Posts by Frumious Bandersnatch

Re: (beer because there's no saki)

You're right... there is no precedent

(先きの無いコンピューターでしょう...)

Re: Shame

"worse than CUDA with a fraction of the performance"

How do you make that out? Care to mention a bit of hardware that uses CUDA (in a more or less similar price range) and that outperforms a PPU/SPU app? I'm not saying it doesn't exist, but your comment is meaningless without looking at hardware--the programming architectures simply aren't directly comparable. On top of that, you also need to consider what type of application you're running. You can't make a blanket statement like that without considering what problem is being thrown at the system.

While I admit I'm not an expert on CUDA, I would consider myself pretty good at programming on the Cell. It seems to me that it would be a lot more difficult to get anything approaching full utilisation of the GPU in a CUDA architecture in comparison. The main reason is that CUDA seems to be designed around throwing many less powerful compute cores at a problem, and due to Amdahl's laws, there's a limit to how effective this can be, depending on the type of problem. CUDA does seem to be an advance over most previous GPGPU attempts, mainly in supporting a richer set of programming primitives. Bitwise operations is the main one I see, but perhaps it also supports branching? I'm fairly ignorant on this, but I know previous GPU shaders didn't support branching, something that seriously restricted their power from a programmer's point of view. Another CUDA downside is the lack of recursive function calls. There are probably more restrictions I'm not aware of.

On the Cell, on the other hand, the SPUs are much more capable computers in their own right, with a full and pretty rich instruction set. It still has a performance penalty for (unhinted) conditional branches and it's really designed as a vector processor so there's a performance (under-utilisation) hit there too if you're writing scalar code for it. But the point is that it's a much more advanced core. Add to that that you've got great inter-core communication possiblities (hardware mailboxes and interrupts and very high bandwidth DMA, though not directly to the GPU--you have to use a flip buffer in main memory) and it's quite possible to think of dedicating some SPUs to specific tasks the way you would threads or concurrent processes in other systems. Or of using a hybrid model, with some SPUs loading code dynamically as needed for small, compute-intensive kernels, while others are statically allocated to certain tasks/threads.

The Cell is also much better at keeping the main CPU free from having to act as a master to slave cores, with all the attendant housekeeping that can entail. In a properly designed application, SPUs can basically act asynchronously and can coordinate work amongst themselves, with main CPU overhead kept to a minimum. My guess is that a CUDA system needs to dedicate a fair amount of main CPU grunt to keep GPU cores singing. And that's power could be spent implementing other parts of your app that the GPU can't help you with at all.

The upshot of all this is that while a CUDA system could very well beat a PS3 at a narrowly-defined task (eg, password cracking, though not things that involve rainbow hashes, since memory/disk bandwidth is the bottleneck there), such applications generally have to be embarrassingly parallel to begin with. So maybe you can write part of your render pipeline by throwing more cores at it in a CUDA system (constrained by your memory/DMA buses), but because of Amdahl's law, there's a limit to how far that will take you. At a certain point, you need to start thinking of apps as distributed programs with complex data interdependencies instead of purely parallel ones with only simple pipelines, and that's where the Cell's architecture really shines, in my opinion.

Re: What?

Meh. An indignant tone is unnecessary: xkcd.com/123/

Re: what

Hmmm.. I foolishly hit reload on the page that had the original url. Now it's gone. Does the Register search box have some sort of silly iframe deficiency that lets search terms embed other pages? Or is it simply a well-placed %40 and a double decoding problem with browsers exclusively? Now that the original post has been edited, I guess I'll never know :(

Re: a huge dose of radiation

Damn, I was going to start a scare thread of "The Triffids are Coming!" a la Kempf. Too obvious?

Re: Re: Prior art

Speaking of digital cameras, quite a few of them have a "swipe to activate" feature too. I'm talking about sliding lens covers that double up as a power switch. Oh, it's also finger-operated. Make of that what you will.

