wow does anyone see how absolutely ridiculous that is... first of all we do not live in a VACUUM (if i was dieing in space because i didnt have a space suit i would definately want an aluminum mp3 player) second if there was oxygen for it to be exposed to it would not be a VACUUM...some one is really confused, and by the way aluminum is really easy to scratch. if you want to get into chemistry ill tell you that Al2O3 (2 and 3 are subscripts) wich is aluminum oxide (in laymens terms aluminum "rust") is not hard at all, you can actually scratch it with you fingernail...
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Will the cell take over the world????
- Thread starter freddiepm61
- Start date
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
http://www.gwydir.demon.co.uk/jo/minerals/hardness.htm
Aluminum oxide would be corundum on that list.
I can't find a proper table yet, but, they do all agree that AL2O3 is 9 of what I've looked at so far.
Aluminum materials CAN be scratched easily. I have yet to find an ordinary heatsink or mp3 player or anything that couldn't be.
Aluminum oxide would be corundum on that list.
I can't find a proper table yet, but, they do all agree that AL2O3 is 9 of what I've looked at so far.
Aluminum materials CAN be scratched easily. I have yet to find an ordinary heatsink or mp3 player or anything that couldn't be.
and why is this...if a console can funtion like this then why cant a pc??? im thinkin as if pc's will develop into somthin similar to a scaleable gameboy. your comment is like sayin that a piston engine is the only way to make an effective internal combustion engine but in actuallity
the wangle (rotary, think rx7) is actually much more efficient.
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
Actually, if you haven't noticed, consoles are moving the opposite direction and becoming more and more PC like. I mean, heck, just look at the X-box. An ultra-customized PC practically.
No, PCs are meant to be used for more than just gaming and it will always be that way. Even simple things like visiting this website could get kind of troublesome if you have to switch back and forth between a PC and a console in a pc case. Who can afford to have two systems to do what one can?
Anyway, like I said. GPU can't do all that stuff alone without becoming a CPU. So, what you are saying is, make a plugable cpu. You know, like, oh, socket a, for example.
No, PCs are meant to be used for more than just gaming and it will always be that way. Even simple things like visiting this website could get kind of troublesome if you have to switch back and forth between a PC and a console in a pc case. Who can afford to have two systems to do what one can?
Anyway, like I said. GPU can't do all that stuff alone without becoming a CPU. So, what you are saying is, make a plugable cpu. You know, like, oh, socket a, for example.
no that wasnt what i was saying at all and i dont see why GPU architecture wont eventually develop to a point were there is no point for a cpu or if there is one that it is actually holding back the GPU (oo kinda like NOW) i mean look at the DS it dosent have a cpu per say and it can do quite a bit for its size imagine a ds 10 years from now thats much much bigger i think your being close minded by saying that the current way is the only way. i honestly think that our current mulitple component systems will be phased out by something similar sort of like the cell processor but more versatile
So just because the XBox is, consoles in general are? Microsoft's first console was built on an architecture that the company was familiar with. It's been made quite clear by Microsoft that the XBox 2 will move away from x86 PC processors. I think it's been publicly announced that the next-generation XBox would use Power-series processors from IBM and a custom OS, moving away from the stripped Windows 2000 and Pentium III platform of old.Nazo said:
No, I think that it would be more accurate to say that consoles are moving towards the CE industry. Already there has been a progression towards playing digital audio and video on consoles. The Sony PSX, available in Japan only, is a PS2 with an internal hard drive, DVD burner, and TV tuner for video recording.
http://www.psx.sony.co.jp/
There was talk about a similar thing for the XBox, coming "who knows when." I think the trend is going more towards CE than PC.
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
Ok, I've been browsing, and I'm not doign it anymore. Best I could find is this. Apparently "soft" aluminum is something to the effect of 2.5 max (this is in the range of possibly scratchable by a mere fingernail.) I don't remember what plastic was, but, it's similarly low. "Hard" aluminum was more to the effect of 4.0, which is higher than plastic I'm pretty sure. I think plastic was somewhere in the range that the soft aluminum is in. Of course, I've seen some plastics made harder than others, so I don't know what the highest is (most of these types of things are made with the higher more brittle plastic.)
Anyway, I can see no reason to follow that idea of yours. The DS sounds to me like it's basically using a specialized CPU essentially if it has no seperate video accelerator. Ok, I don't have one of those and haven't seen much about them, so I don't know where their limits are, but, I do know I could emulate a PSX in Virtual Gaming station using 100% cpu, 0% GPU back when I had a P3-350. Ran smoothly and everything.
