do0d... that's the coolest thing... I can't wait until we can grab some close flybys of other such thing (like planets) with our own digital cameras from the window of our interstellar vehicle:-)
I wish I could moderate today... cause I'd give you a +1, funny on that one... I wasn't gonna say anything, but that comment about the success of Linux and obscure documentation just left me rolling!... but maybe that's because its almost 2 in the morning and i've been trying to figure out how to hack X to get it to do what I want...
i do think that they won't mind so much... seeing as how a wider user base means more software developers will pay sony money to make games for their system... just make sense to me...
Actually, when I was in High School (Western Guilford/Weaver Center) our rockin AP Comp Sci Teacher had groups of people learn something not explicitly covered and teach it to the class.
A friend and I decided that I would be rockin if we learned how to use the Windows API and create graphics. We created a 3D graphics program that manipulated convex polygons. We did translation, rotatation, scaling, zbuffereing, back face culling, lambert shading (flat shading), and wireframe or solid modes. And we were working on gouraud shading but never finished it. Anyway, when we taught the class Windows API and graphics (even though we tried to teach them the API, we were kind enough to provide them with a shell program within which they could insert their code). We had them all write a connect 4 program using WinAPI and graphics. Very fun. Very cool
How many times do we have to have this conversation on/.? Parallelism in hardware won't do any good until parallell code that works well can be written.
Alrighty, then you and your one process can have at it. But I like to listen to some MP3's while I'm playing with the GIMP, and I don't like waiting for filters. Are you such a dumbass as to not realize that different processes (or threads) can be executed in parallel on parallel processors?
Your assertion that everything should be done at once is just plain stupid. If everything could just be that easily made to run all at once, it would have, a long time ago.
Yep, it's dumb to be efficient. I'm stupid for trying to say that things should be written to take advantage of parallelism... HELLO!!!! Wake up, it's not easy to write parallel code for a single process, but it's not hard to write a single application using multiple threads. Obviously you know I didn't mean run all the instructions in one cycle and put together the answers, cause no one is that idiotic. The only thing I meant was to do as much at once as is possible. And not as is easy either. You are right in that if it was easy to write parallel algorithms it would have been done a long time ago. But it is very difficult (one of the hardest CSC topics out there), and it has been done before.
Instead of wasting your time learning all there is to know about the history of running ass loads of no ops through your processor, perhapse you should educate yourself about how processors work. That'd be much more useful and you'd sound less clueless, and much less arogant. And apparently we will have to have these conversations on slashdot until idiots like you get some kind of clue about good programming practice and what that piece of porn downloading equiptment you keep in your bathroom called a computer actually does.
So get back to me when you can tell me all about the dining philosophers problem and how to use semaphores for critical section protection in a parallel algorithm. Cause then maybe you will have a clue as to what parallelism actually is and actually can do. Or maybe you'll give up and realize that stupid hacks like you just can't understand how to write good code.
Or maybe you can come back and tell me what good Trolling you've had today to cover up your idiocy... Either way I'll be happy, cause either way you are the crack whore.
Well, the "parallelism" of the P6 architechture is sorta bound... It's not really explicitly parallel, it's good ole superscalar processing. You are actually executing a handful of instructions in parallel, but you aren't gonna get less-than-one IPC (Instruction per cycle) out of this... here is the reason: you still have to go through the motions. here's a brief explaination of how superscalar crap works so maybe you people can see what I mean.
