A 15% improvement is nothing to sneeze at, but you are right -- you would probably get a better improvement by optimizing your software for locality of reference, and at a much lower cost.
But that is quite an achievement what he has done.
And yet, in the final analysis,in reality he has acheived nothing. Which is exactly the reason why I stopped playing WoW (well, that and it was only a trial account anyway). By the way, why is everyone assuming that this player is a "he"? Statistically this very likely is a guy living in his mom's basement, but let's not reinforce the stereotypes, ok?
That sounds so obvious, not only have I thought of it, I believe I've implemented it in software... several times. Surely a company with the vast resources of Microsoft should be able to find some prior art to fight this with!
Come on geeks... there must be some way of shielding buildings from this device, as well as criminals who are willing to pay top dollar to have that shielding installed! I'm gonna start advertising Faraday cages at all the hydroponic gardening equipment suppliers...
The hot chicks must be crawling all over this guy! Why didn't I think of doing every possible time waster in WoW (at least until the next update) as a way to meet girls? Hey Blizzard -- time to release another expansion pack!
How does it reliably detect the beginning/end of commercials? Sure, there is a video and audio jump at the beginning and end of every commercial, but there is a lot of that in the actual program as well. What if broadcasters get smart and start airing commercials that aren't an integral multiple of 30 seconds in length? I've always thought there was room for a value-added service wherein someone would subscribe to receive commercial start/stop cues over a back channel from someone payed to watch each of the channels. (Now there's a job I would be well-qualified for!)
I have an old Sony Media Center PC running XP that drops frames like crazy, but that might be the fault of the crappy, overpriced Sony hardware rather than the crappy, overpriced Microsoft software. It even drops frames while watching live television; as near as I can tell, in order to enable pause/rewind features it always compresses the content, writes it to disk, treads it back off disk,
Gotta load the data and save the results somewhere. If the entire data set fits into RAM, you could load data and save results directly to the network, so it is possible to do this without disk access. That's the way I would do it; I'm to lazy to look up how the SETI@home guys actually implemented it.
No need to program anything different Actually, I believe performance can be improved by pre-fetching data into cache ahead of time, so you know it will be there when you need it. I was doing this in software on MIPS to improve Linpack performance; I suspect you can do much the same with Intel processors as well.
GPUs are using 256 bit wide data paths now to improve data throughput; I think it is only a matter of time until the memory bus is a whole cache line (256 bits?) in width, enabling read/writing of entire cache lines in a single operation. Seems simple to me, but your pin count and power usage go up, as well as the number of separate DRAM chips you need for a wider memory bus.
You do know why Asynchronous transfer mode uses 48 byte packets don't you? The advocates of 32 byte and of 64 byte packets could not reach agreement, so they compromised. Perhaps the Intel designers reached a similar accomadation. (As a software engineer, I too am frequently puzzled when hardware engineers do things that are not powers of 2, e.g. the triple channel memory that Intel's socket 1366 chips currently use, forcing you to by DDR RAM in multiples of 3.)
Current memory architecture has trouble keeping data fed to just 2 CPUs; unless each of the 48 cores has it's own dedicated cache and memory bus, this is a pretty useless design.
Compared to how inefficient most schools are with energy use, this is trivial. Here in oregon, the school are all brick or concrete -- as near as I can tell, without any insulation. Single pane glass windows everywhere. No motion detectors on the classroom lighting. They are apparently designed to be as cheap as possible to build, not to heat.
As others have pointed out, it sounds like a reasonable estimate for the additional electricity used to run 5000 computers for 10 years when they normally would have been in sleep or suspend mode. But really, I think the $1 million is the price some "consultant" quoted to remove SETI@home from all the district's computers -- which brings up the question: wouldn't it be cheaper to just leave it running on the old computers until they fail, then replace them with new computers without it?