The idea of PC is you seperate everything into upgradable components. This way you can upgrade one thing at a time. Your way makes it impossible to upgrade just one. On a PC, if you don't like how fast your CPU performs, get a better CPU. Don't like your video card, get a better video card. On your desktop console, if you need a faster CPU, you have to upgrade the CGPU at a cost of say $500 for a mediocre one. (That's $100 for a CPU like mine + $200 for the cheaper video cards.) Want something better? $1500 for the latest and greatest. Or, alternately, you can just seperate them, spend a bit on a cpu one year, a bit on a video card another year, etc. When you find yourself CPU limited, get a faster CPU. When you find yourself GPU limited, get a faster video card. No need to do both at once.
EDIT: Xonik, that's still closer to a PC. Ever seen any info on a SNES? A bunch of chips all running at very low speeds. The central processor was something like 15MHz (or was it 12.5?) It wasn't until I had a 486/66 that I finally started seeing my PC start to catch up on many of the SNES's capabilities in some areas. However, at the same time, I could run Doom on a 486/25 and the SNES just couldn't pull it off decently. Ever seen it? It looks like crap compared to the PC version. They even elected to remove floor/ceiling textures entirely.
Consoles used to be HIGHLY specialized systems. Now, they are dangerously close to PCs. Sony has been caught on several occasions calling the PSX a PC!
PS, when you say "integrated dvd burner" and things like that, you aren't exactly arguing your case very well. Ok, tv tuner is a little more specialized, though PCs have had these for a VERY long time (think I got my first video card with a tuner built in back when I had a 350MHz P3, which was still in the time of the PSX before dreamcast,) but, it's just sounding to me like it's turning more PC like. Consoles moved UP to CE, and now they are moving slowly towards almost PC.
Anyway, I can see no reason to follow that idea of yours. The DS sounds to me like it's basically using a specialized CPU essentially if it has no seperate video accelerator. Ok, I don't have one of those and haven't seen much about them, so I don't know where their limits are, but, I do know I could emulate a PSX in Virtual Gaming station using 100% cpu, 0% GPU back when I had a P3-350. Ran smoothly and everything.
The idea of PC is you seperate everything into upgradable components. This way you can upgrade one thing at a time. Your way makes it impossible to upgrade just one. On a PC, if you don't like how fast your CPU performs, get a better CPU. Don't like your video card, get a better video card. On your desktop console, if you need a faster CPU, you have to upgrade the CGPU at a cost of say $500 for a mediocre one. (That's $100 for a CPU like mine + $200 for the cheaper video cards.) Want something better? $1500 for the latest and greatest. Or, alternately, you can just seperate them, spend a bit on a cpu one year, a bit on a video card another year, etc. When you find yourself CPU limited, get a faster CPU. When you find yourself GPU limited, get a faster video card. No need to do both at once.
EDIT: Xonik, that's still closer to a PC. Ever seen any info on a SNES? A bunch of chips all running at very low speeds. The central processor was something like 15MHz (or was it 12.5?) It wasn't until I had a 486/66 that I finally started seeing my PC start to catch up on many of the SNES's capabilities in some areas. However, at the same time, I could run Doom on a 486/25 and the SNES just couldn't pull it off decently. Ever seen it? It looks like crap compared to the PC version. They even elected to remove floor/ceiling textures entirely.
Consoles used to be HIGHLY specialized systems. Now, they are dangerously close to PCs. Sony has been caught on several occasions calling the PSX a PC!
PS, when you say "integrated dvd burner" and things like that, you aren't exactly arguing your case very well. Ok, tv tuner is a little more specialized, though PCs have had these for a VERY long time (think I got my first video card with a tuner built in back when I had a 350MHz P3, which was still in the time of the PSX before dreamcast,) but, it's just sounding to me like it's turning more PC like. Consoles moved UP to CE, and now they are moving slowly towards almost PC.
THANK YOU xonik thats very similar to what im sayin i mean you see it all over the place (think centrino, goforceFX,...cell processor) stuff is all starting to blend together. i wouldnt be suprised if one day there was one chip that incompassed a cpu, gpu and ram that just plugged into a universal motherboard wich tied it into the input and storage devices...thats deffinately not in the near future though, but shit will continue to be dual purposed... think this mobo http://directron.com/epiamii12000.html with the power of the system in my sig
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
Let me put it this simply. I and just about everyone I know prefer to have an upgradable system instead of an integrated system. Ok? What you want to do is turn the PC into a console with little to no upgradable capabilities short of buying a whole new system... And you intend to write a report for a class/work on this console PC? No, you say to buy a seperate system. So, now you have TWO systems to keep semi up to date. The PC will be less of a problem now that you mostly use it for stuff like word processing and all (though god help you if you want to do encoding on your ultra-weak system, and just pray your sony brand "PC" (console) hasn't been protected from using DVDRs because how else are you going to watch the DivX2 you just spent a week encoding since your PC can't handle it?) Nonetheless, you're going to have to update it everynow and then to keep up with that latest MS word...
