There are no easy answers. It's hard enough to compare even a Celeron and a PII, which are more similar than what you described. No one can tell you, for a given hardware configuration, how many frames per second Quake will render (for example) without actually building that system. I think a lot depends on how well your OS and applications take advantage of the extra processor(s). I would look at Tom's Hardware for informative, nicely written, discussions.
My personal choice: I got a board capable of dual Intel processors, but only put one Celeron 333 in it. I figure if I get really antsy for CPU speed, which hasn't been a problem yet, I'll either get a much faster single chip or buy a pair of moderately faster matched chips. The board was $235 about a year ago and was one of the cheapest dual-capable boards that supported 1 GB RAM.
Good luck with whatever you decide, and don't forget to make a web page describing your achievements!
Congrats on S10K. Maybe now it's time to have a moderation category for the stories themselves. The rating could appear on the front page beside the story, for example. Unlike regular moderation, it would always be an option, but like regular moderation, it would preclude posting comments on that story.
Such a system would let all those people who post saying "What does this topic have to do with nerds?" or "This doesn't matter" could instead let their opinion be heard by reducing the rating of the story. What a relief it would be to the rest of us.
political liberals are described as being "on the left" and conservatives "on the right"
I heard that was because in the House or Senate seating was arranged so that the more conservative party sat on the right side while the more liberal party sat on the left.
If you were VA, why would you want to own Slashdot and Freshmeat so badly?
If I were VA, I'd want to own Slashdot and Freshmeat because I'd want to take advantage of all that IPO capital and do some diversification while I still could.
Family income and educational level of parents are the biggest predictors of how well students score on traditional college admissions tests, Swenson said.
I also heard this stated in college, and I'm inclined to believe it. In this country, being black, and/or Hispanic, and/or Native American, is positively correlated with lower family income and educational level of parents, as far as I know. That goes a long way to explaining lower average SAT performance amongst those groups. It may even account for all of the measured differences. Anyone know?
Just because they use Linux doesn't mean jack shit.
That's definitely debatable, though someone else already has. My secondary point (aside from tossing out a Linux tidbit) was that lots of us get caught up in our Linux zealotry to the point where we would consider something like this to be a mitigating factor.
For a reverse of this example, consider Slashdot. Most of us think they're a pretty cool site, but some have pointed out that in spite of promoting open source, they are not terribly forthcoming with their source. (Yes, I know they just released v.9 - damn irony!)
Here's my sociological hypothesis about the quote above (are you listening, Jon Katz?): We've gotten so tightly enmeshed with our digital, binary, logical world, that we can't see that qualities like good and bad are on a continuum.
You didn't say anything about what OS, but here's a page about running Linux on SMP motherboards.
FWIW, I had more or less the same idea as you. I bought a SuperMicro P6DGE for the following reasons:
Cheap - $235
Supports SMP
Supports 1G RAM
This model doesn't have on-board SCSI (that's why it was cheap), but it does have two IDE controllers. I didn't want to lay out a lot of cash at once, so I started with a single slot 1 Celeron 333, and one 128 MB RAM chip.
Since I was upgrading from a 486 DX2/66 20 MB RAM laptop, I figured it would be a while before I would be wishing it were faster. I'm still pretty pleased with its performance - it takes maybe 5 or 10 minutes to build a kernel. (I don't know - it's too long to sit there and stare at it though.)
There are a couple of ways I considered upgrading the processors:
Determine the fastest PII/PIII that the MB will handle, and as soon as Intel stops making them, buy a pair. I figure that'll be the cheapest/fastest processors I can put on that MB.
Get some Socket 370/Slot 1 converters at about $15 each, and a couple of fast Celerons.
Anyone else have any other suggestions for cheaply going to dual-processors?
a "health" meter that slowly decreases unless you interact with the ad bar
Am I the only one that thinks this is really an evil Tamagochi plot here?
Ways to Evade this Technology
on
Author Unknown
·
· Score: 3
My vague understanding of how this technology works is that it measures a correlation between an author's works. Then works that are outside some threshold of that correlation are considered from another author, while works that are within some threshold are considered likely to be by the same author.
This suggests two ways to evade the technology. One is to apply some filter that makes everybody's work look the same. A crude example would be if everyone used the Zippy filter - that style may be so strong as to mask the differences among the original authors, as well as the similarities between original works by the same author. It is somewhat analogous to writing in block print to hide one's handwriting.
The second way is to make ones own work look different every time. In other words, make your own work have very little internal correlation. Again, a crude example would be if sometimes one wrote in Hemingway's style, then Steinbeck, then Maya Angelou, etc. This is analogous to cutting and pasting words and letters from magazines, "ransom-note" style, to hide one's handwriting.
The fascinating thing to me is that someone has come up with a way (perhaps crude so far) to statistically recognize writing style, something that we people are relatively good at already - the Unabomber's brother fingered him upon reading the manifesto, for example. The thing to remember is that it's only as good as his algorithms, which we smart hackers can surely outwit.
I think what makes this difficult to do manually is that our writing style is probably a fairly low-level characteristic, like our gait, handwriting, or voice. Just as those can be overridden with training, surely one's writing style can be as well.
We sometimes get free Pop-Tarts at work, and in the rush, the microwaves are swamped, so I heat mine on top of my monitor. It takes a while, but I'm in no hurry...
Patent/shmatent - as long as they don't turn into a (shudder) portal.
I like Google because they
Load quickly
Have good results
Run on Linux (anyone know if this is still true?)
though not necessarily in that order. I used to like AltaVista's text-only page for the first two reasons, though I think Google has better results. If Google starts to suck (note that my evaluation of whether they suck might be different from yours!), I'll go back to AltaVista.
