Intel Cuts Chip Prices by up to 53 Percent

babbage1815 writes: "Intel Corp. has cut prices on some of its microprocessors by as much as 53 percent as the world's largest chipmaker's investments in manufacturing over the past two years are starting to pay off." Most of the cuts are at the very high end of the line -- it'll be interesting to see what happens to the prices of the competing AMD offerings.

Comments

Sweet News

[CovertSquirrels]

Who cares about Intel, what's AMD going to do to get me into a new processor? 0%APR and no payments for 90 days? You've got yourself a deal, pal!

Watch for Falling prices??

[Brynath]

What is this is Intel being bought by Walmart?

Re:Watch for Falling prices??

[Jucius Maximus]

"What is this is Intel being bought by Walmart?"

As long as they don't have a bouncey happy-face on bootup.

Even the office clippy is not that bad. (Hint: Kill the office clippy by renaming the 'actors' directory under \program files\microsoft office\ to actors.f00 or such)

Re:Watch for Falling prices??

[Snover]

Ow! Dammit! No, really, watch for falling prices! They really hurt!

Finally

[Anomolous Cow Herd]

I'm happy for this. Not only does the consumer win with every price cut that happens in the chip industry, but one of the market leaders will be able to keep up with AMD.

I mean, sure, AMD's chips are dirt cheap, but sometimes I just want to have a chip that I can be sure to depend on over the years. Certainly, the newest offerings from Intel are the coolest running in the competitive gaming market (not like an AMD, which I could probably cook my breakfast over). I'm sick of my room getting all stuffy and hot just from leaving my Athlon machine on for more than 10 minutes, despite the best efforts of the air conditioning unit and the ceiling fan.

Also, I have a DDR SDRAM motherboard for my Athlon, and I've figured that it'd at least work as a stopgap measure until I could afford something better. Fortunately, now that the final price barrier is gone on the alternative, I can finally get some nice Quake III framerates with an RDRAM-based board. That extra memory bandwidth sure is nice.

So, score one for Intel, and score one for my power bill. My wallet will thank me later.

Re:Finally

[Lord Jester]

And why would you remove the heatsink???

*shakes head*

Re:Finally

[Transcendent]

I can finally get some nice Quake III framerates with an RDRAM-based board. That extra memory bandwidth sure is nice.


AMD TBird 900, 640mb PC133 SDRAM, RADEON 8500 and I get 70fps+ on quake3 with the graphic settings on max. Tune it down just at bit and easily 100+ (even with anisotrophic filtering)

...get a good graphics card and then you'll be able to get good fps... its not your ram's bandwidth...

Re:Finally

[Hoser McMoose]

Hmm.. AMD's highest powered Athlons maxed out at roughly 70W, though typically they were consuming 60W or less.

In other words, I sure as hell hope you don't have the lights on in your room, because they're heating the place up as much or more then those Athlons are! We all KNOW how a single 60W light bulb turns ANY room into a sauna.. uhh.. do we?

The difference in power consumption between AMD's Athlon and Intel's P4, if run full-out, would amount to about 10W (if they're sitting idle they're essentially equal), assuming equal Athlon model number vs. P4 clock frequency.

Given a price of electricity of $0.10/KWH (probably about typical for the U.S.? I dunno, I pay $0.08 CDN, or about $0.05 U.S.). If we assume that the systems run 24/7 for an entire 30 day month, that amounts to 7.2KWH for the month, or $0.72. You would have to run these chips full out, 24/7 for over 5 years before you would save $50 on the Intel chip due to it's lower power consumption. Ohh, and since you use RDRAM, which consumes more power then DDR SDRAM, the difference for the system as a whole is likely to be even smaller.

Long story short, heat and power consumption are NOT an issue when comparing AMD's Athlon and Intel's P4. People may have valid complaints one way or the other, but saving money on your electric bill is not one, and claiming your room gets hot?! Well, I'll write that one off as a poor excuse of a troll post.

Re:Finally

[wheany]

So you are saying that the flourescent bulbs that have an power rating of 15W are dimmer than the 60W incandescent bulbs? That is interesting.

Re:Finally

[wheany]

Well, as the processors are getting hotter, the heatsinks are getting bigger. Bigger means heavier. Heavier heatsinks and a big number of people using them mean that those poor retaining clips will fail on someone. And to my understanding it is not always all that clear that the heatsink is attached securely.

And also the fan might fail. If that happens when no-one is watching, the results could be one fried processor, maybe even a fried mobo. Just today at work a collague's fan failed just after he left. He booted the machine before leaving and for some reason the cpu fan stopped. Another worker noticed the strange smell coming from the machine, and we quickly turned it off. The machine seems to work still, though. So we were lucky today...

Re:Finally

[wheany]

They release about 10% of their energy as light, so it's pretty far from "all".

Re:Finally

[Hoser McMoose]

Where exactly do you think that the heat is going to go from a light then? Sure, it will stay slightly more centralized around the lamp, but the difference is going to be REAL small. Free convection DOES work you know. Yes, forced convection does give you a lot more movement, but only for about 1 foot with most of these fans, maybe 2 feet at most. These are not some huge industrial fans that are going to blow hot air all over the room or anything like that.

High school physics is fine for starting to learn the concepts, university physics is better for figuring out when these concepts apply and when they can be ignored because they aren't going to change thing by any meaningful amount.

Re:Finally

[Hoser McMoose]

Your average incandescent lightbulb releases about 5% of it's energy as light, and 95% of it as heat. So, technically I suppose we're talking about a 60W light bulb producing ONLY about 57W of heat, while those "AMD space heaters" produce about 60W of heat. There's also the fact that a light bulb uses ONLY 60W of power, while a 60W processor get's it's power through a supply that's only about 85-90% efficient.

Still, all that being considered, the end result is that two light bulbs will heat up a room a LOT more then a single AMD Athlon processor will, and a single light bulb makes FAR more of a difference then one Athlon vs. one PIII, let alone when compared to a P4. In short, if you see anyone complaining about how much their Athlon heats up their room, they're either trolling, of just dumb (or more than likely, they're both).

Is this supposed to surprise us?

[digital_freedom]

Moore's Law predicts that this will happen which is that we will see the doubling of hardware capacity per dollar every 18 months. It looks like Intel has just finally decided to shift down it's prices. I guess the gravy train of overcharging on processers is over for now, until they release another model chip (which is really based on the last one). What will they call this one? Pentium V? Pentium Squared? Pentium Pentium?

Re:Is this supposed to surprise us?

[djmcmath]

No, the Pentium Pentium is from Little Caesar's Discount Computer Parts -- the first company to offer uniquely round processors...

Re:Is this supposed to surprise us?

[MonkeyBoy]

I suggest Pentiumed. The past-tense indicates that the next chip Intel releases will be a new architecture for the first time in 20 years.

Re:Is this supposed to surprise us?

[smyle]

1) You are adding extra 't's.
2) You are mixing the languages of your roots

Sexium (sounds even better than sextium) wouldn't be right - it would have to be 'hexium' (so you can see why they didn't go this route)
Septium
Octium
Nonium

Very Aggressive

[JM]

A university I know is building a 1000-node Beowulf (yeah, I said the B word) and called both Intel and AMD.

Intel dispatched a suit and an engineer right away, and was very aggressive on price.

They're still waiting for the AMD guy to show up.

I think Intel is trying to push every resource it can to dominate the market, and they had very good results so far.

AMD: Wake Up! ;-)

Re:Very Aggressive

[binaryDigit]

I think Intel is trying to push every resource it can to dominate the market

But they already do dominate the market. Unfortunately, I've heard other stories of AMD not quite being "with it" on the buisness side of things. I always chalk it up to the fact that they have to put all their resources into designing better (i.e. faster) chips and so can't spend money on fancy stuff like marketing and sales.

Re:Very Aggressive

[mattdm]

I don't understand why they called the chip makers -- wouldn't it be more appropriate to call a systems vendor? It's not like you take a bunch of CPUs, put them in a pile, and have Beowulf cluster.

Re:Very Aggressive

[ivan256]

That's probably because intel makes systems, and AMD doesn't. A pile of chips doesn't make a cluster.

Re:Very Aggressive

[JM]

Lemme see... you want 1000 machines, probably dual or even quad cpu, so 2000-4000 processors. Isn't that big enough to deal with the chip manufacturer?

Imagine the PR that the chip manufacturer would have by powering the biggest supercomputer in Canada.

Do you really want to call a systems vendor that's gonna charge you an arm and a leg to integrate your systems?

This is a University... they have all the brains and students to build it, just ship them a truckload of CPUs, Motherboard and cases, they'll save a million or two.

Re:Very Aggressive

[mattdm]

I work for a university. Anything this big is funded by a grant. You can't get a grant that says "yeah, we're going to build all this stuff ourselves"; the people giving out the money won't believe that you can do a decent job (especially if you say "we're going to get a bunch of students to put it together"). If you're doing research into the theoretical possiblity of building a big machine out of parts, maybe, but if your goal is to actually build a functional system for some purpose, you'll have much better success going with a vendor -- preferably a big-name one -- even if it costs a lot more.

Re:Very Aggressive

[cheezedawg]

Intel makes systems? Go to www.shopintel.com and see how many systems they have for sale...

Re:Very Aggressive

[jmv]

Imagine the PR that the chip manufacturer would have by powering the biggest supercomputer in Canada.

AFAIK, the biggest supercomputer in Canada is here and it's built by NEC (SX-5)

Re:Very Aggressive

[JM]

Don't forget, yesterday's supercomputers are today's pocket calculators...

Re:Very Aggressive

[jmv]

Don't forget, yesterday's supercomputers are today's pocket calculators...

I'd say today's supercomputers are tomorrows giant toasters...

Re:Very Aggressive

[mattdm]

I'm just saying: "We're going to get a huge supercomputer from IBM or SGI" sounds a lot better than "we're going to have a buncha kids go down to Fry's and buy some stuff and make us a big ole computer", when it comes to the actual asking to get the money.

Re:Very Aggressive

[Raindeer]

You seem to forget that Intel is a pretty cool company when it comes to investing/helping out with a great research project. This can be done in many forms, shapes and sizes, but in my experience at my old university, you could almost always be sure that if it was cool and in line with Intels products, they would be interested in having their name attached to it. Funny thing is, the university loves Intel in the game, even if it is partially for the greater glory of Intel. Intel has so much clout it attracts other sponsors as well. Having Intel in a project is often a win-win situation both financially, scientifically and pr-wise.

Re:Very Aggressive

[ivan256]

Yes, Intel makes systems. They just don't market them because they don't want to compete with their customers. Many Intel based servers sold by Dell or HP are in fact manufactured by Intel. I'm sitting next to two racks full of intel made machines. There are some 1U duals, so 2U duals, some 2U single CPU machines, and some 8U quads. They all arrived here directly from Intel.

