Dude, most people will do anything for a buck. It's why we have shitty products on the shelves, crap service at every turn, etc, etc. I'm certain SCO could score an entire division of developers within weeks if they simply offer cash money.
that doesn't mean we have to buy what SCO is selling though!
You listed I/O schedulers. I think the multi-core bit talks about a PROCESS scheduler. Two different things. Linux already has specific support for Intel's HTT bullshit and understands NUMA. Understanding multi-core is a good move up.
If you have a 2P dual-core setup the best performance for two independent tasks would be spread to both chips. Specially in the AMD camp. That means each task gets a full memory bus to themselves. The trick is to pick up when two tasks have shared memory between each other and schedule that for one chip. Specially on the Intel side of things with their massive shared L2 cache.
To be fair it's half and half... well one feeds into another.
At the beginning "programmers" were hobbyists who learned it because they were interested and took it seriously. But more and more as things got commodidized managers looked for people who got things in quicker. And by quicker I mean cut corners. So that in turn bred the generation of really shitty programmers [who often call themselves "developers"].
Now you got both shitty "coders" and shitty managers who just won't take "it'll be ready when it's ready" as an answer.
Final result: We the paying customers get shit products.
I realize that, but any half-way competent developer would be bounds checking everything they do. Oh wait... this is $BIGBUSINESS so managers dictate how long a program takes to write...
I mean it's simple, before you copy or read data you make sure your destination is the right size. It helps not to write spaghetti code too I guess...
Water can store a lot more heat than air can. It's also easier to put a small waterblock on the cpu and have it pump into a huge external take with a huge surface area.
Or just spammed relentlessly. I opted to download a beta of SUSE and I still get "gee whiz you could be a SLES admin too!" spam from Novell to this day. Even though I distinctly remember checking the "fuck off, no spam please" option.
The problem is sales/marketting people rarely understand their product let alone the culture it targets.
You don't see many Windows people really clamouring over crypto accelerators. It's usually something that is custom and the people buying it are technically inclined. At the last crypto-hw firm I worked at we targetted custom OSes and Linux mostly. It was easy to setup a demo package on a cheap POS box + Linux and show the customer that the stuff works.
So having sales people piss off hippie free-range OSS zealots like say the maintainers of OpenBSD is a bad thing. [I'm kidding though, I think *BSDes are neato. I don't mean to disparage them].
What I meant is my ISP couldn't copyright stuff I download that is public domain.
Note that you could take public domain stuff, throw it on a CD and copyright the "collection". That's actually legal and therefore could be subject to DMCA laws if you have anti-piracy devices on it.
1. Massive amounts of bandwidth 2. Media to represent message 3. Ability to gather audience 4. ??? == Something to say 5. Audience [and/or profit!]
Same problem with Blogs you'll have with this [on top of the bw problem]. I'm sorry but just because you have a webcam and a net connection doesn't mean I want to waste part of my lifetime listening to retarded 12 yr olds emo about how life is so cruel and what not. Or how the latest iGizmo from Apple is all the craze and Vista will crush Linux or whatever...
If people just spent half the time they spend on chatting and blogging and actually DID something [write a book, play, read a book, write a program or improve one, play music, WHATEVER] then there would actually be blogs worth reading...
How does TPM fit into this dicussion? It's about peering rules not desktop security.
Here's my point. Company puts its site on ISP A, ISP A then starts random peering rules, people in country X boycott ISP A because they won't pay extortion fees. Company moves on to ISP B, ISP B also starts to do this, and so on. Basically the customer of the site will have a bitch of a time, their ISP won't pay the extortion fee and the company on the other end will suffer.
There are more people [and networks] in the world than just what is between the boundaries of the USA. You'll start seeing entire firms off-shoring [more than they do now] their services if net-neutrality isn't enforced.
I agree though it shouldn't be a legal battle but simply a peering one.
We're all upset that the Chinese want to introduce their closed-door proprietary standard...
