PRMan handles polygons *much* better these days, to the point that in the current architecture it sometimes converts high level geometry to polygons immediately. VFX studios keep blowing stuff up, explosions sims tend to output polygons, it had to adapt.
Well, I can't speak to the evidence to or against the protocol for massive insulin overdose. Family members of diabetics are drilled on how to deal with hypoglycemic events, certainly moderate insulin overdoses (from e.g. underestimating the bolus for a meal) just come with the territory and that's about all I'm equipped to deal with. I think the protocol is simply based on: low brain blood sugar = seizure/brain damage/coma/death, avoid at all cost. Just based on that, delivering sugar directly to the bloodstream makes sense.
I'll point one relevant complication: diabetics often have neuropathy, sometimes this neuropathy affects the digestive system. This is a doubly cruel issue: low blood sugar causes neuropathy in the long term, and if it affects the digestive system then trying to ward off low blood sugar simply by eating is hard. I doubt protocol assumes potential stomach nerve damage in all patients but it's something worth considering.
We do have some experience with hospital diabetic protocol: they err heavily on the side of keeping your blood sugar high, like over 150 mg/dL high. And yes, you can still run into nurses who are ill-equipped to deal with diabetic patients - we ran into one who refused to push dextrose even though my wife's blood sugar hit 15 (I have no idea how she was still able to argue with the nurse at that point).
Re: your question on glucagon - as I mentioned elsewhere in this thread, yes, the liver glucagon cache depends heavily on your last meal and exercise level. I believe it fluctuates from 50-200 g or so. If you've just run a marathon, a glucagon shot is not going to help you.
We do have glucagon rescue shots around, I mentioned it in another post on this thread and it would definitely be something to use in a theoretical pump overdose emergency.
But keep in mind that glucagon is not a fixed dose of sugar: it's a hormone that triggers the liver to convert stored glycogen into blood glucose. So, it all depends on how much stored glycogen your liver happens to have, which is a function of your recent meals, time of day, whether you've been exercising or not, etc. Dumping your liver's entire store of glycogen may not be enough to counteract a massive insulin dose. Basically, it'll just tide you over until emergency services arrive.
The problem here is that insulin dose tends not to be linear to carbohydrate consumed, so while my wife knows e.g. that if she eats 40 g of carbs she'll only have to give 2 units of insulin, she may actually have to cover 60 g of carbs with 5 units of insulin. I'm making up these numbers but the point is they're numbers worked out over years of trial and error, and also vary at different times of day.
So basically there's no way she would know how much glucose she would need to consume for very high doses of insulin; it's not something you try to experiment for. In a situation where her pump catastrophically failed and dumped all 6-7 days worth of insulin (that's the capacity of her unit; that's probably typical for a external pump user), her immediate priority would be to start eating as much glucose as she can: high blood sugar is better than low blood sugar. Even then, there's no way she could digest it fast enough without having her blood sugar drop precipitously low, leading to seizure etc, so emergency services would need to be called and told that she needs a dextrose IV drip. At this point it's a race between her digestive system and the rate at which the insulin kicks in, and the insulin is likely to win. If I happened to be around I'd stand by with a glucagon shot - this is an emergency injection that tells the liver to dump glycogen, essentially also raising blood sugar - but even then? Unlikely to be sufficient.
tl;dr I would say it's impossible to eat your way out of a catastrophic insulin overdose - you really need IV dextrose and a hospitalization. My earlier comment about box of cookies notwithstanding. The pump manufacturers know this and one would hope they have umpteen safeguards against it in, both in the hardware and software.
Yes, dried fruit is very concentrated carbohydrate. Re: glycemic index - not sure how typical this is for diabetics but my wife ignores glycemic index in her diet and focuses entirely on carbohydrate content. As far as she's concerned, the type of carb only matters by a few minutes; she can back this up by looking at her blood glucose chart after a meal vs her CGMS. In a life or death situation a few minutes might matter, so your point re: honey vs cookies is a valid point, but actually honey IS a better substitute than cookies because of the fat content in the cookies: the fat gets in the way of fast digestion of the carb content. I was being a bit facetious about the cookies earlier.
My wife is a long term Type One Diabetic and has worn a pump and CGMS for years, so I'm somewhat qualified to answer here:
If the insulin pump fails to deliver insulin, won't the diabetic notice with their next glucose test?
