I hate to break it to you, but AMD is not doing so well. Just 6 months ago, Toms hardware pretty much declared the Athlon dead ("Our extensive tests give the impression that the Athlon design is already a bit outdated and is now reaching its limits."). It took a kludge for them to get up to the 2800, and judging by how long it is taking, they are having problems producing at quantity.
Last I checked, the Athlon XP 2800 was pretty on-par with the P4 2.8 GHz even beating it out on many common benchmarks
And last I checked, the XP 2800 still wasn't in stores (check pricewatch if you don't believe me), and Intel has had a 3 GHz chip with HT on the market for 6 weeks.
The prohibitive cost of P4's, especially the higher end ones, has pretty much kept AMD processors as the choice for home system builders.
If you compare the prices of the highest speed Athlon on pricewatch vs the equivalent performing P4 (in this case, the Athlon 2600 and the 2.6 GHx P4), you might notice that the P4 is actually about $10 cheaper.
As other people have pointed out, you have to remember the timeline of these products. According to AMD's corporate history, they were founded just to take other peoples stuff and try to improve it- they didn't innovate anything. So Intel would release a chip, AMD would fart around with it for a while to try to improve it, and by the time AMD released their 'improved' product, Intel was ready with the next generation. It wasn't until recently that AMD started to branch out a little on their own (with PC processors, that is).
K6 line... beat intel's pentium line in Mhz handily The K6 line never got close to the pentiums in terms of actual performance. And what happened to the "MHz doesn't matter!" drum that AMD has been beating for the past few years?
AMD should still be able to undercut Intel's outrageous pricing
Have you checked prices lately? Lets look at some current chip prices. An Athlon 2600 (yes, the 2800 still is not on the shelves yet) runs for about $280. The 2600 competes roughly with the P4 2.6 GHz (AMD didn't just pull the model number 2600 out of nowhere), and that chip costs about $270- Intel is about $10 cheaper.
chips that have a detailed instruction set, not chips that just do nothing fast
Well, aside from some of the non-standard instruction set extensions that have no effect on 90% of the applications out there, AMD uses the exact same instruction set as Intel. I thought that was pretty obvious, though.
especially if this could hurt Intel and insure that AMD will be able to compete with them for years to come
This is not going to hurt Intel- they have similar technology sharing agreements with IBM already (on top of the billions that they already spend researching manufacturing tech).
Unfortunately, yes- thats how it works here south of your border. And somehow people are trained to think that they are getting some big bonus when the IRS sends their money back.
I did know somebody once that was horrible with saving money, so he had huge deductions taken from his paycheck so he would get a large refund every spring (kind of a forced savings program). In practice it didn't work, though, because he went crazy and spent the whole refund check anyway.
You do realize that if you get a refund when you file your taxes, that basically means you just gave the government an interest free loan, right? The trick is to withhold as little as possible during the year so you either break even or owe a little bit on April 15 (not enough to get an underpayment penalty, of course). Then your money is YOURS throughout the whole year, and you can save it and invest it anyway you like the whole time.
RAMBUS has always and still does offer the best performance combined with a P4. And although this is largely irrelevant now, the price difference has almost disappeared (RAM prices between PC3500 DDR ram and 1066 Mhz RDRAM are within $5, and boards with the Intel i850 chipset are usually about $20-25 more than boards with an 845 series chipset).
Doesnt make a difference now since I think Intel is dropping RDRAM support.
That is true for the next couple of months (no integrated controllers available), but SATA controllers that are integrated into the chipset (due 1H 2003) will most definately not sit on the PCI bus and therefore will not be limited by PCI bandwidth. And look for PCI's replacement (3GIO, or PCI Express, whatever you want to call it) to come around in the next year or so.
Privacy experts said they feared the log file could be used by investigators, divorce lawyers, snooping family members, marketing companies or others interested in learning about a person's entertainment habits.
NO! Anything but that! People can know what my entertainment habits are? What is this world coming to!!!
