SpaceX may be a health hazard

All this drinking to their success may start to stress livers (and brain cells) in many places

Brilliant stuff. Takes me right back to my childhood memories of the Apollo program.

I will wait to see the scientific paper on this. A problem with these tests is that it mixes hardware and software performance measurement. Gaining speed by increasing communication bandwidth (and decreasing latency, for preference) just get the "duh" response it deserves.

The only ways to see if two algorithms differ is to (i) do a proper complexity analysis (computing time and memory/bandwidth use) to see how it should scale theoretically (both in terms of data size and number of processors), and (ii) time optimized versions on the same hardware (or different sets of hardware), using a variable number of processors or nodes.

I will raise a dram to Scotty

Not the frothy kind! Distill the stuff I say. A glass of single malt to my fond memories of Scotty saying

Ye cannae break the laws of physics capt'n!

Parallel code easily transfered to very different architecture?

Let me guess, they can easily parallelize adding two arrays together, or doing matrix vector stuff optimally. This covers some very important bases, but some parallel code needs to be rethought rather than just recompiled when porting to a very different architecture.

We have code which does not use matrix-vector stuff, and works best (40x speed up on 64 cores) on fairly coarse grained, shared memory, parallel architectures. We still have not managed to make a distributed memory version (working on it), and are struggling with an OpenCL version for GPUs (working on it with GPU gurus).

Every time I have heard people claim to have tools that take all the hard work out of parallel programming, they show me examples like "add these 10^9 numbers to another bunch of 10^9 numbers". These tools can indeed take a lot of the hard work out of parallel computing, but not all, by quite a long way.

Brilliant, just brilliant!

Few people can say their knitwear was used by NASA, someone give her medal!

100 "stunned" peopled attended the nuptials

Did he have to stun them in order to get them to attend?

Sorry, couldn't resist

"Apple will also take heart from the fact that, as of yet, there are no reports of Windows Phone fanatics in China offering to exchange organs, or even their virginity, for a handset."

This could mean people are less enthusiastic for Windows Phone, it could mean the Windows phones are sufficiently cheap to be affordable without such extreme measures, or it could mean a certain proportion of the populace needs their brain seen to (or a combination of the three).

For those in need of having their brain seen to, I hear prof Gumby is a very well-known brain specialist (or was that bwain specialist?)

"I can't let you do that, Dave"

Wasn't that what the computer really said?

Cue Q

"Please be careful with it Bond!"

I bet Simon Travaglia would like these. Paint the green button red and vice versa.

Fermi, Tesla, and Maxwell would all want one

Just one?

Physicists (and not just physicists) always want more compute power than currently available.

Are the honourable members of the senate just jealous?

Being rich guys who cannot renounce citizenship without losing their jobs.

A few bottles of Château de Chasselas from the four Yorkshiremen? To wash down the spam?

So if I a bunch of them

could I build a Kim cluster? Might even catch up with a 80486 if you put together a few hundred.

Interesting developments, in particular the having multiple MPI tasks able to run simultaneously on a single Kepler chip. Great added flexibility.

Debugging thousands of threads simultaneously

Now there is an excuse for a HUGE monitor (or two, or three) if ever I heard one.

Seriously, nice toolkits are coming out for this kind of work. Much needed too, as parallel computing allows you to get things wrong MUCH quicker.

So why will cars in the US be smartest?

Some will say it is because the Merkin goverment wants to spy on them

Others will say it is because many American drivers cannot even handle a stick-shift.

Which is it?

Any suggestions?

Now that is rocket science!

Still like blowing things up though

Interesting

I would be willing to try the chips (and other AMD/Radeon graphics), and the newer Linux drivers. Only problem is we would have to port quite a bit of stuff from CUDA to OpenCL (which might be a good idea anyway, similar performance and no vendor lock in).

Regarding the binary nVidia drivers, I have no problems there. I rather like the fact that after inserting a new nVidia card in my PC to replace the old nVidia card, Linux runs without any adaptation, whereas Windows needs a new driver.

Stupid rule!!

I must be allowed to keep up with the register anywhere, even whilst walking across the <SPLAT>

You have got to admire these guys!

They are really getting the excitement in space exploration back to (near) Apollo levels.