keyboard shortcuts, built-in bibtex database creator. a sane setup and formating system.... I could go on.
Ah, I guess you mean lyx in general, and in that sense I agree; lyx could certainly be improved in many ways. But what I specifically meant with my comment was why you want some bibtex editing stuff in lyx?
I'm sure it's possible to make a good bibtex GUI, but believe me, I tried lots of those which are available for Linux, and without exception they all suck in major ways. That is, without the exception of bibtex-mode in emacs, which I reluctantly settled on as I'm no huge fan of emacs. But once you get used to it (which is really pretty easy since all the functionality can be accessed from a couple menus, no need to remember countless C-M-X-Y-Z key commands), it's pretty powerful and complete (AFAIK it supports every nook and cranny of bibtex, which all the GUI tools I tried do not).
Personally, I see no need for further bibtex integration in lyx. Sure, it's a volunteer project so the people working on it are free to do whatever they fancy, but IMHO there are plenty of more useful and important stuff that could be done first.
Yes, the new ssa architecture in GCC 4.0 allows for autovectorization, but at the moment the focus is on getting GCC 4.0 sufficiently stable for release in a few months. Because of this, IIRC, some of the fancier vectorization passes were deferred until GCC 4.1.
So yes, you might see some performance improvements due to vectorization in 4.0, but you'll have to wait until 4.1 or maybe even 4.2 before you'll see the full potential of it.
-joib, occasional GCC contributor (although I have absolutely zilch to do with the mid- and backend stuff, were most of the optimization passes are done)
since they have a large audience now that can take advantage of them maybe LyX will start accelerating development and adding in some nice features that will make document creation much more productive.(integrate a bib database for god sakes!!!)
Personally, I'm very happy using a combination of LyX and BibTeX-mode in emacs. What more do you need?
On August 1 2005, a new EU directive will come into force. It will make manufacturers/resellers of consumer electronics responsible for recycling their own product. That is, the customer is entitled to return his product for recycling for free.
I think this sounds like a better idea than yet another inefficiently run government program, as it gives some incentive to the manufacturer to reduce recycling costs, e.g. by running the recycling center as efficienty as possible but also by designing products that are as easy to recycle as possible.
IMHO, FBDIMM is just Intel's hedge against RAMBUS going bust. The point of RAMBUS was to reduce pincount per chip by reducing the width of the channel between the memory controller and the chip, and to decouple the notion of a memory controller controlling specific banks of memory. FBDIMMs are to solve the same problem, except RAMBUS is shipping already.
Reducing pincount is one important reason for FB-DIMM, but the real reason for it is to get out of the capacity/speed tradeoff game. See, many systems need lots of memory. However, with current DDR-400 or DDR2-667 you can only put two devices per channel. If you want more RAM than what fits in two devices, you have to reduce the speed. FB-DIMM gets around this problem by using point-to-point links between the devices.
Yes, this increases latency a little bit, but there really isn't any other practical way to increase speed without reducing capacity. However, FB-DIMM compensates for the increased latency by allowing many outstanding transactions on each channel; because of this, latency under high load is actually supposed to be lower than for traditional RAM tech with the same specs.
No matter which version you use, if you never VACUUM dead row versions will accumulate and eventually kill your performance.
Actually, for 7.4 and newer that isn't true, since they include the aptly named "autovacuum" daemon. And yes, autovacuum also does "VACUUM ANALYZE", so no need to worry about that either.
AFAIK, all multi-cpu AMD Opteron systems are NUMA. Each processor connects to its own memory (since the processors sport on-die memory controllers), access to memory attached to other processors is via the hypertransport link.
Actually, most of the work SGI did was against the 2.4 kernel (in fact, Altix systems still ship with a SGI patched 2.4 kernel). The vast majority of these patches were included in the 2.6 kernel, which probably is the major reasons that the standard 2.6 kernel scales so well up to at least 64 cpu:s, as this test showed.
$79*200~=$16,000 or enough for a few extra servers per year =) Most IT directors will take that anytime! Not only that but it also means a few tons less sulphur dioxide and quite a few tons less CO2 produced, which if you're even a little interested in the environment is a good thing. All of that benifit at basically zero opertunity cost, why not?!?!
Probably because the Pentium M and its motherboard is significantly more expensive than a comparable A64/P4 setup. Companies are interested in minimizing total cost over the lifetime of the computer, of which yearly electricity costs are only a part.
