A well informed linux user wouldn't, but a well informed windows user who has access to XP (probably cracked) would. Both the linux and windows cost the same amount, but you get a larger hard drive with the linux version to compensate for cost of windows....
I know a fair few people who've gone for the windows version (because linux one was unavailable), then replaced the hard drive (which they were going to do anyway), then installed linux.
I'm more of a 'highend' graphics coder (read: not games). Lets say we want to do some complex soft body animation. We need to be able to access a coherant data structure that represents the entire geometry mesh to be able to do that. You can't do that on the GPU - triangles and vertices are all you get.
Lets say we want to use the numerous deformation techniques that do not work by transforming normals via a matrix (i.e. they must re-compute the normals because the deformation is non linear - FFD's for example). You have only been able to do that on the GPU since the geforce 8800 came out (it requires geometry shaders) - but even then isn't perfect (can't obey soft/hadr edges, and a few other things I'd like to do).
Lets say you want to start doing Global Illumination rendering (of which Ray tracing is one technique - but not the only, or most useful one), you'll need to access a scene database - which you can't do on a GPU.
I've never really liked the design of the GPU's we currently have (though i realise why they've evolved in the way they have) - It all feels like nasty hack after nasty hack (which changes with every new Geforce card). I've always wished that instead of a GPU we just had an add in highly vectorised CPU which we can use for anything. There are literally hundreds of things i can see this being useful for.
.... and is also very restrictive in *how* you can access data, not to mention that all 960 cores have to run the same instructions. If you happen to have code that can be parallelised, but not in a way that would suit a GPU (i.e. seperate tasks, not 100's of the same task), then Larrabee starts to sound quite exciting.
One language that is being used in the sceintific community right now is CUDA - which runs on a GPU and is C based.
In addition, Fortran to C tools have been around for some years. To say that Fortran is the only scientific language is BS. R, S Plus, Octave, matlab, perl and CUDA to name a few. Taking R as an example - it provides an code interface that allows you to write optimised C/C++ routines and utilise those in the language itself.
being video recorded on every street we walk/drive
I'll give you this one; we have more CCTV. However if you think it's every street you are much mistaken.
That only applies to the UK.
Can you show me where people have been arrested for peaceful protests?
For the record, I agree primarily with what you say, but unfortunately I have seen people arrested at peacful protests in the UK. At a Labour party conference a few years ago, we went to protest about top up fee's for students. One student shouted "Cock-A-Doodle-Doo. Wake up Mr Blair, we're all very poor" and was promptly jumped on by about 10 coppers and arrested.
Though the incomes may just meet the expenditures, Europeans have much more free time because they do not take work as seriously as their American counterparts.
Rubbish. I've worked in the the US and the EU, and it's true americans work longer hours, but from experience i can tell you they are no more productive during that time. For the record, Europeans are very serious about their work.
(just as a fisherman who caught enough to eat his fill every day is typically happier than the workaholic investment banker who owns 3 houses but has no time to sleep).
Is that because the investment banker is seeing the amount of negative equity in his 3 homes increase daily with the slowdown in the housing market?
That was evident already. He posted a valid point, that 389 USD is not the same as 399 Euros - and localisation is not a valid reason for the (considerable) price difference in guitars.
You've repeated his point twice, have agreed with it twice, and yet you claim that he is wrong to point out there is a 380 usd price difference? To use your words. Wow.
I developed for SGI machines quite a few years ago, and the FPU performance of the Mips processors is where they really shone (They were after all used primarily for 3D work in an era pre-GPU). A 200Mhz mips chip would still outperform an 800Mhz Athalon (my PC at the time) when it came to raw FPU power. That of course required quite a lot of work to get that kind of performance. The big speed killer was always memory access, but if you got that right, they were astonishingly quick at FPU computation
The R16000's were awesome - i miss the old SGI Fuels:( The benchmarks on that page seem pretty low to me - The FPU performance of the 800/900 Mhz chips i used to use were closer to that of a 3Ghz P4....
It's an irritating piece of shit that runs visual studio quite well. Go use anything else for a month and you'll be glad your back on windows - It's all about the apps you need to run, and not the OS.
Hell, i've met a fair few people who think their OS is firefox......
I used to work in a uni, and we once caught a student running a porn server through an open printer port (on an SGI irix machine). The uni would have kicked him off the course for such an offense, which would have only served to give him a bad university record, and no real job prospects. Since the student was obviously bright, and had an awful lot going for him, we decided that wasn't the best course of action
Instead we chose to log the incident as 'the student has found an exploit in our network security and brought it to our attention', and gave the student a verbal warning over the incident (to hopefully steer him on the right path in future).