The processing is broken up in to descrete stages and when an instruction moves from one stage to the next, another instruction enters the stage that was vacated. The instructions are moving through a pipe, and all the parts of the processor are being utalized at once (ideally). This is a rough generalization for a generic processor with steps divided into IF(instruction fetch), ID(instruction decode), EX(execute), MM(memory), and WB(writeback). An instruction is only complete when it exits all of these stages and at it's fastest the instructions move through one stage for every CPU cycle. so if we have 5 instructions entering a superscalar processor with 5 stages, all noops, we end up with something like this (where each division takes a cycle to complete):
IF | ID | EX | MM | WB | -----IF | ID | EX | MM | WB | ----------IF | ID | EX | MM | WB | ---------------IF | ID | EX | MM | WB | --------------------IF | ID | EX | MM | WB |
Each instruction enters at IF and leaves after WB. since we are talking superscalar, it's looked at like this: each no op takes 5 cycles to complete, but since we are pipelining, each instruction effectively only takes one cycle after the inital 4 cycles to start the process. No instruction takes one cycle to go from fetch to finish. It has become standard practice though to post execution cycles in a pentium class processor (and for the most part superscalar in general) as 1 + (stalls needed). Stalls happen when you need to go to memory or when you have a string of data dependancies or are doing an operation that just takes a freaking long time in the EX stage. And sometimes they just go by how long it takes in the EX stage, ignoring any memory access problems.
And so, yes we have parallelism in a P6 on a superscalar level (and as far a branch predicting goes, but that is irrelevant for this case), but it will not have the effect of acheiveing less than 1 IPC... It doesn't matter how fast you can issue instructions, only completion matters for this case. And that's part of why bogomips are useless pieces of benchmark.
And that's also why a 450MHz P6 can only acheive a theoretical max of 450 million instructions per second.
well actually, bogomips and mips and all that don't really mean jack and a half. If you look at ratings like that, they come from frequency. You're cpu is a 450MHz, it can theoretically execute 450 million instructions per second. Yeah, 450 million no ops. But what good does that do you? to tell you the truth, most of the performance of any x86 processor is lost in branch prediction and memory latency. As far as I can tell, bogomips = bogus mips anyway... I don't know why 2.3.99 would be reporting 2xMHz for your bogomips unless they have something screwed up in the kernel they need to fix. Ain't no way a standard superscalar 450MHz processor is gonna execute one instruction every 0.5 cycles. Not an x86 anyway...
And even if you benchmark an actual program to try and see how many instructions per second it's actually getting, it only means anything for that program. It's totally dependant on how many branches are in the code, and how much of the time memory is being accessed.
These people's idea is excellent because it focuses on the direction the computer industry needs to go: parallelism. Forget doing everything sequentially, do it all at the same time! I'm not talking about in one program though, I'm talking about throughout the system. You've got your websever and you fileserver and all your device drivers and your os and all that good crap to run while you play quake, and the only way to really help out is to tack on another processor to run other programs at the same time. This is kind of like what SUN is doing for the MAJC architechture with thread level parallelism.
And finally, I don't see how you people can honestly think that one processor running at something like 450 is going to be as good as 8 all on the same die all running different processes in parallel.
Speed isn't about megahertz or gigahertz or instructions per second. It's about the time it take to run something. Who needs benchmarks when I have my analog wall clock to tell me what's best.
So you've got MPEG1 layer 3 for music. MPEG1 being the first iteration of the MPEG attempts at digitally encoding and compressing movies... But why can't we use MPEG4 layer 3? If the compression for an MPEG4 movie is several times better than the compression for an MPEG1 movie, than wouldn't the same apply for the layer 3 audio? Or do I just totally misunderstand?
Come ON! Where's El Hazard!!!
on
Essential Anime
·
· Score: 1
I mean, yeah, Akira, and Macross, and Robotech and all that. Great anime, but your list just isn't complete without El Hazard. I don't get the name though, cause I didn't see anything else spanish-like goin on... And what about The Record of Lodoss War?
Coming out in the hopefully near future will be the anime version of Chrono Trigger, and that should be truely awesome, if they do it right and keep true to the snes game.
Anyway, I also really liked Revolutionary Girl Utenea, but I have only been able to find the first volume, and I don't know what the heck happens!!! Anyone know anything?
lovely insight! beautiful... And let me throw in my 2 cents...
