Anecdote time:
I used to have a dual core Athlon 64, and had a GF 9800GT, and Crysis had to be run at a really low res, with all graphics at bottom settings, and I always attributed that to the GPU.... Then I upgraded to a i5-2500.... And suddenly I could play Crysis at 1920x1080 and even pull up some of the settings to medium or even high... Because the CPU could suddenly keep the GPU fed...
That's not to say that it's rare either. Fibre accounts for roughly a third of all subscriptions for wired internet access here in Sweden. xDSL is a bit over 40%, and cable taking up the slack.
Decompression is, in general terms, computationally insignificant. However, the hardware acceleration support in DX was not for decompression, it was for 3D spatialization and effects. And those are NOT computationally insignificant for interactive scenes, especially not when you have a significant number of audio sources.
It was claimed that uncompressing the audio would tie up an entire core. The large amount is also because they stupidly install all languages at once, even if you select a specific language at installation time.
Industry also has another thing that leads to increased safety: Avoidance of solvent stills for example. Another factor is that in industry, you often work in scale, which leads to avoiding highly exothermic reactions that work fine in academic lab scale, but goes BOOM when you scale it up above 10cl or so...
"a) students are quite gungho when it comes to their work and will quickly take shortcuts because they don't know any better or don't have the right tools, example: I didn't see a wire stripper till I got to industry, I used to do it by pressing the wire to a knife using my thumb and I got many cuts as a result."
Students being the only gung-ho ones? Bwahahahahaha.... Students being gung-ho is a result of PI's and others not having a proper safety mindset, or even deliberately pushing students to ignore safety, or not teaching it at all...
RTLinux as a solution is not (hard) realtime as they falsely claim it to be. Only the Hypervisor is realtime, but the Linux kernel running as a process is not, since it can't guarantee that it will respond in a deterministic manner. That's the problem with the PREEMPT_RT patch too.
With the Linux kernel, you get an average timeframe for an interrupt. With QNX or other proper realtime OS's, you get a Max timeframe.
Yeah, Cogent really is trying every dirty trick they can to go past any contract limits etc and freeload, and then they cry loudly to the media when they get told to stick to the contract.
Yeah, a specular map alone won't do anything. One of the techniques used to try and simulate skin is Subsurface Scattering http://en.wikipedia.org/wiki/S...
Unless you count Greenland(which is sort of autnomous) into Denmark, none. If Greenland is folded into Denmark, since they have the same head of state, and Denmark is still handling defense, finances and foreign policy, it'd be larger than Alaska. If Turkey finally joins the EU, it's larger than Texas.
But if you look at it from the end of the perspective, only Alaska and Texas are larger than either France, Spain or Sweden(Sweden's slightly larger than California), and after that, you get a bit of mix and match: 4 EU states above 300k but below 400k square kilometers, compared to 2 US states above 300k but below 400k square kilometers for example.
The three banks in Sweden I have accounts with do it sort of in that way:
Chip/pin card, hardware token...
Bank website shows a string of numbers, you put the card in the hardware token, press Login, type in the string, then your pin, get a string from token to type into website login.
There's another button for a Signing process, to authorize transactions. Another option(that very few card processors support so far, due to the prevalence of US stone age cards....It's mostly western/Northern europe card processors that currently use it) is the Buy option: You get a string from the card processor, insert card in token, press buy, enter string, then pin, enter new string from token into website.
Likewise, they also gave one to Menachim Begin, who was just as nasty a terrorist as Arafat.
Actually, Jabber and Teamspeak/Mumble are the big tools for coordination. Jabber for strategic communication, and Mumble/TS for tactical(Ventrilo is dead outside the smallest corps who keep on using Vent because that's what they've been using since they started...)
And yes, fleet commanders in these large battles usually have 4-5 voice channels active at any one time: Fleet commander group, scouts channel, capital ships group, supercapitals/titans group etc etc, and also speak to general fleet chat. Also maintaining an overview of spies etc, because a POS password might be changed etc.
In small-scale PvP, you usually don't have the same amount of channels, but otoh, you also don't have the same chain of command, so as a commander there, you have to keep track of more tactical stuff such as tactical target calling, keeping track of cyno use etc, which in a large fleet you have other officers do for you. Many commanders who shine in small-scale harasser group PvP can't handle commanding a large force, and vice versa.
They didn't fight just willy-nilly, it was part of a large campaign. N3/PL were fighting to hold on to PL's staging system while CFC went in to try and take it from them. As is, one of the alliances on the side of CFC managed to grab sov in the system.
Well, as the joke goes here in Sweden:
Why are there no mental asylums in Norway?
It'd cost too much to put walls and a roof over the whole country....
Happiness is a hard disk.
