The hammer line can handle 32-bit apps, even OS's I imagine. The problem with the itanic is that it doesn't. No, Itanium can handle x86 applications as well.
The problem is the current IA-64 processors run x86 very, very slowly. I imagine Intel would simply improve the x86 performance before trying to attack the desktop. It's not that big a deal for the high end server market they're aiming for right now.
And exactly why is this OK to you? Do you like being marketed at? Do you like being fed shite and being told it's ice cream? All CPU companies do it. It's part of the game. If you're not going to have another "speed burst" in the core design for 8 months, you spread out the releases in those 8 months. AMD does it, Intel does it.
If you were not so busy singing the praises of P4 you might also notice that the Tualatin core is overclocking as well as the Athlon, and surpasses them both in some benchmarks. No, the Athlon still wins against the Tualatin due to its triple-x87 FPU and higher FSB. But it is pretty close, you're right. But an overclocked Pentium 4 beats out an overclocked Tualatin so I'm not quite sure what your point is...
Do some damn research before you post or start on about "fanboy blah blah fanboy" while being a fanboy. Sorry, I'll try to refrain from posting anything to defend Intel from trolls who don't know what they're talking about.
You're right, I'm an Intel fanboy. That's precisely why I'm on an Athlon XP system right now.:)
You should do your own research. I've done mine, you've yet to counter it.
You must have meant to show this article [anandtech.com], where it is clearly shown that the VIA P4X266 has twice the memory bandwidth of the crippled i845. Or this page [anandtech.com], that clearly shows the P4X266 outperforming the crippled i845 by 12%, on par with the Intel's RDRAM solution.
Um. Maybe that's because in the benchmarks you're looking at, the i845 is using SDRAM and the P4X266 is using DDR?
Go figure that DDR has twice the bandwidth of SDRAM?;)
You said it yourself, Intel had an existing chipset in june of last year supporting DDR but would not allow motherboard manufacturers to use DDR with it. That means they crippled it themselves to make RDRAM look better. No, it means Intel was under contract with Rambus not to release a DDR solution. That contract expired on Jan 1, 2002.
Their deal did not end Yes, yes it did. That's why there's a DDR 845 now.
Dammit I told you to post Intelligently. Would it kill you to take your own advice?:)
A big misconception about Intel is that it designed the Pentium 4 around marketing.
It has the longer pipeline because the target programs (encoding, multimedia, etc) don't really take a big hit from the longer pipeline stalls, as there's not much branch predicting. So they get a fairly large boost.
You should also check out this article: http://www.anandtech.com/cpu/showdoc.html?i=1584 (PS: Never cite the emulators.com article. It's been debunked so many times now it makes me dizzy!;))
(PPS: The "double pumped" ALU in the Pentium 4 does 2 x 16-bit integers per tick, which equals 1 32-bit. All it needs is to be fed 3)
Intel castrated it their selves. Compare its performance to VIA's P4X266 Chipset's performance and you will see that Intel crippled it to prevent it from competing with Intel's Rambus chipset. Can you explain to me the logic in thinking Intel wants to have everyone use RDRAM? They don't make any money on every RDRAM RIMM sold. They have a contract with RAMBUS, which is why they pushed it. The contract is expired now.
Your entire post is full of inaccurate information and typical anti-Intel garbage. Don't take me as pro-Intel, but anyone can see right through your crap if they looked at it.
You base most of your rant on the fact that Intel's part of this elaborate conspiracy to make RDRAM look way better than DDR. They were, that's true, back when they had a contract with Rambus. Now that expired (Jan 01, 2002). They're releasing dual-channel DDR SDRAM solutions, and made the 845 allow for DDR too.
It's outrageous the parent got modded up like it did. Obviously a lot of blind zealotry going on here as usual.;)
IMHO, i think the post is right in that the comparison was unfair, because a running product is beign compared to a will-be-product. It just ISN'T FAIR. The comparison was unfair, but it's unfair for him to blast the Pentium 4 for being an "inefficient" and overall worse processor, as he was strongly implying.
Second point i think the original poster is right, is that the comparison itself was BIASED, because he is using expensive RAM and doing biased tests running intel optimized stuff. Ie: the test by NATURE is biased. I guess you skipped the section on how the RAM is the same price?:) Also, the majority of those tests are actually more Athlon optimized than Pentium 4. They came out long before the Pentium 4, so are optimized for 12 stage pipelines with strong x87 FPU. The Pentium 4 is much different.
Third point he made, and i think he is right, is that a 10/20% improvement in speed for a 2666 mhz core and 533 mhz RAM is not GOOD ENOUGH to call that the NEXT GENERATION desktop. True, but nobody argued that.:)
It is certainly scaling in clockspeed, but its inefficient design is holding it back in actual performance. So surely the Athlon's inefficient design, as compared with the G4, is holding it back as well?
Not its 8k L1 cache--the same size as that of a 486. Are you aware of how radically different the L1 cache functions between the Pentium 4 and conventional processors?
Note its three execution units, only two of which can possibly be fed by the instruction scheduler at any one time. That's not really true. What are your sources? A lot of the design of the P4 looks weird at first, but was designed to allow for more efficient SMT.