Look at the following:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item =2028399075
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item =2028282610
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item =2028048472

All these are made and sold by intel.

Re:Very Aggressive

[mattdm]

I didn't say all universities are the same. I'm just giving my observations.

Unrealized speed

[essdodson]

Hopefully as price drops and more people purchase the pIV chip the power of the chip will become more and more applicable. At first look the pIV may look like a bomb, a dud, a slow chip. But, the chip has great potential which is yet to be realized. As more and more applications are made available which are optimized for the pIV we'll really start to see this chip shine.

Re:Unrealized speed

[Transcendent]

too bad that you have to specially make the program for the chip for it to actually start to "shine".... intel sucks....

Re:Unrealized speed

[ncc74656]

As more and more applications are made available which are optimized for the pIV we'll really start to see this chip shine.

That does absolutely nothing to improve the performance of older apps that you might have...apps for which you might well have forked over a considerable amount of money. Maybe it wouldn't have been so bad if, in those early benchmarks, the P4 had been able to at least keep up with the P!!!, let alone the Athlon. In particular, I recall how people ragged on the K6-series processors for their FPU performance. I wonder why similar noise hasn't been made regarding the P4's subpar x87 FPU performance.

Cheaper prices are all good, but I still don't see any reason to switch away from AMD.

Re:Unrealized speed

[homer_ca]

Yup, by the time SSE2 optimized compilers become widely used, the P4 will be old news. For plain-old i686 code, it's a dog. A 1.6GHz P4 runs as fast as a 1.2GHz PIII. Suddenly, that 2.4GHz P4 don't sound so fast any more.

Re:Unrealized speed

[Junks Jerzey]

Hopefully as price drops and more people purchase the pIV chip the power of the chip will become more and more applicable. At first look the pIV may look like a bomb, a dud, a slow chip. But, the chip has great potential which is yet to be realized. As more and more applications are made available which are optimized for the pIV we'll really start to see this chip shine.

I'm sure you have a heart full of hope when you say that, but don't hold your breath. Speaking as a developer, you get much bigger increases from across the board optimizations than twiddling around with chip specific tweaks. It just doesn't pay off. "When the code is optimized for processor XXX" has been a standard line since the Pentium II days, and it just hasn't been true in practice. Heck, nobody is optimizing for the Pentium III yet!

Re:Unrealized speed

[ljaguar]

What? GeForce 4's got some feature that has to be _utilized_? Man, Nvidia sucks...

What? There's a new fancy-ass features in 386? What? Altivec? What? T&L?

Progress needs change.

Fortune says:
-----
Imagine that Cray computer decides to make a personal computer. It has a 150 MHz processor, 200 megabytes of RAM, 1500 megabytes of disk storage, a screen resolution of 4096 x 4096 pixels, relies entirely on voice recognition for input, fits in your shirt pocket and costs $300. What's the first question that the computer community asks?

"Is it PC compatible?"
-----

It's because of people like you that we are stuck in x86 in the very first place.

Re:Unrealized speed

[EnderWiggnz]

>It's because of people like you that we are
>stuck in x86 in the very first place.

and what a wonderful, well-thought out, coherent instruction set, designed for a friggin' /calculator/ it is...

Re:Unrealized speed

[Junks Jerzey]

No amount of P3 opts will speed up gcc though, what we need are string functions in the CPU, we have had enough matrix/mmx/3d shit added to the cpu, how about some real optimized memory type funcs()s

You mean like the string functions that were in the original 8088 and 8086? Heh. They're still there, just generally slower than emulating them with simpler instructions. movsd is still lightning fast, though. What more do you want?

Software Companies

[GNUman]

Will software companies ever get to do this, they seem to be always charging more for their work... and it's cheaper to copy a CD than to copy a processor...

Then again, that's a two way blade, it's easier for people to pirate their software than to pirate their chips...

Re:Software Companies

[FrostedWheat]

and it's cheaper to copy a CD than to copy a processor

Not true .. we copied an i386 for a project once. Yea, it ran at about 15 Hz - and it only had about 6 instructions. But it only cost a fraction of what the i386 cost at the time! >:0)

8 bytes of memory to, if I remember correctly.

Some minor modifications of your programs may be required....

When CPU manufacturers fight over you...

[Kargan]

...you win!

There is no winning.....

[diorio]

....the minute you buy something the price drops! I have spent a lifetime battling this. AND, if you wait....the technology will change.....arrghghghghg!

.

Re:There is no winning.....

[nil_null]

....the minute you buy something the price drops! I have spent a lifetime battling this. AND, if you wait....the technology will change.....arrghghghghg!

So true, but the best way to go is to stay behind in technology (for consumers anyone, not business). That's why I'm still running on a K6-2+. I basically set a price at which I'm willing to buy something, and I set it very low. For example, for the last year I waited for an 80 gig EIDE hard drive to hit $100, and when it did, I bought it. Similarly I'm waiting for a good $100 motherboard and processor combo to come a long.

Granted, I feel the difference when I use faster PCs. But hey, I'll have that same speed when the time comes, and I'll have it cheaper!

Re:And now #1

[essdodson]

Wrong article? The eminem article is next door...

Good News

[quantaman]

Aren't Intel's prices almost twice as much as AMD's already for mostly equivalent processors? I take this to mean that Intel has decided that AMD is now a veritable threat and as such is no longer pricing like they are the only option. This will take a chunk out of AMD's sales for sure (even if they make similar price cuts) but I suspect that its main purpose will be that knowledgable comsumers will now consider Intel a viable option again.

Re:Good News

[Hoser McMoose]

Perhaps, but it's rather important to look at what chips they cut prices on.

The desktop P4's were cut by, at most, 29%. This means that it's likely that AMD desktop chips will still be cheaper. These P4 price drops were very typical of what Intel and AMD do every 3 months or so.

Where they really cut the prices were in their mobile P4-M chips, where were just released. This isn't too surprising for two reasons. First, they're brand new and this is their first price cut (ie right after the "I gotta have it NOW at any cost" buyers have been taken care of). Secondly because they just aren't very good chips for notebooks. The P4-M consumes a LOT of power for a notebook chip, particularly the higher speed ones where Intel cut the prices the most. All of these are in the 30W+ range. I'm sure that OEMs like high GHz numbers for their marketing efforts, but designing the cooling for one of these chips must be a bitch! Meanwhile AMD just brought out their new notebook chips which consume 20-25W of power for comperable performance, they have more advanced power saving features and they're cheaper to boot.

In short, Intel probably just couldn't keep going charging the rather ridiculous prices for these laptop chips (they were QUITE expensive). When your competitors have a cooler running chip that performs the same or better and sells for less, marketing and clock speeds can only last you so long.

AMD price cuts expected tomorrow

[Seth Finkelstein]

... it'll be interesting to see what happens to the prices of the competing AMD offerings.

From more coverage at ZDNET:

Advanced Micro Devices, Intel's rival in processors, will likely cut prices to match Intel's cut. AMD typically announces price cuts a day or so after Intel. The Sunnyvale, Calif.-based company is also expected to soon release "Thoroughbred," a faster version of its Athlon chip, made on the 130-nanometer manufacturing process, for desktop computers. The company is currently shipping the chip to PC makers, a spokesman confirmed, and will release the chip to the public shortly.

Sig: What Happened To The Censorware Project (censorware.org)

Re:AMD price cuts expected tomorrow

[nexthec]

I hate to be a troll, and I know I will get flamed or down modded for this but here it goes:

I uncheck sigs, so that I dont see redundant information in posts, and see peoples innane(usually) comments. why put it in every single time? I guess my options are to scrolll by every one of your posts or set you to permantly be hidden. But that really doesnt get to the questions, why are you so obsessed with posting this link?

Re:AMD price cuts expected tomorrow

[discstickers]

Did you read the link? I think that explains it ;)

If you don't like him, you can make him a foe and block his posts.

Re:AMD price cuts expected tomorrow

[Sgs-Cruz]

For pity's sake, finally someone refers to the manufacturing process as 130-nanometre. (or meter).

I was wondering when we'd stop seeing .13 micron, etc.

...a faster version of its Athlon chip, made on the .00013-millimetre manufacturing process...

:)

-Cruz

I'd love to upgrade my CPU, but...

[stefanlasiewski]

I'd love to upgrade my CPU, but I've got one of these Slot A things that Intel abandoned so many years ago.

So really, to upgrade my CPU, I need to get a new motherboard. To get a new motherboard, I probably need to get a new case & power supply, maybe some new RAM... and hell, at that point I might as well get a new computer and plug in some of my old peripherals.

Either way I'm out $500-1000 ... think I'll just stick with my Celeron 366, it functions well enough...

Re:I'd love to upgrade my CPU, but...

[acidblood]

Slot A was the first packaging for the 0.25 and 0.18u AMD Athlon CPUs, soon abandoned in favor of Socket A, still in use today.

What you actually have is a Slot 1 motherboard.

Re:I'd love to upgrade my CPU, but...

[Jimmy_B]

So really, to upgrade my CPU, I need to get a new motherboard. To get a new motherboard, I probably need to get a new case & power supply, maybe some new RAM... and hell, at that point I might as well get a new computer and plug in some of my old peripherals.

You shouldn't need to replace your case and power supply, unless you have an old AT case (ATX is now standard). As for complaining about buying a new motherboard and RAM...well, it'd be stupid to put a fast, new CPU in a machine with 66MHz RAM, so really, you just keep the motherboard, CPU, and RAM together. You don't have to upgrade your video card, hard disks or monitor (all similarly expensive components) if you don't want to.

Either way I'm out $500-1000 ... think I'll just stick with my Celeron 366, it functions well enough...

If it does function for what you do with it, fine, keep it. But the high end is lead by early-adopters who buy hardware so they can run games, and an old PC won't cut it for those. You're not in the market this is targeted at.

Re:I'd love to upgrade my CPU, but...

[VVrath]

Slot A was also used for some 'thunderbird' cores (.18 micron), but these only went to OEMs.

I recently replaced my slot A board to upgrade, and only had to replace memory, mainboard and cpu - cost me around £250. Certainly cheaper than buying a new PC :)

Liam

Re:I'd love to upgrade my CPU, but...

[sheetsda]

I scrapped my old computer a couple weeks ago and upgraded my processor, RAM, motherboard, and video card for around $510. Since you didn't mention a video card I'll assume you have one, mine was $240. That leaves $270 for Athlon XP 1800+, Asus A7S333 motherboard, 256 meg PC2100 DDR DIMM. Throw in a decent case/power supply and I'd say your total would come out more like $350 than $500. You might want to check Pricewatch if you haven't lately.

Re:I'd love to upgrade my CPU, but...

[ljaguar]

I assume you mean slot 1.

Actually, I did the calculations a few days ago. So I can say with confidence.
For the record, I have P-III 450Mhz the slowest ever manufactured. It's of course, slot 1.

