Does RAM have anything to do with the proccessor or MB?

[alternatefutures]

But you're making the mistake that I'm arguing against, and that is Efficiency != Superiority. A Honda Civic hybrid is more efficient than an M-1 Abrams tank, does that make it superior to the tank? You need to factor in purpose when you look at architecture, otherwise how do you know what you're supposed to build? The P4 has the superior architecture for multimedia, the Athlon has the superior architecture for scientific calculations. If you fall outside of these two camps, it doesn't really matter as you don't need all that power just to check your e-mail.

[dwchang]

But you're making the mistake that I'm arguing against, and that is Efficiency != Superiority. A Honda Civic hybrid is more efficient than an M-1 Abrams tank, does that make it superior to the tank? You need to factor in purpose when you look at architecture, otherwise how do you know what you're supposed to build? The P4 has the superior architecture for multimedia, the Athlon has the superior architecture for scientific calculations. If you fall outside of these two camps, it doesn't really matter as you don't need all that power just to check your e-mail.

No, I see what you're getting at, but even within these specific instances, Efficiency ON AVERAGE will determine how well you perform on these programs as well. Most programs use the same set of instructions and thus if you are EFFICIENT within these confines, you are probably efficient with these programs. Obviously specific programs that make use of SSE2 and whatnot will perform better on a P4 than an Athlon, but for the most part, most programs are built around the standard x86 library and thus efficiency within these confines is superiority in both design and use. I hope that makes sense.

I do in fact acknowledge that either is superior within specific programs like say CAD tools or Quake3 (FPU), but at the same time...like I said for the MOST part, efficiency with the standard x86 instruction set will generally equate to superior performance (per frequency...as stated earlier about speed) within MOST programs. Therefore, when I say superior, I mean ON AVERAGE. There is always specific instances where one is superior (as stated).

Again I hope this all makes sense...I am basically going off averages and majority of programs/code.

[alternatefutures]

The problem with going by most programs in the x86 instruction set is that most of those don't need the power. Honestly, I don't care if a CPU can run Word a thousand times faster than the PIII I'm writing this on as I am the bottleneck, not the CPU. That leaves benchmarks (I ignore synthetics, just real world), and honestly, since the 2.8Ghz the P4 has been gaining ground over the Athlon in terms of number of benchmarks "won"(for the record, I give no credibility to any benchmarks coming from AMD or Intel, only third parties), which, by your rating would make Northwood superior to Barton, even though it is less efficient.

[dwchang]

The problem with going by most programs in the x86 instruction set is that most of those don't need the power. Honestly, I don't care if a CPU can run Word a thousand times faster than the PIII I'm writing this on as I am the bottleneck, not the CPU. That leaves benchmarks (I ignore synthetics, just real world), and honestly, since the 2.8Ghz the P4 has been gaining ground over the Athlon in terms of number of benchmarks "won"(for the record, I give no credibility to any benchmarks coming from AMD or Intel, only third parties), which, by your rating would make Northwood superior to Barton, even though it is less efficient.

Well say what you want about power, but I sure as hell use it as does anyone who makes AMVs or plays 3D games. If you want just a CPU that can handle mail, word and so on...go buy something else. There's no reason to even argue if it's just about practicality. Hell under your argument, if a CPU can run Word at "your speed", that means it's sufficient since it well...runs right? Maybe that's correct, but at the same time that doesn't mean one wouldn't WANT for it to be able to do more...hell there's things like multi-tasking, more complicated programs and so on. If you want a computer that can run Word or e-mail at your speed, go buy a PI 60 Mhz computer. Now I know that's not what you mean, but you are basically downplaying that people want performance. I do care if Word can run 1000x faster than PI since it means I can do other things and so on.

Now that's the not the meat of my 'argument.' I do agree with you on the benchmarks and when either of us do it since well..it's obviously biased. However, I don't even trust third-party benchmarks. I didn't wanna get into this, but the "trusted" benchmarks like SysMark ARE Intel. Most consumers THINK they're a 3rd party...they're not. I don't need to remind you that these benchmarks are the ones that are called "official" and that many magazines use. Make no mistake when I say they're intel...there is a 12 person board that determines what tests are done, how many times and so on...all 12 members are from Intel and even funnier..the mailing address for Sysmark is that of Intel's.

Now if by third party you mean say...a PC magazine doing their OWN benchmarks that THEY wrote...yeah I would trust those since they are REAL performance. I don't trust SysMark are almost all the "official" benchmarks since all of them are run by or have a majority of Intel workers. Want evidence? From 2000 -> 2002, if you look at the changes in the benchmark, you'll notice that in 2000 the Athlon clobbered the Pentium in benchmarks and as the years progressed it started losing. Now I do acknowledge that a large portion of this is well...they got faster and so on, but at the same time, it is a known fact that the benchmarks themselves changed. Look at the changes from 2000 to 2002 and you'll see that *every* test the athlon outperformed the Pentium on were either taken out or done less and the few things the Pentium did better on were increased.