Yes, the diabetic will notice. In fact, the CGMS (continuous glucose monitoring system) itself will notice and should alarm (and the article is about a pump with a CGMS built in). Keep in mind that the pump can actually fail to deliver insulin because the tubing is kinked, or the injection site is occluded - so this can be a more common occurrence than you think and isn't actually directly due to pump failure.
Don't most travel with spare insulin and a few syringes, in case of pump failure, at least for long distance travel?
Yep. My wife's diabetic medical supplies easily take up half of her carry on baggage.
Conversely, what happens if the insulin is all delivered at once? An insulin pump holds what, typically, a few days of insulin? Can a diabetic keep ingesting enough sugar with juice or soda or candy to keep their blood sugar up for brain function, even if the insulin dose is overwhelmingly high?
This is the scary scenario. It's never happened to my wife, and the pump manufacturers had better go to great lengths to ensure it never happens. The type of insulin used in a pump is fast acting, so if all the insulin is delivered at once she will pretty much immediately need to eat a lot of sugar or go into a coma. We're talking entire bags of cookies within minutes.
As I pointed out to another commenter, his explicit use of the phrase "whatever means is made possible or necessary" makes it pretty clear he wasn't just talking about overthrowing the government via the democratic process. There is nothing unclear about that. As you say: he's a writer, and the fact that he deliberately decided to add this phrase makes it very clear where his intentions lie.
And somehow you don't think his statement "whatever means is made possible or necessary" covers the possibility of violent overthrow? What reason would he otherwise have to add that blanket modifier to his sentence?
He never advocated overthrowing the government for it.
Yes he did. From an article he wrote for the Mormon Times:
Faithful sexual monogamy, persistence until death, male protection and providence for wife and children, female loyalty to children and husband, and parental discretion in child-rearing.
If government is going to meddle in this, it had better be to support marriage in general while providing protection for those caught in truly destructive marriages.
Because when government is the enemy of marriage, then the people who are actually creating successful marriages have no choice but to change governments, by whatever means is made possible or necessary.
And more:
Regardless of law, marriage has only one definition, and any government that attempts to change it is my mortal enemy. I will act to destroy that government and bring it down, so it can be replaced with a government that will respect and support marriage, and help me raise my children in a society where they will expect to marry in their turn.
I don't know about you, but "I will act to destroy that government and bring it down" is a pretty clear advocation of overthrowing the government.
Wow, TFA is really glossing over an inherent limitation:
the "shiftability" of a Lytro image is a function of the width of the image sensor
If the goal of this is to produce useful stereo content that replicates the parallax seen by humans, then the image sensor needs to be at least as big as the average distance between two human pupils. That's roughly six centimeters. The Lytro's sensor is around six millimeters. Somehow I doubt they're going to increase their form factor by ten times in each dimension, and since the point of a Lytro is to avoid fancy lenses they can't bend the light path to compensate.
No one in line ahead of me. Filled out paper ballot, fed ballot into Sequoia machine. My ballot was three pages, double sided - over ten propositions to vote on in California! Spent more than a hour doing research on them last night.
Oh yeah, this is my first election! Newish American citizen - proud to exercise my right to vote:)
But it's not speech recognition. Watson is getting its input via text, it's not doing any speech recognition. And lest you think this gives the computer an unfair advantage, it's nominally the same advantage championship Jeopardy players can pull: they can read the text off the monitor screen faster than Alex Trebek drones it out.
They're called "Lego". I'm sure you've never heard of 'em, but since every last brick is made in Denmark or the Czech Republic
Being a die-hard Lego fan, I wish this was still true. Alas, if you've bought kits lately (for example, try the new Castle line), you'll notice some of the packaging now reports "Components made in DENMARK, HUNGARY, CHINA, and the CZECH REPUBLIC". Given the Lego company's financial troubles in the last couple of years, it's a sad state of affairs that this move is not entirely surprising.
Why don't NBC's stockholders revolt against the kind of mismanagement that throws away free money and turns content-distribution power over to pirates?
NBC's stockholders have other, perhaps more compelling reasons besides this to revolt. NBC's Nielsen ratings have been terrible for at least two years. There have been entire weeks in 2006-2007 where NBC doesn't even have a top-ten show in the ratings - not exactly great for attracting advertisers, which is really what the whole business model is based on. Their top-rated show in April was "Deal or No Deal" - coming in at #23. At this point coming up with shows that attract eyeballs on television should probably be the more important priority. Yes, that still doesn't excuse NBC from being short sighted when it comes to pulling their few shows that actually are popular from iTunes.
Use a different Logitech model if you can find it
on
Mouse or Trackball?