There is actually an adapter from flipchip->slot 1 (http://www.pricewatch.com/1/239/2143-1.htm)($14), so you can upgrade your computer to around a gig. If you have slot 1, you most likely have 100Mhz FBS. These new pentium III have 133mhz, so you will either underclock or look for the rare 100Mhz chips. If you insist on slot one, the Pentium III 1000Ghz 100Mhz slot 1 can be attained for around $183.

The catch is, you can't use the most recent P-III called "taulatine," I believe featuring flipchip 2 or something or the other. Taulatines include some 1.0 Ghz and all 1.13 Ghz and above.

Now the _real_ catch is, a decent motherboard + recent AMD Athlon XP costs just as much.

For example, this ECS-K7S5A (which is nice, because it still can use non-DDR DIMM's) costs under $60 and ATHLON 1700+ costs $110. Good deal, if you ask me.

Re:I'd love to upgrade my CPU, but...

[(H)elix1]

Not quite so fast... you can get a slocket - slot 1 to socket - adapter cheap. I use an Abit adapter on my SuperMicro SBU with an 800 mhz celeron, a fair jump from the PII 350 it started life as.

You want hosed, you should see my old slot 2 xeon board. It will never see anything faster than the dual 450... With slot 1, you have a chance to upgrade cheap.

Re:I'd love to upgrade my CPU, but...

[_ph1ux_]

ya but - to be out 500 or 1k will give you one kick ass machine these days. how much was your cel 366? when i worked for intel doing the celeron pushes et al - they were supposed to be the sub 1k (999.99) machines for home users. now anything over 1k is considered rather overpriced - and the machine you get is expected to be quite powerful. but for a few hundred you have the perfect gameing/workstation and in some cases server boxen you could want.

its interesting that we now see the most expensive single components of the machine are the video card (if youre getting a nice new spiffy military grade Gf4) and the monitor (if youre getting some nice new spiffy large scale flat screen.

I love it - I just wish the economy was good enought to give me a job so i could upgrade my slow POS 800 with a weak ass Gf2 64mb.

Re:I'd love to upgrade my CPU, but...

[slashhot]

I've come to a conclusion that upgrading a processor is a myth. In the old times (486 and before) all processors from the same family used the same socket, so upgrading was possible. Nowadays it is not. Yes, you may buy a better soundcard or graphics card, you may add some RAM, or even a larger and faster hard disk. But you'll probably never upgrade the processor, because by the time it's so old you must upgrade, you won't find a suitable replacement.

Re:I'd love to upgrade my CPU, but...

[ealar dlanvuli]

I did the same thing two weeks ago, it was 380 for a p4 1.7gig and the works. I spent another 190 on a graphics card (I'm not that big of a gamer). It was significanlly cheaper than my system origionally was, because I didn't have to buy a new case/hd/dvd/floppy/nic/sound card, you will be amazed at how much those save.

Re:I'd love to upgrade my CPU, but...

[Xoro]

Well, for the adventurous there is now a hack to mod a tualatin chip to fit a slotket. The great thing is that you can use it with the 1 or 1.1 GHz Tualatin Celeron, which is at 100MHz FSB (and about $70). Since these Celerons are a .13 mu part, they can easily be overclocked by setting your bus to 133MHz.

I was just about to do this, but found *another* catch: all the cheapo RAM uses these new 64x4 chips, but my BX only handles the 16x8 kind, which is twice as much at the low end. So the gap is narrowed. Of course, for all the time I've spent mooning over the options, I could've worked flipping burgers and earned enough to buy two new PCs. Trapped again.

Re:I'd love to upgrade my CPU, but...

[klieber]

You shouldn't need to replace your case and power supply

Actually, that's not true. With the P4, Intel now requires an ATX12V PSU, instead of just a "regular" ATX PSU. The main difference is the +12VDC connector on the newer PSU.

http://www.intel.com/home/tech/components/power_su pply.htm

Re:I'd love to upgrade my CPU, but...

[athakur999]

Same here. I have a v1.0 Abit BH6 board. I bought an MSI slotket and a 1.1gHz Celeron chip, and things are going fine. Just make sure you get a non-Tualitan version of whatever chip you get.

Re:I'd love to upgrade my CPU, but...

[shepd]

As a system builder, I'd say he needs to:

- Upgrade his case. Most older (and cheaper, and therefore popular) ATX cases were simply not long enough, and most newer good motherboards will have RAM and/or capacitors blocking all but your top CD-ROM.

- Upgrade his power supply. New AMD processors require _much_ more power than most of the standard power supplies in most older cases. New P4 motherboards have new P4 power connectors that have to be hooked up, or you'll need to find the rare motherboard with a substitute +12v molex connector.

- Buy new RAM, for the reasons you stated.

- Upgrade his video card. Some AGP cards from that time turn out to be incompatible with newer motherboards (due to either voltage or current problems, I can't remember which). Read the manual for your new motherboard and you may notice a list containing all kinds of old AGP cards that may ruin the new motherboard if they are used with it.

- Upgrade his hard drive. What's the point of having a new system if its bogged down by a slow 5400 RPM hard drive? Or, for that matter, why have a new system if your shiny new O/S takes up half your drive?

- Get a new ATX panel to cover the back of the motherboard (unless he wants to cut out the space for an ethernet jack some new boards are coming with).

BTW:

Re:I'd love to upgrade my CPU, but...

[Dr. Spork]

Well, he could keep his case, but only if he's not buying Intel, because the P4 heatsink requires a special P4-compatible case. Anyway, it's no big deal. Get a new case--a case with a modern power supply isn't much more expensive than a modern power supply alone, and you'll need one of those for sure. I recommend you buy a bare-bones system from someone. You can get decent DDR motherboard in a case with a power supply and an Athlon 1800+ starting at about $200 (+ shipping). That's not so terrible. Actually, I think this is a good time to upgrade, I mean, after the chip price cut and after a significant drop in DDR prices.

However, if you're happy with your Celeron, hold on to it as long as it serves you and then jump straight to the Hammer. I'm sure you'll find some use for that sort of horsepower. I know three years ago nobody would have thought that you'd need anythig more than a 2GHz chip, but that's before everybody started encoding their own movies. Then, all of the sudden, the previous generation of chips, which had seemed totally adequate, just couldn't do the job. I'm not sure what we will ask of the Hammer generation, but I'm sure we'll think of something.

Re:I'd love to upgrade my CPU, but...

[David Jao]

Just make sure you get a non-Tualitan version

Actually, a little known secret is that there exist Tualatin-compatible slot 1 slockets, and they work quite well. I've been using one for months and I am very happy with it. (Disclaimer: I have no relationship with PowerLeap other than that of satisfied customer.)

What's the formula?

[d5w]

If anyone actually wants this much information, here's the full pricelist.

The article says that Intel is attributing the price cuts to higher yields, which in turn are due to large investments in its foundries. I'm a little puzzled by this, since this is suggesting that mass-market chip cost actually has something to do with supply, whereas I'd generally assumed that most chip prices were determined by some combination of development cost and demand (i.e., you'll have enough chips; just charge as much as the market will bear and if development is expensive enough you won't have enough competition to bring the price down). The latter is almost certainly true for many server chips. How much is the price of high end mass-market chips actually determined by supply limitations these days?

Re:What's the formula?

[tjb]

I work for a chip company that does somewhat mass-market chips (around 2 million various chips per quarter), and based on what I know (I don't work much with the manufacturing side, but do deal with LSI designers who bring up their concerns), here's some back of the envelope calculations:

Our single-core chips cost around $3 each for a good chip. Die size is around 80 mm^2, yield is ~70%, process is .18micron.

Intel processors are ~400mm^2, so they can get at best 1/5 the chips on the same size wafer. This brings cost to $15 using my starting point. Their yield would be 0.70^5 or 16% - but lets assume that since Intel runs their own foundry, they have better kung-fu than the fabs we use, so maybe 35% yield. That brings us to $30, and I'm talking out my ass at this point.

However, if they didn't have the improved yields I was assuming, you're looking at $60/chip or more just in fabbing cost. Semiconductor companies tend to like at least 60% gross margins, so a $60 chip would sell for $150 minimum, not counting cost of sales.

Tim

Re:What's the formula?

[ntillery]

How much is the price of high end mass-market chips actually determined by supply limitations these days?

A lot of it actually. When a new chip is manufactured it takes time to optimize the line to produce more high speed chips. You're going to produce more 2Ghz chips then you do 2.4Ghz because its so much more difficult to manufacture die that can function at high speeds. The amount of 2Ghz chips you produce vs 2.4Ghz chips is called sort yield and it determines your average selling price. To compete in the semiconductor industry your high end sort yields must be continuously increasing on pace with Moores law. Since the P4 die is so much larger then that of the P3 Intel had to upgrade its older 200mm wafer fabs to 300mm (hence the factory investments the article talks about) to keep pace. This increase in wafer size allowed them to increase the amount of die per wafer, which increases their yields. Higher sort yields allow them to lower their ASP while still maintaining the same level of profit. This also puts pressure on AMD who must now cut prices to in order to maintain its market share.

Rinse and repeat. This is nothing new and not unexpected. It's how the semi industry has always worked and will continue till we stop manufacturing ICs on silicon wafers. A textbox case of supply and demand.

Re:What's the formula?

[eechuah]

That's not true. If you can make a chip for cheaper (ie. higher yields), you automatically can sell it for cheaper. It's simple math, really. Intel recently moved to 300mm wafers, giving it higher yields per wafer, causing each P4 die to be cheaper.

Re:What's the formula?

[rlangis]

The article says that Intel is attributing the price cuts to higher yields...

I have some friends who work at Intel, and they said that the chips that they get in award plaques are *functional*. Used to be, when you got an award plaque at Intel (I used to work there too) the chip you got on the plaque was a non-functional garbage-bin chip. Good for looking at, ONLY. Nowadays, it seems the chips they are giving out are _very nearly_ consumer-level chips. I heard that the only reason they are put on the plaques instead of sold to the general public was because they failed ONE test at below acceptable margins. Doesn't mean they won't work, but they might not take a beating.

But then again, what do you expect for FREE?

and i'd just bought an athlon!

[Matthew Luckie]

I just bought myself an athlon XP 1800+, and a day later i am regretting it. AMD markets these processors as 1800Mhz CPUs, when in actual fact they are just 1500Mhz CPUs.

Intel clearly labels the P4 line with its true clock speed. No false advertising there! Just think that a P4 2.2Ghz machine is 50% faster than an athlon 1.5Ghz, now for the same price! Not only that, I don't have to worry about my heatsink falling off!

Re:and i'd just bought an athlon!

[Transcendent]

Actually they dont.... if it was advertised like that, then blame the morons at the company you bought it from.... or blame yourself for not knowing what you were buying... how is it false advertising when "XP 1800+" says NOTHING about the clock speed???

