The potential speed increase isn't seen when comparing 1G RAM vs a 2G RAM system. Its comparing a 1G RAM system with a 1G RAM system with swap.
The gist of it is: with swap you can put things that aren't being used (like mingetty, gdm, etc) into swap to free up space for things that are running now. Without swap you have to keep the little-used processes in memory and you don't have as much 'free' space to use for things like caches.
Its also important to note that the kernel will swap out code segments regardless of whether or not you have a swap partition: they get swapped out to nowhere. When they need to be swapped back in, the executable file itself is read.
Yeah, its going to suck, that's not the question anymore. The real question is "By what method will it suck?" I'm thinking its either going to be "Deep Hurting" or "Sand storm"
The source RPMS are available from their servers (though I can't check right now...server is too busy), so you can build it yourself. They're charging for the packaging (which is the.iso as well since it all has to work together), the support, and for any non-free software, but not the free software, which is the right thing to do.
Hear hear! I paged through the whole thing hoping to find some _actual_ information, but was really disappointed on page 6 which was basically the summary over again.
Its not an article, its a powerpoint presentation by HP/Intel. Far from objective...they gloss over many important things to make the Itanium appear to shine.
All of the spec numbers are related to the EV68, not the new EV7. Its an apples/oranges comparison for performance.
"Itanium...has 30% fewer branches than Alpha" is also comparing apples to oranges because the branches are hidden with predicates.
Any time that HP/Intel claim a big win because of "compiler technology", their missing the point: most of these advances can be applied to any architecture.
There's nothing superior architecture-wise about Itanium. If anything, its a step backwards because a compiler cannot know the path software will take in advance. Period. Systems which adapt to the data stream at run time will have better performance overall no matter how much "compiler technology" advances.
Just think if this had been available during the CIA's Operation Acoustic Kitty program? They could have made kitty dentures be the antenna, instead of that "hard to maintiain" tail.
Of course, it'd still have problems with wandering away & getting hit by traffic...
Lots of little patterns in there, providing you cross byte boundaries. 4 1's in a row happens 4 times. 3 zeroes in a row come up 7 times.'10' comes up 17 times. '100' comes up 12 times. '0100' comes up 8 times.
Can this be encoded in a way that takes less than 10 bytes? Don't know. Don't care really, but there are patterns in there.
A thought just occurred to me: If you can do 100:1 compression and compress something down to, say, 2 bytes, what would 'ab' expand to? My thought is "ZeoSync Rulz, Suckas"
Well, I can think of two ways that "random" data might be compressed without an obvious pattern:
* If the data was represented a different way (say, using bits instead of bytesize data) then patterns might emerge, which would then be compressable. Of course, the $64k question is: will it be smaller than the original data?
* If the set of data doesn't cover all possibilities of the encoding (i.e. only 50 characters out of 256 are actually present), then a recoding might be able to compress the data using a smaller "byte" size. In this case, 6 bits per character instead of 8. The problem with this on is that you have to scan through all of the data before you can determine the optimal bytesize...and then it still may end up being 8.
If the Compaq/HP merger goes through, I suspect that VMS will end up on the heap too. MPE was the pro-vms people's rallying point that HP might keep VMS alive after the merge.
Its the best kept secret in the IT industry. Sure, its not going to make you rich overnight, but you're not going to find yourself jobless the next day, either:)
The other nice thing about the uni environment is the surplus machines...I've picked up vaxes, old suns, RS/6000, sgi machines, etc so there are plenty of toys to play with!
Slashdot Mirror
No discussion unless its immortalized into the body of a monkey. That's entertainment!
If I had mod points, you'd get a +1 Funny.
Sort of reminds me of a Spongebob episode:
Patrick: Hey! Pioneers couldn't afford cannons!
Spongebob: They couldn't afford station wagons either...
[Spongebob shoots Patrick with a station wagon coming out of the cannon]
I call bullshit on that: the L500 tape was 130 minutes, L750 tape was 195 minutes.
That's really funny! Unfortunately, I doubt many non residents will understand :)
The potential speed increase isn't seen when comparing 1G RAM vs a 2G RAM system. Its comparing a 1G RAM system with a 1G RAM system with swap.
The gist of it is: with swap you can put things that aren't being used (like mingetty, gdm, etc) into swap to free up space for things that are running now. Without swap you have to keep the little-used processes in memory and you don't have as much 'free' space to use for things like caches.