I think that MicroCrap should be split up into 3 (count them THREE) companies:
1. Operating Systems 2. Applications 3. Window Managers ("shells" as MS likes to call them)
Afterall, it's obvious that Microsoft thinks that a bulky bloated ugly nasty memory hog of a window manager constitutes a solid OS... Well, they would actually have a solid OS (in my opinion) if they took the Win2k kernel, opened up the API and some of the more important source, and created a framework on which to build a window manager. That is, provided you want or need one... Hell, I think they should open up their entire kernel source, but that'll never happen, because then someone ELSE could do what I have just proposed and make a better product than microsoft using their own software... But MS are experienced in doing that to other people (i.e. xerox / apple / ibm / whoevertheystoleqdosfrom).
actually, it's really not EPROM "burning"; it's programming... Generally though, at least in my experience, an EPROM or EEPROM or whatever will come with it's own proprietary programmer, and it is this programmer with which the computer must interface. There isn't a really good way to write a program to tell the programmer to program PROMs because of the fact that programmers have a generally unique interface (wow... that was almost as confusing to write as it will be to read). It is actually possible to program the things manually, and it really doesn't take a sophisticated program to preform this operation. Just build your own programmer to which you know the interface, and write a program to interface with the serial port of you puter. Personally though, I think that something like this is pointless, because anyone that really needs to program a PROM will most likely know how to do it themselves, and you can't expect something universal to come out and be supported under linux when hardware (the physical programmer you would need to build) is involved.
As far as the embedded market goes, EEPROMs again have a specific proprietary interface such as the one in your computer that holds your BIOS information. programs would need to be written for each specific proprietary implementation. And this isn't supported any better or worse under DOS/Win. Infact, it's probably easier to write the needed programs under linux because of it's openness and extensibility. Which would make me think that Linux should be a better choice for the embedded market anyway: not because it has an extensive library of programs which support intricate things such as programming strange and wonderful hardware, but because it is much easier to write this kind of software on an open unixlike system than under DOS or Win...
But all that is just my $0.70710678118654752440084436210485... So take it as you will...
When/. posts a bogus story like this one... uPs aren't gonna produce any harmful "radiation" such as this story suggests. Sure, every electrical device emmits a multitude of interferance, but to claim that when processors start getting above 1GHz I can cook my TV dinner in my case is rediculous. Power has a great deal to do with such things, and so does the appication. I mean, come on... Let's be serious, a 2.4GHz coordless phone meets FCC specs, and it's *transmitting* it's 2.4GHz signal... RF stuff can't hurt you, and it's moving through all the space around you. Hell, the military use spread spectrum encoded wireless transmissions whose carrier frequencies are 10GHz and UP! And if RF doesn't hurt you, ain't no clocked digital logic circuit gonna byte you. Especially at those exceedingly low frequencies of under 100GHz. It's all about power guys... Do some reaserch before you start spewing your crap all over me. It's all about the power...
I think it doesn't matter whether or not Beozozozzz is going to *do* the right thing, but that he has hit on the point of what is the right thing to fix.
Jeff has given us the opportunity to fix what's broken at the point at which it is broken. If he and Tim can get some actual reform happening at the source of the problem, the future will be much brighter.
I do have a suggestion though. I think that the patent laws for cyberspace stuff should be extraordinarilly limited. This is what I would propose for INet patents:
1) patents last a short 2 or 3 years.
2) patents are to be used defensively only.
An important extension to my second point is that, if patents can only be used defensively, they can only be used against the patent holders competitors. In other words, Amazon's patent could only be used against someone like B&N, other online booksellers. But all other types of industries and sales and markets and services should be allowed to use the technology. So if I'm an online retail computer store, and I want to one-click my way to heaven, I should be allowed to do so, because I'm not competing with Amazon. Restrict the patent to the competition on a business, and this leads into my third point for what an internet patent should be:
3) Publish the methodology and completely open it to the public, and to other non-competing businesses.