Dangerous animals

Dear Shakira,

I get letters telling me since I moved away

you've taken to hanging out on that rock about a mile from shore

given what I know about that rock mainly that it's populated by seals

I strongly suggest to you that you not hang out there anymore

'cause the seal is a wily and a vicious creature

and the seal will bite you if you give him half a chance

yeah the seal has a mind set on violence

and the seal is the sworn enemy of man

now when I say that the seal is vicious I use the term advisedly

according to webster's 9th new collegiate definition 4b.

which states that vicious means marked by ferocity

and offers as a synonym...savage

'cause the seal is a vicious and a wily creature

and the seal has a mind full of evil designs

and the seal will harm you and laugh about it

yeah the seal is not a creature you want to toy with

yeah the seal is not a creature you want to toy with

If you have artificial gravity, why do you need Newtonian thrusters?

Maybe because their artificial gravity system is effectively in a bubble with no net change to gravitational forces outside the ship? So like, maybe, the gravity in all the levels pulls in a direction they decide to call "down", but up at the "top" of the ship they've got the reverse pull to balance things out?

TBH though, invoking "artificial gravity" explanations kind of bugs me. We know it's all made up, but there's no need to lampshade it. The one exception: inertial systems that actually work, ie spinning ships and stations, a la 2001.

reading passwords with debug is even earlier

The first password hacking I ever tried was to extract netbios (iirc) login passwords from memory with debug on IBM PS/2s. Must have been in the late 80's. It turned out to be surprisingly easy as the password often remained in memory even after the user had logged out. Security was a bit of a joke back then, though, and there wasn't much practical use for the networking except to play snipes.

Address?

With a designation GJ 667C, it looks to be a neighbour of the beast.

@Yes it matters

I did think I noticed some difference when I wired them up and picked the wiring that I thought sounded better. I'll do the mono test later and see what that sounds like. I'm not looking forward to resoldering them when I'm still not 100% sure that I can aurally tell the difference between the two polarities :(

You wouldn't happen to have any links that explain exactly why phase matters for headphones, would you? My searches didn't turn up anything conclusive, and I still can't quite understand what makes out of phase headphones sound wrong. Thanks :)

Plato?

Not sure exactly what you're trying to get at. It would probably be wrong of me to point out that the Camera Obscura was invented in Greek times (not exactly a "modern" camera, as you put it). But I'll match your quote with a contradicting one:

"when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely, in your thoughts, advanced to the stage of science, whatever the matter may be." -- Lord Kelvin

perhaps I misspoke ...

I meant that to specifically refer to magnetic repulsion, as in "build a big enough electromagnet that will interact with the Earth's magnetic field and knock the thing off course". Not the fundamental force of attraction/repulsion between charged bodies. OK, I know at least that Maxwell's equations unify "electricity" and magnetism (and light), but what I was thinking of was whether an electromagnet (using a purely magnetic motive force) could deflect the thing enough when you take the relative size of the Earth and the attendant gravitational attraction between it and the asteroid. I meant that the electromagnet would probably have to be huge in order to overcome Earth's specific gravitational attraction, and not a statement about the relative size of forces in the abstract.

It would be quite interesting to see whether such a magnet-based solution would work, or whether building and powering a large enough one would even be possible.

+1 for the Back to the Future reference

Strangely apt, too.

or maybe the reverse is true

I've just (in the last two days) updated one of my ps3s to firmware 3.55 from 2 point something and have managed to retain (rebuild) my OtherOS capability thanks to custom firmware signed with Sony's "private" package-signing keys. Now that I know that I can continue to use it/get back into using it for development I might actually be tempted to buy another one (even a lower-wattage slim model that never had OtherOS to begin with). Don't discount the power of pragmatism over principles.

upvote

It's such a pity that that's about the only thing I took from the article.

as I understood it

No, the photon doesn't have any expiry date on it just because it came from the soup(*). Also, energy is conserved because the mirror's momentum is sapped by an amount equal to and opposite of the energy of the created photon. Or, since the mirror isn't physical in this case, the creation of photons means that the power supply for the apparatus has to pump more joules in in order to achieve the same acceleration curve.

(* Although I'm sure the Thomas Edison Electric Light Company (aka General Electric) would love to be able to sell photons with a "use-by" date, a quick search shows that photons are thought to have an infinite half-life)