Obviously you don't know much about processors if you think that the 2.2ghz is 50% faster than the 1.5 because of the clock speed... when speed really has nothing to do with that. Take a G4, for example, at 1ghz (which does a gigaflop... ::drooll::)... the P4 is over %200 "faster", yet the G4 performs twice as good in bechmarks...

yea, most people don't really think about their heatsink just falling off.... I really don't worry about that when buying a processor....

Re:and i'd just bought an athlon!

[captain_craptacular]

Where the hell have you been living? I think even my mother knows that AMD markets the XP line using a "performance rating" and not actual MHZ numbers. If you're going to bitch about what you buy, maybe you should try a little research ahead of time...

Re:and i'd just bought an athlon!

[Kargan]

Um, yeah, the P4 is labeled with its true clock speed, but guess what? If you take an AMD 1.53GHZ and match it up with a P4 1.5, the AMD will outperform it in every single category, except possibly some programs optimized for SSE2 instructions.

The (*still* less expensive, even with the latest Intel price cuts) AMD 1800+ is more suited to compete with Pentiums quite a bit faster than itself, starting at about 1.8GHz.

Clock speed doesn't mean anything as compared to just other manufacturers' clock speeds. Are you going to tell me that your P4 2.4 is faster than the fastest Alpha or Sparc processor, just because the clock speed is higher??

The architecture is completely different in each case, and can't be compared so simplistically.

Re:and i'd just bought an athlon!

[droyad]

Ah, the gigahertz trap, you've fallen in it even when trying to disprove it. It is fair to compare AMD and Intel by MHz (to an extent). Both these chips are functionally identical
Comparing a G4 and P4 does NOT work, as they use a different archetecure. a P4 uses "Complex Instruction Set" while the G4 (motorola) uses a "simple Instruction Set". This means the G4 can execute more instructions per clock cycle than the P4 due to things like pipelining being easier. The downside is that compilers for the G4 are complex to take advantage of this piplining

Re:and i'd just bought an athlon!

[ncc74656]

I just bought myself an athlon XP 1800+, and a day later i am regretting it. AMD markets these processors as 1800Mhz CPUs, when in actual fact they are just 1500Mhz CPUs.

Bullshit. AMD has been more than forthcoming with its view that "megahertz über alles" is a Bad Idea. 1800 isn't the speed at which the processor runs. It's a performance metric that happens (for the time being) to track rather closely with what a P4 at X MHz will deliver, but the processor can deliver that performance at a slower clockspeed. Get a clue before you post next time, go back under your bridge, and consider yourself LARTed.

Re:and i'd just bought an athlon!

[Petrus]

Because your AMD can run 9 unstruction at the same time unlike Intel that can run only 4, it will be in average as fast as 2GHz Pentium and sometimes as fast as 9/4*1.5GHz Pentium, that is about 3.3GHz Pentium. See Distributed.net or POWbench benchmarks at linuxhardware.org So AMD has more than double of parallelism of Pentium4 and it can be vastly faster in code that can be highly parallelised. What I call cheating is when Intel says that they have higher clock speed therefore they have faster processor. Petrus

Re:and i'd just bought an athlon!

[toast0]

actually amd's latest processor work is to do more in a clockcycle, whereas intel's latest processor work is to deepen the pipelines, and avoid pipeline stalls, so they can boost the clock way up, but the end up doing less per clock.

so in the end, the amd and intel chips are about as different on performance as an amd and a motorolla

Re:and i'd just bought an athlon!

[shepd]

All highend processors are now RISC.

The difference is that 6x86 has more instructions stored interally that are basically translated to a set of RISC calls through microcode. :-)

This means that Photoshop was faster on a Mac not because of a faster processor, but because the restrictions on what the processor can do for you are tighter and they therefore had to write better code for that processor.

If they had limited themselves to Motorola-style set of instructions on the 6x86 (of course some are simply missing and one would have to code around that little problem), I have no doubt that photoshop would have kicked ass on the 6x86.

Re:and i'd just bought an athlon!

[Hoser McMoose]

AMD replaced one TOTALLY MEANINGLESS measure of performance (clock speed) with another totally meaningless measure of performance (their model numbers).

Think of clock speed like the displacement of a car engine. If everything else is equal, a 3.0L engine will give you a lot more power then a 2.0L engine. However there are plenty of 3.0L engines out there that only produce 180HP or less, while there are some 2.0L engines that produce over 200HP, and that's without taking into account things like turbochargers. What's more, if you're interested in the actual speed of the car, there's a TON of other factors to consider outside of the engine itself.

So, does having a larger displacement make the car "faster"? It might, but it's only one factor in determining the power that the engine will give you, and that in turn is only one (relatively small) factor in determining how fast the car is.

Same thing with processors. Having a higher clock speed might mean a faster computer, but it's only one of MANY factors in determining how powerful the processor is, and that in turn is only one of MANY factors in determining how fast the overall system is.

Ohh.. And if your heatsink falls off, either your cursed with insanely bad luck, or your a moron. If you happen to run a website with the initials THG, then it's certainly the latter. Hint: Tom found the P4 running at 29C without a heatsink. The P4's thermal throttling kicks in at 60-70C. Do you see a problem? There's a reason why Tom has a degree in medicine, and not electrical engineering, computer science, or anything of the like.

Re:and i'd just bought an athlon!

[dolby2]

mmm I will vouch for that I run distributed.net and a 1800+ athlon is faster in pure processing power then a 2.0 pentium almost as fast a 2.2..

amd *drool* am I in heaven?

Re:and i'd just bought an athlon!

[dolby2]

Well put.

Intel has the support chips

[cardshark2001]

I have bought Athlon the past two times I built a computer.

I hate to say it, but both computers suffer from problems such as lock-ups, random reboots, and other compatability issues, especially when playing directx games. I bought the second board (and chip) because the first one did not work. I even bought the board that TomsHardware recommended as the best athlon board at the time (MSI K7-Master S).

The AMD chip is faster, but my Intelly friends have had NONE of the problems I have had when running the very same programs. Therefore, no matter how much more it costs, or how much slower it goes, I will buy Intel in the future, and recommend that my friends do the same.

It is a real shame, because I think the Athlon is a better chip. I just won't trust Athlon boards anymore. If they made a chip that was compatible with an Intel board, I'd buy it.

Re:Intel has the support chips

[SystemFork]

Perhaps it's not the chip, but that you're reusing a crappy component each time you build the machine. Like a sound card and network card that refuse to play nice together.

To be completely fair though, I had a problem once with a $20 Montego Turtle Beach sound card, it wasn't compatible with AMD processors. Weird. I returned it and bought a cheap Sound Blaster Live. The Montego wasn't much good anyway.

Re:Intel has the support chips

[cardshark2001]

I thought of that. The only component I reused was the Geforce 2 MMX. I had a friend try it out for a week just to make sure that was not the problem. It wasn't.

Re:Intel has the support chips

[gmack]

tried a larger power supply?

Although AMD really did get burned by the motherboard makers this round. It looks like they learned from their mistakes on the Opteron by simplifying things to a design not even VIA can screw up.

Re:Intel has the support chips

[SystemFork]

Smart man, it sounds like you covered all the bases.

I'm running a similiar setup: AMD 1600+ XP & GeForce 2 MX-400. It seems as stable as most Windows ME running computers. (It's never crashed in Linux.)

Sorry to hear that you got stuck with a lemon, I once had a Cyrix 686 and well -- we all know about those.

Re:Intel has the support chips

[cardshark2001]

I even tried a larger power supply. I changed EVERY component, down to the case.

I am not the only one who's had these problems. Diablo II, Dungeon Siege I could write off as bad programming. But Quake 3? I will not badmouth Dr. Karmack. :)

Re:Intel has the support chips

[tshak]

I have bought Athlon the past two times I built a computer.

Maybe my emphasis points to why you are having problems :-)... j/k.

Seriously, with all of the potential points of failure how can you know that it's not the OS, App, Cheap RAM, the Chipset, etc. Plus, I've never been a huge fan of MSI (no concrete reason). If you want rock solid stability go with Asus. I've been running on an Athlon1.2Ghz with Win2K and I have incredible uptime with the box - the only "hard" lockup I've ever had is when I tried overclocking it to 1.6Ghz just for fun. I had an Intel pII350 before that and it wasn't nearly as stable, but I can tell you that it was most likely due to my Win98 install and not the CPU. Finally, I have an old AMD K6 266 that's still running great as my MP3 server/CD Burning station. Really, with all of the "Math" bugs found in the Pentiums, and after the K6's proved AMD's stability, I don't think that there's any objective data that shows that one CPU is more stable then the other.

Re:Intel has the support chips

[jonbrewer]

"suffer from problems such as lock-ups, random reboots, and other compatability issues"

I'd say you have software issues. I've been running an Athlon 800 since October 2000, and it's absolutely rock solid. When I bought the machine from SYS, it ran WinME and locked up a few times a day. Less than a week later I blew ME away and replaced with 2000, everything was fine. I'm now running XP, and reboot about once a month when security patches require it.

The processor in your machine doesn't have squat to do with compatibility. It's all about the software.

Re:Intel has the support chips

[ncc74656]

I hate to say it, but both computers suffer from problems such as lock-ups, random reboots, and other compatability issues, especially when playing directx games. I bought the second board (and chip) because the first one did not work. I even bought the board that TomsHardware recommended as the best athlon board at the time (MSI K7-Master S).

The AMD chip is faster, but my Intelly friends have had NONE of the problems I have had

Sounds like you more than likely have crappy components somewhere in your systems. I have a bunch of Athlon systems at work and one at home, and they've never given me any trouble at all. They've actually been more stable than some of the P!!! systems that we also have at work. That the Athlons are mostly systems I built myself from carefully-selected parts (chipsets involved are the AMD 760, AMD 760MPX, and nVidia nForce) and the P!!!s are mostly HP Pavilions reinforces this point.

(The home system is a 1.0-GHz Athlon (Thunderbird) on a Biostar M7MIA with a mix of Crucial and Mushkin DDR SDRAM, ATI All-In-Wonder Radeon, 3Com 3C905C, Tekram DC315U Ultra SCSI, SIIG UDMA100, and no-name FireWire controller. Two of the work systems are 1.4-GHz Athlon XPs (1600+) on MSI K7N420 Pros with generic DDR SDRAM and onboard everything (added no-name FireWire to one and a generic RTL8139 to the other). Another work system is a dual 1.6-GHz Athlon MP (1900+) with Crucial registered ECC DDR SDRAM, ATI Radeon, integrated 3Com 3C920, and a PCI sound card. Three of the four systems run Win2K Pro SP2, while one of the nForce systems runs Linux From Scratch. I've also got the home system set up so it can load SuSE from a FireWire hard drive. Note that neither VIA nor Creative Labs appear anywhere in the descriptions above...well, the M7MIA uses the VIA 686B southbridge, but that's all, and it hasn't given me any problems. Come to think of it, the VIA chipsets I've used with various K6-* processors in the past haven't given me much grief either.)