Note its use of FSB bandwidth being terribly inefficient, noteable in that the 3GHz P4 with a 100MHz bus *4 is slower than the 2.666GHz P4 with a 133MHz * 4 bus. Huh? Where do you get that? The reason the two were so radically different was PC1066 RDRAM. The 3GHz P4 used PC800, the 2.666GHz P4 used PC1066. The P4 is very bandwidth-hungry, it was designed to be like that. It has very aggressive prefetching. That's not inefficient, it's different. And again, it looks like it works fine...
The performance figures indicate that the P4 is very highly dependant on a huge amount of FSB bandwidth, and how that the FSB must scale almost linearly at a very early point to be able to keep up. I guess we'll see what happens when the P4 exceeds 3GHz and saturates its 133MHz bus. (Of course, Intel has some damn good engineers so they will surely think of something). You're mistaken from the very beginning of your post because you are talking about efficiency and scaling without any real figures to back it up.
Yes, the P4 is less "efficient" than the Athlon. Yes, the Athlon is less "efficient" than the G4. Does that make the G4 better than the Athlon? Depends.:)
What we're trying to balance is how well it scales with how efficient it is. As it stands today, the P4 is scaling so well it definitely offsets the efficiency for most applications. In other applications, where SMT can be applied, the P4 is not really taking any per-MHz penalty for its huge pipeline compared to the Athlon, and that raw clockspeed comes to a huge advantage.
Wait a sec. So you say it's ok to compare an overclocked P4 to a standard Athlon. Why not OC the Athlon too? It's not that difficult. If you aren't going to have a level playing field, you can write a P4 to K6-3 (with 3d now!). I never said that at all. I said it's stupid to compare clock speeds and then at the same time blast people for comparing clock speeds.;)
As for why the Hz matters, there are several reasons. 1st - a large component of the power disappation in a cmos chip increases with the frequency. Second, there is a maximum Ft for a transistor process. Next, it gets more difficult to design motherboards that work at higher frequencies. Another issue is that higher frequencies are more likely to radiate, causing EMI issues. All of your points are only really valid when the clockspeed reaches 30GHz+, by then the Pentium 4 core will be retired. Why do you think the IA64 chips are so low in clock speed?
As for 'Intel could easily release those 2.6GHz processors today' is a bunch of crap. Remember the P3 1.13GHz processors which were released and then unreleased? I'll bet you don't know the first thing on an IC release procedure. It isn't simply 'it runs at this speed - let's sell it'. First, you need to run a complete fault coverage at that speed to ensure that there is no dynamic problem. Next, you need to modify your testing procedure (and maybe buy lots of new test equipment at $2000000+ each). And then you need to estimate yield. And I net I'm missing a few dozen steps too. That's all fine and good, but it misses the point. The 2.6GHz P4s with Northwood use less power and are cooler than the original 2GHz Pentium 4s. Every 2.2GHz Pentium 4 reviewed that's been overclocked has hit at least 2.6GHz stable, and some have hit > 3.0GHz. Intel could certainly ramp up and release 2.6GHz today, but it doesn't make sense to do so. Your argument basically hinges on having to test to make sure it's stable. They could have tested 2.6GHz as well as 2.2GHz, but they didn't. Why? Because they don't want to skip those valuable steps inbetween. They sell more processors like this, so they keep their CPUs roughly on par with the competition, but with the Intel brand, and sell them accordingly. And it works for them.
So, the bottom line is a slower processor that can equal the performance of a higher frequency processor is usually a 'better' design. Of course, you can say the same thing about power dissapation - and the Athlon really burns a lot of power. You can also look at ceiling frequencies and how well they scale: Compare how well Athlons scale to Pentium 4s. The core is so antiquated in design they are losing IPCs as the clockspeed rises, the Pentium 4's core is scaling much better. Further, it's capable of much faster speeds (not just clockrates, but overall performance) on the same process. In that respect the Pentium 4 is better.
The Athlon is better in some cases, the Pentium 4 is better in others. There is no "better" processor.
AthlonXP 2000+ runs at 1,666MHz at a bus which is the equivalent of 266MHz.
The P4 is running at 2666MHz (a full Gigahertz higher frequency) with a bus at the equivalent of 533MHz. How come so many people rant and rant about how clockspeed isn't everything, then they go and use the same argument in a different way to establish the "clear superiority" of the Athlon? Who cares how many Hz one is than the other? (Don't argue about consumers here, that's for another discussion...). Sorry, but if you're going to paint it as an achievement that the Athlon performs so well 1000MHz slower than the 2.6GHz P4, then why can't the Intel fanboys paint the fact that the P4 runs at 2.6GHz as an achievement?
The (essentiually overclocked) Pentium 4 has a full SIXTY PERCENT CPU clockspeed advantage and a ONE HUNDRED PERCENT front side bus (FSB) advantage, yet look at its real-world performance: "Essentially overclocked" Pentium 4? It's not a new Pentium 4 chip, it's a new motherboard. Of course it's an "essentially overclocked" Pentium 4. Why add in the negative connotations?
I just thought I'd point out that the only conclusion that you can really draw from these tests is that, as many in the hardware community know, the P4's architecture is designed for high clockspeed, with zero regard to actual real-world performance. Which matters more to you? I dunno, looking at these benchmarks I'd say the Pentium 4's architecture is damn fast. It's scaling up incredibly fast. Remember when it was first released and everybody called it a disaster?
