Keeping in mind that all this discussion is in the context of that high-performance language, JavaScript:-)
Actually, thanks to JS's popularity, it gets quite nice performance from JIT compilers today.
It has enough speed to write, f.e., game engines in browser JS - and you don't need precision there. Just as you don't need precision, but do need speed in other common javascript FP use cases, such as builidng/manipulating graphics.
Well, this status quo is how the world MUST work if world wants to get any decent performance from existing CPUs, you know.
There are rationals and decimal floats - you can find libraries for any language and some languages include them in the core - but they are not commonly needed, as noone wants to trade speed for (unnecessary in most cases) precision.
It's not cross platform -- The amount of conditional cruft you have to add to ATTEMPT a cross-browser solution is rediculous
That's not the problem of javascript - that's the problem of implementation. Do you really think it would be any better if someone invented SomeBetterScript back then - and then MS made EvenBettererScript, which would be almost, but not completely, unlike the SomeBetterScript, and then Mozilla added their own extensions, and then other browsers implemented those extensions in incompatible way, adding some of their own in process, and...
You see what I'm talking about? JS by itself is quite nice language - web client bindings for JS is awful thanks to all the implementors.
And yes, libraries and frameworks are good thing and they do make browser JS crossplatform - think about how AWT/Swing/SWT makes Java crossplatform and what would happen if you had to have your own bindings and workarounds.
It's not sand-boxed -- Modern browsers compile JS to machine code and run that.
And that's pure bullshit. "Compile to machine code and run that" has nothing to do with sandboxing - that's what all the languages that give a bit of concern about performance do, after all. Please come back when you learn the difference between "sandboxing vs non-sandboxing" and "interpreting vs JIT compiling".
either you're just trolling or you don't have an idea what are you talking about.
"doesn't have types" can be said about half of modern languages - you mention ruby and perl, f.e.
"+" doesn't append _two numbers_, but it can append _number to string_ - which you can have in any language with operator overloading.
"looks object oriented, but has no real notion of classes" and "no inheritance" comes from your ignorance, you should read up on prototype-based OO, really.
P.S.: Insightful? Mmmm, yeah, that gave me an insight in the mind of ignorant haters a bit.
Basically, it goes like this - CPU load is approximately constant for same scene in various resolutions, as it just gives GPU initial geometry to work with and sets up shading states, then come vertex shaders (and optionally geometry shaders), which should be constant as well, and then come pixel shading and rasterization - both of which scale linearly with screen resolution in Mpix.
So, raising the resolution for _same_ scene gives a linear raise to GPU load.
But it's more complex than that, as stretching the scene to 3 monitors wouldn't make sense - cpu and vertex shading costs rise as well, because you're going from, say, 90 degrees horizontal field of vision to 270.
In other words, yes, obviously, you want 3 monitors - you need 3 times the power required for running it on single monitor.
You say that like there have to be only one side is to blame in this.
Hacker should be sued by Sony, as he illegally accessed Sony's computes, and Sony should be sued by customers, as they entrusted Sony with their data which was given away by negligency.
You see, it depends on assumption that it was geohot's and rebug's work that led to the breach - which I have trouble believing is true, as it would imply colossal stupidity on Sony's part.
If it was true, it would be like bank defending it's security by beating people who tried to use the WC, because there is unlocked safe vault door on the way there.
Sony was trying to stop people from doing that to prevent EXACTLY what happened.
Wait. So, you say they knew EXACTLY their security is so bad, some guy could hack them with just a bit of modding of the console.
And they tried to prevent EXACTLY that by going after that one guy instead of, you know, securing the data so it couldn't be stolen by just anyone with google and soldering iron?
It's either: a) Sony is really, _really_ stupid and you could really fondle their privates^H databases just by logging into dev network, or b) Hacking into dev network and hacking into their customer database was 2 separate things just clumped together by media.
Hard to guess, as there's no official information on attack vector anywhere.
How would a keyboard and mouse/trackball work for, say, Street Fighter series or Smash Bros. series? And given that the input APIs on Windows combine all connected keyboards into one virtual device and all connected mice and trackballs into one virtual device, what do players 2 through 4 use?
Gamepads?.. Like, you know, any of those in $5 to $100 range that you can buy in any store.
It's not that mice/keyboards are limited to PC and gamepads to consoles, it's that you can grab any USB gamepad and stick it into your PC with reasonable chances it will work everywhere, and you couldn't - as far as i can remember, correct me if i'm wrong - just grab any cheap mouse/keyboard for your console.
And it seems you didn't take into account that a geek doesn't have much influence outside the circle of geeks like himself. So even if we count occasional parent/sibling/high-school friend/a nice guy/girl from sales dept. number won't go all that high.
Anyways, cell phones, even smart- and smartish-phones, drifted from 'high-tech' to 'consumer electronics' now. If someone would go to his geeky friend for advice when buying one of those new cell phones 8 or 10 years ago, today he wouldn't, just like he wouldn't ask an advice when buying a fridge or an oven. He'd just go to the shop and ask for a phone with prettier buttons, icons and wallpapers, and iphone has lots of those.
Slashdot Mirror
You obviously don't know what are you talking about.
Keeping in mind that all this discussion is in the context of that high-performance language, JavaScript :-)
Actually, thanks to JS's popularity, it gets quite nice performance from JIT compilers today.
