Slashdot Mirror
AMD Introduces the Athlon XP 2200+
NevDull writes "AMD introduces the Thoroughbred core in the Athlon XP 2200+. Tom's Hardware Guide has a review of the new CPU based on the 0.13 micron core, and subsequently declares the current CPU war to have been won by Intel." Update: 06/10 12:48 GMT by T : DavoHH writes "To add to the list of reviews and benchmarks around the net for the new Athlon XP 2200+,
HotHardware.com has one and also
and also Anand's
and AMDMB." Update: 06/10 13:45 GMT by T : One more: Johan contributes a link to an Ace's Hardware review which tries to answer the question "Does the 0.13 Athlon XP run well an on older motherboard, and does it provide good value as an upgrade?"
My 1.1ghz processor is plenty fast enough for me, and will remain so for quite some time.
In terms of tech nuts, AMD has a strong, strong following and lots of brand loyalty - as much, if not more, than Intel.
In terms of people who shop at Staples/Best Buy/etc... They buy what's in the box and tend not to care what's inside. Last time I was at either of those stores, there were more AMD-based boxes on the shelves than ever before.
If we're talking technology alone, it depends what facet you're looking at. Intel processors do better in some areas, AMD in others. With AMD, you always get more bang for your buck, so to speak, as well.
You see, in the middle of the article there is a list of comparative prices ($ per chip when buying 1000). The prices for a xx00 P4 are almost exactly the same as for a xx00+ Athlon, except for the highest end chips ($600 for the P4 2500).
So it seems as if Intel is finally challenged enough by AMD that they actually have to have the same prices for the same 'PR' in the mid-range. In my view that is a win for the consumer.
My current system that I run at home is between a 450 and a 233. Though this does not include the two system that I have both running dual 533. Considering where I was before, upgrading from an overdriven 486/pentium 83mhz to the current 450, that is a 542 percent increase in the clock rate. Ironically though, I aleady had 128MB in my old 83mhz and even my 450 only has 128mb... quite sad.
:-)
The only justification that I will see in getting a new computer anytime soon is when I see some 3GHZ machines. (That is again about a 550 percent increase in clock rate).
At the rate new chip designs are coming out, I think I will have more chips to choose from than underwear. I can keep waiting.
I'll bite.
The numbering scheme currently used by AMD for their Athlon chips reflects the speed of the equivalently-performing Intel chip. ie the Athlon XP 2200+ performs at the same level as the P4 2.2GHz, despite running at a lower frequency.
If AMD just quoted the raw GHz figures, everyone would assume that they are significantly slower than they actually are. They would also lose out on price comparisons - on dabs.com, the Athlon 1900+ is only about £8 cheaper than the P4 1.5GHz. The P4 1.9GHz, however, is almost £60 cheaper, a much larger difference (just over a third of the total price of the P4, in fact).
(Note that I am in no way affiliated with dabs.com - they're just the first website I think of when looking at PC component prices.)
Cheers,
Tim
It's official. Most of you are morons.
The 'throttling' solution has existed for some months now. In the BIOS (build 1007) of my Asus A7V133 motherboard I can select various actions for when the CPU temperature passes a (user-selectable) threshold: 1. Audio Alert on PC Speaker, 2. Do nothing, 3. Shut Down, 4. Throttle the CPU to a lower speed.
should they be advertising with: "well, nobody of you will need that amount of GHz now, Pentium's sucks compared to AMD's... but what the heck, buy our stuff!).
If at first you don't succeed, then sky diving definitely isn't for you.
Still, it's going to be exciting, because the arrival of the Barton, with its larger L2 cache, is imminent. And VIA is working on the KT400 chipset, which is supposed to bring DDR 400 with 200 MHz to the Socket 462 platform. Thus, the race has not yet come to an end - the means are ultimately the goal!
Intel may have won a battle, but the war is far from over.
At least here the reviewer make sure that both CPU work with the same memory.
Tom's gives the P4 PC1066, while 95% of the P4 systems are sold with DDR.
I already see this in an automobile mag:
In order to get a fair comparison, the Ferrari was equipped with a 1600cc VW engine.
BTW, I think that 95% is more than a bit pessimistic. Somewhere between 50 and 75% maybe...
If my heatsink fell off on a server, I would not want the system turning off, I would want it staying on. I mean, it won't do too much good being on in that state, but at least there is no data loss in that situation.
If a heatsink falls off the CPU, I would prefer my system to turn off the power as soon as possible. Know what a copper/aluminium piece can do to the many Amperes running in a mainboard? don't want to find out.
