If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
I once actually learned something from this group
I came here a bit before the P4. The architecture of P4 seemed right
to me: highly-pipelined, clocked fast, and with big bandwidth to memory. I didn't really understand about the trace cache, but I soon learned. I was an early adopter, and I was disappointed. The reasons for my disappointment are too many to go into. For one thing, I didn't understand what a bodge the Camino chipset was. What perplexed me most, though, was the P4's abysmal performance on tasks like compiling gcc. The responses to my perplexity could be divided into two classes: Intel bashers, and those who actually knew something. As far as I can remember, the only one who actually knew anything was David Wang (Dean Kent was probably knowledgeable, but I don't remember learning much from him). David's initial response to me was "you idiot," but after some private communication, I got on track to learn some things. I renewed my membership in the ACM. I started reading papers. The P4's problem (as at least some know) is its endless pipeline that doesn't buy it as much as Intel had hope in clock speed overhead. How much it costs and what can be done about it produced a lot of interesting work, until Intel finally just gave up and went back to a P3 design, which they probably never should have abandoned. I haven't learned any of this from anyone here but David Wang. I've listened to the actual P4 architect come clean on the subject and I've talk to people who know a *lot* more about P4 programming than anyone here. Everyone else is stroking their own ego or perhaps some body part. The problems of the P4 were my entry into learning something about computer architecture other than the Cray 1, and it started here. Because of all of the screaming and bellowing about John Corse and Rambus, I never learned as much from David Wang about Rambus as I think I could have. More generally, my knowledge of the actual mechanics of memory and memory controllers and what might be done about it is still weak. The railing here about AMD and Intel has produced only venom, personal accusations, billowing clouds of smoke, and exercises in ego that would make a professional tennis player blush. As a result of my investigations into the P4, I became interested in Itanium, whose problems are related. Neither chip is what you would call agile. They do well with predictable code (as do stream processors), but they cope very badly with surprises. As a result of my public musings on Itanium, I've given invited talks on the subject to real computer architects, not just to those who play them on usenet. Because I once learned something here, I stupidly keep coming back. As a result of much more reading in the professional literature and talking to really smart people (none here), I have actually learned a few things. The time has come, really, to give up. I don't need the wisdom of people who know how to use a desoldering station. Nor do I need their contempt. It's too bad, really. Real World Tech has its own problems, but, if I'm going to find David Wang, for example, he won't be here. The people who are bent on showing me something or other about my personality or some other personal thing about me are just wasting my time, theirs, and everyone else's. If you don't get the way in which I am bright, and I am, then stop showing how clueless you are by trying to tell me how witless I am. Robert. |
#2
|
|||
|
|||
I once actually learned something from this group
Robert Myers wrote:
If you don't get the way in which I am bright, and I am, You're just too smart for us, man. After all, no one else figured-out that we'd all be better-off if it not for AMD, since they drained Intel of resources that otherwise would have allowed them to do IA64 right. No one else figured-out that the world really hasn't realized massive cost and performance gains, directly attributable to the Intel/AMD competition and the success of AMD64. |
#3
|
|||
|
|||
I once actually learned something from this group
On Sep 30, 4:10*pm, chrisv wrote:
Robert Myers wrote: If you don't get the way in which I am bright, and I am, You're just too smart for us, man. * After all, no one else figured-out that we'd all be better-off if it not for AMD, since they drained Intel of resources that otherwise would have allowed them to do IA64 right. * No one else figured-out that the world really hasn't realized massive cost and performance gains, directly attributable to the Intel/AMD competition and the success of AMD64. You're stuck in a mental rut. Who cares if PC's go 50% faster? The problems are elsewhere. So, no, your insight doesn't impress me. Even were Itanium and NetBurst the worse architectures, we'd be making more fundamental progress by continuing to struggle with the problems they create. The approach that has won out is about "installed base," more than anything else. We'll continue to get badly-written code, so- so compilers, and general cluelessness about the nitty-gritty of computing. If the goal is to design a chip for existing awful code, x86 with a short pipeline and the lowest possible hardware latency wins. That's the way of the world, which is different from being smart. In any case, I''m *really* tired of hearing about it. You're like a broken record. Robert. |
#4
|
|||
|
|||
I once actually learned something from this group
Robert Myers wrote:
chrisv wrote: Robert Myers wrote: If you don't get the way in which I am bright, and I am, You're just too smart for us, man. * After all, no one else figured-out that we'd all be better-off if it not for AMD, since they drained Intel of resources that otherwise would have allowed them to do IA64 right. * No one else figured-out that the world really hasn't realized massive cost and performance gains, directly attributable to the Intel/AMD competition and the success of AMD64. You're stuck in a mental rut. Who cares if PC's go 50% faster? It's price/performance that is the "bottom line". So, "who cares" if Intel has no serious competition, so are allowed to feed us overpriced, mediocre products? We care. The world cares. Sheesh. The problems are elsewhere. What "problems" are you referring to? The "problems" you find technically interesting, or the "problems" the real world faces, trying to get the job done in the most efficient manner? So, no, your insight doesn't impress me. Even were Itanium and NetBurst the worse architectures, we'd be making more fundamental progress by continuing to struggle with the problems they create. Says you. That's absurd, says me. For a few academics, the challenge of making a k00l-enough compiler to make IA64 kick ass may be very interesting. Meanwhile, the hundreds of millions of PC users are making *real* progress with great products like the C2D. You might say "maybe in the long run we'd be better-off with the kool compiler technology". I would respond "maybe not". Maybe it's a complete dead-end, like Netburst (which, in hindsight, was a *really* bad idea). Maybe, when CPU's with 64 cores are being made, IA64 will look to be similarly wrong-headed. (snip) In any case, I''m *really* tired of hearing about it. You're like a broken record. Well, the issue of "how terrible it is that we are locked in to X86" keeps coming-up. |