wow your a douche bag...i didnt say i wanted it like that. you just assumed i did, i was actually just a posing a solution to the question that was initially posed in the thread "will the cell take over the world????" it is just my oppinion and i absolutely cant stand it when close minded people STOMP on other peoples ideas, in fact i actually prefer the idea of a component sytem i just dont think they will be around much longer. i believe the cpu is going to reach a point where it no longer needs to get faster and will be integrated into the gpu...ohh kinda like cell wich is what the topic is about... i dont believe the cell itself will reach into the pc market but its predecessors most definately will.
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
What I'm saying is that there's no way the PC will turn into a console. We have consoles for that. Not my fault that you kept insisting that it will happen this way. Now, calm yourself and try to dispense with the childish insults.
what im sayin is that there both getting so close in specs and purposes that they cant help but merge eventually, and i think it would be awesome to have a top of the line gaming pc for the cost of a console but id still rather build my own. childish insults....you started the condescending remarks and just because you think or i think somthing will go a certain way doesnt mean it will thats why i think its ridiculous to totally dismiss and or make fun of someone elses ideas
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
Say. Tell me, what's so different in the system for gaming/etc PCs use today versus back when people had things like a P2-233 and a Voodoo? I mean, you say that they are growing more alike, so PCs must have changed rather a lot more than just processing power of the CPUs and GPUs if consoles aren't getting more PC-like, but, instead, PCs are getting more console-like... Don't bother with the integrated into the mb argument. Still a seperate system, just stuck onto the same board, and still upgradable by using a non-integrated device, giving you something better potentially in every case.
I don't dismiss your idea, I'm saying that you aren't talking about a future PC, you are talking about the Playstation 3.
BTW, I almost forgot about FMDs. That company name sounds familiar, but, that's a different technology. Kind of interesting actually. FMDs are a little simple of a concept. Instead of bouncing the laser off of a reflective surface, shine it through all the surfaces and where the bits are each individual layer will emit a different wavelength back.
I don't dismiss your idea, I'm saying that you aren't talking about a future PC, you are talking about the Playstation 3.
BTW, I almost forgot about FMDs. That company name sounds familiar, but, that's a different technology. Kind of interesting actually. FMDs are a little simple of a concept. Instead of bouncing the laser off of a reflective surface, shine it through all the surfaces and where the bits are each individual layer will emit a different wavelength back.
Wow, this is getting ridiculous.
Um, what are you talking about? DVD recorders aren't only for the PC. Just about every consumer electronics manufacturer has a standalone DVD recorder.Nazo said:
Okay, and VCRs have had TV tuners for ages. Then there was TiVo, which the PC mimics to this day. And now, we have CE devices that record to hard drive or DVD. What I see is an update to the VCR, not a spinoff of a PC innovation.
pcs are a getting more and more gpu dependent thats what im gettin at, eventually we wont need cpus. especially when operating systems like longhorn come out wich will be very graphically oriented.
considering the title of this thread i dont see why thats out of place
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
VCRs aren't applicable since they are analog. Otherwise you might as well say an old black and white TV counts since it has a tuner.
TiVo came after capture cards for the PC. I don't know when either first came out, but, remember, I got my first ViVo video card was a PCI crappy thing and I'm thinking now it was actually on my old 233 P2. The technology had been developed for a LONG time before they stuck it onto a video card that was affordable. Anyway, the TiVo is basically a highly specialized PC (probably more CE) using the technology first developed for PCs. In other words, PC isn't emulating TiVo, TiVo is making it easier to do what the PC could do for the average user. Same more or less goes for the others.
PCs are GPU dependant for now, yes. However, more and more? That's a current short-term trend. Look at the big picture. It ALWAYS has been switching between one and the other as games are developed. Watch and you'll see. Games will start to use CPU more now that they can count on better CPUs, so they will. They hit a sort of ceiling in the GPU line, so, they stop pushing limits quite so hard there. This is how it has gone since even the days where the only way a GPU removed limits from the CPU was by drawing the screen faster without accelerating anything beyond that. Even back then it worked like this.
TiVo came after capture cards for the PC. I don't know when either first came out, but, remember, I got my first ViVo video card was a PCI crappy thing and I'm thinking now it was actually on my old 233 P2. The technology had been developed for a LONG time before they stuck it onto a video card that was affordable. Anyway, the TiVo is basically a highly specialized PC (probably more CE) using the technology first developed for PCs. In other words, PC isn't emulating TiVo, TiVo is making it easier to do what the PC could do for the average user. Same more or less goes for the others.
