wadetemp writes: "CNET has a story with details on the release of Intel's newest chips. Supposedly 1.4Ghz P4 will be out in time for the holidays, with numerous other models at some random date in the near or far future." And just when I finally got a P3!
"New, from Intel. The [something] will be coming out in X months and will be way faster than [competitor's chip]. Not to mention the other versions of [something] that will be [superlative] and really [synonym for kick-ass]. We really will ship it when we say. No, it's not just another turn of the x86 crank, it's really new architecture [sweat]. Please, just leave us alone, we haven't had competition for 15 years, and now this is starting to wear on us. Yes, it's the [oldchip +1], but really, it's all sorts of new and different. No our coppermines don't actually use copper, but we really thought we were going to... don't be mad... Dave? What are you doing, Dave?....... Daisy... Daisy...."
--
--
"It's tough to be bilingual when you get hit in the head."
All processor upgrades seem to give you very little performance gain compared to what you expect them to give you because the CPU is not the major performance bottleneck of the machine. The biggest offender is the hard disk drive which explains why things run so slow when you have to rely on virtual memory. The amount of idle time my distributed.net client gets on a 266MHz machine when in normal use is amazing, you may need faster machines to play games on but for normal use (at least in Linux) it still performs well. So should I buy one? Perhaps if I want better distributed.net stats but that's about it. We need something more than increased clock speed to make it worthh upgrading.
Maybe I'm just too tired to be reading these sorts of things.. perhaps I feel like pissing people off with my worthless commentary, but either way I think this can all be summarized as:
Intel announces plans for faster chips in the next 6 months, even faster next year.
yeah, the 1.4GHz will be out, but to Dell and Compaq. Curse me if i ever buy from them...
we've had enough of Intel "releasing" processors to two-three vendors, and we're tired of it.
just got my AMD Athlon 1GHz last night.
tell me when Intel can get me a GHz for less than $2000...
The real message (by omission from the C|Net article) is that Intel has no clue when their 64 bit chip will be ready.
The other real message (again by omission) is that Intel won't or can't tell you their forthcoming chips' interface plans. Are we talking Slot 1? Slot 2? PGA? Flip chip (ugh!).
So, if you want to buy a computer now that you can upgrade when the P4s and beyond come out, good fscking luck guessing which interface, which memory type, which bus speed, etc. The main thing we're getting out of Intel's "diversification" in CPU interfaces, cache, bus speed and memory types is screwed out of the ability to just upgrade the CPU when a new one comes out. Instead, we're mostly stuck needing at least a new mobo to go with it, and (while we're at it), why not just get a whole new system?
The delay in shipping Intel's next processor will double every six months.
--
-- Sheesh, evil *and* a jerk. -- Jade
yeah, they do these things to steal from you...
by
Anaplexian
·
· Score: 4
It's nothing but downright robbery.
In the old days you had a 386, and a major leap to the 486. Now every other day they release a processor only *Marginally* Better than the one before. and Spend huge amounts on marketing and ads, and you have people switching to the newer one. I mean 1GHz, 1.2 or 1.4; how much difference is it going to make to anyone who uses only the wordprocessors?
Some thing similar to the Cars in the 70's and 80's, where the Big Three spent a lot of money in Style than in Safety.
Hope people come to their senses and act sensibly.
"What looks like a good thing, might just be a Beta Version."
[More stuff at iotaspace.net]
There is a difference between the x86 architecture (actual silicon), and the x86 ISA (more like an API).
The actually technology behind x86 processors has moved one, and updated. Transmeta is, prehaps, the most extreme example of this, but all modern processors use microcode to 'emulate' x86 (Or, at least, the least commonly used instructions).
So, what has happened is that the 'good' instructions have got faster, but the old cruft, whilst it still works, is slow. Like, I could probably write out some Z80 machine code, and expect it to work on a PIII. But not to maximum efficency.
Slashdot Mirror
That was my thought when I read it.
"New, from Intel. The [something] will be coming out in X months and will be way faster than [competitor's chip]. Not to mention the other versions of [something] that will be [superlative] and really [synonym for kick-ass]. We really will ship it when we say. No, it's not just another turn of the x86 crank, it's really new architecture [sweat]. Please, just leave us alone, we haven't had competition for 15 years, and now this is starting to wear on us. Yes, it's the [oldchip +1], but really, it's all sorts of new and different. No our coppermines don't actually use copper, but we really thought we were going to... don't be mad... Dave? What are you doing, Dave?....... Daisy... Daisy...."
--
"It's tough to be bilingual when you get hit in the head."
All processor upgrades seem to give you very little performance gain compared to what you expect them to give you because the CPU is not the major performance bottleneck of the machine. The biggest offender is the hard disk drive which explains why things run so slow when you have to rely on virtual memory. The amount of idle time my distributed.net client gets on a 266MHz machine when in normal use is amazing, you may need faster machines to play games on but for normal use (at least in Linux) it still performs well. So should I buy one? Perhaps if I want better distributed.net stats but that's about it. We need something more than increased clock speed to make it worthh upgrading.
Maybe I'm just too tired to be reading these sorts of things.. perhaps I feel like pissing people off with my worthless commentary, but either way I think this can all be summarized as:
Intel announces plans for faster chips in the next 6 months, even faster next year.
yeah, the 1.4GHz will be out, but to Dell and Compaq. Curse me if i ever buy from them... we've had enough of Intel "releasing" processors to two-three vendors, and we're tired of it. just got my AMD Athlon 1GHz last night. tell me when Intel can get me a GHz for less than $2000...
The real message (by omission from the C|Net article) is that Intel has no clue when their 64 bit chip will be ready.
The other real message (again by omission) is that Intel won't or can't tell you their forthcoming chips' interface plans. Are we talking Slot 1? Slot 2? PGA? Flip chip (ugh!).
So, if you want to buy a computer now that you can upgrade when the P4s and beyond come out, good fscking luck guessing which interface, which memory type, which bus speed, etc. The main thing we're getting out of Intel's "diversification" in CPU interfaces, cache, bus speed and memory types is screwed out of the ability to just upgrade the CPU when a new one comes out. Instead, we're mostly stuck needing at least a new mobo to go with it, and (while we're at it), why not just get a whole new system?
The delay in shipping Intel's next processor will double every six months.
--
Sheesh, evil *and* a jerk. -- Jade
It's nothing but downright robbery. In the old days you had a 386, and a major leap to the 486. Now every other day they release a processor only *Marginally* Better than the one before. and Spend huge amounts on marketing and ads, and you have people switching to the newer one. I mean 1GHz, 1.2 or 1.4; how much difference is it going to make to anyone who uses only the wordprocessors? Some thing similar to the Cars in the 70's and 80's, where the Big Three spent a lot of money in Style than in Safety. Hope people come to their senses and act sensibly. "What looks like a good thing, might just be a Beta Version." [More stuff at iotaspace.net]
Not really.
There is a difference between the x86 architecture (actual silicon), and the x86 ISA (more like an API).
The actually technology behind x86 processors has moved one, and updated. Transmeta is, prehaps, the most extreme example of this, but all modern processors use microcode to 'emulate' x86 (Or, at least, the least commonly used instructions).
So, what has happened is that the 'good' instructions have got faster, but the old cruft, whilst it still works, is slow. Like, I could probably write out some Z80 machine code, and expect it to work on a PIII. But not to maximum efficency.
This is the curse of bakwards compatability.
Ars Technica have a review of this here.
A new architecture could probably do many things better. But would it be sufficently better to make the cost of getting _everything_ rewritten?
Market forces suggest not. (See the Alpha - newer architecture, but not exactly everywhere).