Now the only reason this is REALLY bad is that...the jobs the Pentium did better in were not applicable in the real world and the things taken out that the Athlon did better on were. Things like....the P4 can handle 30+ layers in Premiere better than an Athlon, but an Athlon could run Premiere better with say 6 video layers. You as video editors know which is more realistic. So...therefore, the Premiere benchmark had TONS of 30+ layer stuff and almost none of the Athlon's stuff. Another example is that the Pentium can sort a GIANT excel sheet faster than the Athlon. The Athlon can in turn sort a reasonabe spreadsheet faster. In the 2002 Sysmark Benchmark, the reasonable sort was taken out and the GIANT one added and given more weight.

Now you tell me with these changes if you can "trust" SysMark or any other "official" benchmark. The tests have been changed to favor everything the Pentium does better and everything AMD does better has been taken out.

Now again, I *do* acknowledge that there have been improvements in architecture and that can easily be a large reason. However, I am only stating facts since this DID occur and one can infer that it's a reason for some of the benchmark skewing. And, no, I'm not making this up...I believe some HP and Compaq customers are even sueing intel over this and AMD even tried to do something and we got a whopping 1 person (out of 12) on the board. It was all over Anandtech and ExtremeTech (I think) when we released our papers on it (to garner support). So for the most part, I am only regurgitation what happened almost a year ago.

Wow...reading over this...I'm not at all going into the argument and am only bashing the benchmarks...whatever....knowledge is power and this is something where Intel did us VERY wrong. Can anyone say Monopoly (oh yeah did you know Intel pays companies NOT to buy AMD....that's also monopolistic).

Wow you sure got me in a fervent mood to bash Intel .

Anyone since I wanna TRY and resemble some degree of non-bias, I will admit that even if the P4's architecture is more inefficient, the reason it can win in benchmarks IS the speed. This is in reply to your comment about superior to barton and whatnot. Basically, if you remember the equation about performance = speed * efficiency (instructions/clock)...if one raises speed, one also raises the performance. Basically Intel is "faster" but has a more inefficient core. AMD is slower, but has a more efficient core. If Intel continues to increase in speed while AMD does it slower...Intel would eventually equal or even surpass the performance of the Athlon. It makes sense mathematically.

This is just a theoretical...not fact...just to illustrate:

AMD
Performance = 2.25 Ghz * 3 Instructions/2.25 Ghz clock

Intel
Performance = 3.06 Ghz * 2 Instructions/3.06 Ghz clock

Well...even though Intel operates at 2 instructions per clock, the clock is faster and thus they do those 2 instructions more often while the AMD does 3 instructions which is more efficient, but at a slower core. Now I'm not sure the exact mathematics (which I guess I could do...and again these are NOT the real #s), but...you can see that with increase of frequency, you can increase performance and hence, one of the reasons for Intel closing the gap and even surpassing it (obviously the benchmark tweaking too).

Anyway...sorry for that LONG post of which 90% is irrelevant, but I felt like posting it and perhaps someone will read it and understand why I dislike Intel's practices AND design.

[kthulhu]

But you're making the mistake that I'm arguing against, and that is Efficiency != Superiority. A Honda Civic hybrid is more efficient than an M-1 Abrams tank, does that make it superior to the tank?

Now, are we talking about a riced out Honda, or stock from the factory?

Anyhow, for me, performance is less of an issue than price, which is why I have a Duron 850, and also why our house is only AMD. The price is good, and the performance is an extra, of sorts.

[alternatefutures]

Well say what you want about power, but I sure as hell use it as does anyone who makes AMVs or plays 3D games. If you want just a CPU that can handle mail, word and so on...go buy something else. There's no reason to even argue if it's just about practicality. Hell under your argument, if a CPU can run Word at "your speed", that means it's sufficient since it well...runs right? Maybe that's correct, but at the same time that doesn't mean one wouldn't WANT for it to be able to do more...hell there's things like multi-tasking, more complicated programs and so on. If you want a computer that can run Word or e-mail at your speed, go buy a PI 60 Mhz computer. Now I know that's not what you mean, but you are basically downplaying that people want performance. I do care if Word can run 1000x faster than PI since it means I can do other things and so on.

