There are plenty of articles like this, from years ago. Every time again people say 'it won't be possible to make it much faster, because we are at the limits of nature'.
Then, one month after I buy a new computer, one that is about twice at fast comes out;-)
OTOH, now even M$ can't come up with Windows/Office versions slow enough to justify a GHz computer for desktop use, there may just be no very large market need for faster processors at the moment.
Look at it this way: you could program the BIOS to always erase the memory on POST, *unless* there was a power faillure (modern ATX supplies can already detect this I believe)
So when you reboot on purpose, everything will be erased, but when power fails, you'll lose nothing!
This also makes creating a hibernation function much, much easier - no more need for a large image file on your harddisk, just let the BIOS know it should *not* erase memory contents after next reboot.
I know I could use a spare machine as a firewall and run Linux's IP masquerading. My only spare machine, however, is an old 486dx2-66 with an NE2000 ethernet card. Not exactly a speed demon, and speed is exactly why I got a cable modem
A DX2-66? I think that's fast enough for a masquerading box, you just have to put in a second ethernet card. I have used a 50 Mhz 386 (8 MB RAM) as a IP Masquerading server for a long time. We only have 60KB/s downstream and 7K upstream though (also cablemodem)...
It's not like you're running Windows, so you don't necessarily need a PII and 128 MB of memory just to run IP masquerading...
Firstbrook objects to the very feature that most tout as Linux's number one asset--the fact that anyone can tweak the code--because it creates a situation in which an IT staffer may make changes that no one else knows about, and that probably go undocumented.
Yeah, and when installing just standard <insert favorite distro here>, you don't have this problem.
Unlike with Micro$oft, where you have undocumented 'features' in about *every* program I know of (not even talking about bugs, just stuff you can't find anywhere in the documentation)
Alcohol is a really flammable liquid. I'm wondering whether it isn't rather dangerous to use it for cooling?
Imagine when circulation of the liquid stops and it starts to boil or something. Then, with many electric equipment around it, it looks to me like you're really *asking* for nasty accidents (computer on fire etc.)
It is about time that Intel saw that Rambus does actually suck.
The release of their 815i chipset already pointed in this direction (Rambus didn't really like that move:)
Also, the 815i chipset seems to be faster then the 820i chipset (which uses expensive Rambus memory).
Now, it looks like they're indeed going to drop it. Look at some of these articles:
This article posted earlier today mentions that the new 1.4 Ghz Pentium IV will also support SDRAM, not only Rambus memory.
This article at Tom's Hardware talks about the performance of 820 and 815 chipsets.
That contest could create some real good compiler tests.
Like, how about a solution to Towers of Hanoi that let the compiler solve the problem by recursively including itself, eventually creating one huge 'printf()'-statement?
And it used a compiler switch for the number of pegs. At the time I tested it, gcc broke (or my computer ran out of memory, I'm not really sure which of the two...) at 15 pegs.
The solution to 14 pegs would create a over 1 MB executable containing just an MB of printf("really long string"); text.
If I find a link to it somewhere I'll let you know. The International Obfuscated C Code Contest (IOCCC) has it's homepage at www.ioccc.org, but I can't seem to remember the name of the program.
VLIW means Very Long Instruction Word, which means that the processor can execute
several instructions at one time (one Instruction Word can contain multiple instructions).
This is a good idea because this way the processor does not have to worry about
running instructions concurrently. Instead, the compiler has to worry about that,
which makes it really hard to write a compiler that handles this right, while still
optimizing as much as possible.
Since the VLIW technology has not been tested very much in microprocessor designs,
Intel is doing some really hard (maybe even innovate!) work, but that also means
many problems/bugs/performance issues, as with almost every 'new' technology.
VLIW is successfully used in (e.g.) DSP processors, but those have very simple
instruction sets, where the order in which things are processed often does not
really matter, so this is rather simple to implement.
However, in a microprocessor it's much harder, which may be part of the reason why Intel's Itanium is delayed so much (performance problems, and problems running high clock frequencies).
According to Tom's Hardware, the problem with using VLIW in complicated microprocessors is writing a fast compiler, which is really hard since instruction have to be executed concurrently (or if this is impossible you'll lose much performance). Read the article for further information.
