My wife and I both had LASIK surgery a couple of years back. We both spend the bulk of each workday on-line.
Neither of us have any issues working. I have a slight bit of halo effect at night, but she does not. Her astigmatism was worse, and was thus eligable for inclusion in a study of the Nidek laser (it operates on a wider patch of the eye). In her case, no halo and 20/15 eyesight.
It should be noted that *eyestrain* is something which as a seriously nearsighted person I never really experienced. Now, if I read until I pass out my eyes hurt. But a little sleep and a little eyedrops takes care of the issue.
We have no regrets (other than it not having been available 20 years earlier).
If you are speaking of integer overflow, the langauge standard doesn't really support the notion. In addition, it's commonly used for (pseudo) random number generators.
If you are speaking of floating point overflow, it can (and has been) added to the language (in an ISO Technical Report, and in the draft of the next rev of the standard). Many platforms have provided it for years as an option (e.g. IEEE machines often provide routines which check the hardware status).
Many vendors do a significantly better job on their math libraries than the example you cite.
3) the language itself lacks complicated constructs, so it is very simple to optimize. This, with (1) makes fortran still outperform c, thanks to the compilers. ------------------- The key thing which makes C (and it's family) difficult to optimize, and Fortran (modern dialects, modern spelling) or FORTRAN ('77 and before) has nothing to do with control constructs (simple or complicated).
It's the fundamental semantics which make it difficult for a C compiler to prove that two entities aren't really the same (blessed or cursed be pointers, depending on context;>) and in Fortran if things look different, the compiler is free to assume they are (oversimplification, but not much). Of course, programmers have long abused some constructs so that optimizers will sometimes break their code, years and many processors later.
Fortran 95 is a much more complex language, but it's still optimizer friendly. In addition to the core optimizer friendly features, the FORALL statement was imported from HPF.
Some C compilers have non-standard ways (pragmas, options, etc.) to assert that the semantics are closer to Fortran... and clever optimizers can get close to the same performance when given that extra information.
I must confess to being unable to recall the name, but I recall reading a "martial arts novel" by you (the hero was a judoaka). Aside from Science Fiction and Fantasy, was this the only foray you made into other genres, or do you pursue other fields under other names?
Do you intend to write/publish any followups to that martial arts novel?
Where automated optical feedback would be (IMHO)most helpful would be in *home roasting* the beans. One can get very nice results with the alpenroast, various fluid bed devices, or even a stovetop popcorn popper... but it's hard to get commerical level consistency... a varient of this guys LED device might do the trick.
Most of the people I see on the freeway commuting to work are also driving solo. Do you usually take your SO to work?
As a motorcyclist for 25 years, I don't see this as less safe. On the plus side, the material is remarkably similar to helmets, so it functions as a full body helmet. On the downside, it can't handle as well as my motorcycle; but is much more visible.
Price is modestly above that of a full dress motorcycle these days.
"...few places for a few hours" any place with 110v sockets, which is just about everywhere in this country.
"overtake petrol based cars in terms of miles/gallon "... using the PEF (petroleum equivalency factor) adopted by the US federal government in june 2000, my corbin sparrow gets over 500mpg (160watts/mile plugged into their formula). The "lameless filter" barfed on the details. See http://www.ott.doe.gov/legislation.shtml#rules
Before I found that site, I'd worked out the following from various web sources:1 US gal of gas (unclear octane rating) 125,000 BTU;`1 kWh 3,413 BTU Thus yielding 36.64 kWh/gal.
Which allowing some generous rounding up to 200watts per mile, yielded a somewhat more conservative 180mpg.
Using today's electric rates and gas prices, it's more like 100mpg if we use a $ for $ comparison.
How much better does it have to be?
jawtheshark said "A lesser known point is that electric cars do use fuel for the heating system"
Lesser known, and not entirely correct. It is true that some EV's do this, but no means all. The corbin sparrow, for example uses a small ceramic heater. I've measured this to consume about 7amps and change (10 to start, but it goes down as it reaches a steady state). Typical freeway driving requires 50-100-amps. So a 14% or less range reduction worst case (actually, the sparrow lacks a thermostat, and the heater gets too hot for nearly anyone, so it's usually used for only a few minutes at time, so the net effect in my usage has been more like 2-5%.