PCs are GPU dependant for now, yes. However, more and more? That's a current short-term trend. Look at the big picture. It ALWAYS has been switching between one and the other as games are developed. Watch and you'll see. Games will start to use CPU more now that they can count on better CPUs, so they will. They hit a sort of ceiling in the GPU line, so, they stop pushing limits quite so hard there. This is how it has gone since even the days where the only way a GPU removed limits from the CPU was by drawing the screen faster without accelerating anything beyond that. Even back then it worked like this.
so basically what your sayin is that digital tuners became so basic that they were able to just stick it onto the video card and consolidate the two...wow thats remarkably similar to what im sayin about cpu's and gpu's, see its happening all over the place!
Decelerate
Limp Gawd
- Joined
- Jul 29, 2004
- Messages
- 467
Ok read the replies to my posts.
Smaller not innovative? Dang, never knew the process of miniaturization was so benign.
What a utopic statement. Go ahead and tell that to Intel researchers trying to reach 65nm and below. Do you know what the hurdles are? Do you not notice how Intel had problems with 90nm? Maybe their miniaturization machines just broke
Stupid-proofing is a feature. Isn't that what Apple is shouting for in their products? How do you think the Shuffle is marketed? Nothing to learn, press a button, two perhaps and voila. Windows is the best product for average use, user-friendliness, compatibility and support. Apple doesn't have that. Unix doesn't have that. Linux doesn't have that.
You make it sound so easy. Why are we stopping at blue? Why not go for the smallest now? Why are we still in 90nm? Why not directly go 66nm? 60? heck 1?
Making something bigger/smaller, faster, stronger, that's innovation.
Mini-cd isn't a cd, it's a mini-cd (and no, I'm not talking about a minidisc). Can you play a full-sized CD in it? If not it's not a smaller cd player.
With your ideals we'd be at nanotechnology and FMD discs a century ago, if not more. Making things small isn't a matter of putting the product in a shrink ray.
Smaller not innovative? Dang, never knew the process of miniaturization was so benign.
Triple_B said:
What a utopic statement. Go ahead and tell that to Intel researchers trying to reach 65nm and below. Do you know what the hurdles are? Do you not notice how Intel had problems with 90nm? Maybe their miniaturization machines just broke
Nazo said:
Stupid-proofing is a feature. Isn't that what Apple is shouting for in their products? How do you think the Shuffle is marketed? Nothing to learn, press a button, two perhaps and voila. Windows is the best product for average use, user-friendliness, compatibility and support. Apple doesn't have that. Unix doesn't have that. Linux doesn't have that.
You make it sound so easy. Why are we stopping at blue? Why not go for the smallest now? Why are we still in 90nm? Why not directly go 66nm? 60? heck 1?
Making something bigger/smaller, faster, stronger, that's innovation.
Mini-cd isn't a cd, it's a mini-cd (and no, I'm not talking about a minidisc). Can you play a full-sized CD in it? If not it's not a smaller cd player.
With your ideals we'd be at nanotechnology and FMD discs a century ago, if not more. Making things small isn't a matter of putting the product in a shrink ray.
Decelerate
Limp Gawd
- Joined
- Jul 29, 2004
- Messages
- 467
cell_491 said:
CPU = Central processing unit
GPU = Graphics processing unit
GPUs take care of graphics. You won't be seeing it compute AI behavior anytime soon for example.
... And if they do at some point, then they just become CPUs.
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
Didn't say it was as simple as that, did I? I just said you can't call it a terribly innovative thing to just follow 1 to 2 and then decide, hey, let's make 3. It's not necessarily easy, and will cost a lot of money sometimes (R&D as well as development) but, is it really innovative to just take the next step?Decelerate said:What a utopic statement. Go ahead and tell that to Intel researchers trying to reach 65nm and below. Do you know what the hurdles are? Do you not notice how Intel had problems with 90nm? Maybe their miniaturization machines just broke
I don't deny this fact. I don't call it a better os because it's made more "user friendly" (eg dumbed down to hide all the horrible complicated things such as file extentions so the user wont' know what is going on later when they run into an unknown file...)
Why do people keep thinking I'm saying that going to the idea of taking the next step is easy? I'm merely saying the next step isn't that innovative compared to, say, skipping a step. Did Intel jump straight to 65 or whatever it is they've supposedly reached these days? Or, did they maybe go one step at a time? 1, 2, 3 instead of 1, 3, 5?