You're missunderstanding what I am saying. The comes a point when just seeking speed in the x86 arena results in diminishing returns overall, and it begins to make more sense to choose a specific area that NEEDs that speed and skew your performance to that. If tomorrow Intel made a processor that ran Word at the speed of a PI yet due to that change it also poored more effort into multimedia apps than a current P4 could offer, I'd hardly scoff at it. Essentially that's what Intel started doing with MMX (alright, so it didn't sacrifice performance, but you get the idea)

Back to benchmarks, you'll note that I narrowed myself down to 3rd party REAL WORLD benchmarks. So, these are people who go out and run 3DStudio Max, Photoshop, AutoCAD, stuff that a good deal of power users actually use, and then measure the performance. Games, yeah, sure, but the video card is more important there (Microsoft doesn't want me running Freelancer on my PIII 550, yet my video card lets that machine play it nice and smooth with all the effects turned on) so I really don't care. So, yeah I don't care about Sysmark, or Sandra. Generally if a benchmark looks too skewed I move on.

[alternatefutures]

Oh, kthulhu, it's a Civic hybrid, so no, it's not riced out.

[kthulhu]

Ahh, ok. Because if it was riced out, I think the tank would have the advantage. The spoiler and giant exhaust pipe and Type R stickers add a lot of weight, you know.

Anyhow, my view is, buy what you want. This is America, damn it, and the reason we beat back the heathen, godless commies is because of choice. We can buy a million different types of shoes in all colors, instead of drab, gray ones that smell like fish and potatoes. If you're informed about what you buy, more power to you. If you're not, it's your own damn fault if you get screwed. Take it back, get an exchange or refund, and be glad the store will let you do so. This applies to everything from fast food to cars to clocks to drugs to hookers!

[the Black Monarch]

Significantly overclocked Pentiums and Athlons can be maintained at near freezing and still be so unstable you can't boot to Windows. Cooling is not the only limit to a chip's design

Oh. No one told me that until now. Thanks.

That would be the MOBILE Pentium 4, which is a lower voltage P4. The desktop version wouldn't be in a laptop unless there was a market for a laptop/heating blanket

No dude, I really did mean the DESKTOP version. All 15 volts. Have you ever been to the Alienware website to look at their laptops? They don't offer the mobile P4s anymore. The Area 51-m is well known for its extremely short battery life (75 mins doing nothing, 30 if you're playing music in Windows Media Player) and large weight and volume because of the hugeass cooling system.

I might also add that the backbone of the Pentium-M architecture was the PIII

The newspaper said that the Centrino used a completely new architecture...

At the same time, one will probably wonder why is Intel clobbering us? That answer is easy....MOST consumers equate bigger numbers with better performance. This is due to Intel's marketing department and even I will say that they do a GREAT job in convincing the consumer that Ghz = performance

Umm... no... Intel has done nothing of the sort. Their commercials make absolutely no mention of clock speeds, and some of them (specifically, the ones with the Blue Man Group) don't mention the chips at all. If consumers equate bigger numbers with performance, it's because of their own stupidity and not Intel.

Personally, I think Intel is winning because of its "monopolistic practices" that you mentioned. If I could have gotten an AMD in my laptop instead of a P4, I would have.

not as laughable as say the Itanium (We've sold more Opterons in one deal (Cray Super Computers ~10,000) than Intel has Itaniums in 3 years haha).

The Itanium was not made for the mainstream. I can't remember what the hell it was supposed to be, though.

A Honda Civic hybrid is more efficient than an M-1 Abrams tank, does that make it superior to the tank?

Yes. It goes faster, too. But the Insight kicks both their asses.





dwchang, I think you should mention cache memory. If I remember correctly, the P4 has like twice as much L2 cache as the biggest Athlon, making the Athlon much more likely to choke when faced with particularly cache-intensive applications (the Quake III engine, for example, was specifically designed to take advantage of the P4's superior L2 cache).


I noticed something very interesting on the AMD website a few months ago. They like to portray their side-by-side comparisons and benchmarks as very fair and unbiased, showing where the P4 is better (like L2 cache) and where the Athlon is better (like instructions/Hz). However, I noticed that in their benchmarks, they used high-end Nvidia or ATI video cards in their own machines and used low-end Intel video cards for the Intel machines. Hmm...



Hey, let's all make fun of Cyrix!

[alternatefutures]

Actually, both the P4 and the Barton have 512K L2 caches. The Barton has a significantly larger L1 cache (not necessarily a good thing as it also has twice the latency of the P4's).

Anyway, the Civic doesn't go faster off road... can't even handle half the terrains the M1-Abrams can. You get in an accident in the Civic you're far less likely to walk away than in the tank. In fact, you get in an accident with the tank you can keep on going, and you don't run into trees, you run over them. But with such a blanket statement I guess that means that we should replace all tanks in the US Army with Honda Civics, right?

Oh, and the Insight blows. I'd prefer the Prius if I could fit in it. Hopefully Lexus will actually make their hybrid...