Actually, though he clearly exaggerates quite a bit, the latest Nvidia consumer chip (the GeoForce) does compare to the more expensive thousand dollar offerings. Nvidia is marketing a 64 meg version of the Geoforce for exactly those purposes, replacing those 'overpriced' graphics workstations. I think they may be targeting linux specifically too, which should be nice.
remember, VOLUME brings PRICE DOWN. With lots of 13 year olds buying graphics accelerators, the price drops, and with id and a few other forcing the chip makers to fully implement openGL, your going to get some stuff that compares to the big SGIs for much much less.
come on, its an arnold movie, they are automatically good, cause he has the neat accent and says "COME ON!"
Re:This commentary is misleading, what was said wa
on
Another Software Spy
·
· Score: 1
huh? the point is this.
If carmack told 3dfx that they are getting 10 times the support questions from 3dfx owners vs. nVidia owners, 3dfx may just assume that 10 times as many people are using 3dfx cards and then do nothing about their drivers.
On the other hand, if carmack has hard data showing that in fact there ARENT more 3dfx cards out there, then 3dfx has some clear data saying "you suck" and hence will be more likely to get off their butts and fix those drivers.
but perhaps I misinterpreted your message. I must admit it didnt really make sense to me.
there was something in a scientific american or some similar magazine about the intermediate representation idea. Someone has thought that out and actually developed such a representation, dont know if any implementations using it exist or not. definitely the right way to go.
I see it as different. More stable, less ram hungry, much harder to use, many missing features.
these things are improving, sure, but its taking a damn long time too.
remember not everyone is techie, and the goodness of software cant be evaluated by techies alone.
My guess is that if open-source translator is written better or not depends not so much on wether its open source, but on how talented the main contributors and/or designer(s) are.
This commentary is misleading, what was said was
on
Another Software Spy
·
· Score: 1
Carmack said this "It has mostly been for tracking the amount of support we give by video card vendor. For instance, 3dfx and nvidia are about equal in players, but we get 10x the support email for 3dfx users.
Mac has held steady at 5% of the players. Linux was about equal to the mac, but dropped off, probably due to players dual booting into windows. "
The commentary states that user emails can not be linked to the information packets being sent in, thus carmack is lying. But read what carmack says, it seems to me the purpose is to get data so they can go to 3dfx and say
"Look, 1 million people are playing with your video card, 1 million people are playing with nvidia's video card, but 10 times as many people are emailing us with support questions about YOUR card, so you ARE not doing well and should improve"
not that this makes the behavior of the program particularly friendly, I think they should tell you before they send data, but carmack was not lying, as it seems he never does.
you were either misinformed or you misinterpreted the information you heard.
AMD chips greatly outperform Intel chips in floating point calculations IN GENERAL. (ie math)
They also outperform it a bit in integer calculations IN GENERAL (ie more math)
Both intel and AMD now have SIMD instructions which AMD happens to market as 3DNOW (pretty pictures) but Intel has the same thing, and both can be use to do (yep thats right, math)
Now, that said, you can probably find real world instances of both Math, and pretty pictures, for which either processor will do a bit better clock for clock than the other. Take Quake3 (pretty pictures, which by the way, require a lot of math) The new coppermine intel system mysteriously do fabulous on that game, better than the athlon clock for clock, while every other game the athlon does better.
But, in general, the athlon is faster clock for clock.
is to write your next application in both Perl and C++, and see which one is actually faster. If your task includes a lot of string manipulation, then perl's built in code may indeed be more efficient than you can manage to accomplish with your own C++.
Then, if C++ is faster, and you *need* the code to run faster, go with it.
Re:can anyone explain the coppermine jump in gamin
on
Coppermine vs. Athlon
·
· Score: 1
yeah but the coppermine is supposed to be the same design. so the optimizations should work for the other P3s too...seems weird.. ah well
can anyone explain the coppermine jump in gaming?
on
Coppermine vs. Athlon
·
· Score: 1
The benchmarks were odd, for everything else, the coppermine chips acted more or less like the other P3 chips, then in gaming stuff all of a sudden they jump way up. why?
Slashdot Mirror
yeah
sweeny doesnt know anything about technology
word
=)
Actually, though he clearly exaggerates quite a bit, the latest Nvidia consumer chip (the GeoForce) does compare to the more expensive thousand dollar offerings. Nvidia is marketing a 64 meg version of the Geoforce for exactly those purposes, replacing those 'overpriced' graphics workstations. I think they may be targeting linux specifically too, which should be nice.
remember, VOLUME brings PRICE DOWN. With lots of 13 year olds buying graphics accelerators, the price drops, and with id and a few other forcing the chip makers to fully implement openGL, your going to get some stuff that compares to the big SGIs for much much less.
fun fun fun =)
this is also exactly the kind of thing which is an illegal monopolisitic practice. Intel will have to watch themselves or end up like Microsoft =)
Arnold throws satan out of a window
how cool is that?
come on, its an arnold movie, they are automatically good, cause he has the neat accent and says "COME ON!"
huh? the point is this.