Actually, a mini-cd is a cd. It's the PRECICE same technology. The same exact density and everything. Uses the same sort of single laser, fits inside any PC cd and quite a number of cd players and will work in any it fits in. It's a smaller cd. Still a CD, just a different type. Not like comparing a DVD to a CD or something where they are the same size but the DVD holds more due to a different technology.
Define playing a full sized cd. I was joking before, but, if a full sized cd is 74 minutes of cd-quality audio, then, yes, it can play a full sized cd. Anyway, it was just a joke, so calm down about the mini-cd thing. I was merely making a point with a little humor attached, and I'm sorry I dared to do that.
wow something we actually agree on... bluray or sumthin similar is the next step i can actually see optical disks replacing magnetic disks as permenent storage in the very near future the blue ray disks will have 25gb per layer and 8 layers per disk thats 200gb however the drives will only have a transfer rate or 35mbps so there definately not perfect.
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
You forgot solid state. It actually looks like that will be the future replacement of them all in the LONG run. As in seriously long run.
I personally can testify to being one of the millions these days who carries a 256MB flash drive in his pockets rather than a 1.44MB floppy disk.
Anyway, the only point I was making with all that is next step isn't innovative. Finding new ways to do things people hadn't thought of, that's innovative. Making the same old thing smaller and faster (or storing more or whatever the particular comparison may be) isn't. The mini-cd player was a little bit innovative because it came up with the idea of using current technology in a way not done before (at the time.) I'm not saying it's VERY innovative, just a little.
I personally can testify to being one of the millions these days who carries a 256MB flash drive in his pockets rather than a 1.44MB floppy disk.
Anyway, the only point I was making with all that is next step isn't innovative. Finding new ways to do things people hadn't thought of, that's innovative. Making the same old thing smaller and faster (or storing more or whatever the particular comparison may be) isn't. The mini-cd player was a little bit innovative because it came up with the idea of using current technology in a way not done before (at the time.) I'm not saying it's VERY innovative, just a little.
yea solid state in the very very very very very very very very very very long run...when they have 1terabyte usb 5.1 jump drives with a 100gb per second transfer rate then we will have truly arrived....lol. but in the near future i believe optical will take the forefront magnetic is reaching a limit i have the fastest 7200 rpm hdd on the planet and it leaves alot to be desired (see sig for details)
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
I agree with you on everything but the RPMs. RPMs aren't a measure of the speed of the thing. However, if you must look at those alone, take a look at the Raptors. Thing is, HDs are limited by things such as fragmentation and the fact that even when there's no fragmentation you have to make it jump back and forth to different files and such. They improve seek time all the time, but, it's not easy. The limitation doesn't lie in the spinning discs faster, that's just to read the data faster at max speed and is almost kind of cheating really.cell_491 said:
Anyway, back to the whole CPU/GPU thing, all I'm saying is we already have consoles. We don't need to convert PCs into consoles and keep two seperate desktop systems. There are reasons we have chosen to stay with a deintegrated system when it could have been integrated in many ways years ago. CPU can do so many of those things. Sound (yes, PC speaker can play "cd-quality" audio,) video, etc. As I said, you are just looking at short-run when you see that the GPU is the main limit today. History (generations in technological terms, merely a few years in human terms) has shown this same trend.
Decelerate
Limp Gawd
- Joined
- Jul 29, 2004
- Messages
- 467
Nazo said:
AH HA! Ok now I see where you're going. We're talking about the same stuff but from a different scale.
Different leagues altogether. Maybe it's my engineering backround. No idea what your field is and not downplaying it in any way, offensive or benign. I'm just saying that I'm probably closer to the field than you are (I can't fathom a computer/IT engineer downplaying miniaturization).
I can see how you perceive that going from 1 to 2 to 3 is nothing innovative, but you seem to forget that the gap between them gets greater and greater.
Well actually you're probably aware of that. What I'm saying is that to go from 1 to 2, people have to innovate not the 2 itself, but what it takes to build it. Basically the 1.1, 1.2, 1.3 etc. that makes the 1 can be very different from the 2.1, 2.2, 2.3 etc that makes the 2.
I guess what irritated me and/or xonik (can't speak for him) was that you (actually a lot of people here) only see the 1 and 2 as one and two, whereas I (and probably xonik, but I'm assuming) take in consideration the underlying innovation; Silicon-on-Insulator, strained-silicon, etc.
Here's an example:
One can see the Airbus A380 as simply a Boeing 747-400, but only bigger. However, without the drive-by-wire technology implemented, amongst others, Airbus couldn't have done what they did.
Now here's the deal:
Drive-by-wire is innovation over conventional mechanical controls, and I doubt you can argue that.
Now, if the A380, arguably not an innovation (bigger plane), required an innovation like drive-by-wire, then isn't the A380 itself an innovation by association/heredity of concept?