If you are so worried about people knowing your entertainment habits, then maybe you are watching some stuff that you shouldn't be watching...
Well, in Q1 2002, desktop processors accounted for 76% of AMD's revenue (according to their S&P stock report). Thats a little more than just a "cog in the AMD wheel" if you ask me. Their flash memory products only accounted for about 17% of their revenue during that same time period.
The fact is AMD really isn't that big of a company. I think they have around 13,000 employees (compare with ~80,000 for Intel), and their revenues have been dropping like crazy ($500 million in Q3 of this year, compared to ~$6 billion for Intel). I don't think either AMD or Intel could really survive if they lost thier PC processor revenue. If you want to see a company that could survive losing a major product, look at Motorola or IBM. Now those are truly HUGE companies.
You don't see Itanium chips because they cost a couple of thousand dollars each and are targeted for the high end server market only (Itanium was never meant to compete with Opteron). According to Intel, Itanium 2 is selling according plan right now.
Bull crap. There is nothing wrong with people not voting. All that means is that people are happy enough with the way things are that they don't care enough to want to change anything - it is a pretty good sign that people are happy if you ask me. Deciding not to vote is voicing just as much of an opinion as going to the polls to cast a vote.
That doesnt seem like a big plot hole to me because the 1985 Marty and Jennifer did make it back to 1985 at the end of the movie. Now what confused me was why the future family didn't know the 1985 Marty and Jennifer were there. Did the future family just forget that they had travelled forward in time 20+ years ago? If I went forward in time, I think I would mark it on my calendar to go visit myself on the day that I get there.
when the most user friendly desktop distro is in financial trouble - you scorn them.
No, its a company that has a crappy business plan and is not making any money. A company asking for cash donations so it can survive a little longer is one of the dummest things I have ever heard. Maybe they should rethink their business plan instead(what a strange idea- a business that actually makes money without donations?). You are not helping out the "movement" at all- Linux does not rely on Mandrake for survival. If you want to donate your money, that is fine, but you should realize that by donating you are just throwing your cash down the toilet to enable a crappy business to survive another week.
And you should leave Microsoft, DRM, and all of the other/. buzzwords out of this -- they are not relevant at all to the topic.
Well, there is no federal speed limit law, and it looks like about 11 states currently have speed limits higher than 70 (including Montana). So "unofficial" or not, there is no enforcement.
I think you are missing the point- you increase the pipeline depth so you can increase the clock speeds. Of course it will do less per cycle- that's the definition of a pipeline. But it gives you a much more scaleable design and the end result is better performance. The paper says that given the current timings and branch prediction accuracies, you can increase the performance a lot by playing with the pipeline depth and cache sizes.
It's pretty appearant from comparing P3 to P4 that longer pipelines are only better if you can manage to crank out a factor of speed greater than the factor of increase in pipeline length.
Look at the clock speeds of the P3 (architecture maxed out around 1.5 GHz) and P4 (3 GHz with lots of room to grow). They have "cranked up the speed" by more than enough to compensate for the lower IPC. If you are comparing a 1.5 GHz P3 to a 1.5 GHz P4 then you have missed the whole point entirely.
Longer pipelines need better branch prediction.
Branch prediction is already very good. Are you suggesting that you should slow down the rest of the chip so the less than 5% branches that are mispredicted won't be as expensive? That's a pretty dumb trade-off.
Don't take my word for it... go clock a P4 against another CPU and see how well it performs in sorting with RAMBUS memory
Still, the actual Rambus technology leaves no room for complaint: RDRAM offers a large bandwidth of up to 4.2 GB/s and offers the best performance, particularly when used together with the Intel Pentium 4.... Here, you should keep in mind that the Intel Pentium 4 has a maximum bandwidth of 4.2 GB/s. In the near future, this will reach well over the 3 GHz limit. Only Rambus memory in the form of PC4200 (533 MHz) is capable of taking full advantage of this bandwidth. By using DDR SDRAM, such as DDR266 or DDR333, the bandwidth remains restricted to 2.1 GB/s and 2.7 GB/s, respectively.