#5
|
|||
|
|||
I once actually learned something from this group
On Oct 2, 10:02*am, chrisv wrote:
Robert Myers wrote: You're stuck in a mental rut. *Who cares if PC's go 50% faster? * It's price/performance that is the "bottom line". *So, "who cares" if Intel has no serious competition, so are allowed to feed us overpriced, mediocre products? * We care. *The world cares. *Sheesh. In reality it matters to hardly anyone. Those who really need muscle have only one way to go, which is multiple cores. In that case, 50% more cores isn't all that different from 50% better performance from a single core. The problems are elsewhere. * What "problems" are you referring to? *The "problems" you find technically interesting, or the "problems" the real world faces, trying to get the job done in the most efficient manner? The real problems have very little to do with hardware. The hardware that has evolved simply enables the bad habits of the world of computer software. So, no, your insight doesn't impress me. Even were Itanium and NetBurst the worse architectures, we'd be making more fundamental progress by continuing to struggle with the problems they create. Says you. *That's absurd, says me. For a few academics, the challenge of making a k00l-enough compiler to make IA64 kick ass may be very interesting. *Meanwhile, the hundreds of millions of PC users are making *real* progress with great products like the C2D. * You might say "maybe in the long run we'd be better-off with the kool compiler technology". *I would respond "maybe not". *Maybe it's a complete dead-end, like Netburst (which, in hindsight, was a *really* bad idea). *Maybe, when CPU's with 64 cores are being made, IA64 will look to be similarly wrong-headed. The only thing you can say about Itanium at this point is that it didn't work the way it was expected to. We'll probably never really understand why. There was more than one dead-end in the pursuit of the ideas around Itanium. If you never encounter dead-ends though, I say it means you are not being sufficiently adventurous. The chief architect for P4 has said that he expected to backfill the original design into something more respectable but that, yes, it was rushed out only to get the clock rate up. For some really fundamental reasons, running much faster to get the same or only slightly better performance seems on the face of it to be a really, really dumb thing to do, except possibly from the POV of selling chips, at least in the short haul. Better compiler technology is not a dead end. Robert. |
#6
|
|||
|
|||
I once actually learned something from this group
On Oct 2, 5:02*pm, chrisv wrote:
Robert Myers wrote: Well, the issue of "how terrible it is that we are locked in to X86" keeps coming-up. More like locked to Windows. I don't have that x86-lock-in issue with any UNIX / BSD / Linux box I've ever had. It just keeps coming up that the x86 offers a good value, anything comparable on other architectures would be more expensive with less hardware support for peripherals and add-on codec's and other sugar coatings. |
#7
|
|||
|
|||
I once actually learned something from this group
Well, the issue of "how terrible it is that we are locked in to X86"
keeps coming-up. Forgot also that x86 isn't the best selling processor on the market.. just on laptops and desktops, I recall seeing that the ARM and such sell a lot more units. Could be wrong.. anyone cares to contest that? |
#8
|
|||
|
|||
I once actually learned something from this group
On Oct 2, 1:06*pm, aku ankka wrote:
Well, the issue of "how terrible it is that we are locked in to X86" keeps coming-up. Forgot also that x86 isn't the best selling processor on the market.. just on laptops and desktops, I recall seeing that the ARM and such sell a lot more units. Could be wrong.. anyone cares to contest that? If you include embedded processors, I think the 8051 or 8051- compatible chips might (would?) win. Robert. |
#9
|
|||
|
|||
I once actually learned something from this group
On Oct 2, 1:05*pm, aku ankka wrote:
On Oct 2, 5:02*pm, chrisv wrote: Well, the issue of "how terrible it is that we are locked in to X86" keeps coming-up. More like locked to Windows. I don't have that x86-lock-in issue with any UNIX / BSD / Linux box I've ever had. It just keeps coming up that the x86 offers a good value, anything comparable on other architectures would be more expensive with less hardware support for peripherals and add-on codec's and other sugar coatings. In any case, the relevant monopolist is Microsoft, not Intel. Even Microsoft no longer has control, though, as it no longer has the capacity to order the entire world to rewrite everything for a new version of Windows. Robert. |
#10
|
|||
|
|||
I once actually learned something from this group
aku ankka wrote:
Well, the issue of "how terrible it is that we are locked in to X86" keeps coming-up. Forgot also that x86 isn't the best selling processor on the market.. just on laptops and desktops, I recall seeing that the ARM and such sell a lot more units. Could be wrong.. anyone cares to contest that? The ARM is the basis of most cellphones, so just by that platform alone it is the biggest selling chip architecture family. However, that does not mean that it is the architecture with the most applications. X86 architecture has the most, likely followed by the Sparc architecture. For that matter most general purpose processor architectures have more applications than the ARM, because ARM is used with a lot of proprietary platforms, since every cellphone maker wants to be different than their competition, so they use customized OS and apps. Yousuf Khan |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Learned sumthin' new | AJ[_2_] | Homebuilt PC's | 0 | July 3rd 07 07:58 PM |
box killing drives?-What I've learned | Rod Speed | Storage (alternative) | 0 | September 23rd 05 03:22 AM |
Lessons learned: Proliant Memory | VinceV | Compaq Servers | 0 | December 14th 04 05:48 PM |
Learned a hard lesson a few days ago | Fred Smith | Asus Motherboards | 14 | August 2nd 04 04:20 PM |