This is also why the Crusoe is not as fast as a Pentium III or Athlon (the Crusoe uses VLIW too). However, in the Crusoe case it might not be a problem since Transmeta is targeting another market: PDA's, laptops and such, requiring low power usage and long battery live.
Obviously, the Itanium (stupid name!) is targeted at the server market, where power consumption does not really matter (except when running into cooling problems ofcourse!), so it had better be fast.
And until now it looks like it isn't! Little problem for Intel that has yet to be solved!
You obviously never read Tom's Hardware. Look at this article describing how Tom thinks about Intel's IA64 architecture (Itanium, formerly Merced) in relation to AMDs SledgeHammer technology.
Why do you think it's called SledgeHammer in the first place???
I always wonder how much power goes into these kind of beasts.
Let's try to estimate it: 682 systems each containing 4 processors. I guess that they will need a 300 W power supply. So that makes about 204 KW just for the computers (when working at full speed only, OK)!
At 110 V this thing would eat 1860 Ampere, not something you'd like to try at home or something (imagine the electricity bill:-)
I don't have/want a creditcard, and since I live in the Netherlands it would be a lot of hassle to send $1 by check or money order. I guess the transaction itself would cost me (a lot?) more than $1.
Too bad because I'd really like to read the story but don't really want to mess up this experiment.
If we ever hear a proposal from the FBI in which it plans to install Carnivores at all 6000 ISPs in the U.S., we'll be giving the government the power to do something it can't do right now.
Shut the Internet down.
How many times do you Americans have to make this mistake? THE U.S. ARE NOT THE ONLY COUNTRY THAT ARE CONNECTED TO THE INTERNET!!!
Okay, so if the FBI does this, they can shut down all sites in the USA. Although many sites will stop working then, how about the rest of the world? Not a problem there!
So I think speaking of 'Shutting down the internet' is a little exaggarating.
Looks to me that binary logging can be much faster then logging in the W3C extended logfile format. Please correct me if I'm mistaken, but I think this can really cause rather different results
Slashdot Mirror
There are plenty of articles like this, from years ago. Every time again people say 'it won't be possible to make it much faster, because we are at the limits of nature'.
;-)
:-)
Then, one month after I buy a new computer, one that is about twice at fast comes out
OTOH, now even M$ can't come up with Windows/Office versions slow enough to justify a GHz computer for desktop use, there may just be no very large market need for faster processors at the moment.
Guess I am wrong, am i????
So, this time, CowboyNeal will 'scratch an itch'?
Have fun guys :-)
Yes, we do.
Look at it this way: you could program the BIOS to always erase the memory on POST, *unless* there was a power faillure (modern ATX supplies can already detect this I believe)
So when you reboot on purpose, everything will be erased, but when power fails, you'll lose nothing!
This also makes creating a hibernation function much, much easier - no more need for a large image file on your harddisk, just let the BIOS know it should *not* erase memory contents after next reboot.
Yeah, just like Golems who have Holy Words in their heads (read Terry Pratchett - Feet of Clay, it's real fun :)
A DX2-66? I think that's fast enough for a masquerading box, you just have to put in a second ethernet card. I have used a 50 Mhz 386 (8 MB RAM) as a IP Masquerading server for a long time. We only have 60KB/s downstream and 7K upstream though (also cablemodem)...
It's not like you're running Windows, so you don't necessarily need a PII and 128 MB of memory just to run IP masquerading...
Yeah, and when installing just standard <insert favorite distro here>, you don't have this problem.
Unlike with Micro$oft, where you have undocumented 'features' in about *every* program I know of (not even talking about bugs, just stuff you can't find anywhere in the documentation)
Alcohol is a really flammable liquid. I'm wondering whether it isn't rather dangerous to use it for cooling?
Imagine when circulation of the liquid stops and it starts to boil or something. Then, with many electric equipment around it, it looks to me like you're really *asking* for nasty accidents (computer on fire etc.)
I am not sure, but I believe Rambus stock is dropping quite a lot today. At this moment, it's down by some 5% compared to yesterday :-)
:-)
Of course, I may be mistaken. Look for yourself at Nasdaq:RMBS
Let's see wath a little slashdot effect can do to stock quotes now
The release of their 815i chipset already pointed in this direction (Rambus didn't really like that move :)
Also, the 815i chipset seems to be faster then the 820i chipset (which uses expensive Rambus memory). Now, it looks like they're indeed going to drop it. Look at some of these articles:
I just found it. It's called 'vanschnitz' and can be found here
Of the Obfuscated C Code Contest.