Intel could easily release those 2.6GHz chips today, but they aren't doing it for marketing reasons. The architecture of the Pentium 4 is incredibly fast, but the management of the company is spreading out the releases over time. You can get a 2GHz today and overclock it to 2.6GHz. People are doing that all over.
The Athlon is a different design: It's very fast. The Pentium 4 is another design: It's very fast. The Athlon is cheaper, by a fair margin, especially at the highest end chips. But painting the picture that the Pentium 4 is so very much slower than the Athlon, especially with benchmarks like this, are just plain stupid.
The only hardware you'd need is the tuner. And that tuner is almost always built into your digital cable receiver or satellite receiver, which you needed anyway.
The reason why they don't have tuners built into TVs is because there's still no official standard yet, so you wouldn't want to be screwed with an obsolete tuner in your nice HDTV.:)
By default Winamp logs "anonymous usage statistics" unless you turn it off during the install.
You can also turn off WMP's unique identifier thing if you're worried about privacy.
Honestly though, set down your tinfoil hats for a second: Why do we really care? Really?
Maybe it's just me but I honestly don't care if some site logs that I viewed porn from so and so site for so many minutes. Why should I?
I also have very serious doubts that MS would ever sell the information it'd collect from it. The money from that is absolutely tiny and the feedback from the public would be absolutely horrible. What I see instead is a more personalized music service, kind of like Launch.com, where it personalizes and gives you music and movie picks based upon what you watch. Amazon does this too when you're logged in, keeping track of recently viewed items, etc.
WHY? I mean, come on... If you want two processors, shouldnt you have 2 processors in the systems???
Maybe because SMT makes the die 5% bigger, while 2 processors is upwards of 100% bigger? This is where a thing called "cost" comes in.
SMT essentially allows for the CPUs to be used more efficiently. A lot of the time an ALU will sit idle while the FPUs work, and with SMT both can work at the same time on different threads.
What are you talking about? Its already better than Netscape, Its also better than IE at loading pages, its more secure than IE, its more stable than IE, the only thing IE has left is the program loads faster and thats mainly due to it being tied into windows itself.
That's entirely subjective and I wouldn't figure worthy of a +3 mod, but I'll reply anyway.:)
I still don't think Mozilla is better at IE at loading pages. Mozilla strictly adheres to the W3C standards, which is good, except the real world standard these days turns out to be IE. W3C's standard still lacks some pretty basic things like having an image as a background in a table cell, which IE handles fine. Until all of those little nuances are added, I won't agree that Mozilla renders better. Strict compliance to the W3C doesn't mean anything when 90% of the browsers out there are strictly compliant to IE anyway. Mozilla should try to stay IE compatible, as well as being W3C compatible, as much as possible. ie: adding in "background" as an element of the "td" tag. It won't break anything...
More secure than IE: I'll give you this one for now, but I guarantee that if Mozilla ever gets 90% marketshare you'll be funding security bugs in that thing too.
More stable than IE: I'll definitely have to disagree with you on this. I'm using IE6 and I honestly can't remember the last time it crashed on me, whereas Mozilla 0.9.8 crashes a couple of times a week on me still.
IE hasn't really been in active development over the past four years, there's not really much you can add to it compared to what Mozilla has done. So sure, in 4 years Mozilla has passed IE, but Mozilla's will also soon start slowing down once it's finally "acceptable". That should be common sense.
MS doesn't force you to register Windows XP at all. It's a seperate process from activation, and is completely optional.
Activation sends a hash value, which is one-way encrypted, to the MS servers to keep track which CDs have been "turned on" and associates that with the encrypted hash value.
The ONLY way MS could ever find out who activated it would be to go by IP (if you're worried about this, go by phone), or the MAC address in the hash value. But seriously, why do they care?
Somebody didn't read the article
on
Inside Intel
·
· Score: 1
I think Intel does need to jump off it's approach to sales by clock speed. They don't work on that assumption. Do you see the Itanium working at 100GHz?
The reason the Pentium 4 has a longer pipeline is because you need a longer pipeline for SMT to work properly with x86 processors. That, and the chip was designed for multimedia, where most of the time the long pipeline is far more beneficial than a deteriment to performance.
I don't care how many GHz it is, I don't care what its IPC is, what I care about is what the GHz * IPC is.
Maybe instead of constantly worrying about clock speeds they spend more research into being able to add larger amounts of cache or try to achieve one clock cycle access to main memory They already have the technology to add larger amounts of cache. It's just not economical to do so, simply because of all the transistors it takes...
Improve memory speed Last time I checked, Intel didn't design system memory.
Although it does (by far) have the most bandwidth available to the CPU out of any consumer CPU out there.
Make instruction set more efficient (ie make alu more efficient, the intel is no RISC) You mean like ditching x86 and designing IA-64?
reduce production costs Sure, Intel will magically wave a wand and make their production costs drop like a rock. Doesn't work like that.;)
Re:AMD Still Has Upper Hand
on
Inside Intel
·
· Score: 2, Troll
Intel chips, while more commonplace in store-bought computers, still do not measure up to the performance and reliability of AMD. Let's drop the act.:)
Intel CPUs are just as reliable as AMD, if not more. AMD CPUs rely on motherboard logic to stop overheating and subsequent fires/melting, Intel CPU's have them built in. AMD CPUs are FAR more likely to crack/shatter installing HSFs due to how the sockets are designed, too. Your reliability argument is simply nonexistant.
