Slashdot Mirror


User: CSMoran

CSMoran's activity in the archive.

Stories
0
Comments
481
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 481

  1. Re:Turing complete on US Reveals Details of $500 Million Supercomputer (nytimes.com) · · Score: 1

    So since all "computers" are Turing complete, they can all produce the same results - it's just a matter of processing time and having enough memory to hold the needed data. So this computer can't actually accomplish anything more than before. It can just fulfill a greater workload. Or am I mistaken?

    That depends. If you take into account the projected age of the universe, you might find that doing the same calculations using giant fields of wooden rods, while equivalent, may not be practical.

    So "this computer can't actually accomplish anything more than before" is as true as saying a huge truck is not an accomplishment in transport, because you could equally well divide your cargo into tiny pieces that can be carried atop snails and then reassembled when they reach the destination.

  2. Re:What's Next? on US Reveals Details of $500 Million Supercomputer (nytimes.com) · · Score: 1

    Before we get to zettascale computing we need to improve resilience. When you are running MPI jobs with 50000 CPU cores you need to really start tolerating single-node failures gracefully. Aborting the job is no longer a desirable option and checkpointing can be exceedingly costly for many workloads.

  3. Maybe CTRL-ALT-DEL was disabled, maybe it wasn't (or it's still wired to immediate hard reboot, but your game crashed hard enough that it doesn't work)

    Soft reboot. Hard (cold) reboot is/was the reset button.

  4. Just BLOCK the acccount after letâ(TM)s say.. 10 wrong tries???

    Oh, lookie, a DoS hammer!

  5. Re:You mean just the online backup servers... on Hackers Wipe US Servers of Email Provider VFEmail (zdnet.com) · · Score: 1

    Also, depending on how nasty they were being, they might have lurked long enough to poison the offline backups too. People tend to not actually check them till something goes wrong.

    Perhaps that's when the ransom request shall materialise.

  6. What about this is a good idea again? I mean, I am practically speechless here.

    You missed both the Doom *and* Wolf3D parts?

  7. Re:4 proteins good, 6 proteins better! on Adding New DNA Letters Make Novel Proteins Possible (economist.com) · · Score: 1

    They're nucleobases, not proteins.

  8. Re:What could possibly go wrong? on Adding New DNA Letters Make Novel Proteins Possible (economist.com) · · Score: 1

    Jesus, dat u?

  9. Re:perhaps on Adding New DNA Letters Make Novel Proteins Possible (economist.com) · · Score: 1

    Where will they find a supply of x and y in the wild?

    What's wrong with synthesising them de novo in vivo?

  10. The Bat! FTW on Slashdot Asks: How Do You Manage Your Inbox? (npr.org) · · Score: 1

    Good email software, filters that automatically sort 90% of incoming messages into folders (family, work, friends, marketing, etc), auto-mark-as-read most of the spam. What's left in the Inbox serves as a priority to-do list, next in line are unread messages in remaining folders. Nothing gets deleted, only marked as read, since storage is cheap. Last 11 years' worth of emails ~30G. All attachments always accessible. Containers easily backupable and convertible to, say, mbox or txt. Powerful internal search and grep for this once in a decade needle in a haystack scenario. Flags, parking, tags and colours for messages if really needed.

    The Bat! is about the only thing that keeps me on Windows.

  11. Putting a loop construct inside a branch doesn't make the branch a loop construct. You should have played a JLE or JNZ.

  12. Information only relevant to tachyons, no? on A Bright Green 'Christmas Comet' Will Fly the Closest To Earth In Centuries · · Score: 2

    Considering it's Monday.

  13. Re:Will they beat Musk? on A New Engine Could Bring Back Supersonic Air-Travel (economist.com) · · Score: 1

    30 minutes to any city on earth sounds pretty good, and I am pretty sure SpaceX will be getting at least one such commercial flight out in the next five years or so... they also potentially could carry 100 people at a time, and passengers would potentially get a free sub-space visual to boot.

    Even supersonic would look pokey next to that, should my predictions come true.

    FTFY, you're welcome.

  14. Re:Perfect democrats on California Gives Final OK To Require Solar Panels On New Houses (npr.org) · · Score: 4, Informative

    The US is the only sovereign nation which allows non citizens to own land.