Its also important to note that the kernel will swap out code segments regardless of whether or not you have a swap partition: they get swapped out to nowhere. When they need to be swapped back in, the executable file itself is read.
Yeah, its going to suck, that's not the question anymore. The real question is "By what method will it suck?" I'm thinking its either going to be "Deep Hurting" or "Sand storm"
EST is always GMT - 5 hours. In the summer, when daylight savings time kicks in, its EDT. EDT is GMT - 4 hours.
Kernel 2.6.1 has a USB driver for the IR Tower...so combine that with NQC or BrickOS and you're set!
The source RPMS are available from their servers (though I can't check right now...server is too busy), so you can build it yourself. They're charging for the packaging (which is the .iso as well since it all has to work together), the support, and for any non-free software, but not the free software, which is the right thing to do.
Go RedHat!
With the perpetual delays, they might as well take a page from the "Duke Nuke'em Guide To Naming Future Products".
I, for one, welcome our new Slashdot T-Shirt Wearing Overlords.
Hear hear! I paged through the whole thing hoping to find some _actual_ information, but was really disappointed on page 6 which was basically the summary over again.
My best friend bought me a rubber stamp with that on it...its come in handy several times already :)
News for infidels, stuff that matters.
Seems to be down at the moment, though.
Its not an article, its a powerpoint presentation by HP/Intel. Far from objective...they gloss over many important things to make the Itanium appear to shine.
All of the spec numbers are related to the EV68, not the new EV7. Its an apples/oranges comparison for performance.
"Itanium...has 30% fewer branches than Alpha" is also comparing apples to oranges because the branches are hidden with predicates.
Any time that HP/Intel claim a big win because of "compiler technology", their missing the point: most of these advances can be applied to any architecture.
There's nothing superior architecture-wise about Itanium. If anything, its a step backwards because a compiler cannot know the path software will take in advance. Period. Systems which adapt to the data stream at run time will have better performance overall no matter how much "compiler technology" advances.
AAAAARRRGGHHH! goatse.cx strikes back!
Just think if this had been available during the CIA's
Operation Acoustic Kitty program? They could have made kitty dentures be the antenna, instead of that "hard to maintiain" tail.
Of course, it'd still have problems with wandering away & getting hit by traffic...
Maybe he meant that you had to be a lawyer to produce a summary of a list of violations that long...
>With truly random data [random.org] there's no pattern to find, assuming you're looking at a large enough sample
How big is a 'large enough sample'? Seems the larger the sample, the more likelyhood of getting longer matches.
However, given 10 bytes from random.org: 39, 233, 196, 127, 220, 228, 10, 146, 60, 68.
Strung together as binary they come out as:
00100111 11101001 11000100 01111111 11011100 11100100 00001010 10010010 00111100 01000100
Lots of little patterns in there, providing you cross byte boundaries. 4 1's in a row happens 4 times. 3 zeroes in a row come up 7 times.'10' comes up 17 times. '100' comes up 12 times. '0100' comes up 8 times.
Can this be encoded in a way that takes less than 10 bytes? Don't know. Don't care really, but there are patterns in there.
A thought just occurred to me: If you can do 100:1 compression and compress something down to, say, 2 bytes, what would 'ab' expand to? My thought is "ZeoSync Rulz, Suckas"
Well, I can think of two ways that "random" data might be compressed without an obvious pattern:
* If the data was represented a different way (say, using bits instead of bytesize data) then patterns might emerge, which would then be compressable. Of course, the $64k question is: will it be smaller than the original data?
* If the set of data doesn't cover all possibilities of the encoding (i.e. only 50 characters out of 256 are actually present), then a recoding might be able to compress the data using a smaller "byte" size. In this case, 6 bits per character instead of 8. The problem with this on is that you have to scan through all of the data before you can determine the optimal bytesize...and then it still may end up being 8.
If the Compaq/HP merger goes through, I suspect that VMS will end up on the heap too. MPE was the pro-vms people's rallying point that HP might keep VMS alive after the merge.
Hear hear!
:)
Its the best kept secret in the IT industry. Sure, its not going to make you rich overnight, but you're not going to find yourself jobless the next day, either
The other nice thing about the uni environment is the surplus machines...I've picked up vaxes, old suns, RS/6000, sgi machines, etc so there are plenty of toys to play with!
- skip the meeting where the new sysadmin is to be chosen,
- Show up to that meeting and pay attention,
- Adopt a grumpy demeanor, or
- Draw the short straw
Its fun, its easy, its BOFH.