Re:Intel has the support chips

[Sivar]

I haven't used an Intel chip since a Pentium MMX 200MHz. Not that I have anything against Intel chips, they are engineered very well (with a few exceptions), but AMDs are generally faster. At the moment, other than in very FPU oriented code, the Athlon is not the fastest, though I am sure AMD will leapfrog Intel only to be leapfrogged themselves a few weeks/months later.
With that out of the way, there are many individuals and companies and even Anandtech, possibly the most respected hardware site on the web, running Athlons with no problems. There have indeed been problems in the past, mostly due to a certain company whose name starts with 'V' and ends with 'A' and their chipsets, but even those problems have been worked out.
Anyway, AMD does make chipsets for their platforms, and they are consistantly outstanding in stability. My 1st generation Tyan TigerMP based on the AMD 760MP chipset has never once crashed, at all, even in Windows 2000. I have run it to the ground with three video cards, games, 3D modelling software, compiling big projects (like FreeBSD, Gentoo Linux, KDE3, etc.) and have run two copies of SETI@Home in the background at all times.
AMD chips do not have incompatibility issued. AMD chips do not crash. Nor do Intels. Yes, there have been very obscure problems in the past with both, such as Intel's 1.13GHz P3 and AMD's very old K6's that didn't like a system to have >40MB of RAM, but those are past.
The secret is to use good components. There are many opinions and recommendations out there, most of which are bullshit. Find some sites and friends that you trust, and verify their claims. I have found that most people, including myself, are full of shit and that most websites, even the most respected such as Tom's Hardware, are completely full of BS. Make it a hobby--keeping up with hardware. It helps. For the record, I use Antec and PC Power & Cooling power supplies, Antec cases, Plextor SCSI CD-ROMs, Adaptec low-end controllers and Mylex higher end ones, MSI-ASUS-and Tyan motherboards (though you must go with the specific model, all companies have duds), Corsair memory, Matrox and ELSA video cards (Gainward Geforce cards for gaming systems), Samsung & Maxtor IDE hard drives, Maxtor and Seagate SCSI hard drives, and I have yet to find a soundcard that does not suck. (if you have any suggestions, let me know!)
Do not take my word on ANY of these parts, though. As I said, I am probably full of shit, so you should verify anything under consideration on your own. Just how you verify it is where you need to get creative. :-)

(disclaimer: I'm not saying you do not know what you are doing, but obviously a mistake was made at some point building your system that caused instability. Be it hardware or software, I do not know)

Re:Intel has the support chips

[SystemFork]

You know what? It wasn't the AMD processor that bothered the Montego sound card, it was the VIA chipset used on the cheap motherboard I bought. It just returned to me out of the flash of the ether of memory...

Re:Intel has the support chips

[karnal]

Hey!!! :)

I've got a cyrix m2-300 for a file server... never had a day of downtime other than my own screw-ups... and bad kernel code for the promise ide card (darn thing will never see ata-100 without corrupting a disk... on a READ!)

I've had issues with all of my AMD systems, though, and I feel it has to do with board/chipset selection. For instance, I currently run a tbird 900 with an Abit kt7-raid. Once top rated on toms hardware (of course, that's when I bought it) and then all of a sudden, people started bitching about it.

Me too.

There's just something not right. I've replaced everything around this board, and every now and again it gets finicky.... (sp?) but for the most part, it is solid, and doesn't give me a problem. (p.s. their highpoint raid solution sucks for a home user, but for a server, I'm sure it's a godsend...)

Re:Intel has the support chips

[ealar dlanvuli]

Asus p5a with an amd k6-III 450 3.3v is the most unstable mobo/cpu combination I've ever seen.

I don't buy asus anymore, because they refuse to acknowledge the problem, even though I know 3 other people who have it. I have replaced every single component (even the powersuply) and the random reboots still happened.

I went p4 on my most recent system because of my bad experience with the asus motherboard, and I've seen very few p4 mobo's that were unstable.

Re:Intel has the support chips

[hey!]

Hmmm. Did you buy quality all around? Especially RAM. Sometimes when you see cut rate RAM it's because it's made with reject chips that sort of work but not to their full spec. I've seen systems with mysterious lock up problems that simply went away when quality memory was installed.

Re:Intel has the support chips

[NineNine]

Is there a particular reason that you inventoried every fucking computer component that you own in this post?

Re:Intel has the support chips

[ncc74656]

Is there a particular reason that you inventoried every fucking computer component that you own in this post?

It's not everything, but at least it's stuff that's known to work since the original poster was claiming problems with whatever stuff he has...

Re:Intel has the support chips

[bedessen]

both computers suffer from problems such as lock-ups, random reboots, and other compatability issues

I'm not a betting person, but if I were I'd put my money on cheap ram. Or at least, on anything but the actual processor. Fire up memtest86 some time and let it crank through it's longest tests, and see if you don't come up with some spotty ram issues.

It's interesting how the DIY hardware scene has changed over the years. The components that used to be no-name commodities are now becoming more critial. If building a system today, I would spend more time worrying about the brand of ram, power supply, and cooling than I ever would have five or ten years ago. It used to be that all you needed to know about ram and heatsinks was "it's 80ns" and "yeah it comes with a fan", respectively.

Confusing numbers

[CmdrTaco (editor)]

Intel shaved the price of its second-fastest processor, the Pentium 4 running at 2.4 gigahertz, by 29 percent to $400 from $562

Yet at pricewatch the lowest price listed for the same processor is $395. Does this mean the companies selling them below the list price are selling them at a loss or are they getting an even better deal than this?

Re:Confusing numbers

[bravehamster]

I think that in the article they're probably quoting the price for the 533mhz FSB 2.4Ghz (known as the 2.4B) The pricewatch list you're looking at is for the 2.4Ghz with the 400mhz FSB. Still...that original $562 price they quoted seems way too high.

Re:Confusing numbers

[Lord Ender]

The way companies do business on pricewatch is by listing very low prices, then charging $20 shipping when it actually only costs $5 to ship the product. That makes the $395 become $410. This is not a huge proffit but it is not selling at a loss.

Re:Confusing numbers

[JayAndSilentBob]

you're paying $2 for a pack of cigs!?!?!?!? You're getting screwed, buddy. Try going to the discount tobacco stores in your area. Cigs go as low as $1.10 a pack and "little cigars" for $1.50 / 2 packs. Do some price shopping, buddy.

Re:Confusing numbers

[bedessen]

The press releases of Intel and AMD are always somewhat of a game. They list the "suggested retail price in quantities greater than 1000" but that's not really how it works. There's no single "retail price" of a CPU. The chipmakers are constantly negotiating deals with suppliers for large lots of CPUs. The price fluctuates a lot and is often quite a bit less than what they quote in their press releases. I think the chipmakes use this system because it allows them good inventory control which is critical with such a fast-moving market. I think it also allows them to accurately gauge demand for their products, since the pipeline for producing a chip is many months. You need good info to make the right decisions about production.

what Pricewatch has to say

[Kargan]

Okay, I was curious enough to check Pricewatch (www.pricewatch.com) and found about the same story there as I've found for the last few years.

$395 Pentium 4 2.4GHz
$245 Pentium 4 2.2GHz Sock 478
$195 Pentium 4 2.0GHz Sock 478
$173 Pentium 4 1.9GHz Sock 478

$186 Athlon XP 2100
$146 Athlon XP 2000
$122 Athlon XP 1900
$95 Athlon XP 1800

You can get an AMD 1.53GHz for less than $100 now!!

These price cuts by Intel are long overdue by my reckoning, and while it is a step in the right direction, they've still got a ways to go.

Anyone wanting a CPU upgrade at this point anyway would be wise to wait a bit for the 64-bit CPU price war to begin, it's not far away at all, and then all these chips will look slow and clunky.

Re:what Pricewatch has to say

[ivan256]

It'll be interesting to see if there is any sort of 'war'. Intel isn't going to remove the premium price tag from Itanium, so for there to be a price war some other vendor is going to have to make a 64bit chip, or Intel will have to come out with a low end (not Itanium based) 64bit chip. That would be interesting indeed.

Re:what Pricewatch has to say

[truesaer]

Just a small side note here....I just got a Athlon 1.53 machinge without an OS from walmart this week ($600) and it is so damn fast I can hardly believe it. The memory and everything is bigger and faster too, but this was really a relief to me because my last computer I bought 2 years ago (a dell with a P3) was a dog....it didn't seem any faster than the Celeron 300 I upgraded from.

Anyway, I incidentally read on Tom's Hardware a couple of weeks ago that Intel came up with a new core that is much cheaper to make and is a good performer. I wonder if this is the reason for the cuts. I wish AMD the best of luck though, competition is good and I'm a satisfied AMD customer.

There is just no losing.....

[joshuac]

I mean, hey, everytime I turn around, more and more of these powerful devices are costing less and less! I have now seen this happen first hand throughout the majority of my life, from my awesome 1.023mhz 8 bit 48KB ram/16 KB rom IIe, to nowadays rackable machines far more powerful than the fastest dedicated-room supercomputer's of just 15 years ago. And if you wait...the technology will just get better!!! YEAH!

:)

say goodbye to competition

[Gizzmonic]

It seems that Intel is redoubling its efforts to squash those upstart rebels at AMD. If Intel eventually triumphs, we are looking at a very bleak future in high performance CPUs.

What happened to all the other high performance processors? MIPS, SPARC, PA-RISC? They are/were all attached to high-performance UNIX workstations.

And what happened to those high-performance boxes? Ask the IT dude who's firing up his handbuilt Dual Athlon running Red Hat 7.2.

It's bad enough that the decrepit x86 architecture has lasted this long. With only Intel around, they will extend its lifetime indefinitately, filling our lives with overheating chips that run at twice the Mhz with half the performance...

Re:say goodbye to competition

[Zo0ok]

The others may be coming... just a little slower.

For example, IBMs new G3 cpu (that you find in Apple iBooks) runs at 800MHz consuming less than 10W.

The i386 instruction set forces Intel/AMD to put lots of reasearch an silicon into on-the-fly-compiling of i386-code to something RISC core can handle. This means more expensive research, more heat, higher production costs etc.

When people realise they prefer a silent and cool machine running at 1 GHz to a noisy machine running at 3 GHz the performance advantage of todays i386 cpus may turn into a simplicity/heat disadvantage.

MIPS R14000 at 600 MHz are shipping in SGI workstations (at prices I cant pay), those CPUs are pretty fast when it comes to floating operations (and they are native 64 bit, which means if you need like 64Gb of RAM i386 is not an option anyway). Also, you can put like 512 of those in a single machine: define high-performance ;)

Apple ships PowerPCs at 1GHz. Hopefully something significantly faster will be presented during the summer.

Re:say goodbye to competition

[WasterDave]

It's bad enough that the decrepit x86 architecture has lasted this long.