With higher latencies and all (which are becoming less significant), the P4 has always and still does perform the best with Rambus.
The bulk of the P4's gains on any other CPU comes from SSE2, 400Mhz FSB, and CPU clock speed. Take those away, and it will be slower per clock cycle than any other CPU (including P3)
Once again, nobody disputes that the P4 is "slower" per clock cycle than other CPUs. That is why it is clocked 1 GHz higher than its competition, and it still has room to grow (current Athlon designs can't go much faster). Your statement is analogous to saying "Take away these extra french fries and larger drink from my super-sized value meal, and its just the same as a regular value meal".
Actually, there is a lot of research about pipeline depths, and here is a paper that calculates the optimal pipeline for x86 to be around 50 stages. In fact, they theorize that you could see up to a 90% increase in performance in the P4 by making the pipeline even deeper. So not everybody thinks that the P4 pipeline is "a bit excessive."
think rambus (bad idea... memory width is a good thing!)
I'm a little confused here- until the past few months, Rambus still offered superior memory bandwidth. It wasn't until DDR333 and higher that SDRAM started to catch up. Rambus didn't lose in the market because of performance.
Itanitium (scrap an entire architechture for one that allows you to disable instructions, so that it is gauranteed that part of the processor won't be used at that point)
That is a pretty strange complaint about Itanium. In fact, I think that it is weird that you even think that is a problem.
They are wanting you to "upgrade" to this new great service, but they make it clear that your email address, mothly price, connection speed, and support stay the same. I was confused how they could consider that an upgrade.
1) He no longer has his mac 2) He is out $3000 3) He failed his finals 4) He spent a bunch of extra money tracking the guy down (plane ticket, PI, etc) 5) He is still happy about it
I'd guess he is a "Glass is half full" kind of guy.
Most of Intel's problem is that they are going to long ends [literally!] to get a high Ghz design while not actually thinking in terms of MIPS.
That is not a problem- it is just a different strategy than AMD. And from the current benchmarks, it is a damn good strategy because they can beat the fastest Athlon by quite a bit right now despite the lower IPC. AMD has had major problems ramping up clock freq (and contrary to what you seem to believe, GHz is a very important factor in performance).
If Intel just scrapped the P4 "long pipeline" route and went to a "many multiple pipeline" route like AMD they'd rock.
I recently read a study from the University of Colorado about the optimal pipeline depth for a CPU with an x86-like instruction set (sorry- I don't have the link handy). Their conclusion? The optimal depth is between 40 and 50 stages- about twice the depth of the P4.
The result is that AMD cpus are cheaper
Thats not why AMD cpus are cheaper- the only reason they are cheaper is because of supply/demand. People are not willing to pay as much for an AMD cpu, so the price is lower.
have few transistors [roughly 7 million less]
I think a lot of that difference is just in the different cache sizes of the chips.
and can take the same power
For a chip with fewer transistors and a slower clock speed, they should use a lot less power. The fact that they are about the same does not speak well for AMD.
So tell me again why AMD chips are soooo much better than Intel?
- Grove said basically the same thing you said- if better insulators or other technologies aren't developed, Moore's Law could become "redundant" in 10 years.
- That said, there are other ways to increase chip performance other than increasing transistor density according to Moore's law. Grove cites a few of them in that article (more efficient transistors, multiple cores, etc). So you will still be able to play the latest Quake in 10 years.
It's not about being "freed" from Microsoft. It's about choice. It's about freedom as in speech.
Nobody has lost any freedoms here. OEMs have the freedom to sell the most popular OS, and you have the freedom to shop elsewhere.
For the most part, manufacturers refuse to offer a computer with Windows because their MS license forbids them from doing so. This is horribly, horribly wrong.
There is nothing wrong with this (assuming you meant without Windows). Just because you don't like it doesnt mean it is wrong, especially when 95+% of the public does like it.
However, MS is the devil for locking them into one-sided agreements and threatening them with the loss of ALL Microsoft licenses if they don't sign.
