Sign in with
Sign up | Sign in
Your question

Just think about Intel for a min ok?? Just think

Last response: in CPUs
Share
Anonymous
a b à CPUs
March 19, 2001 9:35:25 PM

For all you AMD bashers just think for a min ok?

Intel had the Pentium 3 right
AMD Brought out the Athlon the same athlon that is out today, same technology..
And the p3 coppermine came out, then the pentium 4 and amd still pulls very strong and holds its ground

You would think in all that time with AMD's Athlon
Intel had the chance to design something WAY WAY better to exceed the athlon, they had the chance and look what happened?

cough.. cough..

Im just saying.
think about it.

You must credit AMD for using the same Athlon Technology all of this time and we are still amazed how much of a performer it is.

Sure makes intel look bad

-- They have found a way to harness the power of a thunderstorm and expell it with great force!--

More about : intel min

Anonymous
a b à CPUs
March 19, 2001 9:43:28 PM

last time you asked us to "just think about it" it was about putting your PC in a freezer, which is equivalent to "why not put your PC outside in a thunderstorm"

but anyway....

the athlon has changed, the cache is now on-die, there has been a core shrinkage from .25 to .18 micron, along with some minor core enhancements and the clock speed is a whole lot faster.

its a lot like the pentium II/III evolution of a few years back.

___________________________________________

Smoking is a major cause of statistics.
Anonymous
a b à CPUs
March 19, 2001 9:49:45 PM

yes some changes with on die cache and a smaller die

MHz is irrelavant now

Im talking about technology here

Athlon is Athlon same technology over something Intel had a long time to come up with something way way better
but ...

-- They have found a way to harness the power of a thunderstorm and expell it with great force!--
Related resources
March 19, 2001 10:22:32 PM

just because the p4 isn't pulling ahead in today's benchmarks doesn't mean that is isn't a greatly redesigned chip with great potential. it hasn't had enough time to mature to prove whether or not it is WAY, WAY better than the t-bird. i know, it is said all the time that software isn't optimized for the p4 yet and that is true but all i'm saying is that the p4 seems to be a good design with a lot of potential. will AMD still be a good competitor when the p4 has a chance to shine? perhaps (probably it will). but to say that intel should be beating the athlon badly because they had some time (not that long if you think about it) isn't really accurate. it has been a progression. the k6-2 wasn't too much of a competitor in performance (maybe it equalled the celeron) but it had a distinct price advantage. the k6-3 was a faster chip and i remember reading articles where the k6-3 beat the p3 in business applications. then came the athlon...AMD hit on a great cpu. then came the progression in in on-die cache and socket design with a smaller die size with both companies going blow for blow. the t-bird outlasted the p3 in the sense that it could handle the higher speeds so intel needed a new chip to compete. along came the p4. it can obviously handle the very high speeds of tomorrow and it will probably outlast the t-bird in the sense that while the p4 is going AMD will need a new design to keep up with the higher speeds. that's the way competition goes.

now, here might come the surprise... i'm a faithful AMD fan. my first real computer was a k6-3 450. i have since gone through many upgrades and currently use a 900mhz t-bird overclocked to 1ghz. i will not let a friend buy a p4 that i spoke somewhat highly about because at the moment it's a dumb choice (unless you use flask 50% of the time and quake3 the other 50%). i have built computers for a few friends and i will not build them an intel system. i don't think i will ever buy intel. but that doesn't mean that they are bad...nor are they bad because they can't beat a good cpu like the t-bird.
Anonymous
a b à CPUs
March 19, 2001 10:48:26 PM

do you have ANY idea how long it takes to design a new CPU?? in the semiconductor industry, its not really a "long time"

heres a hint: CPU's are complicated, theres a whole lot more to it than just, "hey, lets design a faster processor today!"

___________________________________________

Smoking is a major cause of statistics.
March 20, 2001 2:07:24 PM

Not to support Intel, but....

If you get down to brass tacks, the Pentium 3 can trace it's lineage way back to 1995, with the Pentium PRO. All of them use the P6 core architecture. That's what, 6 years now?
March 20, 2001 3:12:26 PM

Exactly what I was thinking. We are told to think. Well, if we do, we realise that the Coppermine's core gone relatively unchanged from the Pentium Pro that it derived from.

I'd say anyone that can use a core THAT old and STILL compete against a newer core has to have done SOMETHING right with their core.

Now let's look at AMD.
Just HOW MANY K6 modifications were there before AMD FINALLY hit on the Athlon? I'm sorry, but it doesn't exactly sound like they instantly hit it off with a great chip and ran with that same chip since. It took AMD a good amount of time just to finally produce a chip that could even compete against Intel chips.

Especially when you consider that even the Athlon itself has now gone through three versions. Remember the Athlon A, B, and C?

AMD has been revising their chip faster than Intel has just to remain competitive. Why? Because the core Intel was using was more scalable and easy to improve the performance of.

And now Intel has the P4. The current chips themselves aren't as amazing as we would have expected. But the mature P4 product lines based on the same P4 core will be. It was designed for scalability and to utilize the higher-bandwidth memory that we're trying to use today.

Meanwhile the Athlon still has bandwidth limitations that keep the chip from being able to see any real performance gains from the higher bandwidth of DDR SDRAM, even though AMD JUST released their Athlon C.