I think it's a bit of a myth that the x86 architecture has actually lasted at all, these days it's really just a way of storing instructions - the instructions themselves get converted on the fly to whatever Intel/AMD really use. Strangely, as memory bandwidth has increasingly become the scarce resource, CISC instruction sets are going to win out over RISC. Not that I'm defending x86's design, mind you.

Dave

Intel's Questionable Math

[geoffsmith]

How they arrived at the number 53% is a little bizarre:

Intel cut prices on its Pentium 4 processor for laptop computers by 26 percent to 53 percent

So they just add all the price cuts they've made on the processor together to come up with 53%? What's up with that? It's not like they just dropped it 53%, they dropped it by 26%.

Websurfing done right! StumbleUpon

Re:Intel's Questionable Math

[Tester]

The number 53% is for some mobile chip that dropper from 600$ to 300$...

Re:Intel's Questionable Math

[LordSah]

The largest price cut was a 53% reduction in price. I think this was on the 1.7 GHz Pentium4M. It went from $508 to $241. The 26-53% is the range of pricecuts that were made (each model dropped by a different percentage).

Re:Intel's Questionable Math

[ozzmosis]

Its legal. I worked at kmart last summer as a price changer, and kmart would do the same thing.

Re:Intel's Questionable Math

[Petrus]

How did they come with 53%?

Easily!
They dropped 26% on one model and 26% on another, using famus Penmtium Floating Point arithmetic that makes together 52.999867%.

What's questionable here?

Petrus

Still not enought to beat AMD

[Petrus]

Not only AMD remains cheaper, but also way faster (on some test as PowRay as much as twice) and with fewer bugs. It looks that Intel is really afraid of losing the market share.

Well, until Intel cuts prices to half of the AMD, I'll keep buying the better chips.

Petrus

Re:Old news?

[huntdwumpus]

ROFL! You made my day, bro.

And as a countermeasure...

[Anonymous Coward]

Apple computer INCREASES prices on iMacs by 53%. Take that Wintel!

P4 vs. PIII prices

[sacremon]

The odd thing from the price cuts is that a 2.2 MHz P4 Xeon Prestonia, w/ 512KB L2 cache, now costs $32 less than a 1.4GHz PIII Tualatin w/ 512KB L2 cache. Both of these chips are intended (by Intel) for servers/high end workstations.

Re:P4 vs. PIII prices

[sacremon]

And of course that should read GHz, not MHz. So much for previewing my posts...

Re:P4 vs. PIII prices

[Tester]

The odd thing from the price cuts is that a 2.2 MHz P4 Xeon Prestonia, w/ 512KB L2 cache, now costs $32 less than a 1.4GHz PIII Tualatin w/ 512KB L2 cache. Both of these chips are intended (by Intel) for servers/high end workstation.

Maybe intel is trying to phase out P3's...

Re:P4 vs. PIII prices

[sacremon]

Except that the Tualatin PIII's with 512KB cache are relatively new. They released them only 2-4 months ago. My thought is that marketing is the motivation, that they are trying to push the new technology, but to make a Xeon cost less than a PIII is just weird.

Re:P4 vs. PIII prices

[jrwyant]

It seems like the Tualatin is being pushed into the low-power (hence cooler-running) dense rack-mount 1U configuration. Where the P4 is having difficulty going due to it's thermal requirements. So for the mainstream, you won't see the Tualatin core. At least that's what I've seen reading several different mobo. vendors sites.

Re:P4 vs. PIII prices

[Anonymous Coward]

Pentium III SMP Xeons faster than P4 SMP Xeons
http://www.theinquirer.net/25050202.htm

Re:P4 vs. PIII prices

[Wesley Felter]

Too bad the Prestonia motherboards look pretty expensive.

Re: Moore's Law

[Anonymous Coward]

So now it's hardware capacity *per dollar*? Moore's Law originally had information/silicon doubling every 12-18 months, then it was processor speed, now it has "per dollar" thrown in there. Moore wasn't just a visionary, but a scalable one at that.

Check out the Jargon File entry on this one.

Next price cut in October?

[DeadBugs]

Many sites have been stating that the next price cut won't be until October. I also found it interesting that Intel is selling some of it's stock in AMD.

This is great news.

[MisterBlister]

This is great news because I like computer chips.

whatever suits him

[Kappelmeister]

Intel dispatched a suit and an engineer right away

Huh? Do Intel engineers usually go to work naked?

:)

Questionable Intel Bashing

[slugfro]

Your questioning of Intel's math abilities intrigued me, so I looked into the new pricing.

As clearly stated on the new pricing table here, the P4M 1.5GHz dropped 26%, the P4M 1.7Ghz dropped 53%, and other P4Ms dropped between those two percentages. Clearly they were stating the range of percentages of the price drops.

Since you might indeed be math impared yourself, I will show you how they got the 53%:

(Orig. Price - New Price) / New Price * 100 = % Decrease
($508-$241)/$508*100=53%

I hope this cleared up the issue. I don't know which is worse, one who spouts off without looking at the facts or one who just bashes a company to get karma. ;-)

Re:Questionable Intel Bashing

[geoffsmith]

I hit the 50 karma cap ages ago. Also, bashing a company doesn't get you karma, negative comments almost always get modded down. How many Intel employees do you think read slashdot? ;-)

I understand now it was just an ambiguous wording, when I saw "dropped 26% to 53%" I figured they meant from a previous discount of 26% to a current discount 53% for all their mobile chips. In fact they meant a varying range for all their different chips. Thanks for the clarification.

Re:Questionable Intel Bashing

[_ph1ux_]

and here is the equation to determine if its time to upgrade based on new prices from vendors:

(Orig. price of machine - New price of machine)= (total savings in purchasing new machine as compared to old machines original price)*(wife nagging about "why do you need a new processor, whats wrong with this machine")*(days having to go without sex for buying new machine regardless of wifes bitching)/((fun value of getting to play on new machine)*(time with wife not talking to you and you sleeping on couch anyway)) - ((kiss ass gift for wife to make up for new machine)/(one week))

I hope this clears things up for you.. YMMV if wife = 1, see above else New machine Whoop!

Damnit!

[curunir]

It would happen literally the *day* after I ordered all the parts for my new computer.

This just ruins the feeling I get from paying significantly less for an Athlon...paying just "substantially less" is far less satisfying.

Re:Damnit!

[smash]

Hehe..

Same deal... just yesterday I ordered -

Athlon XP 1800+, Geforce 4 ti4200, 512mb ddr sdram, msi kt3 mobo

ah well.

thems the brakes.

Was considering an intel system, but to get better performance, you need RDRAM + RDRAM mobo, which will cost a fair bit more.

smash(either system will comprehensively shit on my P3-700 on abit BH6 anyway :)

Re:Damnit!

[Rayonic]

Ditto. Couple of days ago I ordered:

Athlon 1700+
Soyo-Tek SY-K7V Dragon Motherboard ($110!)
512 MB DDR SDRAM
Logitech WingMan Force 3D Joystick ($25!)

Okay, so maybe that last one is unrelated to the topic.

Here's the formula.

[glrotate]

Every profit maximizing firm will produce at the quantity such that Marginal Revenue minus Marginal Costs = 0. The chip market is a monopolisticaly competitive industry. (few players simmilar but not distinguishable goods).

Since Intel has effectively lowered their marginal costs by increasing capital investment, their optimum quantity will be greater.
In addition since their demand curve (and marginal revenue curve) are downward sloping the new equilibrium quantity comes at a lower price.

Basicaly they made x profit at price P and quantity Q. Because of innovations they can sell more at a lower price to make even higher profits.

In addition their demand curve may have shifted in a bit due to the attractiveness of AMD's offerings.

Re:Here's the formula.

[Cyno]

Ok, maybe I missed something. You mentioned x and P and Q but I didn't see any formula. P*Q-cost=x, right? But certainly the cost of manufacturing chips in volume is cheap compared to the cost of developement. I wonder what equations they use to figure out their market potential. I used to do problems in a calculus class that was very similar to this sort of thing, but I doubt anyone in marketting or finance makes use of calculus to maximize profits. I'm certain there's a way to determine the price you should market your chips at to maximize x knowing a function f(Q) that is based on supply and demand economics and competition and stuff. But I'm a sys admin, I don't know anything about economics or chip making.

AMD Pricelist

[Bloody Pulp]

Here's the AMD pricelist for those of you who are interested.

Re:AMD Pricelist

[Dr. Spork]

Those are the "official" prices. Their actual prices are a whole lot lower. Check Pricewatch.

Wrong Math.

[bstadil]

(Orig. Price - New Price) / New Price * 100 = % Decrease

Stating an erroneous equation for calculating percentage does nothing to I hope this cleared up the issue.

Price changes as any other changes is calculated in relation to the original data point, NOT then new.
How would you calculate a 100% price drop by the way?
Do you have one of those original faulty Intel chips.
Math Bug by any chance?

When will those price cuts get down to us?

[Tester]

The main question that I would have is.. How long will it take for the distributors to sell their stock of "expensive" chips befores cutting there prices too and as a consequence how long will it take for those price reductions to reach us? And it is much shorter for companies like Dell?
Anyone in the industry would know?

Re:When will those price cuts get down to us?

[Anonymous Coward]

You would assume that dell would be able to pass along the cheaper cpu prices to the consumer considering the just in time (JIT) inventory ordering system they use. Basically you pay for your computer components before dell orders them, so they have the advantage of buying at current market prices and the option of passing along any savings to the consumer.

Really?

[Penguinoflight]

I seem to recall some stuff about SSE2 that wasn't too plesant, like bad looking graphics, bad sounding audio... Of course, you could be talking about something else, but P4 already has the majority of the new market, and that's all that matters.

amd != unstable

[Monkelectric]

I run my studio on AMD processors and have never been happier. My XP 1700 machine has bluescreened once in its life. What OS are you running?

In other words..

[moosesocks]

In other words, you can get an i486-DX2 for $4.24 (a 53% decrease of the previous price of $8)

I think AMD is in trouble

[0xA]

I admit that this might seem silly but I have never really wanted to get myself an Athlon based computer. I know too many people that have funny crases and wierd stuff happening, probably the fault of the chipset more that the processor (VIA, ewww) but still. My P3 has been great.

So earlier today I went to look for what I would need to upgrade my system. I need CPU, RAM and a motherboard. AMD is supposed to be the price / performance king right? Comparing an Athlon 1600+ vs a P4 1.6 with roughly compareable (feature wise) MSI motherboards and 256 MB RAM I will save 55 Canadian dollars, about 30 US, with the AMD system. Before this price cut.

So, WTF? For fifty bucks I'll buy the Intel thank you. I'll probably have that in the first 3 month's power bills anyway.

Re:I think AMD is in trouble

[Tom Finch]

Please restate using hexadecimal. Thus 100h MB RAM, and convert everything else. If you even know what hexadecimal is, that is. If you don't, then you're stupid.