I love how/.-ers think that they somehow know exactly what goes on in the closed contract negotiations between OEMs and Microsoft.
Slashdot Mirror
I hate to break it to you, but AMD is not doing so well. Just 6 months ago, Toms hardware pretty much declared the Athlon dead ("Our extensive tests give the impression that the Athlon design is already a bit outdated and is now reaching its limits."). It took a kludge for them to get up to the 2800, and judging by how long it is taking, they are having problems producing at quantity.
Last I checked, the Athlon XP 2800 was pretty on-par with the P4 2.8 GHz even beating it out on many common benchmarks
And last I checked, the XP 2800 still wasn't in stores (check pricewatch if you don't believe me), and Intel has had a 3 GHz chip with HT on the market for 6 weeks.
The prohibitive cost of P4's, especially the higher end ones, has pretty much kept AMD processors as the choice for home system builders.
If you compare the prices of the highest speed Athlon on pricewatch vs the equivalent performing P4 (in this case, the Athlon 2600 and the 2.6 GHx P4), you might notice that the P4 is actually about $10 cheaper.
As other people have pointed out, you have to remember the timeline of these products. According to AMD's corporate history, they were founded just to take other peoples stuff and try to improve it- they didn't innovate anything. So Intel would release a chip, AMD would fart around with it for a while to try to improve it, and by the time AMD released their 'improved' product, Intel was ready with the next generation. It wasn't until recently that AMD started to branch out a little on their own (with PC processors, that is).
K6 line... beat intel's pentium line in Mhz handily
The K6 line never got close to the pentiums in terms of actual performance. And what happened to the "MHz doesn't matter!" drum that AMD has been beating for the past few years?
AMD should still be able to undercut Intel's outrageous pricing
Have you checked prices lately? Lets look at some current chip prices. An Athlon 2600 (yes, the 2800 still is not on the shelves yet) runs for about $280. The 2600 competes roughly with the P4 2.6 GHz (AMD didn't just pull the model number 2600 out of nowhere), and that chip costs about $270- Intel is about $10 cheaper.
chips that have a detailed instruction set, not chips that just do nothing fast
Well, aside from some of the non-standard instruction set extensions that have no effect on 90% of the applications out there, AMD uses the exact same instruction set as Intel. I thought that was pretty obvious, though.
especially if this could hurt Intel and insure that AMD will be able to compete with them for years to come
This is not going to hurt Intel- they have similar technology sharing agreements with IBM already (on top of the billions that they already spend researching manufacturing tech).
Sorry to say that, but it is true
No, thats not even close to true. It wasn't until Athlon that AMD could top an Intel processor in performance, and that didn't last very long either.
Unfortunately, yes- thats how it works here south of your border. And somehow people are trained to think that they are getting some big bonus when the IRS sends their money back.
I did know somebody once that was horrible with saving money, so he had huge deductions taken from his paycheck so he would get a large refund every spring (kind of a forced savings program). In practice it didn't work, though, because he went crazy and spent the whole refund check anyway.
like receiving your return in a couple days quick
You do realize that if you get a refund when you file your taxes, that basically means you just gave the government an interest free loan, right? The trick is to withhold as little as possible during the year so you either break even or owe a little bit on April 15 (not enough to get an underpayment penalty, of course). Then your money is YOURS throughout the whole year, and you can save it and invest it anyway you like the whole time.
It's just too expensive and not effective enough.
RAMBUS has always and still does offer the best performance combined with a P4. And although this is largely irrelevant now, the price difference has almost disappeared (RAM prices between PC3500 DDR ram and 1066 Mhz RDRAM are within $5, and boards with the Intel i850 chipset are usually about $20-25 more than boards with an 845 series chipset).
Doesnt make a difference now since I think Intel is dropping RDRAM support.
What? If you don't want other people/companies using your work, then don't make it open source. Otherwise I don't see any room for you to complain.
