I recall reading about the weather issue for solar panel on roofs. The study said the solar panel kept most of its efficiency when it is cloudy. I wish I have a pointer to the article.
well I requested an access to a machine where the procedure to get access are crazy (as in checking you are not a known terorist and making notarized declarations). When I had a problem login onto the machine, I sent a uncrypted/unsigned email to help@service and the admin replied by giving me a password in clear...
I don't know who tagged that "zzzzz"
on
The Apple Two
·
· Score: 4, Funny
but if I meet you, I'll offer you a beer.
Seriously, we have about 3 news on the iPad a day. Am I posting about the new pad my gf is using ?
(follows numerous post on the non existence of a slashdoter's gf)
Re:Free The iPad Of Apple Domination
on
iPad Progress Report
·
· Score: 2, Funny
In ten years, linux will have a support for the kernel operation time_travel which allowed someone to bring a penguiny-tablet to 1985. Hence GP's comment!
you should RTFA. The main goal is to move sanity checks on the free() operation to a separate thread so that they can be done concurrently with the execution of the actual program. The goal is not asynchronous_malloc()
No it is not garbage collection at all, it is allocating in a separate thread. There is no detectino of unreachable memory which characterize garbage collectors. It is just a wrapper to malloc/free.
well, I guess it depends on your memory allocation policy. I used to use numactl to have a firsttouch allocation policy, which is, the physical memory is allocated when first read or writen and it is allocated to the processor that did this access.
The security checks are definitively not random. I travel frequently between europe and the US and all the 10 times I got controlled were in the same month (several trips).
I never was controlled before and I never was controlled after.
My conclusion from reading reading MIT's stuff: "I am not sure they are better scientist than anywere else. What I am sur about MIT is that they are freaking good at marketing!"
that's a stupid reasoning, I would prefer 20 people that I don't know to die than my daughter to be injured. But that's because I am selfish and I do not care that much about people I don't know compared to people I do know.
Here we have not to think personnally but to think globally. From a global perspective, I feel like they are trading 1 fatal injury for 10 permanent injury (made up number). And I can not say the deal is good (or bad). The human cost does not seem really better to me. And on top of that the hardware cost (broken cars, efficiency of the roads) seems to increase.
Of course I am not a civil engineer and my estimation might be way off.
by making roads looking more dangerous, you ARE making them more dangerous as well. For sure people are going to slow down and make accident less fatal, but since the road IS more dangerous (if there are car parked on the side, it is more difficult to see pedestrian) there will be more accident
I feel like you never want your car to change speed on its own. However, presenting the actual speed limit might be an interesting feature.
I am learning driving in Ohio right now and I wonder sometimes what is the actual speed limit. In small streets without sign are they considered as an alley (15mph) or a regular street (25mph) ? Some streets are limited to 30mph which is not a regular speed as far as Ohio BMV is concerned. The definition of "corporation limit" is sometime really blurry and defined by a stamp-sized sign. And some 2.5 lanes (both ways) streets are capped at 35mph while pedestrian are crossing the street between cars.
But you probably also increase the amount of collisions. To me, it looks like trading two fatal injuries with one fatal injuries and ten major injuries. I am not sure it really is better.
PS: For those who wonder, thoses numbers come from the ether.
well, It was not really a CPU at the heart. It was more a complex network of heterogeneous computing unit with classical CPUs but also DSPs, vector float processing units...
It was of course a prototype and never reached the amrket, but merging CPU-type and GPU-type on the same chip seems definitely to be in Intel's and AMD's roadmap.
Slashdot Mirror
please mod parent "+10 damn right". I am getting tired of 2 maxipad news a day
I am sure I can build a machine that leaves my pen touching the paper 24 horus a week! (and write poems on my computer)
I recall wondering whether you were the marcan from team twiizer or not. I guess I am sure now.
PS: you did an awesome job on the wii. thank you for it!
I recall reading about the weather issue for solar panel on roofs. The study said the solar panel kept most of its efficiency when it is cloudy. I wish I have a pointer to the article.
well I requested an access to a machine where the procedure to get access are crazy (as in checking you are not a known terorist and making notarized declarations). When I had a problem login onto the machine, I sent a uncrypted/unsigned email to help@service and the admin replied by giving me a password in clear...
but if I meet you, I'll offer you a beer.