And after all, those server cases (CPU temperature more than 85 degrees celsius, or CPU fan falling off) are extreme cases, which should never happen. If this happens, then something else want wrong. If you check larger servers, you'd find lots of fans, all being redundant, so even if 1 fan stops working, there's still plenty air being push-pulled through. If 2 fans stop, then the machine should do what it wants to do (slow down or turn off), as this is a very, very rare case.
Both AMD and Intel regularly release new silicon rated at higher clock cycles. This isn't really that big of a deal. Tom's Hardware likes to make a statement rather than pull their punches, so it doesn't surprise me to see something meaningless like "Intel has won the CPU war". Many gamers now swear by AMD, and the damage to Intel's reputation will need to be repaired over time. Intel's deaper pockets may be churning out CPUs which are beating AMD's recent releases in Tom's comparisons, but the trust issue with consumers will lag behind the realities of comparitive performance, just as it used to in Intel's favor. AMD is winning the popular war even with their losses in specific battles.
It may not matter if Intel can deal with heat more effectively than AMD. The AMD CPUs are much cheaper and those with big concerns over heat will drop over $100 on a heatsink/fan.
The CPU war isn't nearly over. Even if Intel continues to win these individual skirmishes, they will still have to demoralize AMD's faithful. Intel may have bigger "weapons", but AMD has something that Intel doesn't to the same extent: trust, loyalty, and support as an underdog.
perl -e 'print $i=pack(c5, (41*2), sqrt(7056), (unpack(c,H)-2), oct(115), 10)'
Now, IANALOEVS (I am not a lawyer, or even very smart) but I can't see how these two are the same. Nowhere in AMD's literature to they claim that the number after XP is a measurement of Mhz, only a measurement of it's performance relative to other AthlonXPs. I have a Hewlett Packard 4500 laser printer, but I don't expect it to print 4500 pages per minute, print at 4500 x 4500, or give me $4500 USD for the privelege of being my printer. The number is marketing, and unless they claim that it's something it's not, I can't see calling it false advertising. But again, IANAL...
do not read this line twice.
But x86 has more than a 20-year history, and Windows (in one form or another) has been around for 10+ years. There is a huge library of Wintel x86-compatible shareware and freeware, not to mention old business software.
Many companies still use older business software with their newer computers, and a change in architecture would involve a tremendous effort in converting/reproducing documents.
This is also a huge financial burdern. Normal cost of system upgrade, plus cost of new software to replace the defunct SW, plus cost & lost time of learning the new SW, plus cost & lost time converting old documents. It adds up, especially for smaller companies who barely afford new systems.
Presumably Intel has considered this. Any architecture change will have slow acceptance. COnsider the Macintosh; the change over from 680x0 to PowerPC was slow, and I can still run old 680x0 apps in MacOS X through Classic mode.
Maybe they'll make a P4/Itanium dual processor board that won't have the compatibility problems, just twice the price.
Linux, with it's ease of portability and open source could really help adoption of Intel's new architecture, actually. Microsoft will need emulators like Apple did or risk leaving many faithful users out in the cold.
All considerations aside, leaving x86 isn't really a bad thing. For all the speed we squeeze out of this silicon, it still has one foot firmly in the early 1980's.
--
Of course, if occurs to me now that it probably won't be an issue of any sort...
That what was all this school was for... to teach us how to solve our own problems. -- janeowit
In the mean time AMD and Intel are really talking about two completely different markets for their wares. Intel want to make a serious pile of money out of selling new (and hence overpriced) Itaniums for use in multi-way servers for bespoke applications where compatability is not such an issue. AMD on the otherhand looks to be targetting the consumer who wants to squeeze every last frame out of their Quake sessions and other (less important) "legacy" 32 bit code. Both companies will probably make a big pile of cash out of their respective sectors, so no problems there.
In the long term though, unless AMD is going to make a seriously brave (or rash) departure from Intel compatability, ultimately they are always going to be playing catchup with Intel for compatability. There's a long history of that too, and in that context, I'm just amazed that AMD has lasted as long as they have when other ventures have long since come and gone - best of luck with giving us all a choice I say!
UNIX? They're not even circumcised! Savages!
I think it would be better for the BIOS to signal a shutdown to the OS to permit a graceful shutdown.
If you lose your heatsink, there isn't time for the OS to shutdown before the chip fries. If just the fan fails, there might be. But what if the OS fails to respond to the signal? Oh come on, don't tell me you've never seen a Windows box that has the power saving features enabled but mysteriously doesn't go into sleep mode every other Tuesday?
Hey kids, there's only 5 days left 'til Yak Shaving Day!
Once CPU's passed 300mhz, I stopped looking at the number and more at the price. That's the only really important number to consumers now. Not counting hardcore gamers or people who need the fastest machines on earth. Now if only both of them would focus on lower heat, and quieter designs.