If this NZ kid had ended up with a criminal record, his chances of a university education and his future job prospects would have been severly affected. So here we have a choice. Do you
a) ruin this kids life, which would greatly increase the chances of him becomming a full time criminal (hacking can be more lucrative than flipping burgers after all)
b) attempt to guide this individual back onto the straight and narrow so that he his skills can actually benefit society?
Perforce doesn't differ too much from SVN, but it just works better.
So yeah, perforce doesn't give you much over SVN with a small/medium dev team, but you do start noticing some short-comings, and little quirks with SVN when you start working on larger projects. e.g. Branching and (especially) merging. I'd definately go with perforce for a large scale project - if only for the relatively pain free merging....
How many do you think will opt to run old-Windows on top of Linux or OS X instead of betting on Microsofts unproven new-Windows, especially considering their track record on their previous offerings?/greger
Most people don't know how to install an OS, so it's unlikely they'll be test driving any of them. I know average users who have Vista, and more or less all of them think it's pretty good - it's shinier than XP afterall.
And yes you can run XP in a VM, but since the average user doesn't know how to install an OS to begin with, I can't see too many installing an OS in VMware.
yeah, when we got hold of a PhysX accelerator, our engine actually slowed down quite a lot - It ran roughly half the speed of the software implementation. The big problem was that whilst the card was great at processing a few thousand colliding boxes, it's performance truly sucked if you wanted to do anything with those results. So yeah, I'd be inclined to agree that the value of the hardware is questionable.
There is a bigger problem though - why would you even want a few thousand boxes colliding? (Youtube videos of Crysis barrels aside...) It doesn't actually add a huge amount to the gameplay - The physX demo 'hanger of doom' is a pretty good example of this...
Though on occasions, an analytical solution is prefferred, especially when you need to handle accuracy. Karma (from the now defunt Maths engine), used a matrix solve internally, which was slightly slower than the iterative solvers found in Havok/PhysX, but it was fantastically stable as a result. That engine was parallelised for PS2, SSE and altivec IIRC.
Being able to choose whether to use an iterative or matrix solver was one of the nice things about that engine...
Slashdot Mirror
one word. nLite.
It's called Larrabee :p
easy then, windows ME..... /ducks
A well informed linux user wouldn't, but a well informed windows user who has access to XP (probably cracked) would. Both the linux and windows cost the same amount, but you get a larger hard drive with the linux version to compensate for cost of windows....
I know a fair few people who've gone for the windows version (because linux one was unavailable), then replaced the hard drive (which they were going to do anyway), then installed linux.
I'm more of a 'highend' graphics coder (read: not games). Lets say we want to do some complex soft body animation. We need to be able to access a coherant data structure that represents the entire geometry mesh to be able to do that. You can't do that on the GPU - triangles and vertices are all you get.
Lets say we want to use the numerous deformation techniques that do not work by transforming normals via a matrix (i.e. they must re-compute the normals because the deformation is non linear - FFD's for example). You have only been able to do that on the GPU since the geforce 8800 came out (it requires geometry shaders) - but even then isn't perfect (can't obey soft/hadr edges, and a few other things I'd like to do).
Lets say you want to start doing Global Illumination rendering (of which Ray tracing is one technique - but not the only, or most useful one), you'll need to access a scene database - which you can't do on a GPU.
I've never really liked the design of the GPU's we currently have (though i realise why they've evolved in the way they have) - It all feels like nasty hack after nasty hack (which changes with every new Geforce card). I've always wished that instead of a GPU we just had an add in highly vectorised CPU which we can use for anything. There are literally hundreds of things i can see this being useful for.
.... and is also very restrictive in *how* you can access data, not to mention that all 960 cores have to run the same instructions. If you happen to have code that can be parallelised, but not in a way that would suit a GPU (i.e. seperate tasks, not 100's of the same task), then Larrabee starts to sound quite exciting.
One language that is being used in the sceintific community right now is CUDA - which runs on a GPU and is C based.
In addition, Fortran to C tools have been around for some years. To say that Fortran is the only scientific language is BS. R, S Plus, Octave, matlab, perl and CUDA to name a few. Taking R as an example - it provides an code interface that allows you to write optimised C/C++ routines and utilise those in the language itself.
being video recorded on every street we walk/drive
I'll give you this one; we have more CCTV. However if you think it's every street you are much mistaken.
That only applies to the UK.
Can you show me where people have been arrested for peaceful protests?
For the record, I agree primarily with what you say, but unfortunately I have seen people arrested at peacful protests in the UK. At a Labour party conference a few years ago, we went to protest about top up fee's for students. One student shouted "Cock-A-Doodle-Doo. Wake up Mr Blair, we're all very poor" and was promptly jumped on by about 10 coppers and arrested.
Though the incomes may just meet the expenditures, Europeans have much more free time because they do not take work as seriously as their American counterparts.