That is true for the next couple of months (no integrated controllers available), but SATA controllers that are integrated into the chipset (due 1H 2003) will most definately not sit on the PCI bus and therefore will not be limited by PCI bandwidth. And look for PCI's replacement (3GIO, or PCI Express, whatever you want to call it) to come around in the next year or so.
NO! Anything but that! People can know what my entertainment habits are? What is this world coming to!!!
If you are so worried about people knowing your entertainment habits, then maybe you are watching some stuff that you shouldn't be watching...
Well, in Q1 2002, desktop processors accounted for 76% of AMD's revenue (according to their S&P stock report). Thats a little more than just a "cog in the AMD wheel" if you ask me. Their flash memory products only accounted for about 17% of their revenue during that same time period.
The fact is AMD really isn't that big of a company. I think they have around 13,000 employees (compare with ~80,000 for Intel), and their revenues have been dropping like crazy ($500 million in Q3 of this year, compared to ~$6 billion for Intel). I don't think either AMD or Intel could really survive if they lost thier PC processor revenue. If you want to see a company that could survive losing a major product, look at Motorola or IBM. Now those are truly HUGE companies.
You don't see Itanium chips because they cost a couple of thousand dollars each and are targeted for the high end server market only (Itanium was never meant to compete with Opteron). According to Intel, Itanium 2 is selling according plan right now.
Bull crap. There is nothing wrong with people not voting. All that means is that people are happy enough with the way things are that they don't care enough to want to change anything - it is a pretty good sign that people are happy if you ask me. Deciding not to vote is voicing just as much of an opinion as going to the polls to cast a vote.
That doesnt seem like a big plot hole to me because the 1985 Marty and Jennifer did make it back to 1985 at the end of the movie. Now what confused me was why the future family didn't know the 1985 Marty and Jennifer were there. Did the future family just forget that they had travelled forward in time 20+ years ago? If I went forward in time, I think I would mark it on my calendar to go visit myself on the day that I get there.
when the most user friendly desktop distro is in financial trouble - you scorn them.
/. buzzwords out of this -- they are not relevant at all to the topic.
No, its a company that has a crappy business plan and is not making any money. A company asking for cash donations so it can survive a little longer is one of the dummest things I have ever heard. Maybe they should rethink their business plan instead(what a strange idea- a business that actually makes money without donations?). You are not helping out the "movement" at all- Linux does not rely on Mandrake for survival. If you want to donate your money, that is fine, but you should realize that by donating you are just throwing your cash down the toilet to enable a crappy business to survive another week.
And you should leave Microsoft, DRM, and all of the other
Well, there is no federal speed limit law, and it looks like about 11 states currently have speed limits higher than 70 (including Montana). So "unofficial" or not, there is no enforcement.
It's pretty appearant from comparing P3 to P4 that longer pipelines are only better if you can manage to crank out a factor of speed greater than the factor of increase in pipeline length.
Look at the clock speeds of the P3 (architecture maxed out around 1.5 GHz) and P4 (3 GHz with lots of room to grow). They have "cranked up the speed" by more than enough to compensate for the lower IPC. If you are comparing a 1.5 GHz P3 to a 1.5 GHz P4 then you have missed the whole point entirely.
Longer pipelines need better branch prediction.
Branch prediction is already very good. Are you suggesting that you should slow down the rest of the chip so the less than 5% branches that are mispredicted won't be as expensive? That's a pretty dumb trade-off.
Don't take my word for it... go clock a P4 against another CPU and see how well it performs in sorting with RAMBUS memory
Ok- how about this: With higher latencies and all (which are becoming less significant), the P4 has always and still does perform the best with Rambus.
The bulk of the P4's gains on any other CPU comes from SSE2, 400Mhz FSB, and CPU clock speed. Take those away, and it will be slower per clock cycle than any other CPU (including P3)
Once again, nobody disputes that the P4 is "slower" per clock cycle than other CPUs. That is why it is clocked 1 GHz higher than its competition, and it still has room to grow (current Athlon designs can't go much faster). Your statement is analogous to saying "Take away these extra french fries and larger drink from my super-sized value meal, and its just the same as a regular value meal".
to a certain point, but the P4 is a bit excessive
Actually, there is a lot of research about pipeline depths, and here is a paper that calculates the optimal pipeline for x86 to be around 50 stages. In fact, they theorize that you could see up to a 90% increase in performance in the P4 by making the pipeline even deeper. So not everybody thinks that the P4 pipeline is "a bit excessive."
think rambus (bad idea... memory width is a good thing!)