Re:I think AMD is in trouble

[truesaer]

My experience is just the opposite. I upgraded from a celeron 300 2 years ago to a P3 500....and the system was terrible. Not a bit faster than the machine I upgraded from. This week I got an Athlon 1.53Mhz system, and it is absolutely amazing. Incredibly fast....I know there's a lot more involved here than processors, but I really like my Athlon system.

Re:I think AMD is in trouble

[beamz]

Power?
I didn't know 20watts of power equated to a $10-$15USD price difference in your powerbill. I'm sure you use candles at home because 1 60watt bulb costs $30-$45 to run a month.

So really... What percentage of the cost is 55CDN/30USD? An AMD and Intel MSI based motherboard runs about $65-$70USD on pricewatch, $81USD for an Athlon XP 1600+, $129USD for a P4 1.6 and buy a 256MB stick of DDR for $75USD so what you get is ~$226USD for AMD vs. ~$274USD for P4.

About 20% cheaper. So by my numbers you save ~$50USD. When you're talking about 20% difference, it makes a difference.

Flame away.

Re:I think AMD is in trouble

[Some Dumbass...]

So earlier today I went to look for what I would need to upgrade my system. I need CPU, RAM and a motherboard. AMD is supposed to be the price / performance king right? Comparing an Athlon 1600+ vs a P4 1.6 with roughly compareable (feature wise) MSI motherboards and 256 MB RAM I will save 55 Canadian dollars, about 30 US, with the AMD system. Before this price cut.

Note that the highest speed P4s (2.2Ghz and up) have no Athlon equivalent, yet the high end where the largest price cuts are taking place. The article says a price cut of 29% on a 2.4 Ghz P4 will occur. Does this cut affect AMD at all? Compare that to the 12% cut on a 1.7 Ghz P4. Does 12% close the gap between the P4 and Athlon? Not according to Pricewatch -- presently $85 US for the Athlon 1700+, $141 US for the P4!

Incidentally, the cheapest price for an Athlon 1600+ on Pricewatch is $81. For the P4 1600, it's $129. That's a difference of $48 US -- a bit larger than what you suggested. And what about the difference in memory costs? Does the new "Northwood" P4 perform as well with DDR as RAMBUS, or do you need to spend a bit more and buy RAMBUS memory to get its performance up to Athlon levels?

Also, I have to wonder if most buyers would consider power consumption or stability when buying (even though they should!) Do most buyers know anything about them at all? Even with just a $30 US difference, I bet most people would still buy the Athlon.

Re:I think AMD is in trouble

[PD]

What are you, some kind of base-ist? Hex Power! Is that your slogan?

Jeez man, I think using decimal to specify a memory size is just fine.

Re:I think AMD is in trouble

[jedrek]

Hm... after seeing this article me and my friend started checking out processor prices at one of Poland's top 3 wholesalers.

This is the deal: the Athlon XP 1800 costs as much as a P3-866. A Celeron 1.3/100 costs a bit more than a Athlon Thunderbird 1.4/266 chip.

There is absolutely no reason for me to pay that price difference. None.

Re:I think AMD is in trouble

[eldimo]

I don't know where you got these figures for the cost of the light bulb, but a 60W light bulb cost around 30$ per year!

Here in Quebec, the cost for the kiloWatts per hour is 0,068 $/kWh ($CAN)

0.06 kW X 24h X 365 = 525 kWh
525 X 0,068 $/kWh = 35.74 $

Re:I think AMD is in trouble

[Zathrus]

I admit that this might seem silly but I have never really wanted to get myself an Athlon based computer. I know too many people that have funny crases and wierd stuff happening, probably the fault of the chipset more that the processor (VIA, ewww) but still. My P3 has been great.

And my AMD systems, of which I've built several, have also been great. And cost significantly less than Intel (and I say this holding Intel stock).

The issues with problematic systems are usually due to one of two things.

1) Cheap hardware. All components are not the same. Buy cheapo hardware, get an unstable system. The biggest factors are the motherboard, memory, and video card. Oh, and MSI is not one of the better brands for reliability.

2) Bad drivers. Don't upgrade the drivers unless you need to -- if the system is unstable, you need to. Otherwise leave the thing alone.

So, WTF? For fifty bucks I'll buy the Intel thank you

Clearly this is your choice, and comfort level is a big factor in these things. But I think your prices are off.

The below prices are from Newegg, which isn't the cheapest place (anymore), but I've done business with them a lot now. They have good prices, good shipping, and excellent customer support.

Athlon Thunderbird 1600+, Retail: $79
MSI KT3 Ultra (Via KT333) - $86
Gigabyte GA-7VRX (Via KT333, my preference currently) - $89
256MB DDR333 Memory (Crucial) - $83

Total (MSI/GB) - $248/$251

Intel P4 1.6A, Retail - $137
MSI 6566E (Intel i845 chipset) - $90
Gigabyte GA-8IE (Intel i845, dunno if GB is good for Intel though) - $105
256MB RDRAM 800 MHz Memory (Corsair) - $100

Total (MSI/GB) - $327/$342

Difference: $79/$91 (or $145/$167 Canadian)

That's about 3x what you said... obviously using different suppliers, and the price drop hasn't figured in yet, but still much more substantial, especially when you consider percentages.

Could you go cheaper on the Intel system? Sure. You could use DDR instead of RDRAM. You could use a non-Intel chipset. But you've now nuked performance so much that the Athlon system is performing 20-30% faster. And it's still cheaper.

I'm not saying don't buy Intel. Just realize that a lot of the "funny crashes" are no more than FUD, and people have just as many issues with Intel systems if they buy cheap components. If you're happy to pay the price premium for Intel for peace of mind, then that's fine. But to say that it's insubstantial is incorrect (particularly since an Athlon 1600+ is faster than a P4 1.6A on most benchmarks).

Hrm....

[hage]

Despite this news, I bet the big PC retailers won't be dropping their prices (or at least not right away). Higher profit margins on high-end systems?

Somewhere in Texas, Michael Dell is touching himself.

Amd Chips

[dolby2]

I could never buy intel chips takes a lot longer to type intel then amd in a price seach engine. heh =-P

IN OTHER NEWS...

[mr_gerbik]

In other news..

Theres a sale at JC Pennies!!!

Re:Oh dear, oh dear...

[Transcendent]

the point to be made (which no one got) was that in standard applications there are certian ones that are "optimised" for intel chipsets... its almost like when ATI had the driver cheat which gave them better fps on quake3... lots of people thought it was a dirty trick...

P4's are no good for servers

[DietFluffy]

P4's consume too much power and generate too much heat. This is the reason why P3's are still used in most rackmount servers. This is why Intel is charging so much for the Tualatin 512KB P3 and so little for the P4 Xeon.

Re:CPU prices equal greater levity

[azzid]

Not sure what this has to do with CPU prices, however it was an interesting read. Can I get some of the weed you're smoking please =)

Good Components == Stability. Dell != Stability

[Brian Stretch]

It's that simple. If you do boneheaded things like use cheap, weak 200W power supplies like Dell uses and put one too many drives in the machine (two Seagate Barracuda IV's in the 1GHz Celeron box in this case), you'll have an unstable, flakey system (unplugging the second drive fixed that). If you don't install the current Service Pack, updates, and drivers (like Dell failed to do), you'll get an unstable Windows system (yes, I know, run Linux, but we don't have the source to everything that'd need porting).

If you carried over your 5-year-old ATX power supply to your new Athlon system just because the plug fit and didn't buy an Athlon-certified power supply (the P4's second power plug forced upgrade spared them from that), you'll have a flakey system. If you bought a VIA chipset board (ASUS's A7V333 is great, just so y'know) and didn't install the current 4in1 driver set, you'll risk a flakey system. If you bought an Intel board because you don't like VIA and didn't check out the nVidia nForce boards (which are driving AMD's invasion of the big OEM market), you're an idiot.

Building Athlons requires slightly more skill than building an Intel-based system. If you can't handle it, go buy a prebuilt system from someone who can.

Re:You should try a VIA C3 then

[shepd]

Plus C3s are dirt cheap.

I can buy a motherboard with:

- Integrated 666 Mhz C3 chip
- Integrated Video, Sound, Ethernet, and AMR that actually works

For just $139 CDN locally. That's like $75 US.

It uses a 486 fan and a smaller than 486 heatsink. Runs equivalent to a celeron-I 700 Mhz. Perfect for the quiet home-entertainment machine, and uses well under 50 watts of power (from what I can tell) so you won't be worrying about leaving it on, or using a fanless power supply.

And, for the first time in a long time, I've actually found it stable (its a PC-Chips product, and my opinions on that are a little *ahem* strong).

Whoops...

[shepd]

>BTW:

That's a mistake (typed in from an email by accident)... didn't mean to insult your english. Time to change my sig anyways...

Re:You mean an XP 1800 for less than $100

[Graspee_Leemoor]

" So while it may run at 1.53GHZ, that is not the name of the processor. I find it annoying that you just can't call the processor by the correct name"

Yeah, and if you run GNU software on that processor you'd better get used to calling it the GNU/AMD XP 1800.

graspee

Re:You should try a VIA C3 then

[Bryan Ischo]

What MB are you talking about? I am going to be building a (hopefully) quiet, low-power Linux home router soon and that MB sounds very interesting, as long as it has at least 3 PCI slots so that I can get 3 NICs on it ...

Thanks!

Re:You should try a VIA C3 then

[Bryan Ischo]

Sorry to respond to my own post, but ... I found it:

http://203.161.230.38/product_img/socket_370/m78 7c lr.html

Looks like it's right around $75 US on Pricewatch ... sweet!

Re:Good Components == Stability. Dell != Stability

[mrm677]

A quality 200W power supply is better than a lousy 300W power supply. Dell makes a quality 200W power supply. That is, Dell wrote a specification calling for a quality power supply that some company (probably in Asia) meets.

I have 2 machines: A Dell Dimension 4100 w/ PIII and a 200W power supply and two 7200 IDE drives. My other machine is a newly built Athlon box-- I bought quality RAM (Samsung) and a quality 300W PS (Enermax). It also has 2 7200 drives. Guess which one is more stable?? Yep, the Dell.

Re:You should try a VIA C3 then

[shepd]

Yep, that's the one. :)

BTW: From what I recall, everything on that board works great in linux.

The video is fine for 2D, but useless for 3D (its a Trident-alike video chipset). Note that the video RAM does steal from physical RAM, so for a router change it to 1MB instead of 8 MB. Can't remember what the sound was, but I think it just uses the standard built-in VIA 686 stuff. The onboard NIC is an RTL8139 10/100 dealy.

Oh, you'll love this, the onboard NIC has a bootrom built into the BIOS (Shift-F10 to enable). I think you can make this work for tftpboot in linux if you want absolute minimum power for your box.

IDE is excellent on the board, however, watch for problems with 2 Gig WD drives. I already had one crazy customer that wouldn't believe me that the BIOS on this board is incompatible with some very old hard drives (it locks up on POST with that specific drive). I saw it not boot with my own eyes.