As for performance, there are instances where the Pentium 4 is king, there are instances where the Athlon XP is king. There's no clear performance leader.
Athlons are cheaper, but take some shortcuts. Still damn fast, still damn cheap. Lots of people love that.
Close, but no cigar
on
Inside Intel
·
· Score: 4, Interesting
The 10GHz ALU shown was run at room temperature, and was not actually a Pentium 4 ALU at all. While it is true the Pentium 4's ALUs are double pumped, that's because they're actually 16-bit (16 x 2 = 32-bit, thus double pumped).
The 10GHz ALU was a very limited ALU, not part of any modern processor.
Intel is losing little ground at a time right now Actually, in Q4 2001 Intel gained market share and AMD lost some. But overall in 2001, AMD did gain market share, that's true.
I still think it's because Intel wants to point to AMD and say "See? Competition!".;)
Intel could easily release faster CPUs right now to totally crush the Athlon, but it doesn't make sense to do so.
Intel knew the Merced wasn't a speed demon. An exec actually admitted this in 2000. McKinley is one of the first "real" chips they'll release.
As for your SPEC figures, you could have at least made your post worthwhile by not fudging the numbers to make the Itanium worse than it actually is (although it still isn't very good).
Source: http://www.aceshardware.com/read_news.jsp?id=30000 281
800MHz Itanium
SPECint2000 base: 403 (Your number was 358)
SPECfp2000 base: 701 (Your number was 526)
The McKinley should be much faster. Hell, it's got an 8 stage pipeline instead of 10 stage, an additional 2 integer units (6 total), MUCH more efficient cache, on-die L3 cache, more L2 cache, MUCH faster system bus, etc.
Don't write off an entire architecture because you didn't like how the experimental implementation came out. Itanium was just to get a product out there for IA-64 early adopters to start getting code working on.
I use Shaw (Roger's main competition), and several times now they've called my house and asked me to tone down my bandwidth usage.
I asked them that very question: What's the point of broadband if I can't use it to its full extent?
The license agreement I signed clearly stated there's no bandwidth restrictions for home users, but you can't run servers. I wasn't running any servers, they knew that, and they called me anyway. They actually tried to get me to switch to a business account (more money, bandwidth restrictions), too.
If the ISP can't handle the bandwidth it makes available, it's their loss if people use it too much. It's not my fault I enjoy streaming content and sending movies to friends and all that.:)
First: the study associates third-party software with the operating system, and aggregates all the distributions together into a meaningless "Linux" category. This study is literally just pattern matching against advisories.
If you're including IIS with Windows 2000 bugs, it's only fair to include the third party Linux applications common.
Second: there is no notion of "severity" or "impact" in the study. This is a shame, because SecurityFocus has actually put some real effort into deriving a taxonomy of vulnerabilities from their (enormous) vulnerability database. There is no way to determine whether the N Linux vulnerabilities were equivalent to the K NT vulnerabilities.
The last time that list was released, Unix was still #1.
http://www.sans.org/topten.htm
The actual design of the security in NT and.NET especially are quite excellent, actually.
Check out this interview with Miguel de Icaza (you know, the Gnome guy): http://www.theregister.co.uk/content/4/23919.html
He touches on NT &.NET security:
The I/O model is based on the Win32 model - we have to emulate the behaviour - and it turns out that behaviour is extremely nice. We went and implemented that.
"They have a beautiful security system and we're emulating the whole security infrastructure. It's actually easier to use than the Windows counterpart. We're basically wrapping the Unix functionality inside the Windows functionality.
...
"OK there are two security modeks in place - one is the Windows NTN security model; which is actually a pretty [pauses]... You've seen security holes in Microsoft products - buffer overflows - they're not problems in the security architecture - that happens with Unix too. They happen to be really bad at managing their bugs, and not providing fixes on time, but that's another issue. That's the NT security system.
"But.NET has another security system on top. It's designed to have same sort of security as Java. Instead of the application being the unit of security - you trust the whole app or you don't trust any of it - each portion of the program can be running on a separate trust domain.
"So you can have Gnumeric running and you import a file from 1-2-3. Gnumeric asks the network or service do you have a plug-in for 123 - but the problem is you cannot trust this code, right? So you give it limited permissions. You're not going to let them write to the file, just read, populate the spreadsheet. If they try anything else the plug-in is killed, it throws an exception.
It's a sandbox?
"Exactly - it's a sandbox at any point, and you can define what, which I find really interesting."
Anyway, long story short, MS is pretty good at designing secure code. It's just that the programmers have been exceptionally lazy and left some really common possible exploits in the code. That's what they're going to try to fix up this month.
So if MS doesn't get it, should they follow your advice? Your advice would be for them to rewrite all of Windows... somehow I think you're the one that doesnt get it.
And if you think XP phones-home, wait till you get this baby. It should be a lot easier for Microsoft to control things on YOUR machine via db.
Oh come on. You seriously thinks MS cares about how many gigs of porn you have? Unlike the small spyware companies, MS doesn't need the pathetic income provided by "usage statistics" and whatnot.
They would NOT risk the publicity of them having access to your machine via db remotely, there's almost nothing to gain and everything to lose.