    Except of course for Argentina, Australia, Belize, Brazil, Bulgaria, Canada, Croatia, Cyprus, Czech Republic. All right, I got to C. You can read the rest at https://internationalliving.co...

  15. Re:Supercomputers are mostly useless on US Overtakes China in Top Supercomputer List (bbc.com) · · Score: 1

    Basically you can simulate weather, nuclear bombs and a few other things, but that is it.

    Right. And in silico drug design and functional materials design do not exist. Those rotor blades and aeroplane wings also just simulate themselves.

  16. Re:Been following this stuff on The Future of the Kilo: a Weighty Matter (theguardian.com) · · Score: 1

    anyway, isn't 1 kilogram already defined at 1 L of pure H2O?

    At what temperature? Room? Ambient? 0C? At what pressure? Sea-level? Sea-level where? At what ratio of tritiated water to heavy water to protium water? In what electric field? Zero? Ambient? Ambient where? etc, etc.

  17. Re:An observation on CERN Begins New Antimatter Gravity Experiments (phys.org) · · Score: 5, Informative

    Example 2, think of a crystal forming. At the bind site for the molecule, the force is zero. If you squeeze the crystal the force goes negative and the crystal pushes back. But beyond the nano level, the force is attraction only, and reduces according to the square rules, just like gravity. (Think about this force for a moment, as it gets closer, the force increases, at super small distances it decreases to zero, then goes negative. I could label this force 'crystal strong force' or some other name and model it as if its a real force with magic properties, but to do so would be dumb).

    This is patently untrue. First, repulsive forces are positive, not negative -- they are the negative gradient of the potential after all. It's attractive forces that are negative. Second, as you squeeze the crystal together, it's Pauli repulsion, not electrostatics, that produces the huge positive (repulsive force). Third, attractive forces at the nanoscale are due to van der Waals (dispersive) interactions, where instantaneous dipoles induce dipoles in nearby atoms, and this instantaneous dipole - induced dipole interaction is attractive. This potential decays as R^-6, so unlike gravity.

  18. Re:The headline is bunk, for a start on Microplastics Found In 90 Percent of Table Salt (nationalgeographic.com) · · Score: 1

    Do all brands sell equal volumes of salt? Because only then would your argument work.

  19. Re:Electron resistance to unseen particles on Measurement Shows the Electron's Stubborn Roundness (scientificamerican.com) · · Score: 1

    However, an electron is not just a point carrying charge. For instance it interferes with itself (so, wave nature) and it has spin (this weird property that classical point charges lack).

  20. Re:Does it even make sense? on Measurement Shows the Electron's Stubborn Roundness (scientificamerican.com) · · Score: 4, Insightful

    In many scenarios electrons are not well-described by point particles. Their de Broglie wavelength is in the order of one Angstrom, depending on how fast they move. That means you have to take their wave nature into account and the only reasonable description is via the Schroedinger equation (for non-relativistic velocities), or the Dirac equation (for relativistic velocities).

  21. Re:Which LED bulb ? on Europe To Ban Halogen Lightbulbs (theguardian.com) · · Score: 1

    Dude. You're comparing LED "mean lifetime" with "longest-lasting" incandescents. Seriously?

  22. Of course. Cf. Rick Strassman, "DMT: The spirit molecule".

  23. Re: Yes on Is C++ a 'Really Terrible Language'? (gamesindustry.biz) · · Score: 1

    The only completely useless feature I con think of is C++ arrays - new[] and delete[] don't do anything that can't be done much better with std::vector.

    There are restrictions on elements that an std::vector can hold. You can't have a vector of const-qualified entities, for instance.

  24. Re:Waaahmbulance is coming! on Amazon's Alexa is Getting Clobbered (axios.com) · · Score: 1

    I can't fucking stand statistics. No matter how valid a tool it could be, someone will always find a way to abuse statistics to manipulate facts, ultimately making the tool worthless in the end.

    I suppose you hate painkillers too?

  25. Re:No manual memory management necessary in C++ on Eric Raymond Shares 'Code Archaeology' Tips, Urges Bug-Hunts in Ancient Code (itprotoday.com) · · Score: 1

    You're confusing memory fragmentation with memory leaks.