Rubbish. I've worked in the the US and the EU, and it's true americans work longer hours, but from experience i can tell you they are no more productive during that time. For the record, Europeans are very serious about their work.
(just as a fisherman who caught enough to eat his fill every day is typically happier than the workaholic investment banker who owns 3 houses but has no time to sleep).
Is that because the investment banker is seeing the amount of negative equity in his 3 homes increase daily with the slowdown in the housing market?
That was evident already. He posted a valid point, that 389 USD is not the same as 399 Euros - and localisation is not a valid reason for the (considerable) price difference in guitars.
You've repeated his point twice, have agreed with it twice, and yet you claim that he is wrong to point out there is a 380 usd price difference? To use your words. Wow.
I developed for SGI machines quite a few years ago, and the FPU performance of the Mips processors is where they really shone (They were after all used primarily for 3D work in an era pre-GPU). A 200Mhz mips chip would still outperform an 800Mhz Athalon (my PC at the time) when it came to raw FPU power. That of course required quite a lot of work to get that kind of performance. The big speed killer was always memory access, but if you got that right, they were astonishingly quick at FPU computation
:( The benchmarks on that page seem pretty low to me - The FPU performance of the 800/900 Mhz chips i used to use were closer to that of a 3Ghz P4....
The R16000's were awesome - i miss the old SGI Fuels
Though there's still the current rising cost of fuel to consider before you do that....
If it can't be duct, it's f***ed...
It's an irritating piece of shit that runs visual studio quite well. Go use anything else for a month and you'll be glad your back on windows - It's all about the apps you need to run, and not the OS.
Hell, i've met a fair few people who think their OS is firefox......
at least take the effort to partition the original install and set D:/ to be "My Documents". It's then a relatively pain free process to re-install...
> Anyone who says Windows is easy to install has either used pre-made image CDs, has only done upgrades, or has never actually installed it.
It's called nLite. Windows *is* easy to install....
We have laws, and everyone breaks (some of) them.
Going 5Mph above a speed limit.
Drinking a beer when you're 17.
Smoking a spliff.
If everything was so black and white, most people would be in jail....
I used to work in a uni, and we once caught a student running a porn server through an open printer port (on an SGI irix machine). The uni would have kicked him off the course for such an offense, which would have only served to give him a bad university record, and no real job prospects. Since the student was obviously bright, and had an awful lot going for him, we decided that wasn't the best course of action
Instead we chose to log the incident as 'the student has found an exploit in our network security and brought it to our attention', and gave the student a verbal warning over the incident (to hopefully steer him on the right path in future).
If this NZ kid had ended up with a criminal record, his chances of a university education and his future job prospects would have been severly affected. So here we have a choice. Do you
a) ruin this kids life, which would greatly increase the chances of him becomming a full time criminal (hacking can be more lucrative than flipping burgers after all)
b) attempt to guide this individual back onto the straight and narrow so that he his skills can actually benefit society?
I know which one I'd choose....
That'll all depend on the speed of your ram....
Perforce doesn't differ too much from SVN, but it just works better.
So yeah, perforce doesn't give you much over SVN with a small/medium dev team, but you do start noticing some short-comings, and little quirks with SVN when you start working on larger projects. e.g. Branching and (especially) merging. I'd definately go with perforce for a large scale project - if only for the relatively pain free merging....
How many do you think will opt to run old-Windows on top of Linux or OS X instead of betting on Microsofts unproven new-Windows, especially considering their track record on their previous offerings? /greger
Most people don't know how to install an OS, so it's unlikely they'll be test driving any of them. I know average users who have Vista, and more or less all of them think it's pretty good - it's shinier than XP afterall.
And yes you can run XP in a VM, but since the average user doesn't know how to install an OS to begin with, I can't see too many installing an OS in VMware.
yeah, when we got hold of a PhysX accelerator, our engine actually slowed down quite a lot - It ran roughly half the speed of the software implementation. The big problem was that whilst the card was great at processing a few thousand colliding boxes, it's performance truly sucked if you wanted to do anything with those results. So yeah, I'd be inclined to agree that the value of the hardware is questionable.
There is a bigger problem though - why would you even want a few thousand boxes colliding? (Youtube videos of Crysis barrels aside...) It doesn't actually add a huge amount to the gameplay - The physX demo 'hanger of doom' is a pretty good example of this...
Though on occasions, an analytical solution is prefferred, especially when you need to handle accuracy. Karma (from the now defunt Maths engine), used a matrix solve internally, which was slightly slower than the iterative solvers found in Havok/PhysX, but it was fantastically stable as a result. That engine was parallelised for PS2, SSE and altivec IIRC.
Being able to choose whether to use an iterative or matrix solver was one of the nice things about that engine...
Newtonian/Hamiltonian/Lagrangian are all more or less interchangable, however Lagrangian gives the most numericaly stable results....