That way, we are all happy. Companies have incentive to innovate and patent, and other areas of tech and econ and internet benefit from the openness.
yep, I think it's made up. Why? Cause I design caches. If you push too hard, things don't work. That doesn't make the tech bad, it makes your company stupid. If they have these problems, they are really trying to do something very very wrong. (like save too much money)...
Complete bullshit. I don't think I could have loaded one up any fuller than this. Man, It must have taken a while to come up with all that crap. What did you eat? Let's just say that, if you really work for AMD, I hope you keep the toilets clean, cause only a janitorial worker could pull up all that B.S.
OK, to set the record straight, the L2 cache is a proven mechanism for increasing the average access time for data in a computer system. On die L2 caches can be as simple as you like. The performance gain comes in the high bandwidth availability and the distance from the processor core. The only real problem with on die L2 caches is that they increase the size of the die, and the cost of the chip.
If I were to believe that AMD had such problems, then it would NOT be in the design of the L2 cache, but rather the implementation. And I don't think you could implement such a simple device in such a poor manner. I could build an L2 cache with tinker toys that worked better than something that had these kinds of problems (problems which, by the way, CANNOT physically exist in the way they are described here).
Ok, so like let me overclock my 1GHz to what? 1.1?that's not really neat... the percentage gained in performance by overclocking goes down as the speed increases, because the relationships between die size, speed, and heat are exponential not linear... The faster things get, the harder it will be to overclock, at least without cooling... I see a nice future for kryotech...
Re:Freshmeat whiners, let me see if I get this
on
Linux 2.3.48 Released
·
· Score: 1
You said it! That's exactly how I feel. It seems like if we followed these guys advice, everything would be in a slashbox and there would be no more stories but for Katz posts... But did they know that they can filter out storys that they don't want to see? I don't know how that's news to anyone here, but all you gotta do is look at the preferences page and there you go! People should lighten up around here...
Slashdot Mirror
do0d... that's the coolest thing... I can't wait until we can grab some close flybys of other such thing (like planets) with our own digital cameras from the window of our interstellar vehicle :-)
JDW
I wish I could moderate today... cause I'd give you a +1, funny on that one... I wasn't gonna say anything, but that comment about the success of Linux and obscure documentation just left me rolling! ... but maybe that's because its almost 2 in the morning and i've been trying to figure out how to hack X to get it to do what I want...
JDW
i do think that they won't mind so much... seeing as how a wider user base means more software developers will pay sony money to make games for their system... just make sense to me...
JDW
cool... where do i sign up
would have a supplementary power source to amplify the light... you can't get somethin fer nuthin now adays... Oh well...
JDWilso2
Actually, when I was in High School (Western Guilford/Weaver Center) our rockin AP Comp Sci Teacher had groups of people learn something not explicitly covered and teach it to the class.
A friend and I decided that I would be rockin if we learned how to use the Windows API and create graphics. We created a 3D graphics program that manipulated convex polygons. We did translation, rotatation, scaling, zbuffereing, back face culling, lambert shading (flat shading), and wireframe or solid modes. And we were working on gouraud shading but never finished it. Anyway, when we taught the class Windows API and graphics (even though we tried to teach them the API, we were kind enough to provide them with a shell program within which they could insert their code). We had them all write a connect 4 program using WinAPI and graphics. Very fun. Very cool
so you can tell all your friends to type: "h,t,t,p,colon,slash,slash,slashdot,dot,dot" Man!!! that'd rock man!
How many times do we have to have this conversation on /.? Parallelism in hardware won't do any good until parallell code that works well can be written.
Alrighty, then you and your one process can have at it. But I like to listen to some MP3's while I'm playing with the GIMP, and I don't like waiting for filters. Are you such a dumbass as to not realize that different processes (or threads) can be executed in parallel on parallel processors?
Your assertion that everything should be done at once is just plain stupid. If everything could just be that easily made to run all at once, it would have, a long time ago.