But please, tell me, how many cryptographers were consulted BEFORE the design of WEP? I know of a few who worked on the implementation AFTER the design [e.g. when they couldn't change things]. WEP and WAP [and WiMAX and...] are all essentially closed door standards. Even if you're in the SIG you're only one of many. And the many are usually NOT cryptographers so they'll basically vote for whatever turns into the least amount of VB.NET code for their Windows only drivers.
Like it's so fucking hard to get a shared-secret lossy communication medium secured... AES + CCM + proper rekeying == router that doesn't cost 69.95$ at Fry's but does == a wifi device you can trust.
No rational computer shop will wholesale replace all of their machines for another product line because the competitor is 6 months ahead [or whatever the gap is]. Desktop users may flip/flop but by that logic they could just flip back in six months...
That said, I'll probably be picking up a Conroe part later this year to do benchmarking on. I'd be hard pressed to make the switch completely given that my desktop workstation is um... a bit more powerful than a single Conroe:-)
Anyways, all the power [or lack of I guess] to Intel. Competition is good, specially if companies like Dell start opening their doors a bit more.
I'm sorry, but under no conditions will "compile this C file" vary by an unmanageable amount. If you expect the power to vary by 500 Watts each time you compile something... you're sadly mistaken.
Most likely with the CPU/memory under full load the Wh deviation is less than 10% of the mean usage. On a typical desktop the Wh rating is about 200-250 at full load. If you see a variance of more than say +/- 20Wh something is wrong or the test isn't reproducible. If you think things like differing occurences of interrupts and cache misses will make a difference in Wattage... you're wrong.
So if you can say box A and box B compile [or do work] in T units of time, let's compare the power. Box A takes 100Wh +/- 15Wh and box B takes 175Wh +/- 20Wh. Which one do you suppose takes more power? Is that "numbers-saying-anything-they-want?"
That's the whole MIPS/Watt thing. If AMD and Intel hit an empasse where both are just as IPC efficient the next question is the power they take to get there. Sure if Box A takes 3x the time the results need adjusting (namely it takes 3x the power) but we say MIPS PER WATT for a reason.
Nothing wrong with being a bit thorough. I mean, why then can't I still accomplish the same unit of work at twice the clock, half the time and the same amount of energy?
Yeah, the simple answer is "more shit is happening for a constant unit of time". The more accurate answer is the circuit is less efficient requiring more energy to operate at a higher frequency.
That also explains why you can't scale indefinitely without the chip melting. If was a matter of work you could just duty cycle it. Sure your throughput doesn't change but latency would be awesome. So crack out a 8Ghz chip that has a 50% cycle and we're good?
My point was that scaling a 200Mhz chip doesn't usually make sense since you needed the speed to use the damn box. Sure you could idle it but the savings wouldn't be as important.
Only when we started getting into designs like the K7 and Core processors did the speed become excessive. A 2Ghz K7 core was way more than capable of playing mp3s or video files while not killing the box.
Slashdot Mirror
Dude, most people will do anything for a buck. It's why we have shitty products on the shelves, crap service at every turn, etc, etc. I'm certain SCO could score an entire division of developers within weeks if they simply offer cash money.
that doesn't mean we have to buy what SCO is selling though!
Tom
I dunno, fark has more of a cult following while slashdot is a bit more diverse. ...
/.I read both
/./.not on a plane
/././.slashdotties!
Tom
You listed I/O schedulers. I think the multi-core bit talks about a PROCESS scheduler. Two different things. Linux already has specific support for Intel's HTT bullshit and understands NUMA. Understanding multi-core is a good move up.
If you have a 2P dual-core setup the best performance for two independent tasks would be spread to both chips. Specially in the AMD camp. That means each task gets a full memory bus to themselves. The trick is to pick up when two tasks have shared memory between each other and schedule that for one chip. Specially on the Intel side of things with their massive shared L2 cache.
