You could probably have an LED pressed against the white of your eye (which is likely to actually be quite thin) which you will be able to see when it flashes; you'll get tints of colour at the edge of your field of vision.

Diabetic here too. My experiences with lows are similar to yours. I've spoken to other diabetics that have said that they have been found having an argument with a wall about whether they should have orange juice or not. Your brain can do weird things when starved of glucose.

It doesn't *officially* get support from Nvidia. That isn't the same as not getting support at all.

The problem with basing figures for raytracing on current hardware is that none of it is designed for rays. GPUs are designed to be good at rasterization, and ray tracing has different hardware requirements. CPUs aren't particularly good at it either. The 7-hour scene might render many times quicker if it uses hardware designed for the job it is doing, rather than shoehorning it in to an inefficient architecture.

You mean like the Gamecube was called "dolphin" before it was released? Or the Kinect was called "Project Natal"?

My calculations for GBP -> USD were a bit off, it seems that £25 is closer to $40, but that is still $120/month just on test strips.

The guy in question paid $450 for 3 months health insurance, or $150 a month (less than the $197.15 you suggested).

WRT the $70/mo, the supplies listed are inadequate for a months supply for a type 1 diabetic.

1. 3 glucose test strips a day is insufficient. Maintaining decent control usually requires testing at least 5 times a day - on waking, before each meal and before bed
2. Each glucose test should be done with a sterile lancet - so 150 of those a month
3. Most type 1 diabetics are on between 3 and 5 injections a day, which requires between 3 and 5 new needles or syringes a day, or 120 a month

Buying diabetes supplies is expensive. They may not cost much to produce, but 50 test strips costs about £25 in the UK over-the-counter (or $1 per strip). At 5 tests a day, thats $150/month just for test strips. I've not had to buy any other diabetes supplies myself, so I am unsure of just how much they cost.

I'm not sure having 3 possible preset speeds, where you able to select one of them arbitrarily and engage the cruise control to have your vehicle automatically accelerate or decelerate to that speed is sensible. For example, driving on the motorway and engaging the "I'm outside school" mode (or the opposite) would cause serious accidents.
A better system may be allowing the cruise control at any speed, and if you are within 3 mph of one of the "target" speeds it automatically adjusts to that speed (along with a "no, don't autocorrect" switch).

The limit may not be appropriate, but I would be very surprised if more than about 10% of drivers could accurately assess the appropriate maximum and minimum speeds for any stretch of road, and the vast majority of drivers would reckon that they would be within those 10%.

I wouldn't trust most people to set their own limits, as they don't have a clue what their (or their vehicles) capabilities are.

Surely thats the same speedo you use to tell what speed you are doing when cruise control isn't enabled?

trillion

There are different boards for different regions of the world, to allow for decent timezone coverage; Americas, EMEA and Austrilasia.

I believe that the people who make up the membership boards are appointed by the Ubuntu Community Council.

The members of the Ubuntu CC are appointed by Mark Shuttleworth, and "confirmed" by a vote by people that currently hold "Ubuntu Member" status.

"Intel® Professional Edition Compilers include advanced optimization features, multithreading capabilities, and support for Intel® processors and compatible processors. They also provide highly optimized performance libraries for creating multithreaded applications."

Taken directly from http://software.intel.com/en-us/intel-compilers/.

The difference is MS would be saying "We don't support OS X", and actively enforcing it. Intel are saying that their compiler does work for AMD processors, without any mention of the fact that they put time and money into actively making it worse for AMD processors (instead of just not doing anything to improve performance specifically for them)

There is a difference between not optimising for a competitors processor and deliberately making performance worse for a competitors processor.

You found it down "for a few minutes at least twice", over how many years? Just because financial systems would suffer insane losses if they had that kind of downtime it doesn't mean that it can't happen - within the last couple of weeks the London Stock Exchange was suspended for several hours due to technical problems. Even if you find Google down for a few minutes once a month, it is still doing better than the LSE this year.