It's also interesting that they wrote their own compiler, instead of patching GCC. GCC also works on Windows and 3.1 already has optimizations for different CPU extensions to x86.
Well, that's mostly sad. They cannot extend gcc, because gcc is under GPL -> they'd have to release the source -> many patented & licensed tricks would be exposed.
And to compile the kernel...who will do that? Is there some server vendor planning to sell Intel servers with such kernels on?
Kernel compiling is pretty much 'fire test'. If they can do that, they can do almost anything (save from mplayer;). Also, kernel accounts for pretty much system overhead, so speeding kernel up speeds system. especially linux kernel SW RAID, sweet.
The area Intel compiler will shine is all computing -intensive tasks. Like Maya 4, povray, most math libs and stuff. So there is place and _demand_ for it.
I wonder how this code will perform on Athlon/Hammer-32..
I'll tell you a thing.
Writing code is very much like writing a novel. One day it's flowing out of you. Then on another day, it just stops. And there's no way you can easily tap it back 'ON'.
It's not about 'formal methods'. With all your bag of formal method tricks, if you can't visualize (in some personal way) problem you're facing, then you can't apply anything to it. Nothing, nada.
This is why CAP will never really take off. You need to really understand what is going on, and what one should do. Computer just can't do it.
Some time ago I read an article on newsforge that it's really easy to switch from IE to Mozilla when using it as a html viewing engine -> the calls are all exactly same.
Now I'm thinking.. Is it possible to switch that in already build (binary) application? Like ditch that mshtml.dll or something like that?
You don't know what you're talking about don't you?
using OSS would be cheaper for MS. They're not paying their programmers of sheer goodness of their heart. They need them.
Now, if there's something like libpng, carefully designed and crafted library. It's somewhat nightmarish to go and implement it from scratch. But with libpng you can just make some simple calls and You've got all the functionality you wan't. In a matter of days instead of months. This applies to most bigger OSS projects. They're mostly libraries, things that can be reused till no end.
Your comparision with NC bomb doesn't exactly light a bulb at this end, how do you apply it?
Think about it like a family gathering (well.. 'dream' family anyways).
Everybody brings some food with them, and give to everyone wanting it.
What would you like about some cousin that'd just come and take without ever giving anything (or even saying if it had some hair in it)?
Since the GPL makes it near impossible for an entity to simply produce and sell software as its core business, it is unsurprising that the world's largest software company would be wary of doing
anything that encouraged the spread of this meme.
This is one of the goals too. A SW company must support it's code. If it's only producing it, then it's no good for anyone.
(We're talking about big software projects, not some dinky cout code).
resort to conspiracy theories
about how MSFT wants to steal their code
But this is exactly how it is. Maybe my code is not one MS would steal, but someone other's is. By stealing I mean merely using it without _contributing_ all the fixes/features back.
One of the main features (for me) of GPL is feeding all the code back to the pool.
So you're saying that you do this only for fame/employer karma?
I for one don't want to give up not only mine, but also community's rights to my code. You see, when I write something and release it, it's not only mine, it's also everyone's elses. Everybody scores.
Now, MS doesn't want to play _our_ game by _our_ rules. Instead all they want is to take, profit. One'd think that IBM (the mothership of patenting in US) would have more problems with GPL than MS, but no..
Using my code, and not giving back to community. No that's not my piece of cake that I want to share with MS.
I look more for what I Want.
Let's consider my first system. An IBM PS/1 (486sx/25,4Mb/170Mb,14"). I paid my ass off for that almost $2000 (1994 something..).
Next computer I had, I paid for it with my summerjob, it was about $1500, and it was almost top-of-the-line one (PPro180 when most you could get was PPro200).
I paid much, much less than any corporation for that one. At that time I already got some connections and I bought it in parts.
Most of that joy was to get it assembled;)
Don't know about you, but I ain't gonna get some dell-dimwit computer full of already b0rken parts (i740..). I'm gonna build mine from ground up. And I will not get any support for that.