That contest could create some real good compiler tests.
Like, how about a solution to Towers of Hanoi that let the compiler solve the problem by recursively including itself, eventually creating one huge 'printf()'-statement?
And it used a compiler switch for the number of pegs. At the time I tested it, gcc broke (or my computer ran out of memory, I'm not really sure which of the two...) at 15 pegs.
The solution to 14 pegs would create a over 1 MB executable containing just an MB of printf("really long string"); text.
If I find a link to it somewhere I'll let you know. The International Obfuscated C Code Contest (IOCCC) has it's homepage at www.ioccc.org, but I can't seem to remember the name of the program.
No, I think it is cheaper to just bribe someone at an university to say there's nothing wrong with it than appointing an entire comittee :-)
For the uninformed,
VLIW means Very Long Instruction Word, which means that the processor can execute
several instructions at one time (one Instruction Word can contain multiple instructions).
This is a good idea because this way the processor does not have to worry about
running instructions concurrently. Instead, the compiler has to worry about that,
which makes it really hard to write a compiler that handles this right, while still
optimizing as much as possible.
Since the VLIW technology has not been tested very much in microprocessor designs,
Intel is doing some really hard (maybe even innovate!) work, but that also means
many problems/bugs/performance issues, as with almost every 'new' technology.
VLIW is successfully used in (e.g.) DSP processors, but those have very simple
instruction sets, where the order in which things are processed often does not
really matter, so this is rather simple to implement.
However, in a microprocessor it's much harder, which may be part of the reason why Intel's Itanium is delayed so much (performance problems, and problems running high clock frequencies).
According to Tom's Hardware, the problem with using VLIW in complicated microprocessors is writing a fast compiler, which is really hard since instruction have to be executed concurrently (or if this is impossible you'll lose much performance). Read the article for further information.
This is also why the Crusoe is not as fast as a Pentium III or Athlon (the Crusoe uses VLIW too). However, in the Crusoe case it might not be a problem since Transmeta is targeting another market: PDA's, laptops and such, requiring low power usage and long battery live.
Obviously, the Itanium (stupid name!) is targeted at the server market, where power consumption does not really matter (except when running into cooling problems ofcourse!), so it had better be fast.
And until now it looks like it isn't! Little problem for Intel that has yet to be solved!
You obviously never read Tom's Hardware. Look at this article describing how Tom thinks about Intel's IA64 architecture (Itanium, formerly Merced) in relation to AMDs SledgeHammer technology.
Why do you think it's called SledgeHammer in the first place???
I always wonder how much power goes into these kind of beasts.
:-)
Let's try to estimate it: 682 systems each containing 4 processors. I guess that they will need a 300 W power supply. So that makes about 204 KW just for the computers (when working at full speed only, OK)!
At 110 V this thing would eat 1860 Ampere, not something you'd like to try at home or something (imagine the electricity bill
So it is probably just a bug in the page itself, or in the URL
Makes you wonder which OS it runs, since my connection comes trough without problem, but I just get an 'Internal server error' or something
Too bad because I'd really like to read the story but don't really want to mess up this experiment.
Shut the Internet down.
How many times do you Americans have to make this mistake?
THE U.S. ARE NOT THE ONLY COUNTRY THAT ARE CONNECTED TO THE INTERNET!!!
Okay, so if the FBI does this, they can shut down all sites in the USA. Although many sites will stop working then, how about the rest of the world? Not a problem there!
So I think speaking of 'Shutting down the internet' is a little exaggarating.
So I think the RIAA could still screw you in this case, although they probably won't anyway...
Okay, a little concurrention ain't bad as we always say...
Anyone checked if you can do the same to cats to?
Might come in handy when those mouses get too big....
Win2000: Log Mode W3C: Extended Log File Format
Linux: Binary CLF
Looks to me that binary logging can be much faster then logging in the W3C extended logfile format. Please correct me if I'm mistaken, but I think this can really cause rather different results