And drive-by-wire is to the airplane industry what SoI and SS are to the processor industry.
Sincerely looking forward to the response to that.
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
Ok, since the 1,2,3 vs 1,3,5 thing was somehow misunderstood, let me say it this way. Blue laser is 1,2,3. FMD is 1,2,B. B is similar to 2 in that it is the second letter. It is not, however, the next step after 2. It's a new step based vaguely on 2.
*sigh* Should I stop using parallels?
*sigh* Should I stop using parallels?
Decelerate
Limp Gawd
- Joined
- Jul 29, 2004
- Messages
- 467
Nazo said:
Doesn't make blue laser any less innovative. It just makes FMD different. And I really think that you're associating the two words/concept/definition.
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
Fine, call it a win if you wish, but, I'm tired of repeating myself. I don't know why 1,2,3 is too complicated, but, I'll just give up on it. You win. 3 is innovative.
I have nothing more to say on this matter. May we now return to the original point?
I have nothing more to say on this matter. May we now return to the original point?
Decelerate
Limp Gawd
- Joined
- Jul 29, 2004
- Messages
- 467
I'm not here to win. I'm here to explain.
And if you're not getting it then I guess that I also lost. That or you're bastardizing the whole concept of progress.
Anyhoo, we're at a moot point, I'm also ending it here.
Ok back to the point. I'm not so sure Cell is here to take over the world, but it shouldn't be scuffed at so easily as the emotion engine. IBM is a formidable partner, and their involvement in the project leads to the assumption that they have planned BIG THINGS for it. Doubt I can find any more arguments for/against it until they really show the item and its performance (as in, more than on a sheet of paper).
But you can bet that the whole industry is keeping an eye on it and sees it as an IBM/Sony response to the whole IT image as it is.
It's like Linux & Apple banding together to take on Microsoft.
And if you're not getting it then I guess that I also lost. That or you're bastardizing the whole concept of progress.
Anyhoo, we're at a moot point, I'm also ending it here.
Ok back to the point. I'm not so sure Cell is here to take over the world, but it shouldn't be scuffed at so easily as the emotion engine. IBM is a formidable partner, and their involvement in the project leads to the assumption that they have planned BIG THINGS for it. Doubt I can find any more arguments for/against it until they really show the item and its performance (as in, more than on a sheet of paper).
But you can bet that the whole industry is keeping an eye on it and sees it as an IBM/Sony response to the whole IT image as it is.
It's like Linux & Apple banding together to take on Microsoft.
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
Linux has been on Macs for a long time now and they still haven't dislodged MS from it's grip on the PC world.Decelerate said:
My point on the quality thing earlier is that I feel like Sony will end up draggin the others down. Oh, and I personally have a high opinion of Toshiba as well, so I hate to see this for either of them.
Anyone care to prove/disprove the implication that the cell is going to suck for ACTUAL real life games rather than some kind of game that is 100% graphics, 0% everything else (things like ai/physics/etc included in "everything.")
i never said i think rpms are what make a drive fast all i said was that i had the fastest 7200rpm harddrive available the maxtor 250gb sata 16mb cache its only slighlty slower than the raptors wich run at 10,000...you seem to be good at missinterpreting people...lol. because we were sayin the exact same thing... about data storage at least
Decelerate
Limp Gawd
- Joined
- Jul 29, 2004
- Messages
- 467
I was talking about Linux (as in, the whole community) & Apple banding together, not about getting Linux on macs.
The cell processor is basically a regular RISC type of processor with the most absolutely badass math coprocessor you could think of attached to it... they've got the equivalent of 4 or 8 modern DSP chips on a single chip, with the clock rate almost quadrupled compared to the current offerings from DSP vendors.
For things like SETI, video encoding and processing, MP3 encoding, raytracing, 3D, etc which can be broken into parallel streams of vector data to process, the Cell will absolutely scream. It's even incredibly useful for things like cellphone basestation applications, but I doubt it will replace the DSP/FPGA arrays used in those applications.
It won't take over the world, I don't expect to find a Cell processor in a PC anytime soon. But I'd still love to see a vector processor of some sort in a PC.. unfortuantely that came and died with the NeXT.
For things like SETI, video encoding and processing, MP3 encoding, raytracing, 3D, etc which can be broken into parallel streams of vector data to process, the Cell will absolutely scream. It's even incredibly useful for things like cellphone basestation applications, but I doubt it will replace the DSP/FPGA arrays used in those applications.
It won't take over the world, I don't expect to find a Cell processor in a PC anytime soon. But I'd still love to see a vector processor of some sort in a PC.. unfortuantely that came and died with the NeXT.