Tom
Why punish the people who PAID for the product????????
/me hugs OpenOffice and teTeX
And people thought just running Windows was punishment enough...
Tom
To be fair it's half and half ... well one feeds into another.
At the beginning "programmers" were hobbyists who learned it because they were interested and took it seriously. But more and more as things got commodidized managers looked for people who got things in quicker. And by quicker I mean cut corners. So that in turn bred the generation of really shitty programmers [who often call themselves "developers"].
Now you got both shitty "coders" and shitty managers who just won't take "it'll be ready when it's ready" as an answer.
Final result: We the paying customers get shit products.
Tom
I realize that, but any half-way competent developer would be bounds checking everything they do. Oh wait... this is $BIGBUSINESS so managers dictate how long a program takes to write...
I mean it's simple, before you copy or read data you make sure your destination is the right size. It helps not to write spaghetti code too I guess...
Tom
Do you get executable code in a SPREADSHEET!?!
Seems like another MSFT "feature".
Tom
Water can store a lot more heat than air can. It's also easier to put a small waterblock on the cpu and have it pump into a huge external take with a huge surface area.
Tom
Or just spammed relentlessly. I opted to download a beta of SUSE and I still get "gee whiz you could be a SLES admin too!" spam from Novell to this day. Even though I distinctly remember checking the "fuck off, no spam please" option.
The problem is sales/marketting people rarely understand their product let alone the culture it targets.
You don't see many Windows people really clamouring over crypto accelerators. It's usually something that is custom and the people buying it are technically inclined. At the last crypto-hw firm I worked at we targetted custom OSes and Linux mostly. It was easy to setup a demo package on a cheap POS box + Linux and show the customer that the stuff works.
So having sales people piss off hippie free-range OSS zealots like say the maintainers of OpenBSD is a bad thing. [I'm kidding though, I think *BSDes are neato. I don't mean to disparage them].
Tom
People buying crypto accelerators tend not to be the same "Best Buy Shopping ooh wow 3 GigaHurts" type of people.
If you're doing hardware crypto you're going custom and using BSD wouldn't be a far stretch.
Tom
What I meant is my ISP couldn't copyright stuff I download that is public domain.
Note that you could take public domain stuff, throw it on a CD and copyright the "collection". That's actually legal and therefore could be subject to DMCA laws if you have anti-piracy devices on it.
Tom
Gah you can't make something that is public domain and make it not public domain in terms of distribution.
My head asploded.
Tom
Usually when I get bored and need entertainment I either play videa games, write code/text/etc or just go for a walk.
You don't HAVE to be sitting on your ass staring at an electron gun to be content.
Tom
1. Massive amounts of bandwidth
2. Media to represent message
3. Ability to gather audience
4. ??? == Something to say
5. Audience [and/or profit!]
Same problem with Blogs you'll have with this [on top of the bw problem]. I'm sorry but just because you have a webcam and a net connection doesn't mean I want to waste part of my lifetime listening to retarded 12 yr olds emo about how life is so cruel and what not. Or how the latest iGizmo from Apple is all the craze and Vista will crush Linux or whatever...
If people just spent half the time they spend on chatting and blogging and actually DID something [write a book, play, read a book, write a program or improve one, play music, WHATEVER] then there would actually be blogs worth reading...
Tom
From their outward actions AMD seems to work with Nvidia, VIA and ATI all alike.
An alliance with just one would go against everything they hoped to accomplish with their open platforms.
Of course I'm not part of the "need to know" crowd so don't take my word for gospel...
Tom
How does TPM fit into this dicussion? It's about peering rules not desktop security.
Here's my point. Company puts its site on ISP A, ISP A then starts random peering rules, people in country X boycott ISP A because they won't pay extortion fees. Company moves on to ISP B, ISP B also starts to do this, and so on. Basically the customer of the site will have a bitch of a time, their ISP won't pay the extortion fee and the company on the other end will suffer.