If carmack told 3dfx that they are getting 10 times the support questions from 3dfx owners vs. nVidia owners, 3dfx may just assume that 10 times as many people are using 3dfx cards and then do nothing about their drivers.
On the other hand, if carmack has hard data showing that in fact there ARENT more 3dfx cards out there, then 3dfx has some clear data saying "you suck" and hence will be more likely to get off their butts and fix those drivers.
but perhaps I misinterpreted your message. I must admit it didnt really make sense to me.
there was something in a scientific american or some similar magazine about the intermediate representation idea. Someone has thought that out and actually developed such a representation, dont know if any implementations using it exist or not. definitely the right way to go.
umm hmmm
so linux is BETTER than windows?
I see it as different. More stable, less ram hungry, much harder to use, many missing features.
these things are improving, sure, but its taking a damn long time too.
remember not everyone is techie, and the goodness of software cant be evaluated by techies alone.
My guess is that if open-source translator is written better or not depends not so much on wether its open source, but on how talented the main contributors and/or designer(s) are.
Carmack said this
"It has mostly been for tracking the amount of support we give by video card vendor. For instance, 3dfx and nvidia are about equal in players, but we get 10x the support email for 3dfx users.
Mac has held steady at 5% of the players. Linux was about equal to the mac, but dropped off, probably due to players dual booting into windows.
"
The commentary states that user emails can not be linked to the information packets being sent in, thus carmack is lying. But read what carmack says, it seems to me the purpose is to get data so they can go to 3dfx and say
"Look, 1 million people are playing with your video card, 1 million people are playing with nvidia's video card, but 10 times as many people are emailing us with support questions about YOUR card, so you ARE not doing well and should improve"
not that this makes the behavior of the program particularly friendly, I think they should tell you before they send data, but carmack was not lying, as it seems he never does.
wdfds
its a good thing smart people like carmack actually TRY these things rather than just assuming its impossible :)
and no, you dont need neural nets necessarily. You dont even know what hes trying to do!!!
Imagine a setup where the web cam looks at your face, if you smile, the texture on your model smiles, something like that. That is certainly doable.
I apologize, I should have said Athlon along with AMD each time
:)
true dat yo.
you were either misinformed or you misinterpreted the information you heard.
AMD chips greatly outperform Intel chips in floating point calculations IN GENERAL. (ie math)
They also outperform it a bit in integer calculations IN GENERAL (ie more math)
Both intel and AMD now have SIMD instructions which AMD happens to market as 3DNOW (pretty pictures) but Intel has the same thing, and both can be use to do (yep thats right, math)
Now, that said, you can probably find real world instances of both Math, and pretty pictures, for which either processor will do a bit better clock for clock than the other. Take Quake3 (pretty pictures, which by the way, require a lot of math) The new coppermine intel system mysteriously do fabulous on that game, better than the athlon clock for clock, while every other game the athlon does better.
But, in general, the athlon is faster clock for clock.
can you say EVERQUEST, massively multiplayer adventure game
can you say Tribes!
can you say Thief!
can you say Half Life!
theres more than just quake3 these days
:)
is to write your next application in both Perl and C++, and see which one is actually faster. If your task includes a lot of string manipulation, then perl's built in code may indeed be more efficient than you can manage to accomplish with your own C++.
Then, if C++ is faster, and you *need* the code to run faster, go with it.
a negative score!
nothing better than killing nazis!
deal with the new defition.
yeah but the coppermine is supposed to be the same design. so the optimizations should work for the other P3s too...seems weird.. ah well
The benchmarks were odd, for everything else, the coppermine chips acted more or less like the other P3 chips, then in gaming stuff all of a sudden they jump way up. why?
NO IT DOESNT
maybe in a loop of adding integers, which is REAL uninteresting.
lets try, oh, floating point math or something thats actually HARD.
and its not 3 times as fast either. my guess is you misinterpreted something.
Even in photoshop, the apple speed messiahs most favorite program to quote, a hellah high MHZ pentium STILL RUNS IT FASTER.
the g4 is SO NOT FASTER
god when the hell are the apple people gonna get a grip on reality?
aarrrggg
yeah! :)
maybe if they port Quake3 to an alpha
I mean, do I really care if my computional analysis experiments, or neural nets finish 20% faster? nah, not really.
but if I get a another few FPS or some more polys in Quake3, its like YEEEEHA! yeah! kickASS!
you know?
hehehe
what if we computed FPS/dollar for AMD/Intel and Alpha? hmm? hehe
this is like when you write a recursive function and forget to check for an end state and return;
To buy a chip:
wait for the faster one
then a buy a chip
and of course, to buy a chip, you recur
wait for the faster one,
then buy a chip
and so on.
eventually you run out of stack space and are forced to go back to drooling in a padded corner with an etch-a-sketch
is a coppermine at 700Mhz > AMD at 700Mhz?
cuz if its the same ole P3, an AMD 650 is still gonna win methinks.
motherboard manufactures need to be shot.
:)
I would buy an Athlong for XMAS if there was a motherboard that fit my needs.
ARE YOU LISTENING MOBO COMPANIES!
ah, better to wait for 2Ghz chips anyway