Maynerd Goldstein
Weaksauce
- Joined
- Nov 14, 2003
- Messages
- 81
"We have never seen a leap in performance like this before and I don't expect we'll ever see one again, It'll send shock-waves through the entire industry and we'll see big changes as a result".
Never underestimate IBM. IBM is the company that developed the IBM 360
mainframe, in what's been called the civilian equivalent of the Manhattan Project.
IBM has technology so far advanced of anything on the market today they don't
know what to do with it. They had 21 inch LCD displays with over 3000x2000 resolution back in 2000! Yeah, they cost 25,000 but still. IBM's current mainframe technology and Z/OS operating systems is lightyears ahead of anything you'll find on the PC platform.
Cell Architecture Explained - Part 5: Conclusion and References
Short Overview
The Cell architecture consists of a number of elements:
The Cell Processor
This is a 9 core processor, one of these cores is something similar to a PowerPC G5 and acts as a controller. The remaining 8 cores are called APUs and these are very high performance vector processors. Each APU contains it's own block of high speed RAM and is capable of 32 GigaFlops (32bit). The APUs are independent processors and can act alone or can be set up to process a stream of data with different APUs working on different stages. This ability to act as a "stream processor" gives access to the full processing power of a Cell which is more than 10 times higher than even the fastest desktop processors.
In addition to the raw processing power the Cell includes a high performance multi-channel memory subsystem and a number of high speed interconnects for connecting to other Cells or I/O devices.
Distributing Processing
Cells are specifically designed to work together. While they can be directly connected via the high speed interconnects they can also be connected in other ways or distributed over a network. The Cells are not gaming or computer specific, they can be in anything from PDAs to TVs and all can be used to effectively act as a single system. The infrastructure for this is built into each Cell as they operate on "Software Cells" which contain routing information as well as programs and data.
Parallel programming is usually complex but in this case the OS will look at the resources it has and distribute tasks accordingly, this process does not involve re-programming. If you want more processing power you simply add more Cells, you do not need to replace the existing ones as the new Cells will augment the existing ones.
Overall the Cell architecture is an architecture for distributed, parallel processing using very powerful computational engines developed using a highly aggressive design strategy. These devices shall be produced in vast numbers so they will provide vast processing resources at a low cost.
Conclusion
The first Cell based desktop computer will be the fastest desktop computer in the industry by a very large margin. Even high end multi-core x86s will not get close. Companies who produce microprocessors or DSPs are going to have a very hard time fighting the power a Cell will deliver. We have never seen a leap in performance like this before and I don't expect we'll ever see one again, It'll send shock-waves through the entire industry and we'll see big changes as a result.
The sheer power and low cost of the Cell means it will present a challenge to the venerable PC. The PC has always been able to beat competition by virtue of it's huge software base, but this base is not as strong as it once was. A lot of software now runs on Linux and this is not dependant on x86 processors or Microsoft. Most PCs now provide more power than is necessary and this fact combined with fast JIT emulators means that if necessary the Cell can provide PC compatibility without the PC.
It will not just attack the PC industry but expect it to be widely used in embedded applications where high performance is required. This means it will be made in numbers potentially many times that of x86 CPUs and this will reduce prices further. This will also hurt PC based vendors' desires to enter the home entertainment space as PC based solutions [Entertainment] will be more complex and cost more than Cell based systems.
This is going to prove difficult for the PC as CPU and GPU suppliers will have essentially nothing to fight back with. All they can hope to do is match a Cell's performance but even that is going to be incredibly difficult given the Cell's aggressive Cray-esqe design strategy.
Cell is going to turn the industry upside down, nobody has ever produced such a leap in performance in one go and certainly not at a low price. The CPU producers will be forced to fight back and irrespective of how well the Cell actually does in the market you can be sure that in a few short years all CPUs will be providing vastly more processing resources than they do today. Even if the Cell was to fail we shall all gain from it's legacy.
Not all companies will react correctly or in time, this will provide opportunities for newer, smaller and smarter companies. Big changes are coming, they may take years but the Cell means a decade from now the technology world is going to look very different.
http://www.blachford.info/computer/Cells/Cell5.html
Never underestimate IBM. IBM is the company that developed the IBM 360
mainframe, in what's been called the civilian equivalent of the Manhattan Project.
IBM has technology so far advanced of anything on the market today they don't
know what to do with it. They had 21 inch LCD displays with over 3000x2000 resolution back in 2000! Yeah, they cost 25,000 but still. IBM's current mainframe technology and Z/OS operating systems is lightyears ahead of anything you'll find on the PC platform.