·
· Score: 1
When I came down with RSI due to mouse usage, I did some research and picked up a Logitech Trackman Marble FX. It's the most comfortable trackball I've used bar none and really helped with the RSI. Unlike Logitech's recent offerings the ball is huge and is meant to be manipulated by the index and middle fingers, while the thumb handles left and middle mouse buttons; the ring finger handles middle click. I found this to be easier to handle than any design where the thumb has to manipulate the ball. It's also optical based, so there's no gunk that builds up in rollers that needs to be cleaned. The one con of the design is that it's really hard to click two buttons simultaneously: hence, this is not a trackball you want to use for Maya unless you're prepared to change the default button mappings. Unfortunately, Logitech never came out with a left-hand version of this mouse (it's definitely biased towards right hand usage), and the model has been discontinued - it's obviously in demand since it seems to go for over $250 on Amazon, last time I looked. But if you can find one cheap I highly recommend it.
Perhaps a better claim for first use of a CG character in a live-action film was Young Sherlock Holmes (the stained glass knight), although I agree T2 was a landmark.
Jurassic Park should definitely be on the list. For the first time, we saw humans interacting relatively seamlessly with CG characters - and those effects were *intrinsic* to the movie. For example, you could argue T2 would still be the same movie without the CGI sequences - that's a harder argument to make with Jurassic Park. If you're pointing out Cameron films, the pseudopod sequence in The Abyss is also noteworthy.
I wouldn't hire you to design rendering algorithms.
That's funny. I get to paid to work on a renderer. I'd say 95% of my job involves "actual programming" and less math. (Case in point: I didn't do that well in linear algebra in university.) Most of the work in a renderer involves sorting, designing data structures, managing file I/O, efficient caching - in other words, your typical CS stuff. I'm not saying that there isn't occasionally some heavy math lifting involved, but it's less than you might expect.
And since I couldn't figure out a way to install both on one PC I had the same problem too, until I discovered this last week: Install multiple versions of IE on your PC. Allows installation of IE 3-6 side by side with 7. Works great for me, and I can now test my web site on IE 6 and IE 7 (and 5.5, god forbid).
Uh, no, Spyglass was bought out for $2.5 billion. Granted, this was a stock exchange at the height of the dot com boom, but it's hard to call that going broke.
Renderman is the shading language that tells Pixar's REYES renderer what to do. It is popular, and it is pretty standard in the graphics industry. People have even implemented Ray-tracers capable of using the same shaders that were used for the REYES renderer. But, in all this is pretty inefficent. Shaders are intimately tied to the rendering method that you are using. Therefore, a new shading language needed to be developed.
I disagree. The RenderMan shading language is not at all tied to the rendering method. The original specification allowed for global illumination methods such as radiosity and raytracing. As evidenced by the RenderMan compatible renderers over the years, it didn't care whether the algorithm used was chiefly scanline, REYES, or raycasting/raytracing.
Slashdot Mirror
PRMan handles polygons *much* better these days, to the point that in the current architecture it sometimes converts high level geometry to polygons immediately. VFX studios keep blowing stuff up, explosions sims tend to output polygons, it had to adapt.
Disclaimer: I work on PRMan.
10 EXTERMINATE
20 GOTO 10
Well, I can't speak to the evidence to or against the protocol for massive insulin overdose. Family members of diabetics are drilled on how to deal with hypoglycemic events, certainly moderate insulin overdoses (from e.g. underestimating the bolus for a meal) just come with the territory and that's about all I'm equipped to deal with. I think the protocol is simply based on: low brain blood sugar = seizure/brain damage/coma/death, avoid at all cost. Just based on that, delivering sugar directly to the bloodstream makes sense.
I'll point one relevant complication: diabetics often have neuropathy, sometimes this neuropathy affects the digestive system. This is a doubly cruel issue: low blood sugar causes neuropathy in the long term, and if it affects the digestive system then trying to ward off low blood sugar simply by eating is hard. I doubt protocol assumes potential stomach nerve damage in all patients but it's something worth considering.
We do have some experience with hospital diabetic protocol: they err heavily on the side of keeping your blood sugar high, like over 150 mg/dL high. And yes, you can still run into nurses who are ill-equipped to deal with diabetic patients - we ran into one who refused to push dextrose even though my wife's blood sugar hit 15 (I have no idea how she was still able to argue with the nurse at that point).