Yep, it's dumb to be efficient. I'm stupid for trying to say that things should be written to take advantage of parallelism... HELLO!!!! Wake up, it's not easy to write parallel code for a single process, but it's not hard to write a single application using multiple threads. Obviously you know I didn't mean run all the instructions in one cycle and put together the answers, cause no one is that idiotic. The only thing I meant was to do as much at once as is possible. And not as is easy either. You are right in that if it was easy to write parallel algorithms it would have been done a long time ago. But it is very difficult (one of the hardest CSC topics out there), and it has been done before.
Instead of wasting your time learning all there is to know about the history of running ass loads of no ops through your processor, perhapse you should educate yourself about how processors work. That'd be much more useful and you'd sound less clueless, and much less arogant. And apparently we will have to have these conversations on slashdot until idiots like you get some kind of clue about good programming practice and what that piece of porn downloading equiptment you keep in your bathroom called a computer actually does.
So get back to me when you can tell me all about the dining philosophers problem and how to use semaphores for critical section protection in a parallel algorithm. Cause then maybe you will have a clue as to what parallelism actually is and actually can do. Or maybe you'll give up and realize that stupid hacks like you just can't understand how to write good code.
Or maybe you can come back and tell me what good Trolling you've had today to cover up your idiocy... Either way I'll be happy, cause either way you are the crack whore.
Well, the "parallelism" of the P6 architechture is sorta bound... It's not really explicitly parallel, it's good ole superscalar processing. You are actually executing a handful of instructions in parallel, but you aren't gonna get less-than-one IPC (Instruction per cycle) out of this... here is the reason: you still have to go through the motions. here's a brief explaination of how superscalar crap works so maybe you people can see what I mean.
... It doesn't matter how fast you can issue instructions, only completion matters for this case. And that's part of why bogomips are useless pieces of benchmark.
The processing is broken up in to descrete stages and when an instruction moves from one stage to the next, another instruction enters the stage that was vacated. The instructions are moving through a pipe, and all the parts of the processor are being utalized at once (ideally). This is a rough generalization for a generic processor with steps divided into IF(instruction fetch), ID(instruction decode), EX(execute), MM(memory), and WB(writeback). An instruction is only complete when it exits all of these stages and at it's fastest the instructions move through one stage for every CPU cycle. so if we have 5 instructions entering a superscalar processor with 5 stages, all noops, we end up with something like this (where each division takes a cycle to complete):
IF | ID | EX | MM | WB |
-----IF | ID | EX | MM | WB |
----------IF | ID | EX | MM | WB |
---------------IF | ID | EX | MM | WB |
--------------------IF | ID | EX | MM | WB |
Each instruction enters at IF and leaves after WB. since we are talking superscalar, it's looked at like this: each no op takes 5 cycles to complete, but since we are pipelining, each instruction effectively only takes one cycle after the inital 4 cycles to start the process. No instruction takes one cycle to go from fetch to finish. It has become standard practice though to post execution cycles in a pentium class processor (and for the most part superscalar in general) as 1 + (stalls needed). Stalls happen when you need to go to memory or when you have a string of data dependancies or are doing an operation that just takes a freaking long time in the EX stage. And sometimes they just go by how long it takes in the EX stage, ignoring any memory access problems.
And so, yes we have parallelism in a P6 on a superscalar level (and as far a branch predicting goes, but that is irrelevant for this case), but it will not have the effect of acheiveing less than 1 IPC
And that's also why a 450MHz P6 can only acheive a theoretical max of 450 million instructions per second.
well actually, bogomips and mips and all that don't really mean jack and a half. If you look at ratings like that, they come from frequency. You're cpu is a 450MHz, it can theoretically execute 450 million instructions per second. Yeah, 450 million no ops. But what good does that do you? to tell you the truth, most of the performance of any x86 processor is lost in branch prediction and memory latency. As far as I can tell, bogomips = bogus mips anyway... I don't know why 2.3.99 would be reporting 2xMHz for your bogomips unless they have something screwed up in the kernel they need to fix. Ain't no way a standard superscalar 450MHz processor is gonna execute one instruction every 0.5 cycles. Not an x86 anyway...