I know only one single reason (for consumer, not prosumer) to get a 'branded' computer: Support. 1-3 years will be almost enough for everyone.
no, it's subdivided many, many times. mostly by factor of 2, but probably adaptated (like 800px -> 200px+600px).. it should stop subdividing at about 8.
Only the losers buy systems at DELL, Compaq, IBM, HP or whatever the brands are.
And BIG companies. Like.. umh.. Those that matter. And consumers, those who really pay manuf.'s bill.
Those who can build their own system are not a big market, they're more like marginal. Maybe a PR market, but really, really small.
Interesting stuff. isn't it 'better' to not think about rubberband at outer surface bat at 'outer rim'. At about below surface of apple/doughnut? Then one will see that in apple rubberband (even in 3D) is convexish (I mean infinitely thin rubberband), but in doughnut, there is no way to see some part of rubberband unless it's quantized. Same applies fo 'standard' universe and with one which has a 'pen'-hole which goes straight through rubberband (some odds for that..).
You're a little bit off there. A 100 foot roll of bulk load film is good for about 18 rolls, 36 exposures each. 36*18 = 648 exposures, not 800.
Well, yeh, probably.
I calculated it this way: 30.5*100*10mm/((36mm+2mm)/exposure) ~= 802 (exposures).
Remember that with single back there is no need to 'loose' exposures when reloading. (2-4 exposures/36exposure roll).
I believe AF is not importaint here as I would use a 16/20mm at hyperfocal.
About Canon, all is nice and swell, but I failed my google search for canon large-amount exposure back. D'you know if they have some or what is their solution for that?
Well.. If we had full-frame digiatl cameras, that would be.. Great.
For example some 16mm lens weould be pretty nice thing on digital. but as most digital (serious) SLR's have 1,6x magnification, you need something like 8mm fisheye (which is pretty damn expensive.. and not even full-frame.)
Well.. it's not Kodak that makes the, but maker of camera.
For example Nikon has 250 & 750 frame (check them out -- huge) 'backs'. You need to take back from your camera and change it.
Of course you need lots of film for that too. Pretty much standard is 100ft (30.5m) or 55ft (17m) rolls (with these you can fill standard 36 exposure canisters). That is enough for about 800 exposures.
LOL
I'd like to see some drive that really puts out 66.6 MBytes/sec at ATA/66.
With scsi that's altogether another question. It beats shit out of ATA/66.
There's no point comparing theoretical outputs.
I don't think that there can be something like 'local' ice-age. I think that if Yurop gets colder, then some other part at the other side of the planet has to (New Zealand/Oz).
I think that ice-age is always global. There doesn't seem to be much in scientific literature about what happened in non-Yurop while last ice-age lasted. I remember seeing in some BBC doc that there was wall of ice at about Washington height.
I know that Afica was a lot wetter (like forests in Sahara), and that means that other (south africa) side of africa has had to be cooler than it is now.
Well. Yes.
The problem is that US has something like 500:1 vehicles comared to thos poorer countries.
And US' steel industry (single most pollutive in terms of C2O) is badly b0rked. And Bush' goverment just went on to support it.
And in Finland our goverment has insane (like 150%) taxes on new vehicles, so most (75%) people drive in damn non-catalyzator cars, polluting just more.
Yeah, of course you're absolutely right. The problem seems to be that we should be entering another ice-age any day now;), but it just doesn't look like it.
According to tradition it's got to have a name. More X's, the better. Not three X's thou.
So my bet would be Hurdix. (No GNU, because there's no non-GNU hurd..)
Slashdot Mirror
Sorry? Do you mean they're writing a compiler that'll compile things real fast? Why's that? -- compiling is 'cheap'.
It's also interesting that they wrote their own compiler, instead of patching GCC. GCC also works on Windows and 3.1 already has optimizations for different CPU extensions to x86.
