Oh Tom, silly Tom
Don't know how to take this one, but I do agree with Tom's premise, but not his exact commentary as presented. If AMD doesn't shove a new core out the door in six months, then the Pentium IV will win this so-called "war", atleast in regards to performance. Hardware prefetch and SSE is nice, but AMD needs a new core revision with additional porcessing options (not just more MHz) if it wants to compete in 2002.
The Athlon fails to scale well with a ramp in MHz at 2.0+ GHz, while the P4 is showing positive positive returns with alomost linear results to 3.5+ GHz. Add in hardware dynamic multithreading and 512KB of L2 cache. The P4 will be a tough competitor in the upcoming months.
Finally, before my inbox floods with hate mail, I do own and use an Athlon 1.2 GHz within my primary box! End-user costs and real-world performance in today's non-P4 optimized apps are still predominate motivators.
You'll get no hate mail from me RR. I am looking forward to benchmarks pitting the 2GHz P4 against the Athlon 1.4.
[This message has been edited by daveleau (edited 08-31-2001).]
My opinion? It's all market share. Sure the AMD will generally outperform an Intel (for now, at least) but in the end who will make more money? Intel. And that's why they'll be declared the 'winner.'
While I am a loyal AMD fan, ultimately I think intel will win out in the raw battle. News from inside: P4 architecture is scalable to 10 Ghz! Yeah, that might be straight off the rumor mill, but even some of the big wigs from AMD recognize that this is a huge threat. Personally, I hope AMD gets its innovation on, and produces another high quality product. (Who knows how much of this is true, but it's from a eliable source. More or less, it's something to think about..)
Ok so Intel came out with a 2 GHz. So what? Until recently P4 has only supported Rambus, Good move intel. Also, the AMD T-Bird 1.4 GHz out performs the P4 1.7 in almost every single benchmark out there. And the AMD is about 3x less. Now I would like to know HOW intel is ahead in the processor race, when they have a chip out that is about 5x more expensive than the amd one and only very slightly more powerful. The problem is all the brainless idiots who see 2gh Whooo hoo, and drool over big numbers. Why dont you look at the bottom line. To me paying 562$ for a p4 2ghz is pointless when you can get a amd 1.4 for 130$ or less online and get very close to the same perfomance. Just common sense, but most people dont have much when it comes to computers i guess. they just see bigger numbers and they have to have it.
Everyone should know by now that TOM is an advocate of AMD.
A reviewer's job is to convey the facts without prejudice, which is exactly what he is doing.
Opinions and personnal preferrences are secondary, and belong in a separate provision for such, unlike what I'm seeing around here, today.
Tom knows whats up. You have to admit that in a dead heat, the 2GHZ will outgun the 1.4 Athlon in most benchmarks. Eventually the deciding factor between procs will be less dependent on the chip design and increasingly dependent on mainboard design. One reason for the P4's success in content creation and streaming video/mpeg encoding is due to the bandwidth afforded by its workstation mainboard dual memory bus design and it's implementation of Rambus RIMMs. We can see the diminishing returns of faster processors when paired with slow buses and the substantial speed bonus of an FSB increase on those same chips(Intel's Celeron for example). Any hardware guru understands that a 10% increase in FSB is worth alot more than a 10% increase in processor MHZ. Eventually these great increases in clock speed will become cost prohibitive (espescially in cooling-I don't care what Intel says the 4GHZ at IDF was not conventionally air cooled)and we will see the lack of consumer interest in faster processors continue because of the relatively small gains in performance associated with it. We are seeing the limits of current processor design when attached to these poor buses (AMD's EV6 at 133x2 and Intels P4 bus at 100x4). These are workarounds to adress the real problem with modern chipsets/RAM. As a result, what will begin to see is that cash poor RAM manufacturers will develop new, better, faster chips to fill the performance gap and sell these chips at higher prices to combat current loses (Nanya for example with their 333MHZ DDR). Undoubtedly, others will follow suit with speed solutions to work around the processor/FSB/bandwidth glut (IDE RAID, Serial ATA, GPU's like Nvidia). The company to watch out for is the one who gets their new mainboard technology to market first with a chip to support it (AMD-Hypertransport, Intel-3GIO/Arapahoe). These fabs can spend all day cranking out 4GHZ chips that only support a 100MHZ native bus, but expect them to get thoughorly spanked by the new guy sporting his next gen bus tech. We need to focus on the board design and let the chips follow.