And even if you benchmark an actual program to try and see how many instructions per second it's actually getting, it only means anything for that program. It's totally dependant on how many branches are in the code, and how much of the time memory is being accessed.
These people's idea is excellent because it focuses on the direction the computer industry needs to go: parallelism. Forget doing everything sequentially, do it all at the same time! I'm not talking about in one program though, I'm talking about throughout the system. You've got your websever and you fileserver and all your device drivers and your os and all that good crap to run while you play quake, and the only way to really help out is to tack on another processor to run other programs at the same time. This is kind of like what SUN is doing for the MAJC architechture with thread level parallelism.
And finally, I don't see how you people can honestly think that one processor running at something like 450 is going to be as good as 8 all on the same die all running different processes in parallel.
Speed isn't about megahertz or gigahertz or instructions per second. It's about the time it take to run something. Who needs benchmarks when I have my analog wall clock to tell me what's best.
JDW
So you've got MPEG1 layer 3 for music. MPEG1 being the first iteration of the MPEG attempts at digitally encoding and compressing movies... But why can't we use MPEG4 layer 3? If the compression for an MPEG4 movie is several times better than the compression for an MPEG1 movie, than wouldn't the same apply for the layer 3 audio? Or do I just totally misunderstand?
I mean, yeah, Akira, and Macross, and Robotech and all that. Great anime, but your list just isn't complete without El Hazard. I don't get the name though, cause I didn't see anything else spanish-like goin on... And what about The Record of Lodoss War?
Coming out in the hopefully near future will be the anime version of Chrono Trigger, and that should be truely awesome, if they do it right and keep true to the snes game.
Anyway, I also really liked Revolutionary Girl Utenea, but I have only been able to find the first volume, and I don't know what the heck happens!!! Anyone know anything?
using the old I P Freely joke is freaking awesome!
don't think I believe them... but that's just me...
lovely insight! beautiful... And let me throw in my 2 cents...
I think that MicroCrap should be split up into 3 (count them THREE) companies:
1. Operating Systems
2. Applications
3. Window Managers ("shells" as MS likes to call them)
Afterall, it's obvious that Microsoft thinks that a bulky bloated ugly nasty memory hog of a window manager constitutes a solid OS... Well, they would actually have a solid OS (in my opinion) if they took the Win2k kernel, opened up the API and some of the more important source, and created a framework on which to build a window manager. That is, provided you want or need one... Hell, I think they should open up their entire kernel source, but that'll never happen, because then someone ELSE could do what I have just proposed and make a better product than microsoft using their own software... But MS are experienced in doing that to other people (i.e. xerox / apple / ibm / whoevertheystoleqdosfrom).
Well, I'm done for the day...
actually, it's really not EPROM "burning"; it's programming... Generally though, at least in my experience, an EPROM or EEPROM or whatever will come with it's own proprietary programmer, and it is this programmer with which the computer must interface. There isn't a really good way to write a program to tell the programmer to program PROMs because of the fact that programmers have a generally unique interface (wow... that was almost as confusing to write as it will be to read). It is actually possible to program the things manually, and it really doesn't take a sophisticated program to preform this operation. Just build your own programmer to which you know the interface, and write a program to interface with the serial port of you puter. Personally though, I think that something like this is pointless, because anyone that really needs to program a PROM will most likely know how to do it themselves, and you can't expect something universal to come out and be supported under linux when hardware (the physical programmer you would need to build) is involved.
... So take it as you will...
As far as the embedded market goes, EEPROMs again have a specific proprietary interface such as the one in your computer that holds your BIOS information. programs would need to be written for each specific proprietary implementation. And this isn't supported any better or worse under DOS/Win. Infact, it's probably easier to write the needed programs under linux because of it's openness and extensibility. Which would make me think that Linux should be a better choice for the embedded market anyway: not because it has an extensive library of programs which support intricate things such as programming strange and wonderful hardware, but because it is much easier to write this kind of software on an open unixlike system than under DOS or Win...