User #47 claims "similar to having longer chiminies -- merely moves the pollution elsewhere".
This is neither supported by logic nor measurements. As for the former, fixed plants can easily have much better pollution controls and more efficient burning, use of engine waste heat, etc. (and a large fraction of fixed plants use cleaner fuel). Mobile generators use gas or diesel, don't have the space or weight budget for extensive emission controls, nor do they have any practical use for most of the waste heat.
Depending on the power sources being compared, EV's are measured to produce 97% or less of the pollution of conventional engines.
If you are making comparisons, you might want to make consistent ones. Gas doesn't spring up fully refined in your pump, it has to be refined (which takes a considerable amount of energy and produces considerable pollution) and then shipped (more pollution, more energy) and finally pumped.
See http://www.ott.doe.gov/legislation.shtml#rules for the fed/DOE comparison on pure efficiency grounds. Bottom line: wellhead to pump.83 total vs..924 for electric transmission in the US and.328 for electric generation aggregate (which is biased towards older plants, because there are so many more older ones). State of the art plants are, I'm told, pushing 40%. According to a different fed site, http://www.fueleconomy.gov/feg/atv.shtml a conventional combustion engine is about 13% efficient in moving a car down the road (it details where the losses are).
For a more complete, albeit more opinionated analysis, http://www.evadc.org/papers.html
I *did* buy an electric vehicle, admittedly it's not a car, it's a motorcycle.
http://www.corbinmotors.com has details.
My sparrow is an enclosed 3 wheel motorcycle, and my measured top speed is 80mph.
My daily commute is 50m, recharging requires 110v 20amps. A total charge is a six to eight hour affair, but with my usage pattern (drive to work, plug in, drive home; or drive to work, plug in, drive to alternate site, plug in, drive to... etc.) I'm hardly ever at less than 45% of full. Three hours is enough to be back up to 90% or thereabouts.
Performance is NOT as good as the tzero, not by a mile. However, purchase price is a mere $15K.
Slashdot Mirror
My wife and I both had LASIK surgery a couple of years back. We both spend the bulk of each workday on-line.
Neither of us have any issues working. I have a slight bit of halo effect at night, but she does not. Her astigmatism was worse, and was thus eligable for inclusion in a study of the Nidek laser (it operates on a wider patch of the eye). In her case, no halo and 20/15 eyesight.
It should be noted that *eyestrain* is something which as a seriously nearsighted person I never really experienced. Now, if I read until I pass out my eyes hurt. But a little sleep and a little eyedrops takes care of the issue.
We have no regrets (other than it not having been available 20 years earlier).
If you are speaking of integer overflow, the langauge standard doesn't really support the notion. In addition, it's commonly used for (pseudo) random number generators.
If you are speaking of floating point overflow, it can (and has been) added to the language (in an ISO Technical Report, and in the draft of the next rev of the standard). Many platforms have provided it for years as an option (e.g. IEEE machines often provide routines which check the hardware status).
Many vendors do a significantly better job on their math libraries than the example you cite.
3) the language itself lacks complicated constructs, so it is very simple to optimize. This, with (1) makes fortran still outperform c,
;>) and in Fortran if things look different, the compiler is free to assume they are (oversimplification, but not much). Of course, programmers have long abused some constructs so that optimizers will sometimes break their code, years and many processors later.
... and clever optimizers can get close to the same performance when given that extra information.
thanks to the compilers.
-------------------
The key thing which makes C (and it's family) difficult to optimize, and Fortran (modern dialects, modern spelling) or FORTRAN ('77 and before) has nothing to do with control constructs (simple or complicated).
It's the fundamental semantics which make it difficult for a C compiler to prove that two entities aren't really the same (blessed or cursed be pointers, depending on context
Fortran 95 is a much more complex language, but it's still optimizer friendly. In addition to the core optimizer friendly features, the FORALL statement was imported from HPF.
Some C compilers have non-standard ways (pragmas, options, etc.) to assert that the semantics are closer to Fortran
I must confess to being unable to recall the name, but I recall reading a "martial arts novel" by you (the hero was a judoaka). Aside from Science Fiction and Fantasy, was this the only foray you made into other genres, or do you pursue other fields under other names?