So AMD knows that to remain competetive, they need to release a chip that uses more memory bandwidth. But they haven't done anything to their cores to solve this yet.

But Intel has.

Gee, that sure makes AMD look bad for completely ignoring a known flaw that needs fixing to remain competetive.

Do I really believe that? No. But then, that's not MY logic. That's logic according to tbirdinside.

Because we KNOW that AMD is working on their Hammer series. Just as we KNOW that Intel is working on their Northwood.

My point? Both companies are working to remain competetive. Neither are stupid. Both have interesting future plans. End of story.

-Despite all my <font color=red>rage</font color=red>, I'm still just a <font color=orange>rat</font color=orange> in a <font color=white>cage</font color=white>.
March 20, 2001 4:43:27 PM

I think AMD's real problem is that they need to be a design initiator instead of a follower. They have done this for years

The initial K6 was released to compete with the P55 MMX chips. The initial releases were good, especially when paired up with an Intel chipset like the TX. I remember how some early K6 233 were as fast as the early Pentium II. However, as speeds increased, they couldn't compete because of the slower access to L2 cache through the FSB. They then switched to alternative chipsets to bump up the FSB but a Pentium II 300 with L2 bus speed of 150 would be faster and a K62 300 with L2 bus speed of only 100. To fix that, they integrated the L2 in the K6III line. That helped, but the poor chipset designs hampered overall speed. I know first hand that a 400 Pentium II is 20% on sysmark than a 400 K6III.

The initial Athlons were good, taking another lesson from Intel. The Slot A was so similar to the Slot 1 (physically) that you know that it was just copied. I understand that it is the same piece, just flipped from side-to-side and wired differently. The Atlons were able to beat the Pentium 3's (katmai), just like the initial K6 did. But then, Intel foreseen that the speed of the L2 cache was going to be a problem so they integrated it in the chip with the Coopermine. Once again, AMD didn't act fast enough and when they hit the L2 speed limit, they had to alter the ratio for the L2 cache. Once again, L2 played a role in the AMD chips and the 1st generation Athlons were slower than same speed Pentium 3.

Now, AMD integrated the L2 on the chip, just like in the Coopermine. Sure, the did 4X the L1 cache than the P3, that was a smart move on their part. And that it is mutually exclusive helps as well. But it still is not innovative enough.

I hope that the Sledgehammer is a success. I am a big AMD fan, but I worry that there will be a time where they can't compete with Intel because they can't come up with their own designs. Like Intel, AMD has recently begun to "sit on their laurels" and haven't really released anything new. They been stuck at 1.2 GHz for several months now!!!.

Please AMD, don't go the way of IDT Winchip and Cyrix.
a b à CPUs
March 20, 2001 6:07:22 PM

I must say that the Pentium Pro was not too impressive at it's release, but lended it's core to the phenominal PIII eventually. AMD took a big risk with the Athlon-because of the size of the initial investment, they had to get it right the first time, big pat on the back for them. But Intel relies on suckers for the developement of a new product. They have such a large market share and such a commited consumer base that they can afford to release a chip before it is perfected. Eventually the P4 core will lend itself to a great processor. Poor suckers who bought the first version. But the technology is there anyway, it just needs to be "finished"! Let's hope for all consumers that AMD can keep up!

Suicide is painless...........
March 20, 2001 6:11:46 PM

Those are the exact worries that I have about AMD as well.

That and they may never have a good dual-CPU setup.

Me, I like to push my computers to their limits. And it has become obvious to me that I need a dual-CPU system for home use. I had been happy to hear that AMD was finally going to give us one, until I saw the pricetag on the motherboard.

At first I thought it must have been a joke or a typo.

I want AMD to be innovative and lead the way with new technology.

But so far, they haven't even been able to prove that they can duplicate everything a P3 can do with any success.

And if the Intel comes out with a dual-CPU motherboard for the Northwood, I just may have to go with that. Unless AMD can come out with something that is actually competetive. Their .13micron Athlon/Duron shrinks aren't going to be enough.

I guess we'll have to hope that their Hammers aren't a joke. But since those are supposed to be aimed at being 64bit server chips, I don't know that I could ever afford them.

-Despite all my <font color=red>rage</font color=red>, I'm still just a <font color=orange>rat</font color=orange> in a <font color=white>cage</font color=white>.
Anonymous
a b à CPUs
March 20, 2001 7:56:28 PM

slvr...just watch the price the price of dual mobos for amd's, i reckon its gonna plummet. it isnt even released yet, give it a month or two and itll be a whole lot cheaper. it wouldnt make any sense for the mobo manufacturer to keep it at that price.

___________________________________________

AMD. It gets you chicks.
Anonymous
a b à CPUs
March 20, 2001 8:25:18 PM

Agreed. What helps the Athlon to shine over the P4, despite the P4's higher memory bandwidth, is the size of the L1 and L2 cache. Seriously, what was Intel thinking? An 8 kB L1 data cache? That's no bigger than a 486's cache.

If they'd gone for a few less MHz on the CPU speed and sprung for at least a P3-level, if not Athlon-level cache, it would consistently smoke both of them. Maybe Intel will fix that first before moving on to the next chip...
!