I'm a little confused here- until the past few months, Rambus still offered superior memory bandwidth. It wasn't until DDR333 and higher that SDRAM started to catch up. Rambus didn't lose in the market because of performance.
Itanitium (scrap an entire architechture for one that allows you to disable instructions, so that it is gauranteed that part of the processor won't be used at that point)
That is a pretty strange complaint about Itanium. In fact, I think that it is weird that you even think that is a problem.
They are wanting you to "upgrade" to this new great service, but they make it clear that your email address, mothly price, connection speed, and support stay the same. I was confused how they could consider that an upgrade.
Hmmm- I don't remember. Do you ever say dumb things? That's the most common reason :)
I think its funny that
1) He no longer has his mac
2) He is out $3000
3) He failed his finals
4) He spent a bunch of extra money tracking the guy down (plane ticket, PI, etc)
5) He is still happy about it
I'd guess he is a "Glass is half full" kind of guy.
Most of Intel's problem is that they are going to long ends [literally!] to get a high Ghz design while not actually thinking in terms of MIPS.
That is not a problem- it is just a different strategy than AMD. And from the current benchmarks, it is a damn good strategy because they can beat the fastest Athlon by quite a bit right now despite the lower IPC. AMD has had major problems ramping up clock freq (and contrary to what you seem to believe, GHz is a very important factor in performance).
If Intel just scrapped the P4 "long pipeline" route and went to a "many multiple pipeline" route like AMD they'd rock.
I recently read a study from the University of Colorado about the optimal pipeline depth for a CPU with an x86-like instruction set (sorry- I don't have the link handy). Their conclusion? The optimal depth is between 40 and 50 stages- about twice the depth of the P4.
The result is that AMD cpus are cheaper
Thats not why AMD cpus are cheaper- the only reason they are cheaper is because of supply/demand. People are not willing to pay as much for an AMD cpu, so the price is lower.
have few transistors [roughly 7 million less]
I think a lot of that difference is just in the different cache sizes of the chips.
and can take the same power
For a chip with fewer transistors and a slower clock speed, they should use a lot less power. The fact that they are about the same does not speak well for AMD.
So tell me again why AMD chips are soooo much better than Intel?
Hmmm- I didn't see anywhere in the article where Grove said it is "unsolvable". Lets read what the article actually said:
He said the company' engineers "just can't get rid of" power leakage.
Sounds to me like he is just saying Intel hasn't solved it yet (but neither has anybody else).
A couple of things:
- Grove said basically the same thing you said- if better insulators or other technologies aren't developed, Moore's Law could become "redundant" in 10 years.
- That said, there are other ways to increase chip performance other than increasing transistor density according to Moore's law. Grove cites a few of them in that article (more efficient transistors, multiple cores, etc). So you will still be able to play the latest Quake in 10 years.
Puh-leeese.
/.-ers think that they somehow know exactly what goes on in the closed contract negotiations between OEMs and Microsoft.
It's not about being "freed" from Microsoft. It's about choice. It's about freedom as in speech.
Nobody has lost any freedoms here. OEMs have the freedom to sell the most popular OS, and you have the freedom to shop elsewhere.
For the most part, manufacturers refuse to offer a computer with Windows because their MS license forbids them from doing so. This is horribly, horribly wrong.
There is nothing wrong with this (assuming you meant without Windows). Just because you don't like it doesnt mean it is wrong, especially when 95+% of the public does like it.
However, MS is the devil for locking them into one-sided agreements and threatening them with the loss of ALL Microsoft licenses if they don't sign.
I love how