Of course if I ever get back into gaming, I might change my story and lust for the fastest system. Then again, by that time, I won't need anything faster than a 2ghz anyways. Especially with the way the video card are improving these days, CPU ceases to be a major factor in gaming. I'm no expert, but I'm guessing improvements in bus and bandwidth will do more for realism in 3D than CPU clock speed.
Read here and the author tells you how The Pentium 4 finally takes the speed crown as AMD falls asleep at the wheel
These are really great read.
Not to take sides, because honestly, I really don't care... But Intel is hardly leaving the x86 behind.
Intel's IA64 strategy is banking on a few things.
One, that x86 doesn't have much need for 64bit support. They don't think much of our software would be benifited from a 64bit CPU right now, so they aren't going to give consumers a 64bit CPU to use yet. In general I agree with this, for now, it isn't like Microsoft Word would benifit all that much from being 64bit.
Two, the people that do need the 64bit CPU will be in the price range of the current Itanium based computers. These people are those that do scientific calculations, intense server stuff... not Quake players.
Three, they will have the IA64 technology optimized and proven in the future for when it is time to completely drop x86 from the planet, and the software houses will have a head start developing for it.
Basically I think they are using todays Itaniums to get a head start on what they want to have 5 of 10 years up the road for main stream computers.
Now this isn't to say I particularly like the IA64 architecture... but that is what Intel is doing.
I guess what I am really trying to say is, don't count out the x86 CPU from Intel, they aren't through with them yet. I beleive they were aiming at 5~10Ghz with Pentium 4 like x86 designs... which is obviously way ahead of where they are now.
I guess the real test will come when we see how much x86-64 software comes out. And what kind of performance increase comes with it over 32bit versions. Intel could easily have not extended there x86 to 64 bits just to try and bully the software industry into not shipping much 64bit x86 software thus negating any advantage AMD x86-64 might have over Intel x86-32.
This is a winning war for both Intel and AMD. The only ones who lose are us, standing in line forking over 180 bucks for the XP 2000+
Depending on who you ask, from a few thousand to a few million years of technological development;
for $180.
Doesn't sound like a bad deal at all. : )
Need help treating your acne? Come here!
5% less heat when changing the production process from 0,18 to 0,13 is IMHO a big disappointment.
When AMD announced the new Mobile Athlons the new core seemed to be very promising. Core voltage down significantly, power consumption down significantly. Now the core voltage is just decreased by a mere 0,1 volt and power consumption down by less than 10%.
Very frustrating! Especially when you know, that AMD ships 1,5 Volt-Athlons to NEC for Low-Noise-Office-PCs for Japan.
AMD, LISTEN UP:
Deliver those Low-Voltage-Athlons to all of us. If I could chose between an Athlon 1.8 Gig 1,75 Volt (60 watt) and an Athlon 1,7 Gig and 1,5 Volt (40 watt), I would definetly buy the the 1,5 Volt version. And I am sure, that I'm not the only one who would prefer a quiter PC for 5 % more performance.
Bye egghat.
-- "As a human being I claim the right to be widely inconsistent", John Peel
You assume too much, and I don't think you accurately read my post either.
1) I never said that the 2200+ was 2.2 GHz. AMD has made their point about NOT comparing clock speeds, and I did not. What I said was that the 2200+ was meant to compete with P4s running at 2200 MHz, and thus the labelling as "2200+" by AMD. This is it's approximate speed according to AMD, not me.
2) I did read the articles. Tom has repeatedly shown himself to be biased against Intel in the past year, and that is not going out on a limb. I much prefer AnandTech because he is somewhat more objective -- in favor of AMD when they are on top of speed, and in favor of Intel when they are winning. This is moot, however, because:
3) The 2200+ consistently placed itself in speed around the range from the 2.0 GHz P4 and the 2.4 GHz P4, even on Tom's Hardware. Compare it to the P4s using the 533 MHz FSB in the benchmarks, not just the 400 MHz. If you claim otherwise, you are not looking at the benchmarks carefully. I don't count the overclocked Athlon at 1.9 GHz, because that's no longer the "2200+".
Thus, I think splitting the difference between 2.0 and 2.4 is fair, and calling the 2200+ a competitor with the 2.2 or 2.26 P4 is fair.
No matter what, comparing prices to the 2.53 P4 clearly is not fair. I believe what the original poster intended was that the top of the line athlon is priced much lower than the top of the line p4. That is true.
That being said, I love the fact that AMD is doing well, and I want them to keep doing so. Real competition makes both companies produce better products at lower prices.