Yes, all this and lower power consumption (than P4) to boot. There should be some sweet notebooks and servers coming out over the next few months also, as the true low-power Athlon64s and Opterons roll out.
While I agree that Athlon64 and Opteron are superior to Intels offerings in the desktop and server market, I have to say that Intel really hit the nail on the head with the Pentium M. While the P-M won't match a high clocked Athlon64 performance-wise, it's still plenty fast and uses very little power. I think it's something like 21 W, while the mobile A64 is around 35 W, a significant advantage for notebooks.
Exactly. Remember all the hype about electric vehicles 10-20 years ago? We were all supposed to be using electric vehicles by now. Unsuprisingly, electric vehicles are nowhere to be seen.
To an extent. As such, I see no wrong with private ownership, private entities can probably manage their land more efficiently than the government. However, there is certainly the moral case to be made that as land, or natural resources (land, minerals, spectrum, etc.) more generally, was here long before man, no single man has the right to said resource. Thus the government can fairly tax these resources heavily, up to the rental value of the resource (and in the process reduce other, more distorting, taxes).
And before you go off labeling this as yet another communist conspiracy, consider that this concept has been endorsed by many high brow economists, even right wingers like Milton Friedman.
Google found the following quotes:
"Pure ground rent is in the nature of a 'surplus,' which can be taxed heavily without distorting production incentives or reducing efficiency." -- Paul Samuelson, Nobel laureate in Economics
"In my opinion the least bad tax is the property tax on the unimproved value of land, the Henry George argument of many, many years ago." --Milton Friedman, Nobel laureate in Economics
"It is important that the rent of land be retained as a source of government revenue." -- Franco Modigliani, Nobel laureate in Economics
"For efficiency, for adequate revenue, and for justice, every user of land should be required to make an annual payment to the local government equal to the current rental value of the land he or she prevents others from using." -- Robert Solow, Nobel laureate in Economics
"While the governments of developed nations with market economies collect some of the rent of land, they do not collect nearly as much as they could, and they therefore make unnecessarily great use of taxes that impede their economies -- taxes on such things as incomes, sales, and the value of capital goods." -- William Vickrey, Nobel laureate in Economics and past president of the American Economics Association
First I'd like to say that there has been a record amount of interest in this post. (dont mod me down, I can't help being disapointed in a posibly world saving subject being ignored in favor of less important subjects)
Take it easy. You know, there have been prototypes of these stirling dish things since at least the early 1980:s. If they can be made economically viable in the 2010 timeframe as the article suggest, I'm sure they will be discussed a lot.;-)
There was a project some years ago in Australia where the heat difference between (one of our) desert surface and pipes/heat exchange buried underground was going to power a whole lot of engines. I'm not positive but I think they were Stirling engines that had been changed in some way (possibly day and night running- desert cold at night). They got to the stage of cupple of dozen in a test area to show efficiency workability etc. then nothing. Googling it doesnt show much. Rumors of state govt being bought out by the local energy companies which rezoned and shut them down. The usual Nexus style hippies chanted conspiracy, maybe it just wasnt as efficient as it could have been at that time.
Anyway, plenty of room and plenty of heat and cold differentials in the desert(s).
The efficiency of a heat engine is proportional to the heat difference between the cold and hot sides. In the proposal above, the heat difference is pretty small, so you'd need lots of expensive engines and piping to produce little power. I assume that was what killed it, rather than some Big Oil conspiracy.
Just something bothers me. Wouldnt the loss associated with transporting the energy back to where it is needed (suberbia, industrial and city use) by leakage make it not sustainable?
No worries, mate. Modern high voltage DC transmission lines, often used for long range transmission, have losses of about 4 % per 1000 km.
What are the efficiency ratings of other energy storages such as spliting water into Hydrogen and Oxygen to be stored separately and recombined when needed to make combustion engine generate power?
Appalingly low. Electrolysis is about 70 %, liquefaction 66 %, burning the hydrogen in a combustion engine about 40 % for a medium sized stationary engine. In total, 18 % efficient, compared to 96 % efficient for 1000 km cabling.
Assuming that the guys designing this stuff aren't total idiots, rest assured that they looked into something like your proposal. Apparently they came to the solution that using grid power and software to control the staggered startup is cheaper than adding a battery to each dish.
Ah, good old Archimedes Plutonium is still alive. In case you didn't know, that guy is a known usenet crackpot. I added him to my killfile in 1995 or so.
