He's innocent until proven guilty, regardless of whether or not he resigned. And a little more circumspection would be nice. I know there's "no smoke without fire", but both I, you and everyone else here knows how easy it is for something explicit to land on your drive without you having any idea where it came from. We try to defend against it but we're not always successful. I also wouldn't rule out foul play here, as the Police seem to be all too eager to stitch-up Cabinet Ministers with lies and false representations as we know from "Plebgate".
The funny thing about this comment is that if you rephrase "dark matter" to be "the error in our observations when compared to theoretical calculations", it's not really a surprise that dark matter matches it so well. Dark matter has been specifically hypothesised to fill in the discrepancy here, so OBVIOUSLY it's going to fit any theory that includes it extremely well. Indeed, it is parametrised to do so!
Don't be silly. The mammoth evolved about 1.5 million years ago. Quite a few interglacials between then and now when the temperature was a lot higher, yet they did not go extinct then.
I have all of my passwords in a KeePass database. That KeePass database is written to a TrueCrypt volume and the TrueCrypt volume is in Crypted on my Dropbox. I also have the KeePass database on a USB key on my home PC. I plug it in when I need a password and unplug it when I'm done.
In 15 years as a developer I have never used a trigger and I don't intend to in the next 15 years either. If you need to use triggers, you're either (a) lazy or (b) haven't thought the problem through adequately.
Nobody really talks about false sharing and fine grained control of cache lines that you really need to make multicore actually perform in the kind of scenarios a game makes use of, most of which are totally beyond the scope of a Javascript app anyway.
Yes, that was my point - the GPU has to schedule. It multi-threads at the warp level on the actual metal, but it'll execute one "command" at a time. The benefit here is as you say - lock free queuing of commands and highly optimised state management.
It's not just that, it's the locking behaviour needed for correct concurrency. There's no point in multi-threading a lot of cores that are going to spend most of their time waiting on another thread to release a lock. Even Mantle is single threaded. i.e. There's a queue for the GPU, one for Compute and one for DMA, on different threads. There aren't going to be 8 threads all using the GPU at once. It'll still be serialised by the driver, just more efficiently than you can do it with existing D3D or GL drivers.
Two words, "false sharing". Just throwing cores at a problem doesn't necessarily improve the performance. You have to caress your cache lines pretty gently to get real improvements.
GR doesn't predict it, but the motions of stars in galaxies does not conform to what you would expect from GR. You need to add in a lot more mass, a kind-of halo around the galaxy of invisible "stuff" to get the correct figures.
According to particle physics they could exist, although no theory has predicted their existence in the same way that the existence of the positron was predicted (by Dirac, was it, I forget). WIMPs are postulated because the visible matter in the universe accounts for only 5% of that needed according to current cosmological theories. The invisible "stuff" is required to make the theory work. It isn't itself predicted to exist by the theory.
It's trivial to demonstrate that it doesn't match observations and that is the only valid test of a theory or hypothesis. If at some point in the future observation and theory are reconciled then that's a different matter.
It's trivial to show General Relativity is wrong: It doesn't match observations. At the moment a hypothesis (WIMPs) are presented to explain the discrepancy. If WIMPs are not found by experiment, however, then the proposition becomes somewhat different, does it not? Right now GR and Newtonian Mechanics are correct insofar as they're useful for making predictions that turn out through observation to be true, to some degree of accuracy. With GR there's no need for the planet Vulcan to explain the peculiarities of the orbit of Mercury, for example. But is it true at the largest scales? The jury is still out.
I don't think it is, no. But I do think he overstates his case somewhat. For me, the left eyebrow is raised by the need for huge quantities of "dark matter" in order to account for large scale structure of galaxies even though studies of our local region of space show no such matter exists (or is required to explain it). Now OK, if they discover some WIMPs in future I will hold my hands up, but right now being sceptical is the correct position to take on this. And if WIMPs don't exist, well, the predictions of GR won't match observation so at the very least it will need modification.
This is all notwithstanding the fact that physics simply describes the regularities of experience and apparently gives different answers to the same question depending on how that question is posed. It's still amusing to me that mathematics cannot even deal with the 3-body problem in Newtonian Mechanics adequately without resorting to perturbation methods. Then there's the regularisation issue in Quantum Theory, where infinities magically cancel each other out.
I am not a physicist, but I reiterate the need for scepticism everywhere and at all times. It's possible to be sceptical and also have "wow" moments when physicists come up with genuinely new ideas. I do sometimes wonder just how primitive our current bleeding edge ideas will look to our distant descendents.
The "public option" costs money that has to be raised through taxation. You're effectively asking those who choose private to also pay for those who choose public. Investment isn't provided by the private sector, it's provided by tax payers.
That is ridiculous. IPCC predictions v reality. The journal Nature also published Michael Mann's idiotic hockey stick and Steig's ridiculous Antarctic warming paper. And as we know from this story, it wouldn't publish anything that goes against the paradigm in any case, although recently it has tried to be at least a little more circumspect in its editorials. That's the whole point, which you seem to have missed from the OP. Here's a blog about it.
Slashdot Mirror
He's innocent until proven guilty, regardless of whether or not he resigned. And a little more circumspection would be nice. I know there's "no smoke without fire", but both I, you and everyone else here knows how easy it is for something explicit to land on your drive without you having any idea where it came from. We try to defend against it but we're not always successful. I also wouldn't rule out foul play here, as the Police seem to be all too eager to stitch-up Cabinet Ministers with lies and false representations as we know from "Plebgate".