Do you intend to write/publish any followups to that martial arts novel?
Where automated optical feedback would be (IMHO)most helpful would be in *home roasting* the beans. One can get very nice results with the alpenroast, various fluid bed devices, or even a stovetop popcorn popper ... but it's hard to get commerical level consistency ... a varient of this guys LED device might do the trick.
Most of the people I see on the freeway commuting to work are also driving solo. Do you usually take your SO to work? As a motorcyclist for 25 years, I don't see this as less safe. On the plus side, the material is remarkably similar to helmets, so it functions as a full body helmet. On the downside, it can't handle as well as my motorcycle; but is much more visible. Price is modestly above that of a full dress motorcycle these days. "...few places for a few hours" any place with 110v sockets, which is just about everywhere in this country.
"overtake petrol based cars in terms of miles/gallon " ... using the PEF (petroleum equivalency factor) adopted by the US federal government in june 2000, my corbin sparrow gets over 500mpg (160watts/mile plugged into their formula). The "lameless filter" barfed on the details. See http://www.ott.doe.gov/legislation.shtml#rules
Before I found that site, I'd worked out the following from various web sources:1 US gal of gas (unclear octane rating) 125,000 BTU;`1 kWh 3,413 BTU Thus yielding 36.64 kWh/gal.
Which allowing some generous rounding up to 200watts per mile, yielded a somewhat more conservative 180mpg.
Using today's electric rates and gas prices, it's more like 100mpg if we use a $ for $ comparison.
How much better does it have to be?
jawtheshark said "A lesser known point is that electric cars do use fuel for the heating system" Lesser known, and not entirely correct. It is true that some EV's do this, but no means all. The corbin sparrow, for example uses a small ceramic heater. I've measured this to consume about 7amps and change (10 to start, but it goes down as it reaches a steady state). Typical freeway driving requires 50-100-amps. So a 14% or less range reduction worst case (actually, the sparrow lacks a thermostat, and the heater gets too hot for nearly anyone, so it's usually used for only a few minutes at time, so the net effect in my usage has been more like 2-5%.
User #47 claims "similar to having longer chiminies -- merely moves the pollution elsewhere". This is neither supported by logic nor measurements. As for the former, fixed plants can easily have much better pollution controls and more efficient burning, use of engine waste heat, etc. (and a large fraction of fixed plants use cleaner fuel). Mobile generators use gas or diesel, don't have the space or weight budget for extensive emission controls, nor do they have any practical use for most of the waste heat. Depending on the power sources being compared, EV's are measured to produce 97% or less of the pollution of conventional engines. If you are making comparisons, you might want to make consistent ones. Gas doesn't spring up fully refined in your pump, it has to be refined (which takes a considerable amount of energy and produces considerable pollution) and then shipped (more pollution, more energy) and finally pumped. See http://www.ott.doe.gov/legislation.shtml#rules for the fed/DOE comparison on pure efficiency grounds. Bottom line: wellhead to pump .83 total vs. .924 for electric transmission in the US and .328 for electric generation aggregate (which is biased towards older plants, because there are so many more older ones). State of the art plants are, I'm told, pushing 40%. According to a different fed site, http://www.fueleconomy.gov/feg/atv.shtml a conventional combustion engine is about 13% efficient in moving a car down the road (it details where the losses are).
For a more complete, albeit more opinionated analysis, http://www.evadc.org/papers.html
I *did* buy an electric vehicle, admittedly it's not a car, it's a motorcycle. http://www.corbinmotors.com has details. My sparrow is an enclosed 3 wheel motorcycle, and my measured top speed is 80mph. My daily commute is 50m, recharging requires 110v 20amps. A total charge is a six to eight hour affair, but with my usage pattern (drive to work, plug in, drive home; or drive to work, plug in, drive to alternate site, plug in, drive to ... etc.) I'm hardly ever at less than 45% of full. Three hours is enough to be back up to 90% or thereabouts.
Performance is NOT as good as the tzero, not by a mile. However, purchase price is a mere $15K.