Utility fees, yes. Slowing down other programs, no. It is my understand that SETI@home suspends itself when it detects someone trying to do useful work on the computer, so it should have had negligible impact on educational programs. As far as replacement parts, computers wear out faster when the temperature of their components varies; Sun used to recommend NOT turning off diskless workstations for just that reason. Keeping a steady load on the computer should help it last longer. The exception to this is the disk drives, which should be spun down when idle. I'm not clear if SETI@home causes disk access, but if it does, yes it would cause increased maintenance fees. Of course, the monitors and other peripherals should have gone into sleep mode while SETI@home was running, so it only make a difference for the CPU.
Wait... you expect me to believe that the school district is so mind-bogglingly stupid that it leaves all the computers ON for 3 months while the kids and staff are on vacation?!? WTF?
I agree; it is highly unlikely that SETI will find proof of Alien radio broadcasts. However, it is very likely that they will find interesting "signals" out there that will require more scientific investigation to explain their origin. Many scientific discoveries are made accidentally while looking for something else, and I expected SETI to find _something_ interesting, but not find any Martian Soap Operas.
If it is Open Source, aren't you free to add drivers to talk to local storage, or to a cloud computing vendor other than Google? I wouldn't recommend Chromium for mission-critical data, but for what 90% of users do on their computers (e.g. Facebook), it should be just fine.
In the midst of all these options, do we need another operating system? We just might. If nothing else, it really pisses off Microsoft! In and of itself, doesn't that make it well worth it?
1. If you are on a term plan and Verizon ceases offering service to
your location prior to the end of your term commitment, you will not
have to pay an Early Termination Fee.
My only response was, "Well, duh! They really have to tell you this?" Think maybe they were having a bit of a problem collecting those Early Termination Fees from people they cut off against their will?
Slashdot Mirror
A 15% improvement is nothing to sneeze at, but you are right -- you would probably get a better improvement by optimizing your software for locality of reference, and at a much lower cost.
I don't understand the appeal of doing so many things over and over again.
You've never masturbated, have you?
But that is quite an achievement what he has done.
And yet, in the final analysis,in reality he has acheived nothing. Which is exactly the reason why I stopped playing WoW (well, that and it was only a trial account anyway). By the way, why is everyone assuming that this player is a "he"? Statistically this very likely is a guy living in his mom's basement, but let's not reinforce the stereotypes, ok?
Gotta go... mom's calling me upstairs for dinner!
That sounds so obvious, not only have I thought of it, I believe I've implemented it in software... several times. Surely a company with the vast resources of Microsoft should be able to find some prior art to fight this with!
Come on geeks... there must be some way of shielding buildings from this device, as well as criminals who are willing to pay top dollar to have that shielding installed! I'm gonna start advertising Faraday cages at all the hydroponic gardening equipment suppliers...
The hot chicks must be crawling all over this guy! Why didn't I think of doing every possible time waster in WoW (at least until the next update) as a way to meet girls? Hey Blizzard -- time to release another expansion pack!
How does it reliably detect the beginning/end of commercials? Sure, there is a video and audio jump at the beginning and end of every commercial, but there is a lot of that in the actual program as well. What if broadcasters get smart and start airing commercials that aren't an integral multiple of 30 seconds in length? I've always thought there was room for a value-added service wherein someone would subscribe to receive commercial start/stop cues over a back channel from someone payed to watch each of the channels. (Now there's a job I would be well-qualified for!)
I have an old Sony Media Center PC running XP that drops frames like crazy, but that might be the fault of the crappy, overpriced Sony hardware rather than the crappy, overpriced Microsoft software. It even drops frames while watching live television; as near as I can tell, in order to enable pause/rewind features it always compresses the content, writes it to disk, treads it back off disk,
"Mmmmm.... Brains! (drool)" -- Zombie Homer Simpson
Gotta load the data and save the results somewhere. If the entire data set fits into RAM, you could load data and save results directly to the network, so it is possible to do this without disk access. That's the way I would do it; I'm to lazy to look up how the SETI@home guys actually implemented it.