Ahh, now I see. I can just see the high-ups at Microsoft, "Hey, we can't make an RDBMS as good as Oracle or IBM's, so let's make our OS one, then when people run SQL Server on it it will be like 10 times faster, and SQL Server will capture the high-end database market."
I hope many of you submitted feedback for the Tunney act before yesterday's deadline or we will see a lot more anti-competitive behavior over the next year. I dunno, that sounds competitive to me. God forbid they try to compete more with IBM and Oracle.
Slashdot Mirror
The hammer line can handle 32-bit apps, even OS's I imagine. The problem with the itanic is that it doesn't.
No, Itanium can handle x86 applications as well.
The problem is the current IA-64 processors run x86 very, very slowly. I imagine Intel would simply improve the x86 performance before trying to attack the desktop. It's not that big a deal for the high end server market they're aiming for right now.
And exactly why is this OK to you? Do you like being marketed at? Do you like being fed shite and being told it's ice cream?
:)
All CPU companies do it. It's part of the game. If you're not going to have another "speed burst" in the core design for 8 months, you spread out the releases in those 8 months. AMD does it, Intel does it.
If you were not so busy singing the praises of P4 you might also notice that the Tualatin core is overclocking as well as the Athlon, and surpasses them both in some benchmarks.
No, the Athlon still wins against the Tualatin due to its triple-x87 FPU and higher FSB. But it is pretty close, you're right.
But an overclocked Pentium 4 beats out an overclocked Tualatin so I'm not quite sure what your point is...
Do some damn research before you post or start on about "fanboy blah blah fanboy" while being a fanboy.
Sorry, I'll try to refrain from posting anything to defend Intel from trolls who don't know what they're talking about.
You're right, I'm an Intel fanboy. That's precisely why I'm on an Athlon XP system right now.
You should do your own research. I've done mine, you've yet to counter it.
You must have meant to show this article [anandtech.com], where it is clearly shown that the VIA P4X266 has twice the memory bandwidth of the crippled i845. Or this page [anandtech.com], that clearly shows the P4X266 outperforming the crippled i845 by 12%, on par with the Intel's RDRAM solution.
;)
:)
Um. Maybe that's because in the benchmarks you're looking at, the i845 is using SDRAM and the P4X266 is using DDR?
Go figure that DDR has twice the bandwidth of SDRAM?
You said it yourself, Intel had an existing chipset in june of last year supporting DDR but would not allow motherboard manufacturers to use DDR with it. That means they crippled it themselves to make RDRAM look better.
No, it means Intel was under contract with Rambus not to release a DDR solution. That contract expired on Jan 1, 2002.
Their deal did not end
Yes, yes it did. That's why there's a DDR 845 now.
Dammit I told you to post Intelligently.
Would it kill you to take your own advice?
Athlon XP. With the money I'd save I'd buy a GeForce 4. ;)
A big misconception about Intel is that it designed the Pentium 4 around marketing.
;))
It has the longer pipeline because the target programs (encoding, multimedia, etc) don't really take a big hit from the longer pipeline stalls, as there's not much branch predicting. So they get a fairly large boost.
You should also check out this article: http://www.anandtech.com/cpu/showdoc.html?i=1584
(PS: Never cite the emulators.com article. It's been debunked so many times now it makes me dizzy!
(PPS: The "double pumped" ALU in the Pentium 4 does 2 x 16-bit integers per tick, which equals 1 32-bit. All it needs is to be fed 3)
Intel castrated it their selves. Compare its performance to VIA's P4X266 Chipset's performance and you will see that Intel crippled it to prevent it from competing with Intel's Rambus chipset.
;)
Can you explain to me the logic in thinking Intel wants to have everyone use RDRAM? They don't make any money on every RDRAM RIMM sold. They have a contract with RAMBUS, which is why they pushed it. The contract is expired now.
What're you talking about with the performance? It seems to be about the same as all of the other chipsets, but sometimes a bit slower as the Intel chips usually are as they try to go for stability over raw speed.
Your entire post is full of inaccurate information and typical anti-Intel garbage. Don't take me as pro-Intel, but anyone can see right through your crap if they looked at it.
You base most of your rant on the fact that Intel's part of this elaborate conspiracy to make RDRAM look way better than DDR. They were, that's true, back when they had a contract with Rambus. Now that expired (Jan 01, 2002). They're releasing dual-channel DDR SDRAM solutions, and made the 845 allow for DDR too.
It's outrageous the parent got modded up like it did. Obviously a lot of blind zealotry going on here as usual.
IMHO, i think the post is right in that the comparison was unfair, because a running product is beign compared to a will-be-product. It just ISN'T FAIR.
:)
:)
The comparison was unfair, but it's unfair for him to blast the Pentium 4 for being an "inefficient" and overall worse processor, as he was strongly implying.
Second point i think the original poster is right, is that the comparison itself was BIASED, because he is using expensive RAM and doing biased tests running intel optimized stuff. Ie: the test by NATURE is biased.
I guess you skipped the section on how the RAM is the same price?
Also, the majority of those tests are actually more Athlon optimized than Pentium 4. They came out long before the Pentium 4, so are optimized for 12 stage pipelines with strong x87 FPU. The Pentium 4 is much different.