Seriously, we have about 3 news on the iPad a day. Am I posting about the new pad my gf is using ?
(follows numerous post on the non existence of a slashdoter's gf)
In ten years, linux will have a support for the kernel operation time_travel which allowed someone to bring a penguiny-tablet to 1985. Hence GP's comment!
Shouldn't web browsers be almost embaressingly easy to parallellize, or am I missing something here?
In parallel computing, nothing is embarassingly parallel. We should say pleasantly parallel since it makes our life easier. :)
BTW, in my favourite slugish firefo, rendering is the part that takes time (probably mainly because it depends on cairo) and not fetching pages...
you should RTFA. The main goal is to move sanity checks on the free() operation to a separate thread so that they can be done concurrently with the execution of the actual program. The goal is not asynchronous_malloc()
No it is not garbage collection at all, it is allocating in a separate thread. There is no detectino of unreachable memory which characterize garbage collectors. It is just a wrapper to malloc/free.
well, I guess it depends on your memory allocation policy. I used to use numactl to have a firsttouch allocation policy, which is, the physical memory is allocated when first read or writen and it is allocated to the processor that did this access.
here is what I believe is the review by C. Doctorow P was talking about. It is bashing and enlightening. http://www.boingboing.net/2010/04/02/why-i-wont-buy-an-ipad-and-think-you-shouldnt-either.html
you mean we are reaching the performances we had 15 years ago instead of 20 ?
that's a proof of concept, a technical demo... but I agree with you, I don't need it in my webbrowser.
The security checks are definitively not random. I travel frequently between europe and the US and all the 10 times I got controlled were in the same month (several trips).
I never was controlled before and I never was controlled after.
why is that modeed insightful and not funny ? Is that moderator april's fool ?
untrained humans are incapable of making even the simplest calculations in probability theory correctly
And obviously does not know how to count to 4. :)
Hey, but it's MIT!! It's freaking cool!!!
My conclusion from reading reading MIT's stuff: "I am not sure they are better scientist than anywere else. What I am sur about MIT is that they are freaking good at marketing!"
that's a stupid reasoning, I would prefer 20 people that I don't know to die than my daughter to be injured. But that's because I am selfish and I do not care that much about people I don't know compared to people I do know.
Here we have not to think personnally but to think globally. From a global perspective, I feel like they are trading 1 fatal injury for 10 permanent injury (made up number). And I can not say the deal is good (or bad). The human cost does not seem really better to me. And on top of that the hardware cost (broken cars, efficiency of the roads) seems to increase.
Of course I am not a civil engineer and my estimation might be way off.
by making roads looking more dangerous, you ARE making them more dangerous as well. For sure people are going to slow down and make accident less fatal, but since the road IS more dangerous (if there are car parked on the side, it is more difficult to see pedestrian) there will be more accident
I feel like you never want your car to change speed on its own. However, presenting the actual speed limit might be an interesting feature.
I am learning driving in Ohio right now and I wonder sometimes what is the actual speed limit. In small streets without sign are they considered as an alley (15mph) or a regular street (25mph) ? Some streets are limited to 30mph which is not a regular speed as far as Ohio BMV is concerned. The definition of "corporation limit" is sometime really blurry and defined by a stamp-sized sign. And some 2.5 lanes (both ways) streets are capped at 35mph while pedestrian are crossing the street between cars.
But you probably also increase the amount of collisions. To me, it looks like trading two fatal injuries with one fatal injuries and ten major injuries. I am not sure it really is better.
PS: For those who wonder, thoses numbers come from the ether.
I am removing my hat for you sir!
a non ironical post that gives details and provides sources. Damned, I thought I was reading /. !
well, It was not really a CPU at the heart. It was more a complex network of heterogeneous computing unit with classical CPUs but also DSPs, vector float processing units...
It was of course a prototype and never reached the amrket, but merging CPU-type and GPU-type on the same chip seems definitely to be in Intel's and AMD's roadmap.