Re: your question on glucagon - as I mentioned elsewhere in this thread, yes, the liver glucagon cache depends heavily on your last meal and exercise level. I believe it fluctuates from 50-200 g or so. If you've just run a marathon, a glucagon shot is not going to help you.
We do have glucagon rescue shots around, I mentioned it in another post on this thread and it would definitely be something to use in a theoretical pump overdose emergency.
But keep in mind that glucagon is not a fixed dose of sugar: it's a hormone that triggers the liver to convert stored glycogen into blood glucose. So, it all depends on how much stored glycogen your liver happens to have, which is a function of your recent meals, time of day, whether you've been exercising or not, etc. Dumping your liver's entire store of glycogen may not be enough to counteract a massive insulin dose. Basically, it'll just tide you over until emergency services arrive.
The problem here is that insulin dose tends not to be linear to carbohydrate consumed, so while my wife knows e.g. that if she eats 40 g of carbs she'll only have to give 2 units of insulin, she may actually have to cover 60 g of carbs with 5 units of insulin. I'm making up these numbers but the point is they're numbers worked out over years of trial and error, and also vary at different times of day.
So basically there's no way she would know how much glucose she would need to consume for very high doses of insulin; it's not something you try to experiment for. In a situation where her pump catastrophically failed and dumped all 6-7 days worth of insulin (that's the capacity of her unit; that's probably typical for a external pump user), her immediate priority would be to start eating as much glucose as she can: high blood sugar is better than low blood sugar. Even then, there's no way she could digest it fast enough without having her blood sugar drop precipitously low, leading to seizure etc, so emergency services would need to be called and told that she needs a dextrose IV drip. At this point it's a race between her digestive system and the rate at which the insulin kicks in, and the insulin is likely to win. If I happened to be around I'd stand by with a glucagon shot - this is an emergency injection that tells the liver to dump glycogen, essentially also raising blood sugar - but even then? Unlikely to be sufficient.
tl;dr I would say it's impossible to eat your way out of a catastrophic insulin overdose - you really need IV dextrose and a hospitalization. My earlier comment about box of cookies notwithstanding. The pump manufacturers know this and one would hope they have umpteen safeguards against it in, both in the hardware and software.
Yes, dried fruit is very concentrated carbohydrate. Re: glycemic index - not sure how typical this is for diabetics but my wife ignores glycemic index in her diet and focuses entirely on carbohydrate content. As far as she's concerned, the type of carb only matters by a few minutes; she can back this up by looking at her blood glucose chart after a meal vs her CGMS. In a life or death situation a few minutes might matter, so your point re: honey vs cookies is a valid point, but actually honey IS a better substitute than cookies because of the fat content in the cookies: the fat gets in the way of fast digestion of the carb content. I was being a bit facetious about the cookies earlier.
My wife is a long term Type One Diabetic and has worn a pump and CGMS for years, so I'm somewhat qualified to answer here:
Yes, the diabetic will notice. In fact, the CGMS (continuous glucose monitoring system) itself will notice and should alarm (and the article is about a pump with a CGMS built in). Keep in mind that the pump can actually fail to deliver insulin because the tubing is kinked, or the injection site is occluded - so this can be a more common occurrence than you think and isn't actually directly due to pump failure.
Yep. My wife's diabetic medical supplies easily take up half of her carry on baggage.
This is the scary scenario. It's never happened to my wife, and the pump manufacturers had better go to great lengths to ensure it never happens. The type of insulin used in a pump is fast acting, so if all the insulin is delivered at once she will pretty much immediately need to eat a lot of sugar or go into a coma. We're talking entire bags of cookies within minutes.
As I pointed out to another commenter, his explicit use of the phrase "whatever means is made possible or necessary" makes it pretty clear he wasn't just talking about overthrowing the government via the democratic process. There is nothing unclear about that. As you say: he's a writer, and the fact that he deliberately decided to add this phrase makes it very clear where his intentions lie.
And somehow you don't think his statement "whatever means is made possible or necessary" covers the possibility of violent overthrow? What reason would he otherwise have to add that blanket modifier to his sentence?
He never advocated overthrowing the government for it.
Yes he did. From an article he wrote for the Mormon Times:
And more:
I don't know about you, but "I will act to destroy that government and bring it down" is a pretty clear advocation of overthrowing the government.
My wife is a diabetic and travels with her insulin supplies in her cabin baggage. You throw that into the fire and I'm punching you in the face.