Third point he made, and i think he is right, is that a 10/20% improvement in speed for a 2666 mhz core and 533 mhz RAM is not GOOD ENOUGH to call that the NEXT GENERATION desktop.
True, but nobody argued that.
It is certainly scaling in clockspeed, but its inefficient design is holding it back in actual performance.
:)
So surely the Athlon's inefficient design, as compared with the G4, is holding it back as well?
Not its 8k L1 cache--the same size as that of a 486.
Are you aware of how radically different the L1 cache functions between the Pentium 4 and conventional processors?
Note its three execution units, only two of which can possibly be fed by the instruction scheduler at any one time.
That's not really true. What are your sources?
A lot of the design of the P4 looks weird at first, but was designed to allow for more efficient SMT.
Note its use of FSB bandwidth being terribly inefficient, noteable in that the 3GHz P4 with a 100MHz bus *4 is slower than the 2.666GHz P4 with a 133MHz * 4 bus.
Huh? Where do you get that?
The reason the two were so radically different was PC1066 RDRAM. The 3GHz P4 used PC800, the 2.666GHz P4 used PC1066. The P4 is very bandwidth-hungry, it was designed to be like that. It has very aggressive prefetching. That's not inefficient, it's different. And again, it looks like it works fine...
The performance figures indicate that the P4 is very highly dependant on a huge amount of FSB bandwidth, and how that the FSB must scale almost linearly at a very early point to be able to keep up. I guess we'll see what happens when the P4 exceeds 3GHz and saturates its 133MHz bus. (Of course, Intel has some damn good engineers so they will surely think of something).
You're mistaken from the very beginning of your post because you are talking about efficiency and scaling without any real figures to back it up.
Yes, the P4 is less "efficient" than the Athlon. Yes, the Athlon is less "efficient" than the G4. Does that make the G4 better than the Athlon? Depends.
What we're trying to balance is how well it scales with how efficient it is. As it stands today, the P4 is scaling so well it definitely offsets the efficiency for most applications. In other applications, where SMT can be applied, the P4 is not really taking any per-MHz penalty for its huge pipeline compared to the Athlon, and that raw clockspeed comes to a huge advantage.
Wait a sec. So you say it's ok to compare an overclocked P4 to a standard Athlon. Why not OC the Athlon too? It's not that difficult. If you aren't going to have a level playing field, you can write a P4 to K6-3 (with 3d now!). ;)
I never said that at all.
I said it's stupid to compare clock speeds and then at the same time blast people for comparing clock speeds.
As for why the Hz matters, there are several reasons.
1st - a large component of the power disappation in a cmos chip increases with the frequency. Second, there is a maximum Ft for a transistor process. Next, it gets more difficult to design motherboards that work at higher frequencies. Another issue is that higher frequencies are more likely to radiate, causing EMI issues.
All of your points are only really valid when the clockspeed reaches 30GHz+, by then the Pentium 4 core will be retired. Why do you think the IA64 chips are so low in clock speed?
As for 'Intel could easily release those 2.6GHz processors today' is a bunch of crap. Remember the P3 1.13GHz processors which were released and then unreleased? I'll bet you don't know the first thing on an IC release procedure. It isn't simply 'it runs at this speed - let's sell it'. First, you need to run a complete fault coverage at that speed to ensure that there is no dynamic problem. Next, you need to modify your testing procedure (and maybe buy lots of new test equipment at $2000000+ each). And then you need to estimate yield. And I net I'm missing a few dozen steps too.
That's all fine and good, but it misses the point. The 2.6GHz P4s with Northwood use less power and are cooler than the original 2GHz Pentium 4s. Every 2.2GHz Pentium 4 reviewed that's been overclocked has hit at least 2.6GHz stable, and some have hit > 3.0GHz. Intel could certainly ramp up and release 2.6GHz today, but it doesn't make sense to do so. Your argument basically hinges on having to test to make sure it's stable. They could have tested 2.6GHz as well as 2.2GHz, but they didn't. Why? Because they don't want to skip those valuable steps inbetween. They sell more processors like this, so they keep their CPUs roughly on par with the competition, but with the Intel brand, and sell them accordingly. And it works for them.
So, the bottom line is a slower processor that can equal the performance of a higher frequency processor is usually a 'better' design. Of course, you can say the same thing about power dissapation - and the Athlon really burns a lot of power.
You can also look at ceiling frequencies and how well they scale: Compare how well Athlons scale to Pentium 4s. The core is so antiquated in design they are losing IPCs as the clockspeed rises, the Pentium 4's core is scaling much better. Further, it's capable of much faster speeds (not just clockrates, but overall performance) on the same process. In that respect the Pentium 4 is better.
The Athlon is better in some cases, the Pentium 4 is better in others. There is no "better" processor.
AthlonXP 2000+ runs at 1,666MHz at a bus which is the equivalent of 266MHz.
The P4 is running at 2666MHz (a full Gigahertz higher frequency) with a bus at the equivalent of 533MHz.
How come so many people rant and rant about how clockspeed isn't everything, then they go and use the same argument in a different way to establish the "clear superiority" of the Athlon? Who cares how many Hz one is than the other? (Don't argue about consumers here, that's for another discussion...).
Sorry, but if you're going to paint it as an achievement that the Athlon performs so well 1000MHz slower than the 2.6GHz P4, then why can't the Intel fanboys paint the fact that the P4 runs at 2.6GHz as an achievement?