The funny thing about this comment is that if you rephrase "dark matter" to be "the error in our observations when compared to theoretical calculations", it's not really a surprise that dark matter matches it so well. Dark matter has been specifically hypothesised to fill in the discrepancy here, so OBVIOUSLY it's going to fit any theory that includes it extremely well. Indeed, it is parametrised to do so!
Of course the idiots in the climate change lobby think that. Everything is caused by climate change these days.
Don't be silly. The mammoth evolved about 1.5 million years ago. Quite a few interglacials between then and now when the temperature was a lot higher, yet they did not go extinct then.
It's stuff you think is there because of gravitational effects assuming the laws of physics as currently understood hold at those larger scales.
I have all of my passwords in a KeePass database. That KeePass database is written to a TrueCrypt volume and the TrueCrypt volume is in Crypted on my Dropbox. I also have the KeePass database on a USB key on my home PC. I plug it in when I need a password and unplug it when I'm done.
Disagreeing with someone is not trolling.
There, I fixed it for you.
Not sure why this comment is modded "troll". It's 100% accurate.
In 15 years as a developer I have never used a trigger and I don't intend to in the next 15 years either. If you need to use triggers, you're either (a) lazy or (b) haven't thought the problem through adequately.
Nobody really talks about false sharing and fine grained control of cache lines that you really need to make multicore actually perform in the kind of scenarios a game makes use of, most of which are totally beyond the scope of a Javascript app anyway.
The thing is, if you just use peanuts you can't charge £100 a shot for the drug.
Yes, that was my point - the GPU has to schedule. It multi-threads at the warp level on the actual metal, but it'll execute one "command" at a time. The benefit here is as you say - lock free queuing of commands and highly optimised state management.
Precisely. Mod this dude up.
It's not just that, it's the locking behaviour needed for correct concurrency. There's no point in multi-threading a lot of cores that are going to spend most of their time waiting on another thread to release a lock. Even Mantle is single threaded. i.e. There's a queue for the GPU, one for Compute and one for DMA, on different threads. There aren't going to be 8 threads all using the GPU at once. It'll still be serialised by the driver, just more efficiently than you can do it with existing D3D or GL drivers.
Two words, "false sharing". Just throwing cores at a problem doesn't necessarily improve the performance. You have to caress your cache lines pretty gently to get real improvements.
GR doesn't predict it, but the motions of stars in galaxies does not conform to what you would expect from GR. You need to add in a lot more mass, a kind-of halo around the galaxy of invisible "stuff" to get the correct figures.
According to particle physics they could exist, although no theory has predicted their existence in the same way that the existence of the positron was predicted (by Dirac, was it, I forget). WIMPs are postulated because the visible matter in the universe accounts for only 5% of that needed according to current cosmological theories. The invisible "stuff" is required to make the theory work. It isn't itself predicted to exist by the theory.
It's trivial to demonstrate that it doesn't match observations and that is the only valid test of a theory or hypothesis. If at some point in the future observation and theory are reconciled then that's a different matter.
Yes, absolutely. And when WIMPs are discovered, we will then know.
Can you link to information about this? I'm very interested to read about it.
It's trivial to show General Relativity is wrong: It doesn't match observations. At the moment a hypothesis (WIMPs) are presented to explain the discrepancy. If WIMPs are not found by experiment, however, then the proposition becomes somewhat different, does it not? Right now GR and Newtonian Mechanics are correct insofar as they're useful for making predictions that turn out through observation to be true, to some degree of accuracy. With GR there's no need for the planet Vulcan to explain the peculiarities of the orbit of Mercury, for example. But is it true at the largest scales? The jury is still out.
I don't think it is, no. But I do think he overstates his case somewhat. For me, the left eyebrow is raised by the need for huge quantities of "dark matter" in order to account for large scale structure of galaxies even though studies of our local region of space show no such matter exists (or is required to explain it). Now OK, if they discover some WIMPs in future I will hold my hands up, but right now being sceptical is the correct position to take on this. And if WIMPs don't exist, well, the predictions of GR won't match observation so at the very least it will need modification.
This is all notwithstanding the fact that physics simply describes the regularities of experience and apparently gives different answers to the same question depending on how that question is posed. It's still amusing to me that mathematics cannot even deal with the 3-body problem in Newtonian Mechanics adequately without resorting to perturbation methods. Then there's the regularisation issue in Quantum Theory, where infinities magically cancel each other out.
I am not a physicist, but I reiterate the need for scepticism everywhere and at all times. It's possible to be sceptical and also have "wow" moments when physicists come up with genuinely new ideas. I do sometimes wonder just how primitive our current bleeding edge ideas will look to our distant descendents.
The "public option" costs money that has to be raised through taxation. You're effectively asking those who choose private to also pay for those who choose public. Investment isn't provided by the private sector, it's provided by tax payers.
That is ridiculous. IPCC predictions v reality. The journal Nature also published Michael Mann's idiotic hockey stick and Steig's ridiculous Antarctic warming paper. And as we know from this story, it wouldn't publish anything that goes against the paradigm in any case, although recently it has tried to be at least a little more circumspect in its editorials. That's the whole point, which you seem to have missed from the OP. Here's a blog about it.