Follow Slashdot stories on Twitter


Forgot your password?

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).


Comment: benchmarks (Score 1) 134

by suitti (#37424956) Attached to: River Trail — Intel's Parallel JavaScript

I have a couple benchmarks i've run over the years, in multiple languages. For CPU intensive jobs on the same machine:

C: 1
Java: 1.1-3
javascript: 117

Javascript is in IE8 on win xp.

Javascript on IE has restrictive time limits for execution, though there are work arounds.
But If you have 8 cores, you're still 14.6x slower than C.

IMO, Java, which already runs in the browser, is the better solution. That said, compared to C, Java also has huge memory requirements.
It would make more sense to allow mutli-core execution in Java in the browser.

Comment: performance is important (Score 1) 425

by suitti (#37271748) Attached to: Sixteen Years Later: GNU Still Needs An Extension Language

Performance is important

My benchmarks show that guile is a pretty slow. For example, on one machine, it was 238 times slower than C. So, instead of a 3.6 GIPS machine, we'll all go back to 15 MIPS. Who here wants to use a 486/66 again? Who here isn't old enough to remember when Intel had numbers for chip models? Perhaps the long jump support could be removed and half way decent performance could be put in. I mean, what good is a functional language when it's faster when you code loops?

I was researching LISP a couple years back and ran into a great quote. It talked about how a student had written an elegant matrix multiply function in LISP, which had the misfeature of being incredibly slow. It caught my attention because i'd written a nearly identical function, and was amazed at how poorly it performed. The author made the comment that no C programmer could make such a mistake. However, he totally failed to give the slightest hint on why the performance should have sucked. And, he continued to rant about how (compiled) LISP was as fast as C.

Javascript is finally getting some performance. Firefox 3.5.7's javascript is only 25.6x slower than C. The recent browser wars show rapid progress.

Readability is important
That said. I know dozens of languages. I understand my 30 year old BASIC programs without much effort. I was a bit more clever with my 25 year old C programs, so they're a bit tougher. LISP is one of only two languages where i actually took a class. But none of my LISP or scheme is halfway comprehensible two weeks after i've written it. Forth is a bit better, taking a couple months. Procedural languages are simply easier. At the end of the day, you spend most of your time in maintenance. If it isn't broken, then the requirements are changing. Oh, and time is money, just like Einstein said.

Comment: Math = truth (Score 1) 220

by suitti (#29453765) Attached to: New 'Drake Equation' Selects Between Alien Worlds

If you add 2 + 2 and get 4, you can say that this is true in a way that almost nothing else is true. And people seem to think that this means that math means truth.

But Frank Drake created his famous equation to organise his thoughts and get a handle on what is and isn't known. As time has moved on, we have gotten better estimates of the terms. For example, actually discovering 300+ planets around other stars gives us a handle on the fraction of stars with planets. And the Kepler mission should improve things even more. If Drake's equation did nothing more than inspire the launch of the Kepler mission, it would be very important indeed.

But as the Drake equation is filled in with better data, there's a next step. And it's interesting that people are thinking about what those next steps might be.

Attributed to Mark Twain: "There are liars. There are Damned Liars. And then, there are statistics."

Comment: $20K modern violin vs Strad (Score 1) 210

by suitti (#29453633) Attached to: Fungivarius Beats $2 Million Stradivarius Violin

I'm happy that people are still researching violin making. And there are many new techniques in the last 20 or so years. But I keep hearing that while the Strads are good, a $20,000 modern violin is better. At least this seems to be a consensus view among professionals. The key problem with violin making is consistency. So people have attempted to make graphite violins, and so on. And if all violins could be top notch, they wouldn't cost $20,000.

I, personally, own a very poor violin. A modern $500 violin would be an improvement. My son's $200 3/4 size (it's really just an inch shorter than mine) sounds, in my opinion, better. And for MY violin, pretty much the only opinion that matters is mine. I'm my only audience.

There are people who have publicly said that if they obtained a Stradivarius, it would quickly meet with an accident, for insurance reasons. Not for personal gain. Such money could fund other music endeavours. I'd have thought he could have just sold it. A Stradivarius is an historical artifact, without regard to how good it sounds, and for that reason alone is worth the million+ dollars they often fetch.

Force needed to accelerate 2.2lbs of cookies = 1 Fig-newton to 1 meter per second