;). Also, kernel accounts for pretty much system overhead, so speeding kernel up speeds system. especially linux kernel SW RAID, sweet.
Well, that's mostly sad. They cannot extend gcc, because gcc is under GPL -> they'd have to release the source -> many patented & licensed tricks would be exposed.
And to compile the kernel...who will do that? Is there some server vendor planning to sell Intel servers with such kernels on?
Kernel compiling is pretty much 'fire test'. If they can do that, they can do almost anything (save from mplayer
The area Intel compiler will shine is all computing -intensive tasks. Like Maya 4, povray, most math libs and stuff. So there is place and _demand_ for it.
I wonder how this code will perform on Athlon/Hammer-32..
The idea here is that X has efficiently managed to maintain full backwards compatibility efficiently.
... by not really adding anything new in the last 10 years.
;-)
I see your smiley =)
xinerama, render, shared memory, xv, truetype. To name just a few.
I'll tell you a thing.
Writing code is very much like writing a novel. One day it's flowing out of you. Then on another day, it just stops. And there's no way you can easily tap it back 'ON'.
It's not about 'formal methods'. With all your bag of formal method tricks, if you can't visualize (in some personal way) problem you're facing, then you can't apply anything to it. Nothing, nada.
This is why CAP will never really take off. You need to really understand what is going on, and what one should do. Computer just can't do it.
Some time ago I read an article on newsforge that it's really easy to switch from IE to Mozilla when using it as a html viewing engine -> the calls are all exactly same.
Now I'm thinking.. Is it possible to switch that in already build (binary) application? Like ditch that mshtml.dll or something like that?
Hit 'Reload'. Pic references not allowed at TH.
You don't know what you're talking about don't you?
using OSS would be cheaper for MS. They're not paying their programmers of sheer goodness of their heart. They need them.
Now, if there's something like libpng, carefully designed and crafted library. It's somewhat nightmarish to go and implement it from scratch. But with libpng you can just make some simple calls and You've got all the functionality you wan't. In a matter of days instead of months.
This applies to most bigger OSS projects. They're mostly libraries, things that can be reused till no end.
Your comparision with NC bomb doesn't exactly light a bulb at this end, how do you apply it?
Think about it like a family gathering (well.. 'dream' family anyways). Everybody brings some food with them, and give to everyone wanting it. What would you like about some cousin that'd just come and take without ever giving anything (or even saying if it had some hair in it)?
Since the GPL makes it near impossible for an entity to simply produce and sell software as its core business, it is unsurprising that the world's largest software company would be wary of doing anything that encouraged the spread of this meme.
This is one of the goals too. A SW company must support it's code. If it's only producing it, then it's no good for anyone.
(We're talking about big software projects, not some dinky cout code).
resort to conspiracy theories about how MSFT wants to steal their code
But this is exactly how it is. Maybe my code is not one MS would steal, but someone other's is. By stealing I mean merely using it without _contributing_ all the fixes/features back.
One of the main features (for me) of GPL is feeding all the code back to the pool.
So you're saying that you do this only for fame/employer karma?
I for one don't want to give up not only mine, but also community's rights to my code. You see, when I write something and release it, it's not only mine, it's also everyone's elses. Everybody scores.
Now, MS doesn't want to play _our_ game by _our_ rules. Instead all they want is to take, profit.
One'd think that IBM (the mothership of patenting in US) would have more problems with GPL than MS, but no..
Using my code, and not giving back to community. No that's not my piece of cake that I want to share with MS.
heh.. well. But take a lok inside, there's still plenty of wires.
I look more for what I Want. ;)
Let's consider my first system. An IBM PS/1 (486sx/25,4Mb/170Mb,14"). I paid my ass off for that almost $2000 (1994 something..).
Next computer I had, I paid for it with my summerjob, it was about $1500, and it was almost top-of-the-line one (PPro180 when most you could get was PPro200).
I paid much, much less than any corporation for that one. At that time I already got some connections and I bought it in parts.