Trusty hammer, indeed. I spent a few months maintaining some Fortran code and started calling it blunt-object oriented technology. In fairness, that describes the previous coders' styles as much as the language features.
Slashdot Mirror
My personal choice: I got a board capable of dual Intel processors, but only put one Celeron 333 in it. I figure if I get really antsy for CPU speed, which hasn't been a problem yet, I'll either get a much faster single chip or buy a pair of moderately faster matched chips. The board was $235 about a year ago and was one of the cheapest dual-capable boards that supported 1 GB RAM.
Good luck with whatever you decide, and don't forget to make a web page describing your achievements!
Such a system would let all those people who post saying "What does this topic have to do with nerds?" or "This doesn't matter" could instead let their opinion be heard by reducing the rating of the story. What a relief it would be to the rest of us.
Weak Monty Python reference...
Were you joking, or did you mean "Number nine. Number nine...."?
I play my female cat (Mignon) on Slashdot. What about other people's nicknames?
Yes, but not me. I took a graduate class at Duke University from someone who had been Knuth's graduate student at Stanford.
I heard that was because in the House or Senate seating was arranged so that the more conservative party sat on the right side while the more liberal party sat on the left.
I wonder if he found this book in the Sith Help section?
Actually, for sites of this nature, I believe it's first served, first come.
Put that in your Furby and smoke it!
If I were VA, I'd want to own Slashdot and Freshmeat because I'd want to take advantage of all that IPO capital and do some diversification while I still could.
I also heard this stated in college, and I'm inclined to believe it. In this country, being black, and/or Hispanic, and/or Native American, is positively correlated with lower family income and educational level of parents, as far as I know. That goes a long way to explaining lower average SAT performance amongst those groups. It may even account for all of the measured differences. Anyone know?
I wonder what kind of environmental disasters that would have led to.
That's definitely debatable, though someone else already has. My secondary point (aside from tossing out a Linux tidbit) was that lots of us get caught up in our Linux zealotry to the point where we would consider something like this to be a mitigating factor.
For a reverse of this example, consider Slashdot. Most of us think they're a pretty cool site, but some have pointed out that in spite of promoting open source, they are not terribly forthcoming with their source. (Yes, I know they just released v.9 - damn irony!)
Here's my sociological hypothesis about the quote above (are you listening, Jon Katz?): We've gotten so tightly enmeshed with our digital, binary, logical world, that we can't see that qualities like good and bad are on a continuum.
I have it on good authority that DoubleClick uses Linux. Seriously - no trollin'. They may be evil, but they're not all bad...
FWIW, I had more or less the same idea as you. I bought a SuperMicro P6DGE for the following reasons:
- Cheap - $235
- Supports SMP
- Supports 1G RAM
This model doesn't have on-board SCSI (that's why it was cheap), but it does have two IDE controllers. I didn't want to lay out a lot of cash at once, so I started with a single slot 1 Celeron 333, and one 128 MB RAM chip.Since I was upgrading from a 486 DX2/66 20 MB RAM laptop, I figured it would be a while before I would be wishing it were faster. I'm still pretty pleased with its performance - it takes maybe 5 or 10 minutes to build a kernel. (I don't know - it's too long to sit there and stare at it though.)
There are a couple of ways I considered upgrading the processors:
- Determine the fastest PII/PIII that the MB will handle, and as soon as Intel stops making them, buy a pair. I figure that'll be the cheapest/fastest processors I can put on that MB.
- Get some Socket 370/Slot 1 converters at about $15 each, and a couple of fast Celerons.
Anyone else have any other suggestions for cheaply going to dual-processors?Am I the only one that thinks this is really an evil Tamagochi plot here?
This suggests two ways to evade the technology. One is to apply some filter that makes everybody's work look the same. A crude example would be if everyone used the Zippy filter - that style may be so strong as to mask the differences among the original authors, as well as the similarities between original works by the same author. It is somewhat analogous to writing in block print to hide one's handwriting.
The second way is to make ones own work look different every time. In other words, make your own work have very little internal correlation. Again, a crude example would be if sometimes one wrote in Hemingway's style, then Steinbeck, then Maya Angelou, etc. This is analogous to cutting and pasting words and letters from magazines, "ransom-note" style, to hide one's handwriting.
The fascinating thing to me is that someone has come up with a way (perhaps crude so far) to statistically recognize writing style, something that we people are relatively good at already - the Unabomber's brother fingered him upon reading the manifesto, for example. The thing to remember is that it's only as good as his algorithms, which we smart hackers can surely outwit.
I think what makes this difficult to do manually is that our writing style is probably a fairly low-level characteristic, like our gait, handwriting, or voice. Just as those can be overridden with training, surely one's writing style can be as well.
Why not just encrypt it with CSS?
I wonder if Crusoe will come with a coprocessor called Friday.
We sometimes get free Pop-Tarts at work, and in the rush, the microwaves are swamped, so I heat mine on top of my monitor. It takes a while, but I'm in no hurry...
I like Google because they
- Load quickly
- Have good results
- Run on Linux (anyone know if this is still true?)
though not necessarily in that order. I used to like AltaVista's text-only page for the first two reasons, though I think Google has better results. If Google starts to suck (note that my evaluation of whether they suck might be different from yours!), I'll go back to AltaVista.Trusty hammer, indeed. I spent a few months maintaining some Fortran code and started calling it blunt-object oriented technology. In fairness, that describes the previous coders' styles as much as the language features.
In case anyone's interested, here's the low-graphics, printer-friendly link.
The same reason that's a pain in the ass when dealing with the clueless is why "slash dot dot org" is cool.