the intro CS courses go Scheme, Java, C/assembly. I complained about these courses when I took them, but now I know: choose the language to go with the concepts taught in the course.
Slashdot Mirror
Assuming the things can be manufactured (big if), there wouldn't be enough fuel inside a single buckyball to make igniting it worthwhile.
It's not like you turn up the lasers to 100% and start throwing in fuel. The pulse applied to an ICF pellet is carefully shaped to achieve the maximum neutron yield. Lining everything up properly is also a pain. Plus current laser facilities are only able to do a few shots per day, at most.
I guess the problems with this are the problems with ICF in general: our technology isn't there yet.
No, DDR2-800 runs at 400 MHz.
I'm not sure when they added it; I was first there two summers ago but I can ask the other people in my group. It's called the "Center", and is across the street from the water tower and police station.
http://www.bnl.gov/visitorinfo/onsite_services.asp#center
When I first went to Brookhaven National Lab (New York) I was surprised to find a sort of Bar/Grill on site. You can usually find people chatting there over a few beers. Here in California, we don't have alcohol on site, probably because Berkeley and Livermore are close to urban areas, and Brookhaven is relatively isolated.
An FPGA's configuration bitstream is typically not Turing complete. Only the high-end FPGAs (Virtex-5) can reconfigure themselves on the fly.
Think further back. A few years ago, the Opteron and the Athlon64 took a big bite out of Intel's market share. That happened because Intel was arrogantly chasing higher clocks (P4) and awkward architectures (Itanium). The tick-tock strategy was adopted in response to AMD's success. If AMD hadn't embarrassed Intel so badly, I doubt we'd be seeing such rapid product cycles today.
Though, 45nm processors are currently in short supply. They're usually sold out, and are marked up considerably.
http://techreport.com/discussions.x/14323
I think it's far more important that he was in charge of the Fermilab Permanent Magnet Antiproton Recycler Ring.
"The Recycler Ring incorporates a number of innovative cost-saving features including strontium ferrite hybrid permanent magnets, a low-cost high-vacuum system, and novel stochastic cooling and broadband RF systems."
http://adsabs.harvard.edu/abs/1997APS..PAC..4C01F
Technical subjects are generally not objected to. The mathematics pages are full of some really obscure stuff, for example. The worst you can expect is a "too technical" tag, as long as you back up your contribution with a citation, or ten.
Most of the energy from DT fusion comes out in the form of fast neutrons. What's envisioned (emphasis on envisioned) is to have a lithium "blanket" surrounding the first reactor wall that will 1) be heated by neutrons 2) breed tritium for the fuel cycle. For bonus points make this a molten lithium system and run it through a heat exchanger. The rest is just a standard balance of plant: steam generator and turbine. Nothing exotic.
The main problem is dealing with all these pesky neutrons. Aneutronic fusion avoids them, but is far more difficult than DT fusion.
The truth is, we still don't fully understand how plasmas act in the real world. The article alludes to this, by mentioning turbulence and instability. Fluid models and magnetohydrodynamics just aren't detailed enough, and full-blown simulations are far too complex to be of much use on a fusion-reactor scale.
A key concept is "transport". What a fusion reactor requires is to keep heat bottled up. The ions in particular need to be kept hot so that they can fuse. What happens, though, is that heat gets dumped from the ions into the electrons (which are useless for fusion) at a rate which exceeds theoretical predictions -- one of many "anomalous transport" phenomena. (Great phrase, which you may recognize from HL.)
Bottom line: we need to do more research on fundamental plasma physics for fusion. Yet for whatever reason, fusion funding has been dropping for decades.
Yeah, a 2W PA at 60 GHz would be very impressive.
The guy has some publications. There's a conference paper called "Implementation of a Gigabit Per Second Millimetre Wave Transceiver on CMOS", but I don't have access to IEEE papers, unfortunately.
Here's a 60 GHz PA that's about 50 mW: http://www.eecg.utoronto.ca/~sorinv/papers/rfic_06_tyao.pdf
This one claims 23 dBm out (200 mW): http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=4117364
So maybe it's some class-E deal.
A spring has potential energy of 0.5*k*x^2, where k is the spring constant and x is the displacement. This clock is going to require a huge spring to provide enough energy storage, and it would be hell to wind up.
In general relativity, light follows geodesics -- straight lines in spacetime. Black holes bend spacetime, so they also bend light. (No, I didn't RTFA.)
You'd get one from Kip Thorne too, but he's not convinced yet.
You probably want to mention Noether's theorem by name.
Thanks for sharing. I was running 2003 32-bit up until a few months ago (disk crash) and thinking about going to 64-bit. Maybe I'll give XP-64 a look.
The only hardware site that remains in my daily bookmarks is the Tech Report. One of the few honest sites on the web.
How to build a PC
Christmas 2007 system guide
Note that most hardware sites are geared toward gaming. As you're a graphic designer, your requirements are probably
0) color accuracy (don't know anything about this, really)
1) gobs of memory
2) gobs of storage
3) fast CPU
Then there's software limitations, which are out of the purview of most hardware sites. 32-bit Windows is limited to about 3 GB of usable RAM. Kickass hardware is useless without software that supports its features.
To today's evangelical Christians, Unitarians and Deists might as well be Godless commies.
You do realize that Congress just cut ITER out of the budget? And that fusion funding has been dropping for the past few decades?
They're not really donating it. They are "awarding" hours of processor time on supercomputers that were paid for with taxpayer money. Energy research is the mandate of the DOE. The concept is:
1. DOE buys/builds supercomputers.
2. Scientists write proposals about running programs on supercomputers.
3. DOE chooses proposals that look cool / are scientifically interesting / don't suck
4. DOE divvies up the time pie
5. Scientists run their code and figure stuff out, write papers
Time = money, especially when supercomputers are involved.
Terabytes of scientific data don't purport to hold the answers to life, the universe, and everything. A limit on CP violation probably doesn't have anything to do with getting into heaven.
Yep.
It's pretty sad.
It's likely implementation dependent. In the Flash I've used, the first block has guaranteed 100K or so write cycles, so you generally stick the bad block map there. For a SSD, I'm sure things get very complex.
"Civil disobedience is the assertion of a right which law should give but which it denies."
At Berkeley, the intro CS courses go Scheme, Java, C/assembly. I complained about these courses when I took them, but now I know: choose the language to go with the concepts taught in the course.