There are more people [and networks] in the world than just what is between the boundaries of the USA. You'll start seeing entire firms off-shoring [more than they do now] their services if net-neutrality isn't enforced.
I agree though it shouldn't be a legal battle but simply a peering one.
Tom
Only explanation why a printer driver could be more than 10MB in size :-)
Driver writers are usually the lowest of the low in terms of programming ability.
Tom
We're all upset that the Chinese want to introduce their closed-door proprietary standard...
...] are all essentially closed door standards. Even if you're in the SIG you're only one of many. And the many are usually NOT cryptographers so they'll basically vote for whatever turns into the least amount of VB.NET code for their Windows only drivers.
But please, tell me, how many cryptographers were consulted BEFORE the design of WEP? I know of a few who worked on the implementation AFTER the design [e.g. when they couldn't change things]. WEP and WAP [and WiMAX and
Like it's so fucking hard to get a shared-secret lossy communication medium secured... AES + CCM + proper rekeying == router that doesn't cost 69.95$ at Fry's but does == a wifi device you can trust.
Tom
When americans start finding their packets getting null-routed in the "rest-o-the-world" they may think twice about sitting apathetically by.
There are 270 million people in the states. There are 6 billion on earth.
FUCK YOU.
Tom
My copy of windows was supplied by my employer.
My personal workstation runs Gentoo.
MSFT can do pretty much ANYTHING THEY WANT because I never pay for their stupidity.
Tom
No rational computer shop will wholesale replace all of their machines for another product line because the competitor is 6 months ahead [or whatever the gap is]. Desktop users may flip/flop but by that logic they could just flip back in six months...
... a bit more powerful than a single Conroe :-)
That said, I'll probably be picking up a Conroe part later this year to do benchmarking on. I'd be hard pressed to make the switch completely given that my desktop workstation is um
Anyways, all the power [or lack of I guess] to Intel. Competition is good, specially if companies like Dell start opening their doors a bit more.
Tom
I'm sorry, but under no conditions will "compile this C file" vary by an unmanageable amount. If you expect the power to vary by 500 Watts each time you compile something... you're sadly mistaken.
... you're wrong.
Most likely with the CPU/memory under full load the Wh deviation is less than 10% of the mean usage. On a typical desktop the Wh rating is about 200-250 at full load. If you see a variance of more than say +/- 20Wh something is wrong or the test isn't reproducible. If you think things like differing occurences of interrupts and cache misses will make a difference in Wattage
So if you can say box A and box B compile [or do work] in T units of time, let's compare the power. Box A takes 100Wh +/- 15Wh and box B takes 175Wh +/- 20Wh. Which one do you suppose takes more power? Is that "numbers-saying-anything-they-want?"
That's the whole MIPS/Watt thing. If AMD and Intel hit an empasse where both are just as IPC efficient the next question is the power they take to get there. Sure if Box A takes 3x the time the results need adjusting (namely it takes 3x the power) but we say MIPS PER WATT for a reason.
Tom
Nothing wrong with being a bit thorough. I mean, why then can't I still accomplish the same unit of work at twice the clock, half the time and the same amount of energy?
Yeah, the simple answer is "more shit is happening for a constant unit of time". The more accurate answer is the circuit is less efficient requiring more energy to operate at a higher frequency.
That also explains why you can't scale indefinitely without the chip melting. If was a matter of work you could just duty cycle it. Sure your throughput doesn't change but latency would be awesome. So crack out a 8Ghz chip that has a 50% cycle and we're good?
Tom
My point was that scaling a 200Mhz chip doesn't usually make sense since you needed the speed to use the damn box. Sure you could idle it but the savings wouldn't be as important.
Only when we started getting into designs like the K7 and Core processors did the speed become excessive. A 2Ghz K7 core was way more than capable of playing mp3s or video files while not killing the box.
Tom
Maybe not the K7 semprons but I thought K8s did. Did your BIOS recognize the CPU?
Tom