Re:Netscape backed by firefox??
on
Netscape Reborn?
·
· Score: 3, Informative
Actually, if you go back even longer, you'll find Mosaic, the common ancestor of both Mozilla and Internet explorer.
And before you ask, no, there is no common source code. The source code lineage has been broken many times. For example, the original Netscape was made by the same guys who did Mosaic, but apparently they didn't reuse Mosaic code. Same for Internet explorer, MS licenced a version of Mosaic (Spyglass) but it is doubtful if they actually used any of the source code for IE. And lets not forget that the Mozilla project decided to ditch the netscape codebase they had been given.
And of course, to make it even more complicated, netscape 7.x were/are based on Mozilla.
Wasn't there mozilla back in the early 90's when the net first started?
No. I guess you mean NCSA Mosaic.
Ah, those were the days. Mosaic was pretty revolutionary back then, although I swithed to Netscape 0.9 as it became available, since it could display pages while downloading over my 9600 baud modem. Mosaic needed to download the entire page before displaying anything.
The biggest machine was a huge Cray T3E - I don't recall any blinkenlights on it, but it didn't need them! I recently heard that turing.mcc.ac.uk has since been dismantled, presumably because it was no longer cost-effective for its mere few hundred Gflops. I've no idea what was done with it and its parts, or what (if anything) it has been replaced with, but it's what I thought of when I saw this article.:-)
Yeah I saw a T3E once too, at the local supercomputer center. Quite impressive. When they upgraded to a p690 cluster a couple of years ago they just dismantled it and threw the cards into the bin. Apparently it was quite an expensive system to keep running, using lots of electricity and that funky liquid freon cooling system etc. So despite the 500 peak Gflop/s it produced, nobody wanted it.
Slashdot Mirror
Does anyone know of a nice computational/science benchmark that runs on Linux/Windows/BSD and is free?
Linpack, stream, polyhedron, livermore kernels, you name it. Most of them are freely available and require little more than a Fortran compiler.
keyboard shortcuts, built-in bibtex database creator. a sane setup and formating system.... I could go on.
Ah, I guess you mean lyx in general, and in that sense I agree; lyx could certainly be improved in many ways. But what I specifically meant with my comment was why you want some bibtex editing stuff in lyx?
I'm sure it's possible to make a good bibtex GUI, but believe me, I tried lots of those which are available for Linux, and without exception they all suck in major ways. That is, without the exception of bibtex-mode in emacs, which I reluctantly settled on as I'm no huge fan of emacs. But once you get used to it (which is really pretty easy since all the functionality can be accessed from a couple menus, no need to remember countless C-M-X-Y-Z key commands), it's pretty powerful and complete (AFAIK it supports every nook and cranny of bibtex, which all the GUI tools I tried do not).
Personally, I see no need for further bibtex integration in lyx. Sure, it's a volunteer project so the people working on it are free to do whatever they fancy, but IMHO there are plenty of more useful and important stuff that could be done first.
Yes, the new ssa architecture in GCC 4.0 allows for autovectorization, but at the moment the focus is on getting GCC 4.0 sufficiently stable for release in a few months. Because of this, IIRC, some of the fancier vectorization passes were deferred until GCC 4.1.
So yes, you might see some performance improvements due to vectorization in 4.0, but you'll have to wait until 4.1 or maybe even 4.2 before you'll see the full potential of it.
-joib, occasional GCC contributor (although I have absolutely zilch to do with the mid- and backend stuff, were most of the optimization passes are done)
since they have a large audience now that can take advantage of them maybe LyX will start accelerating development and adding in some nice features that will make document creation much more productive.(integrate a bib database for god sakes!!!)
Personally, I'm very happy using a combination of LyX and BibTeX-mode in emacs. What more do you need?
You assume we all run some form of Unix.
Ok, so there's also planner-mode in emacs, which works just fine under unix, windows, and probably whatever platform you fancy.
This article explains the philosphy behind L4, and how it's different from Mach.
On August 1 2005, a new EU directive will come into force. It will make manufacturers/resellers of consumer electronics responsible for recycling their own product. That is, the customer is entitled to return his product for recycling for free.
I think this sounds like a better idea than yet another inefficiently run government program, as it gives some incentive to the manufacturer to reduce recycling costs, e.g. by running the recycling center as efficienty as possible but also by designing products that are as easy to recycle as possible.