The (essentiually overclocked) Pentium 4 has a full SIXTY PERCENT CPU clockspeed advantage and a ONE HUNDRED PERCENT front side bus (FSB) advantage, yet look at its real-world performance:
"Essentially overclocked" Pentium 4? It's not a new Pentium 4 chip, it's a new motherboard. Of course it's an "essentially overclocked" Pentium 4. Why add in the negative connotations?
I just thought I'd point out that the only conclusion that you can really draw from these tests is that, as many in the hardware community know, the P4's architecture is designed for high clockspeed, with zero regard to actual real-world performance. Which matters more to you?
I dunno, looking at these benchmarks I'd say the Pentium 4's architecture is damn fast. It's scaling up incredibly fast. Remember when it was first released and everybody called it a disaster?
Intel could easily release those 2.6GHz chips today, but they aren't doing it for marketing reasons. The architecture of the Pentium 4 is incredibly fast, but the management of the company is spreading out the releases over time. You can get a 2GHz today and overclock it to 2.6GHz. People are doing that all over.
The Athlon is a different design: It's very fast. The Pentium 4 is another design: It's very fast. The Athlon is cheaper, by a fair margin, especially at the highest end chips. But painting the picture that the Pentium 4 is so very much slower than the Athlon, especially with benchmarks like this, are just plain stupid.
It's not as bad as you make it sound, actually.
:)
The only hardware you'd need is the tuner. And that tuner is almost always built into your digital cable receiver or satellite receiver, which you needed anyway.
The reason why they don't have tuners built into TVs is because there's still no official standard yet, so you wouldn't want to be screwed with an obsolete tuner in your nice HDTV.
By default Winamp logs "anonymous usage statistics" unless you turn it off during the install.
You can also turn off WMP's unique identifier thing if you're worried about privacy.
Honestly though, set down your tinfoil hats for a second: Why do we really care?
Really?
Maybe it's just me but I honestly don't care if some site logs that I viewed porn from so and so site for so many minutes. Why should I?
I also have very serious doubts that MS would ever sell the information it'd collect from it. The money from that is absolutely tiny and the feedback from the public would be absolutely horrible. What I see instead is a more personalized music service, kind of like Launch.com, where it personalizes and gives you music and movie picks based upon what you watch. Amazon does this too when you're logged in, keeping track of recently viewed items, etc.
WHY? I mean, come on... If you want two processors, shouldnt you have 2 processors in the systems???
Maybe because SMT makes the die 5% bigger, while 2 processors is upwards of 100% bigger? This is where a thing called "cost" comes in.
SMT essentially allows for the CPUs to be used more efficiently. A lot of the time an ALU will sit idle while the FPUs work, and with SMT both can work at the same time on different threads.
What are you talking about? Its already better than Netscape, Its also better than IE at loading pages, its more secure than IE, its more stable than IE, the only thing IE has left is the program loads faster and thats mainly due to it being tied into windows itself.
:)
That's entirely subjective and I wouldn't figure worthy of a +3 mod, but I'll reply anyway.
I still don't think Mozilla is better at IE at loading pages. Mozilla strictly adheres to the W3C standards, which is good, except the real world standard these days turns out to be IE. W3C's standard still lacks some pretty basic things like having an image as a background in a table cell, which IE handles fine. Until all of those little nuances are added, I won't agree that Mozilla renders better. Strict compliance to the W3C doesn't mean anything when 90% of the browsers out there are strictly compliant to IE anyway. Mozilla should try to stay IE compatible, as well as being W3C compatible, as much as possible. ie: adding in "background" as an element of the "td" tag. It won't break anything...
More secure than IE: I'll give you this one for now, but I guarantee that if Mozilla ever gets 90% marketshare you'll be funding security bugs in that thing too.
More stable than IE: I'll definitely have to disagree with you on this. I'm using IE6 and I honestly can't remember the last time it crashed on me, whereas Mozilla 0.9.8 crashes a couple of times a week on me still.
IE hasn't really been in active development over the past four years, there's not really much you can add to it compared to what Mozilla has done. So sure, in 4 years Mozilla has passed IE, but Mozilla's will also soon start slowing down once it's finally "acceptable". That should be common sense.
How is the parent post +4?
He links to an obsolete article from Q3 2000 about RDRAM on the Pentium III...
He talks about the "insanely high latency", and it's pretty obvious he's exaggerating slightly.
RDRAM's latency, particularly with the upcoming PC1066, is far better than people give it credit for. See this AcesHardware article.
PC1066 RDRAM latency for 128 bytes: 207 cycles
PC800 RDRAM latency for 128 bytes: 247 cycles
PC133 SDRAM latency for 128 bytes: 229 cycles
Slashdot moderators: Would it kill you to check the links before going points-crazy?
MS doesn't force you to register Windows XP at all. It's a seperate process from activation, and is completely optional.
Activation sends a hash value, which is one-way encrypted, to the MS servers to keep track which CDs have been "turned on" and associates that with the encrypted hash value.
The ONLY way MS could ever find out who activated it would be to go by IP (if you're worried about this, go by phone), or the MAC address in the hash value. But seriously, why do they care?
I think Intel does need to jump off it's approach to sales by clock speed.
;)
They don't work on that assumption. Do you see the Itanium working at 100GHz?