It has enough speed to write, f.e., game engines in browser JS - and you don't need precision there. Just as you don't need precision, but do need speed in other common javascript FP use cases, such as builidng/manipulating graphics.
Yes, that's one of JS design flaws.
They addressed it only in ES5 strict mode, so you can catch this kind of errors, at least.
Well, this status quo is how the world MUST work if world wants to get any decent performance from existing CPUs, you know.
There are rationals and decimal floats - you can find libraries for any language and some languages include them in the core - but they are not commonly needed, as noone wants to trade speed for (unnecessary in most cases) precision.
Well, there's this article, but it's quite outdated and benchmarking technique is not clear.
Hmm, it would be fun to compare browser JS engines' performance growth vs libraries bloat...
It's not cross platform -- The amount of conditional cruft you have to add to ATTEMPT a cross-browser solution is rediculous
That's not the problem of javascript - that's the problem of implementation. Do you really think it would be any better if someone invented SomeBetterScript back then - and then MS made EvenBettererScript, which would be almost, but not completely, unlike the SomeBetterScript, and then Mozilla added their own extensions, and then other browsers implemented those extensions in incompatible way, adding some of their own in process, and ...
You see what I'm talking about? JS by itself is quite nice language - web client bindings for JS is awful thanks to all the implementors.
And yes, libraries and frameworks are good thing and they do make browser JS crossplatform - think about how AWT/Swing/SWT makes Java crossplatform and what would happen if you had to have your own bindings and workarounds.
It's not sand-boxed -- Modern browsers compile JS to machine code and run that.
And that's pure bullshit. "Compile to machine code and run that" has nothing to do with sandboxing - that's what all the languages that give a bit of concern about performance do, after all. Please come back when you learn the difference between "sandboxing vs non-sandboxing" and "interpreting vs JIT compiling".
either you're just trolling or you don't have an idea what are you talking about.
"doesn't have types" can be said about half of modern languages - you mention ruby and perl, f.e.
"+" doesn't append _two numbers_, but it can append _number to string_ - which you can have in any language with operator overloading.
"looks object oriented, but has no real notion of classes" and "no inheritance" comes from your ignorance, you should read up on prototype-based OO, really.
P.S.: Insightful? Mmmm, yeah, that gave me an insight in the mind of ignorant haters a bit.
Then sue them for "tune snatching", not for "enabling infringers".
No need to look at various games.
Basically, it goes like this - CPU load is approximately constant for same scene in various resolutions, as it just gives GPU initial geometry to work with and sets up shading states, then come vertex shaders (and optionally geometry shaders), which should be constant as well, and then come pixel shading and rasterization - both of which scale linearly with screen resolution in Mpix.
So, raising the resolution for _same_ scene gives a linear raise to GPU load.
But it's more complex than that, as stretching the scene to 3 monitors wouldn't make sense - cpu and vertex shading costs rise as well, because you're going from, say, 90 degrees horizontal field of vision to 270.
In other words, yes, obviously, you want 3 monitors - you need 3 times the power required for running it on single monitor.
You say that like there have to be only one side is to blame in this.
Hacker should be sued by Sony, as he illegally accessed Sony's computes, and Sony should be sued by customers, as they entrusted Sony with their data which was given away by negligency.
No. Just no.
You see, it depends on assumption that it was geohot's and rebug's work that led to the breach - which I have trouble believing is true, as it would imply colossal stupidity on Sony's part.
If it was true, it would be like bank defending it's security by beating people who tried to use the WC, because there is unlocked safe vault door on the way there.
Sony was trying to stop people from doing that to prevent EXACTLY what happened.
Wait.
So, you say they knew EXACTLY their security is so bad, some guy could hack them with just a bit of modding of the console.
And they tried to prevent EXACTLY that by going after that one guy instead of, you know, securing the data so it couldn't be stolen by just anyone with google and soldering iron?
It's either:
a) Sony is really, _really_ stupid and you could really fondle their privates^H databases just by logging into dev network, or
b) Hacking into dev network and hacking into their customer database was 2 separate things just clumped together by media.
Hard to guess, as there's no official information on attack vector anywhere.
How would a keyboard and mouse/trackball work for, say, Street Fighter series or Smash Bros. series? And given that the input APIs on Windows combine all connected keyboards into one virtual device and all connected mice and trackballs into one virtual device, what do players 2 through 4 use?
Gamepads?.. Like, you know, any of those in $5 to $100 range that you can buy in any store.
It's not that mice/keyboards are limited to PC and gamepads to consoles, it's that you can grab any USB gamepad and stick it into your PC with reasonable chances it will work everywhere, and you couldn't - as far as i can remember, correct me if i'm wrong - just grab any cheap mouse/keyboard for your console.
And it seems you didn't take into account that a geek doesn't have much influence outside the circle of geeks like himself. So even if we count occasional parent/sibling/high-school friend/a nice guy/girl from sales dept. number won't go all that high.
Anyways, cell phones, even smart- and smartish-phones, drifted from 'high-tech' to 'consumer electronics' now. If someone would go to his geeky friend for advice when buying one of those new cell phones 8 or 10 years ago, today he wouldn't, just like he wouldn't ask an advice when buying a fridge or an oven. He'd just go to the shop and ask for a phone with prettier buttons, icons and wallpapers, and iphone has lots of those.