IMHO, FBDIMM is just Intel's hedge against RAMBUS going bust. The point of RAMBUS was to reduce pincount per chip by reducing the width of the channel between the memory controller and the chip, and to decouple the notion of a memory controller controlling specific banks of memory. FBDIMMs are to solve the same problem, except RAMBUS is shipping already.
Reducing pincount is one important reason for FB-DIMM, but the real reason for it is to get out of the capacity/speed tradeoff game. See, many systems need lots of memory. However, with current DDR-400 or DDR2-667 you can only put two devices per channel. If you want more RAM than what fits in two devices, you have to reduce the speed. FB-DIMM gets around this problem by using point-to-point links between the devices.
Yes, this increases latency a little bit, but there really isn't any other practical way to increase speed without reducing capacity. However, FB-DIMM compensates for the increased latency by allowing many outstanding transactions on each channel; because of this, latency under high load is actually supposed to be lower than for traditional RAM tech with the same specs.
No matter which version you use, if you never VACUUM dead row versions will accumulate and eventually kill your performance.
Actually, for 7.4 and newer that isn't true, since they include the aptly named "autovacuum" daemon. And yes, autovacuum also does "VACUUM ANALYZE", so no need to worry about that either.
Next-generation Pentium M processors are supposed to be dual-core.
AFAIK, all multi-cpu AMD Opteron systems are NUMA. Each processor connects to its own memory (since the processors sport on-die memory controllers), access to memory attached to other processors is via the hypertransport link.
Actually, most of the work SGI did was against the 2.4 kernel (in fact, Altix systems still ship with a SGI patched 2.4 kernel). The vast majority of these patches were included in the 2.6 kernel, which probably is the major reasons that the standard 2.6 kernel scales so well up to at least 64 cpu:s, as this test showed.
$79*200~=$16,000 or enough for a few extra servers per year =) Most IT directors will take that anytime! Not only that but it also means a few tons less sulphur dioxide and quite a few tons less CO2 produced, which if you're even a little interested in the environment is a good thing. All of that benifit at basically zero opertunity cost, why not?!?!
Probably because the Pentium M and its motherboard is significantly more expensive than a comparable A64/P4 setup. Companies are interested in minimizing total cost over the lifetime of the computer, of which yearly electricity costs are only a part.
Yes, all this and lower power consumption (than P4) to boot. There should be some sweet notebooks and servers coming out over the next few months also, as the true low-power Athlon64s and Opterons roll out.
While I agree that Athlon64 and Opteron are superior to Intels offerings in the desktop and server market, I have to say that Intel really hit the nail on the head with the Pentium M. While the P-M won't match a high clocked Athlon64 performance-wise, it's still plenty fast and uses very little power. I think it's something like 21 W, while the mobile A64 is around 35 W, a significant advantage for notebooks.
Exactly. Remember all the hype about electric vehicles 10-20 years ago? We were all supposed to be using electric vehicles by now. Unsuprisingly, electric vehicles are nowhere to be seen.
I just wish it were easier to extract only the packages I like and use them without having to install the whole deal.
With disk space costing less than $1/GB, I don't see the point of your argument.
As for old computers with small disks, well tough. The KDE/GNOME apps are probably too heavy for them anyways.
Oh well, I guess marketing didn't approve of it...
It works for land.
To an extent. As such, I see no wrong with private ownership, private entities can probably manage their land more efficiently than the government. However, there is certainly the moral case to be made that as land, or natural resources (land, minerals, spectrum, etc.) more generally, was here long before man, no single man has the right to said resource. Thus the government can fairly tax these resources heavily, up to the rental value of the resource (and in the process reduce other, more distorting, taxes).
And before you go off labeling this as yet another communist conspiracy, consider that this concept has been endorsed by many high brow economists, even right wingers like Milton Friedman.
Google found the following quotes:
"Pure ground rent is in the nature of a 'surplus,' which can be taxed
heavily without distorting production incentives or reducing
efficiency."
-- Paul Samuelson, Nobel laureate in Economics
"In my opinion the least bad tax is the property tax on the unimproved
value of land, the Henry George argument of many, many years ago."
--Milton Friedman, Nobel laureate in Economics
"It is important that the rent of land be retained as a source of
government revenue."
-- Franco Modigliani, Nobel laureate in Economics
"For efficiency, for adequate revenue, and for justice, every user of
land should be required to make an annual payment to the local
government equal to the current rental value of the land he or she
prevents others from using."