Use PC133 memory for best results, however I think I got one or two working with PC100.

And, a last note, swap the position of the CMOS clear jumper so its closer to the IDE/Floppy ports. The board is shipped with it on the clear position and the board won't boot until you switch it.

Hope it works out for ya. :-)

Upgrade Systems

[cide1]

I view my computer as 4 systems.

1. The processor, ram and motherboard all get replaced at the same time. This one is the expensive one, and the most pertinent to performance, but it is hard to upgrade one without the other 2.

2. Second system is your storage. Your drives and controller. IDE and SCSI are more or less backward compatible, but the newer drives are sooooo much faster, feature less noise, and seem more reliable. Drives make a large amount of the high performance perception. Adding RAM in linux helps cache drives, adding performance. Windows gains less from this addition.

3. The third system is your graphics, audio and network. Im an app developer, and do little 3d. I listen to mp3's, so I touch these components rarely. I buy the consumer level NVidia, and do well with it.

4. The fourth is your case and powersupply. ATX is standard now, but cases wear out, get scratched, I modify them too much. Im on my fourth case in 2 motherboards, so I average about one a year.

Computers have planned obsolesence, make sure you buy at the right point on the price curve, and you come out ahead. I love performance, so I buy dual processors, but I buy a little slower chips. I find that helps prolong computer life without spending too much. I also multitask constantly, for gamers, it is a differant story. Watch pricewatch, read anandtech, save your pennys.

cide1

Sound Card?

[moogla]

Turtle Beach Santa Cruz and...
Hercules Fortissimo II

Good, solid gaming cards based on the excellent CS4236 chipset.
If you want an audio card that rocks harder be willing to spend money on something like an Envy24 based chipset (like the Midiman Dio 66). They are the highest fidelity imaginable, but they lack gaming features. If you need gaming on top of ultra-connectivity, the Terratec DMX line was a popular choice but they've got WDM driver model issues.

Better?

[hendridm]

> And yes AMD processors continue to be vastly cheaper and preform better than their Intel counterparts.

If by "better" you mean stability problems and compatibility issues, then I agree.

I dare you to find a PC device that is incompatible with an Intel processor running on an Intel mainboard + BIOS.

Re:Better?

[Junta]

I've used AMD products since my 286 and have never ever had an issue (except the K6/2400 was a poor perfomrer, but compatible). I keep hearing this stuff and I have yet to have a bad experiencewith my KT266 A motherboard nor my Athlon Xp 1700+. I've seen flaky systems, but this thing has been rock solid. And despite all my personal experience and the positive reviews, the name "Intel" still has a tendency to think "reliable". The two things I hear is that motherboard chipsets for Socket A are unreliable and that AMD processors are space heaters, but the latest line doesn't seem to demonstrate either to me... I agree the K6/2 was truly bad with heat, but the XPs seem to be reasonable..

Re:You mean an XP 1800 for less than $100

[acceleriter]

If the name's deceptive and disingenuous, which it is, he's doing the right thing. It's not our fault that AMD decided to copy Cyrix PR ratings in an attempt to fool people into thinking their clock speeds were as high as Intel's.

Competition is good, despite what MS says.

[Anonymous Coward]

Intel is being forced to cut costs (while maintaining quality) and also to continue innovating, because if they don't, AMD will 'own' them.

Right now, Intel simply puts out overpriced chips, and is surviving on brand recognition in terms of business, and the fact that average home buyers don't care what's in the magic box as long as it works.

Other than power consumption (which is really only a factor for businesses), Intel's chips have little on AMD's. People whine about stability and such, but I've yet to have a problem with an AMD chip in that department. Most of the unstability comes from crappy motherboards/chipsets, which, yes, also is a problem with Intel processors.

This is a pretty weak flame!

[Dr. Spork]

Please write back when you have looked up the energy consumption (~heat production) of a new Athlon compared to a (.18-micron) P4. I think you'll find the difference is negligible, certainly not enough to make a difference in room temperature.

As for getting a good frame rate in Quake3, your comment is pretty stupid. Unless you have an ancient graphics card, you surely get a higher frame rate with your Athlon than the refresh frequency of your monitor. I know I like playing at 1600X1200, and I still get better than 85fps, which is all my monitor can display.

If you've fallen for Intel brainwashing, that's your own problem. Just don't go thinking you're insightful when all you do is repeat their FUD without really taking the time to look at real specs.

But wait, there's more...

[athlon02]

Consider the company AMD recently required (Alchemy)... Imagine an XP 2100+, clawhammer, sledgehammer, or what not running say 10 Watts *MAX* and they'd own the technology. Let's see Intel try to keep up with that. Not to mention keep all the trolls who complain about AMD's chips being hot quiet. As it is, with my modest heatsink/fan combo my Athlon XP 1800+ is so cool that with my bare feet next to the open side of the case I don't feel any heat. The power usage and heat issue is such nonsense, unless you didn't bother spending money on a decent tower with a decent power supply. And if you did that, that's your fault, not AMD's.

Re:Intel the support chips

[Petrus]

So why I have not had any problems with my Asus
P5A + AMD K6-III 3Dnow! snce I have it - and that
is more than 3 years now.

That said, I had a verifiably failing and forgetting memory chip for two years, due to which Windows could not be even installed, althoug Linux had never any problems besides weekly core dumps on Netscape.

I do not remember any reboot in 3 years, however I have not problem iwth the 80GB 7200 rpm Maxtor HD that does not get recognized by the BIOS, only by Linux directly.

It has been my maibn machine for more than 3 years, continually hosting between 120 to 250 processes for 2 to 5 users, each running X-windows.

I can recomment Asus and AMD combination to anybody.

Petrus

Intel is making room for the 2.66GHz P-4

[deaddeng]

...timed for the launch of the t-bred (AthlonXP on 13-um process).

Intel 2.66GHz and 2.6GHz P-4s are already on pricewatch (alternatively for 400Mhz and 533Mhz FSB). Only $633! Such a deal!

100mhz slot 1 upgrades

[David Jao]

If you have slot 1, you most likely have 100Mhz FSB. These new pentium III have 133mhz, so you will either underclock or look for the rare 100Mhz chips.

100 MHz Tualatin Pentium IIIs are actually extremely easy to find. You just have to realize that Intel calls them "Celerons" and not "Pentium III"s.

For a 100 MHz slot 1 upgrade kit complete with Tualatin-compatible slocket and Pentium III based 1.2 GHz Celeron CPU, look for the PowerLeap PL-iP3T.

Re:I like potato chips

[jx100]

.. which would lower the prices for 802.11b antennas for all of us!

Intel is lying (partly)

[cyberformer]

Intel obviously doesn't want to say that the chip costs have fallen due to competition, because that would remind people that AMD exists and is usually better value than Intel. So it attributes the cut to the lower cost of manufacturing.

Manufacturing costs are falling, of course, as is the need to recoup development costs, but this has little to do with Intel's prices. It charges whatever it thinks the market will bear (as does AMD).

Re: 60W for a month

[stu72]

60W * 1KW/1000W * 24h * 30days/1Month = 43.2 KWh/Month

43.2 KWh/Month * C$0.08/KWh = $3.456/Month

Nice try.

Inanium remains expensive

[Animats]

The Intel Itanium is unchanged at $4,227 (!). And that's for 800MHz. You get a 4MB cache, but still, that's incredibly high. The Itanium was slower than the equivalent IA-32 machines when it launched (at 733MHz), and has been losing ground since. Inaniums are made only in the older 0.18 micron technology, too.

That's significant. Intel's "processor of the future" is only made in the old fab. That's a strong indication that the Itanium is moving to the back burner.

The next generation Itanium is supposed to launch at 1GHz this summer. Meanwhile, Intel has demoed a 5GHz Pentium 4, although that's a year or two from production.

Re:Inanium remains expensive

[StandardDeviant]

Finding prices for parts this high is somewhat of a black art becuase all the companies treat this data like goddamned state secrets or something, but from what I can tell that price is roughly equivalent (maybe a little higher) to a 900 Mhz Ultrasparc III (with, I might add, 8mb of cache), which is really the competition that chip is aimed at (along with the power4)... Of course, anyone in their right mind that needs the performance of 64 bit is going to choose something other than Intel's solution which is (IMHO) overpriced and immature. I think the current holder of the specfp/specint crown is IBM, followed fairly closely by Sun, and then a pretty big gap down to the itanium... True, you'll pay considerably more for an ibm sp2 or sun fire server than you would for most Itanium-carrying things, but if you need this class of performance then money isn't really much of an issue (big science labs, huge database instances, etc).

ah, actual prices ;-)

[StandardDeviant]

finally found somebody willing to talk about how much they want for the Usparc3 (allsunplus.com):
X6990A Sun UltraSPARC-III 750MHz CPU module with 8MB L2 cache.
Condition RETAIL BOX $3,400.00

X7000A UltraSPARC III 900 MHz processor module with 8MB L2 cache for Sun Blade 1000.
Condition RETAIL BOX $9,500.00
The 750Mhz part will iirc the specX numbers right beat the 800Mhz itanium across the board in terms of performance... If nothing else, there are mature optimizing compilers for the Usparc3 architecture, which is not AFAIK the case yet with the Itanium.

First rule of buying computer parts....

[icey5000]

Anyone who reads slashdot should know better... never look at PC pricing until at least 6 months after you buy your system

Re:Questionable Intel Bashing - Correction

[slugfro]

I noticed that the text equation has an error while the numeric one is correct. Correction below:

(Orig. Price - New Price) / Orig. Price * 100 = % Decrease
($508-$241)/$508*100=53%

Pentium XP... PXP, Pentium 2K... Pentium ME .. ??

[Programmer_In_Traini]

Maybe they'll want to go away from the numbers and start doing like MS does

Doh....can't wait for my very own Pentium 2k Pro SP 1...

HeHe!

Re:You mean an XP 1800 for less than $100

[acceleriter]

No, I'm not a brainwashed consumer, and do understand that clock speed is not the sole arbiter of performance. However, you and are are not the ones that AMD is trying to fool by making "model numbers" that look a lot like clock speeds to the untrained eye. Those people are the "good brainwashed consumers" of which you speak.

I would have supported AMD had they made a legitimate effort to educate the public, and called their CPUs, say, Athlon Mark III, Athlon Mark IV, or something. But instead, they chose to take the low, deceptive road of faking clock speeds. You can be in denial about it all you want, but that's what they've done.

Today's Toasters are Tomorrow's Supercomputers!

[Behrooz]

"The University of British Colombia today released a long-term research study claiming that discarded toasters showed signs of sentience. By wiring nearly 3500 toasters in paralell, they were able to burn down the administration building. The net result of the toaster parallelization notably increased the intelligence of the campus as a whole, proving that toasters are indeed the building blocks of future genius."