Wow, TFA is really glossing over an inherent limitation:
If the goal of this is to produce useful stereo content that replicates the parallax seen by humans, then the image sensor needs to be at least as big as the average distance between two human pupils. That's roughly six centimeters. The Lytro's sensor is around six millimeters. Somehow I doubt they're going to increase their form factor by ten times in each dimension, and since the point of a Lytro is to avoid fancy lenses they can't bend the light path to compensate.
No one in line ahead of me. Filled out paper ballot, fed ballot into Sequoia machine. My ballot was three pages, double sided - over ten propositions to vote on in California! Spent more than a hour doing research on them last night.
Oh yeah, this is my first election! Newish American citizen - proud to exercise my right to vote :)
The government bailed out General Motors, not General Electric.
But it's not speech recognition. Watson is getting its input via text, it's not doing any speech recognition. And lest you think this gives the computer an unfair advantage, it's nominally the same advantage championship Jeopardy players can pull: they can read the text off the monitor screen faster than Alex Trebek drones it out.
Yes - Pixar's RenderMan supports it already.
They're called "Lego". I'm sure you've never heard of 'em, but since every last brick is made in Denmark or the Czech Republic
Being a die-hard Lego fan, I wish this was still true. Alas, if you've bought kits lately (for example, try the new Castle line), you'll notice some of the packaging now reports "Components made in DENMARK, HUNGARY, CHINA, and the CZECH REPUBLIC". Given the Lego company's financial troubles in the last couple of years, it's a sad state of affairs that this move is not entirely surprising.
NBC's stockholders have other, perhaps more compelling reasons besides this to revolt. NBC's Nielsen ratings have been terrible for at least two years. There have been entire weeks in 2006-2007 where NBC doesn't even have a top-ten show in the ratings - not exactly great for attracting advertisers, which is really what the whole business model is based on. Their top-rated show in April was "Deal or No Deal" - coming in at #23. At this point coming up with shows that attract eyeballs on television should probably be the more important priority. Yes, that still doesn't excuse NBC from being short sighted when it comes to pulling their few shows that actually are popular from iTunes.
Also: NBC's stockholders are GE stockholders, and GE has been doing pretty well.
When I came down with RSI due to mouse usage, I did some research and picked up a Logitech Trackman Marble FX. It's the most comfortable trackball I've used bar none and really helped with the RSI. Unlike Logitech's recent offerings the ball is huge and is meant to be manipulated by the index and middle fingers, while the thumb handles left and middle mouse buttons; the ring finger handles middle click. I found this to be easier to handle than any design where the thumb has to manipulate the ball. It's also optical based, so there's no gunk that builds up in rollers that needs to be cleaned. The one con of the design is that it's really hard to click two buttons simultaneously: hence, this is not a trackball you want to use for Maya unless you're prepared to change the default button mappings. Unfortunately, Logitech never came out with a left-hand version of this mouse (it's definitely biased towards right hand usage), and the model has been discontinued - it's obviously in demand since it seems to go for over $250 on Amazon, last time I looked. But if you can find one cheap I highly recommend it.
Perhaps a better claim for first use of a CG character in a live-action film was Young Sherlock Holmes (the stained glass knight), although I agree T2 was a landmark.
Jurassic Park should definitely be on the list. For the first time, we saw humans interacting relatively seamlessly with CG characters - and those effects were *intrinsic* to the movie. For example, you could argue T2 would still be the same movie without the CGI sequences - that's a harder argument to make with Jurassic Park. If you're pointing out Cameron films, the pseudopod sequence in The Abyss is also noteworthy.
I wouldn't hire you to design rendering algorithms.
That's funny. I get to paid to work on a renderer. I'd say 95% of my job involves "actual programming" and less math. (Case in point: I didn't do that well in linear algebra in university.) Most of the work in a renderer involves sorting, designing data structures, managing file I/O, efficient caching - in other words, your typical CS stuff. I'm not saying that there isn't occasionally some heavy math lifting involved, but it's less than you might expect.
Spyglass then went broke.
Uh, no, Spyglass was bought out for $2.5 billion. Granted, this was a stock exchange at the height of the dot com boom, but it's hard to call that going broke.
This is known as the wheel of reincarnation, and has come up several times in the last forty years of graphics hardware.
I disagree. The RenderMan shading language is not at all tied to the rendering method. The original specification allowed for global illumination methods such as radiosity and raytracing. As evidenced by the RenderMan compatible renderers over the years, it didn't care whether the algorithm used was chiefly scanline, REYES, or raycasting/raytracing.
Why do you think this is inefficient?