Cell Architecture Explained - Part 5: Conclusion and References
Short Overview
The Cell architecture consists of a number of elements:
The Cell Processor
This is a 9 core processor, one of these cores is something similar to a PowerPC G5 and acts as a controller. The remaining 8 cores are called APUs and these are very high performance vector processors. Each APU contains it's own block of high speed RAM and is capable of 32 GigaFlops (32bit). The APUs are independent processors and can act alone or can be set up to process a stream of data with different APUs working on different stages. This ability to act as a "stream processor" gives access to the full processing power of a Cell which is more than 10 times higher than even the fastest desktop processors.
In addition to the raw processing power the Cell includes a high performance multi-channel memory subsystem and a number of high speed interconnects for connecting to other Cells or I/O devices.
Distributing Processing
Cells are specifically designed to work together. While they can be directly connected via the high speed interconnects they can also be connected in other ways or distributed over a network. The Cells are not gaming or computer specific, they can be in anything from PDAs to TVs and all can be used to effectively act as a single system. The infrastructure for this is built into each Cell as they operate on "Software Cells" which contain routing information as well as programs and data.
Parallel programming is usually complex but in this case the OS will look at the resources it has and distribute tasks accordingly, this process does not involve re-programming. If you want more processing power you simply add more Cells, you do not need to replace the existing ones as the new Cells will augment the existing ones.
Overall the Cell architecture is an architecture for distributed, parallel processing using very powerful computational engines developed using a highly aggressive design strategy. These devices shall be produced in vast numbers so they will provide vast processing resources at a low cost.
Conclusion
The first Cell based desktop computer will be the fastest desktop computer in the industry by a very large margin. Even high end multi-core x86s will not get close. Companies who produce microprocessors or DSPs are going to have a very hard time fighting the power a Cell will deliver. We have never seen a leap in performance like this before and I don't expect we'll ever see one again, It'll send shock-waves through the entire industry and we'll see big changes as a result.
The sheer power and low cost of the Cell means it will present a challenge to the venerable PC. The PC has always been able to beat competition by virtue of it's huge software base, but this base is not as strong as it once was. A lot of software now runs on Linux and this is not dependant on x86 processors or Microsoft. Most PCs now provide more power than is necessary and this fact combined with fast JIT emulators means that if necessary the Cell can provide PC compatibility without the PC.
It will not just attack the PC industry but expect it to be widely used in embedded applications where high performance is required. This means it will be made in numbers potentially many times that of x86 CPUs and this will reduce prices further. This will also hurt PC based vendors' desires to enter the home entertainment space as PC based solutions [Entertainment] will be more complex and cost more than Cell based systems.
This is going to prove difficult for the PC as CPU and GPU suppliers will have essentially nothing to fight back with. All they can hope to do is match a Cell's performance but even that is going to be incredibly difficult given the Cell's aggressive Cray-esqe design strategy.
Cell is going to turn the industry upside down, nobody has ever produced such a leap in performance in one go and certainly not at a low price. The CPU producers will be forced to fight back and irrespective of how well the Cell actually does in the market you can be sure that in a few short years all CPUs will be providing vastly more processing resources than they do today. Even if the Cell was to fail we shall all gain from it's legacy.
Not all companies will react correctly or in time, this will provide opportunities for newer, smaller and smarter companies. Big changes are coming, they may take years but the Cell means a decade from now the technology world is going to look very different.
http://www.blachford.info/computer/Cells/Cell5.html
Nazo
2[H]4U
- Joined
- Apr 2, 2002
- Messages
- 3,672
Don't judge by PR and theories. Judge by practice. In practice, I have seen on my own TV through composite (that's right, not even s-video) the crawling ants effect of those horribly jagged edges and I've noticed that while characters have plenty of detail, if you look away from the characters for but a moment you tend to find that a lot of simple things were skimped out on. (Xenosaga anyone?)
PS. Anyone else dissapointed by the fact that next to nothing actually uses > stereo after they make all these boasts of the audio capabilities? The optical connection is almost only good for DVD.
EDIT: BTW, weren't PC graphics cards up to 256-bits at the time the PS2 was made? Regardless, I know it was typical for a mediocre card to pull off at least very basic FSAA by that time.
If you read the article on Ars that I linked above you will see that the Cell is missing most of a CPU's front end, even the PPC core is without the things most CPU's have these days. So it will have to depend on software to do much of what current CPU's do in hardware.. sound familiar? Transmeta anyone? Sure that works, as Transmeta proved.. But performance? On a Dumbed Down PPC? Heck the Transmeta is a VLIW processor under it's software, (like Itanium) and it can't keep up with a P3.. What makes people think a PPC601/Power5 hybrid will beat whatever PC's will be using by the time this is out?
==>lazn
==>lazn