But all that is just my $0.70710678118654752440084436210485
When /. posts a bogus story like this one... uPs aren't gonna produce any harmful "radiation" such as this story suggests. Sure, every electrical device emmits a multitude of interferance, but to claim that when processors start getting above 1GHz I can cook my TV dinner in my case is rediculous. Power has a great deal to do with such things, and so does the appication. I mean, come on... Let's be serious, a 2.4GHz coordless phone meets FCC specs, and it's *transmitting* it's 2.4GHz signal... RF stuff can't hurt you, and it's moving through all the space around you. Hell, the military use spread spectrum encoded wireless transmissions whose carrier frequencies are 10GHz and UP! And if RF doesn't hurt you, ain't no clocked digital logic circuit gonna byte you. Especially at those exceedingly low frequencies of under 100GHz. It's all about power guys... Do some reaserch before you start spewing your crap all over me. It's all about the power...
I like hardware... it makes me happy
and one day DVDs will be here too... Oh, I hope that day comes soon...
and do I get the FP too?
JDW
I think it doesn't matter whether or not Beozozozzz is going to *do* the right thing, but that he has hit on the point of what is the right thing to fix.
Jeff has given us the opportunity to fix what's broken at the point at which it is broken. If he and Tim can get some actual reform happening at the source of the problem, the future will be much brighter.
I do have a suggestion though. I think that the patent laws for cyberspace stuff should be extraordinarilly limited. This is what I would propose for INet patents:
1) patents last a short 2 or 3 years.
2) patents are to be used defensively only.
An important extension to my second point is that, if patents can only be used defensively, they can only be used against the patent holders competitors. In other words, Amazon's patent could only be used against someone like B&N, other online booksellers. But all other types of industries and sales and markets and services should be allowed to use the technology. So if I'm an online retail computer store, and I want to one-click my way to heaven, I should be allowed to do so, because I'm not competing with Amazon. Restrict the patent to the competition on a business, and this leads into my third point for what an internet patent should be:
3) Publish the methodology and completely open it to the public, and to other non-competing businesses.
That way, we are all happy. Companies have incentive to innovate and patent, and other areas of tech and econ and internet benefit from the openness.
J Derek Wilson
yep, I think it's made up. Why? Cause I design caches. If you push too hard, things don't work. That doesn't make the tech bad, it makes your company stupid. If they have these problems, they are really trying to do something very very wrong. (like save too much money)...
Complete bullshit. I don't think I could have loaded one up any fuller than this. Man, It must have taken a while to come up with all that crap. What did you eat? Let's just say that, if you really work for AMD, I hope you keep the toilets clean, cause only a janitorial worker could pull up all that B.S.
OK, to set the record straight, the L2 cache is a proven mechanism for increasing the average access time for data in a computer system. On die L2 caches can be as simple as you like. The performance gain comes in the high bandwidth availability and the distance from the processor core. The only real problem with on die L2 caches is that they increase the size of the die, and the cost of the chip.
If I were to believe that AMD had such problems, then it would NOT be in the design of the L2 cache, but rather the implementation. And I don't think you could implement such a simple device in such a poor manner. I could build an L2 cache with tinker toys that worked better than something that had these kinds of problems (problems which, by the way, CANNOT physically exist in the way they are described here).
Ok, so like let me overclock my 1GHz to what? 1.1?that's not really neat... the percentage gained in performance by overclocking goes down as the speed increases, because the relationships between die size, speed, and heat are exponential not linear... The faster things get, the harder it will be to overclock, at least without cooling... I see a nice future for kryotech...
la la lalalala
i want access...
You said it! That's exactly how I feel. It seems like if we followed these guys advice, everything would be in a slashbox and there would be no more stories but for Katz posts... But did they know that they can filter out storys that they don't want to see? I don't know how that's news to anyone here, but all you gotta do is look at the preferences page and there you go! People should lighten up around here...
--We are among you...