The reason the Pentium 4 has a longer pipeline is because you need a longer pipeline for SMT to work properly with x86 processors. That, and the chip was designed for multimedia, where most of the time the long pipeline is far more beneficial than a deteriment to performance.
I don't care how many GHz it is, I don't care what its IPC is, what I care about is what the GHz * IPC is.
Maybe instead of constantly worrying about clock speeds they spend more research into being able to add larger amounts of cache or try to achieve one clock cycle access to main memory
They already have the technology to add larger amounts of cache. It's just not economical to do so, simply because of all the transistors it takes...
Improve memory speed
Last time I checked, Intel didn't design system memory.
Although it does (by far) have the most bandwidth available to the CPU out of any consumer CPU out there.
Make instruction set more efficient (ie make alu more efficient, the intel is no RISC)
You mean like ditching x86 and designing IA-64?
reduce production costs
Sure, Intel will magically wave a wand and make their production costs drop like a rock. Doesn't work like that.
Intel chips, while more commonplace in store-bought computers, still do not measure up to the performance and reliability of AMD. :)
Let's drop the act.
Intel CPUs are just as reliable as AMD, if not more. AMD CPUs rely on motherboard logic to stop overheating and subsequent fires/melting, Intel CPU's have them built in. AMD CPUs are FAR more likely to crack/shatter installing HSFs due to how the sockets are designed, too. Your reliability argument is simply nonexistant.
As for performance, there are instances where the Pentium 4 is king, there are instances where the Athlon XP is king. There's no clear performance leader.
Athlons are cheaper, but take some shortcuts. Still damn fast, still damn cheap. Lots of people love that.
The 10GHz ALU shown was run at room temperature, and was not actually a Pentium 4 ALU at all. While it is true the Pentium 4's ALUs are double pumped, that's because they're actually 16-bit (16 x 2 = 32-bit, thus double pumped).
;)
The 10GHz ALU was a very limited ALU, not part of any modern processor.
Intel is losing little ground at a time right now
Actually, in Q4 2001 Intel gained market share and AMD lost some. But overall in 2001, AMD did gain market share, that's true.
I still think it's because Intel wants to point to AMD and say "See? Competition!".
Intel could easily release faster CPUs right now to totally crush the Athlon, but it doesn't make sense to do so.
Intel knew the Merced wasn't a speed demon. An exec actually admitted this in 2000. McKinley is one of the first "real" chips they'll release.
0 281
As for your SPEC figures, you could have at least made your post worthwhile by not fudging the numbers to make the Itanium worse than it actually is (although it still isn't very good).
Source: http://www.aceshardware.com/read_news.jsp?id=3000
800MHz Itanium
SPECint2000 base: 403 (Your number was 358)
SPECfp2000 base: 701 (Your number was 526)
The McKinley should be much faster. Hell, it's got an 8 stage pipeline instead of 10 stage, an additional 2 integer units (6 total), MUCH more efficient cache, on-die L3 cache, more L2 cache, MUCH faster system bus, etc.
Don't write off an entire architecture because you didn't like how the experimental implementation came out. Itanium was just to get a product out there for IA-64 early adopters to start getting code working on.
I use Shaw (Roger's main competition), and several times now they've called my house and asked me to tone down my bandwidth usage.
:)
I asked them that very question: What's the point of broadband if I can't use it to its full extent?
The license agreement I signed clearly stated there's no bandwidth restrictions for home users, but you can't run servers. I wasn't running any servers, they knew that, and they called me anyway. They actually tried to get me to switch to a business account (more money, bandwidth restrictions), too.
If the ISP can't handle the bandwidth it makes available, it's their loss if people use it too much. It's not my fault I enjoy streaming content and sending movies to friends and all that.
If you're including IIS with Windows 2000 bugs, it's only fair to include the third party Linux applications common.
The last time that list was released, Unix was still #1.
http://www.sans.org/topten.htm
Check out this interview with Miguel de Icaza (you know, the Gnome guy): http://www.theregister.co.uk/content/4/23919.html
He touches on NT &
Anyway, long story short, MS is pretty good at designing secure code. It's just that the programmers have been exceptionally lazy and left some really common possible exploits in the code. That's what they're going to try to fix up this month.
So if MS doesn't get it, should they follow your advice? Your advice would be for them to rewrite all of Windows... somehow I think you're the one that doesnt get it.
And if you think XP phones-home, wait till you get this baby. It should be a lot easier for Microsoft to control things on YOUR machine via db.
Oh come on. You seriously thinks MS cares about how many gigs of porn you have? Unlike the small spyware companies, MS doesn't need the pathetic income provided by "usage statistics" and whatnot.
They would NOT risk the publicity of them having access to your machine via db remotely, there's almost nothing to gain and everything to lose.
How much did you pay for your tinfoil hat?
Ahh, now I see. I can just see the high-ups at Microsoft, "Hey, we can't make an RDBMS as good as Oracle or IBM's, so let's make our OS one, then when people run SQL Server on it it will be like 10 times faster, and SQL Server will capture the high-end database market."
I hope many of you submitted feedback for the Tunney act before yesterday's deadline or we will see a lot more anti-competitive behavior over the next year.
I dunno, that sounds competitive to me. God forbid they try to compete more with IBM and Oracle.