No need to program anything different Actually, I believe performance can be improved by pre-fetching data into cache ahead of time, so you know it will be there when you need it. I was doing this in software on MIPS to improve Linpack performance; I suspect you can do much the same with Intel processors as well.
GPUs are using 256 bit wide data paths now to improve data throughput; I think it is only a matter of time until the memory bus is a whole cache line (256 bits?) in width, enabling read/writing of entire cache lines in a single operation. Seems simple to me, but your pin count and power usage go up, as well as the number of separate DRAM chips you need for a wider memory bus.
You do know why Asynchronous transfer mode uses 48 byte packets don't you? The advocates of 32 byte and of 64 byte packets could not reach agreement, so they compromised. Perhaps the Intel designers reached a similar accomadation. (As a software engineer, I too am frequently puzzled when hardware engineers do things that are not powers of 2, e.g. the triple channel memory that Intel's socket 1366 chips currently use, forcing you to by DDR RAM in multiples of 3.)
Current memory architecture has trouble keeping data fed to just 2 CPUs; unless each of the 48 cores has it's own dedicated cache and memory bus, this is a pretty useless design.
Compared to how inefficient most schools are with energy use, this is trivial. Here in oregon, the school are all brick or concrete -- as near as I can tell, without any insulation. Single pane glass windows everywhere. No motion detectors on the classroom lighting. They are apparently designed to be as cheap as possible to build, not to heat.
As others have pointed out, it sounds like a reasonable estimate for the additional electricity used to run 5000 computers for 10 years when they normally would have been in sleep or suspend mode. But really, I think the $1 million is the price some "consultant" quoted to remove SETI@home from all the district's computers -- which brings up the question: wouldn't it be cheaper to just leave it running on the old computers until they fail, then replace them with new computers without it?
Utility fees, yes. Slowing down other programs, no. It is my understand that SETI@home suspends itself when it detects someone trying to do useful work on the computer, so it should have had negligible impact on educational programs. As far as replacement parts, computers wear out faster when the temperature of their components varies; Sun used to recommend NOT turning off diskless workstations for just that reason. Keeping a steady load on the computer should help it last longer. The exception to this is the disk drives, which should be spun down when idle. I'm not clear if SETI@home causes disk access, but if it does, yes it would cause increased maintenance fees. Of course, the monitors and other peripherals should have gone into sleep mode while SETI@home was running, so it only make a difference for the CPU.
Wait... you expect me to believe that the school district is so mind-bogglingly stupid that it leaves all the computers ON for 3 months while the kids and staff are on vacation?!? WTF?
I agree; it is highly unlikely that SETI will find proof of Alien radio broadcasts. However, it is very likely that they will find interesting "signals" out there that will require more scientific investigation to explain their origin. Many scientific discoveries are made accidentally while looking for something else, and I expected SETI to find _something_ interesting, but not find any Martian Soap Operas.
If it is Open Source, aren't you free to add drivers to talk to local storage, or to a cloud computing vendor other than Google? I wouldn't recommend Chromium for mission-critical data, but for what 90% of users do on their computers (e.g. Facebook), it should be just fine.
So true... That Rosa Parks bitch should have just sat in the back of the bus where she belonged!
His DVD collection consists of only 100 discs?!? How big is his collection of movies downloaded via bittorrent?
It's all good fun until you get that bill from your electric company... (Yes, I know it is very low current, hence not really that much power.)
The manuscript is not written in anagrams at all; it is simply the work of a medieval Italian dyslexic who also spells poorly.
In the midst of all these options, do we need another operating system? We just might. If nothing else, it really pisses off Microsoft! In and of itself, doesn't that make it well worth it?
1. If you are on a term plan and Verizon ceases offering service to your location prior to the end of your term commitment, you will not have to pay an Early Termination Fee.
My only response was, "Well, duh! They really have to tell you this?" Think maybe they were having a bit of a problem collecting those Early Termination Fees from people they cut off against their will?