Most of that joy was to get it assembled
Don't know about you, but I ain't gonna get some dell-dimwit computer full of already b0rken parts (i740..). I'm gonna build mine from ground up. And I will not get any support for that.
I know only one single reason (for consumer, not prosumer) to get a 'branded' computer: Support. 1-3 years will be almost enough for everyone.
afaik .ogg uses wavelets, and gets better sound at lower bitrate (subjective..) than FFT -based mp3.
no, it's subdivided many, many times. mostly by factor of 2, but probably adaptated (like 800px -> 200px+600px).. it should stop subdividing at about 8.
Only the losers buy systems at DELL, Compaq, IBM, HP or whatever the brands are.
And BIG companies. Like.. umh.. Those that matter. And consumers, those who really pay manuf.'s bill.
Those who can build their own system are not a big market, they're more like marginal. Maybe a PR market, but really, really small.
Interesting stuff.
isn't it 'better' to not think about rubberband at outer surface bat at 'outer rim'. At about below surface of apple/doughnut?
Then one will see that in apple rubberband (even in 3D) is convexish (I mean infinitely thin rubberband), but in doughnut, there is no way to see some part of rubberband unless it's quantized.
Same applies fo 'standard' universe and with one which has a 'pen'-hole which goes straight through rubberband (some odds for that..).
You're a little bit off there. A 100 foot roll of bulk load film is good for about 18 rolls, 36 exposures each. 36*18 = 648 exposures, not 800.
Well, yeh, probably.
I calculated it this way: 30.5*100*10mm/((36mm+2mm)/exposure) ~= 802 (exposures).
Remember that with single back there is no need to 'loose' exposures when reloading. (2-4 exposures/36exposure roll).
I believe AF is not importaint here as I would use a 16/20mm at hyperfocal.
About Canon, all is nice and swell, but I failed my google search for canon large-amount exposure back. D'you know if they have some or what is their solution for that?
Well.. If we had full-frame digiatl cameras, that would be.. Great.
For example some 16mm lens weould be pretty nice thing on digital. but as most digital (serious) SLR's have 1,6x magnification, you need something like 8mm fisheye (which is pretty damn expensive.. and not even full-frame.)
Well.. it's not Kodak that makes the, but maker of camera.
For example Nikon has 250 & 750 frame (check them out -- huge) 'backs'. You need to take back from your camera and change it.
Of course you need lots of film for that too. Pretty much standard is 100ft (30.5m) or 55ft (17m) rolls (with these you can fill standard 36 exposure canisters). That is enough for about 800 exposures.
LOL
I'd like to see some drive that really puts out 66.6 MBytes/sec at ATA/66.
With scsi that's altogether another question. It beats shit out of ATA/66.
There's no point comparing theoretical outputs.
I don't think that there can be something like 'local' ice-age. I think that if Yurop gets colder, then some other part at the other side of the planet has to (New Zealand/Oz).
I think that ice-age is always global. There doesn't seem to be much in scientific literature about what happened in non-Yurop while last ice-age lasted. I remember seeing in some BBC doc that there was wall of ice at about Washington height.
I know that Afica was a lot wetter (like forests in Sahara), and that means that other (south africa) side of africa has had to be cooler than it is now.
Well. Yes.
The problem is that US has something like 500:1 vehicles comared to thos poorer countries.
And US' steel industry (single most pollutive in terms of C2O) is badly b0rked. And Bush' goverment just went on to support it.
And in Finland our goverment has insane (like 150%) taxes on new vehicles, so most (75%) people drive in damn non-catalyzator cars, polluting just more.
Yeah, of course you're absolutely right. The problem seems to be that we should be entering another ice-age any day now ;), but it just doesn't look like it.
What's the name?
According to tradition it's got to have a name. More X's, the better. Not three X's thou.
So my bet would be Hurdix. (No GNU, because there's no non-GNU hurd..)