I agree with MrMiguel. I've been reading Tom for a while now and he is just calling it the way he sees it.
The P4 is ahead in the race right now and they are in a position to really take off. The P4 does ramp up in speed at a sacrifice of instructions per clock. To stay up with AMD they have to have about a 600MHz advantage to do the same work. If Intel wanted to they probably could spit out a few 2.5GHz CPUs, but they want to milk each 100MHz jump for all they can.
We all root for the underdog (AMD being the underdog), but I wouldn't say that Intel is the winner just yet. I'd just say AMD needs to slowly erode Intel's market dominance and that is slowly what is happening. Hopefully they can keep it up.
I'm not so sure that Thomas Pabst loves AMD so much for itself as he dislikes Intel for their sloppy work (ie the 1.13Ghz Pentium), their poor employee relations situation, and their shady marketing & customer service policies.
Everyone should know by now that TOM is an advocate of AMD.
But I do think he loves the "hunt": the thrill of the chase, the bay of the hounds, the victory of the underdog, and all that.
Remember: when David slew Goliath it was big news, and news is Tom's business.
You guys may not want to hear this...but as long as I have known of Tom Pabst ( from the days before Tweakit ), he has demonstrated a strong commitment to pleasing those who pay his bills. So much to the point that if he has advertisements for one product, he will bash the other product regardless of fact or fiction.
No, that was not a compliment to Tom.
Tom has also proven himself to hate and despise America and it's citizens. If half his reader base was to read some of the stuff he wrote blasting Tweakit members, you would want to boycott his site. That is why I believe his opinion is totally irrelevant. He is not interested in the truth at all.
I will have to agree with BBA on that. Tom tends to support the bill payers and if you have noticed lately, intel is all over his place.
With OS like WinXP the P4 will no doubt shine. As long as programers hand optimise for SSE and finally render FPU useless, so be it. It's such a irony, given the fact intel always beat the competition if FPU to completely abandon it.
Unix and non-proprietary users will stick to the raw FPU of the Athlon though. It will be a big shame if AMD dumbs FPU as well to catch up the MHz race. What will be left to fpu hungry code then (and there will always be some!)?
I think amd is just fighting off the inevitable (intel crushing them someday). Even though amd athlon's are faster now than the p4, how bout in the future with new software optimized for the p4. and intel is still the stock giant over amd. intel has many plans on keeping amd in bay, but amd is just blasting thru now. intel also knows that almost 90% of the market only look at mhz. if amd wants to fight back now, they need to be close or ahead in the mhz race and show some serious tactics. i would say introducing larger cache models with full support for all the instructions, and consider 256 bit cache. if they could get all that and sell it at low cost, they will win. you pple know intel's p4 is not really great even with the mhz, but many out there do not. intel has the resources to lose a lot of money by selling p4's just as cheap as amd athlons and persuasive relations to all the other computer companies.
Jeffy, you're wrong.
Intel has NO interest in crushing AMD, as crushing them completely would lead immediately to charges of monopoly being filed against Intel. The FTC, in fact, was investigating this possibility up to early 2000, when AMD had gained such strength it was clear that Intel was NOT a monopoly.
Secondly, AMD is now in a far stronger position, historically, than they have ever been. They hold far more market share and cash on hand then they did five years ago, when the K6 processor was just making it to market, and the Athlon is far more competitive against the P4 than the K6 was against the PII.
With the upcoming K8 core be prepared to watch AMD strike back at Intel.
Advertising not only pays the bills...it also creates sales.
What put AMD in real contention? The 1st 1G Athlon that they advertised the hell out of. When was that? Fall of '99 ?
I haven't seen an AMD ad since.
Advertising also makes the consumer price higher, though.
New Security Features Planned for Firefox 4
Another Laptop Theft Exposes 21K Patients' Data
Oracle Hits to Road to Pitch Data Center Plans
Microsoft Preps Array of Windows Patches
Microsoft Nears IE9 Beta With Final Preview
Simplified Analytics Improve CRM, BI Tools
Android Passes RIM as Top Mobile OS in 2Q
VMware Updates Hyperic System Management
File Monitoring Key to Enterprise Security
LinkedIn Snaps Up SaaS Player mSpoke