-- Robert Solow, Nobel laureate in Economics
"While the governments of developed nations with market economies
collect some of the rent of land, they do not collect nearly as much
as they could, and they therefore make unnecessarily great use of
taxes that impede their economies -- taxes on such things as incomes,
sales, and the value of capital goods."
-- William Vickrey, Nobel laureate in Economics and past
president of the American Economics Association
First I'd like to say that there has been a record amount of interest in this post. (dont mod me down, I can't help being disapointed in a posibly world saving subject being ignored in favor of less important subjects)
Take it easy. You know, there have been prototypes of these stirling dish things since at least the early 1980:s. If they can be made economically viable in the 2010 timeframe as the article suggest, I'm sure they will be discussed a lot.
There was a project some years ago in Australia where the heat difference between (one of our) desert surface and pipes/heat exchange buried underground was going to power a whole lot of engines. I'm not positive but I think they were Stirling engines that had been changed in some way (possibly day and night running- desert cold at night). They got to the stage of cupple of dozen in a test area to show efficiency workability etc. then nothing. Googling it doesnt show much. Rumors of state govt being bought out by the local energy companies which rezoned and shut them down. The usual Nexus style hippies chanted conspiracy, maybe it just wasnt as efficient as it could have been at that time.
Anyway, plenty of room and plenty of heat and cold differentials in the desert(s).
The efficiency of a heat engine is proportional to the heat difference between the cold and hot sides. In the proposal above, the heat difference is pretty small, so you'd need lots of expensive engines and piping to produce little power. I assume that was what killed it, rather than some Big Oil conspiracy.
Just something bothers me. Wouldnt the loss associated with transporting the energy back to where it is needed (suberbia, industrial and city use) by leakage make it not sustainable?
No worries, mate. Modern high voltage DC transmission lines, often used for long range transmission, have losses of about 4 % per 1000 km.
What are the efficiency ratings of other energy storages such as spliting water into Hydrogen and Oxygen to be stored separately and recombined when needed to make combustion engine generate power?
Appalingly low. Electrolysis is about 70 %, liquefaction 66 %, burning the hydrogen in a combustion engine about 40 % for a medium sized stationary engine. In total, 18 % efficient, compared to 96 % efficient for 1000 km cabling.
So you mean that a horny teenager wouldn't be able to find pr0n on the internet otherwise? Riiiight..
Any reason why not?
Assuming that the guys designing this stuff aren't total idiots, rest assured that they looked into something like your proposal. Apparently they came to the solution that using grid power and software to control the staggered startup is cheaper than adding a battery to each dish.
Ah, good old Archimedes Plutonium is still alive. In case you didn't know, that guy is a known usenet crackpot. I added him to my killfile in 1995 or so.
Actually, if you go back even longer, you'll find Mosaic, the common ancestor of both Mozilla and Internet explorer.
And before you ask, no, there is no common source code. The source code lineage has been broken many times. For example, the original Netscape was made by the same guys who did Mosaic, but apparently they didn't reuse Mosaic code. Same for Internet explorer, MS licenced a version of Mosaic (Spyglass) but it is doubtful if they actually used any of the source code for IE. And lets not forget that the Mozilla project decided to ditch the netscape codebase they had been given.
And of course, to make it even more complicated, netscape 7.x were/are based on Mozilla.
Wasn't there mozilla back in the early 90's when the net first started?
No. I guess you mean NCSA Mosaic.
Ah, those were the days. Mosaic was pretty revolutionary back then, although I swithed to Netscape 0.9 as it became available, since it could display pages while downloading over my 9600 baud modem. Mosaic needed to download the entire page before displaying anything.
The biggest machine was a huge Cray T3E - I don't recall any blinkenlights on it, but it didn't need them! I recently heard that turing.mcc.ac.uk has since been dismantled, presumably because it was no longer cost-effective for its mere few hundred Gflops. I've no idea what was done with it and its parts, or what (if anything) it has been replaced with, but it's what I thought of when I saw this article.
Yeah I saw a T3E once too, at the local supercomputer center. Quite impressive. When they upgraded to a p690 cluster a couple of years ago they just dismantled it and threw the cards into the bin. Apparently it was quite an expensive system to keep running, using lots of electricity and that funky liquid freon cooling system etc. So despite the 500 peak Gflop/s it produced, nobody wanted it.