AMD Moving to a 400MHz Bus?

An anonymous reader writes "According to this tantalizing Infoworld Scoop, AMD soon introduce a 400 Mhz bus. Seems that SiS's big announcement at CEBIT is the SiS748 chipset, which supports both 400 MHz DDR & AGP 8X, and is targeted at the upcoming Athlon 3200+."

Comments

little penis

[Reikk]

I like to rub my little weeweee

Yummy

[visgoth]

Argh... do i wait for athlon64 or opteron, or do I get one of these bad boys?! Decisions, decisions...

Re:Yummy

[Anonymous Coward]

I've got a dual 1800+... definately worth investigating since they're inexpensive!

Re:Yummy

[gmack]

" Argh... do i wait for athlon64 or opteron, or do I get one of these bad boys?! Decisions, decisions..."

I think you just put your finger on why AMD sales are down. Opteron is so hyped up people are waiting for that. I'd feel sorry for them but I'm also waiting for the opteron before replacing my PC.

Re:Yummy

[emotionus]

Most kt400 boards CAN support 200mhz(effictivlly 400) FSB. Mine is a soyo and with trip into the BIOS you can make the FSB 200(X2 = 400) without messing up PCI, AGP, etc clock speeds.

Re:Yummy

[lost_it]

" Argh... do i wait for athlon64 or opteron, or do I get one of these bad boys?! Decisions, decisions..."

I think you just put your finger on why AMD sales are down. Opteron is so hyped up people are waiting for that. I'd feel sorry for them but I'm also waiting for the opteron before replacing my PC.

I'm facing the same dilemma, but the conclusion I reached is: When Opterons are first released they will be overpriced (like all new technologies) and possibly still have some kinks to work out; likewise with the motherboards. No matter how hard AMD and the mobo manufacturers work to get all the kinks out, nothing finds bugs like the first real-world release.

I figure I'll upgrade now and take advantage of the low prices caused by everyone holding out. By my next upgrade cycle the Opterons (and related hardware/software) should be very nicely matured.

Of course, that's the opinion of someone is trying to avoid the "bleeding" aspect of the bleeding edge.

Re:Yummy

[zbuffered]

I think you just put your finger on why AMD sales are down. Opteron is so hyped up people are waiting for that.

I blame:
1) non-gamers/power users don't need a new PC
2) economy sucks
3) Athlon kicked butt when it came out, but Intel came back with some nice, fast chips. And hyperthreading. Mmm, delicious hyperthreading...
4) ???
5) No profit!

Re:Yummy

[Ramze]

yes, but most AMD processors have integrated DDR memory controllers that can't run at 400. If you have a Barton core AMD processor, the DDR 400 mhz memory and run the system at 400, you should , in theory, be able to run them synchronously... but, most people don't have the barton core AMD processors & when they run the RAM at 400, the AMD memory controller is still running at 333, so the asynchronous operation actually hurts performance in some areas. (we're talking fraction of a percent here, but still... it doesn't boost performance like a barton core would).

Re:Yummy

[Ramze]

I'm in the same boat you are & I think I've come to the same conclusion. I figure... if I go out and buy only what I NEED right now (which isn't much), then I could buy a system fairly cheap. Later, I could buy an Opteron system when they've matured & Serial ATA is more commonplace (I want a RAID w/ Serial ATA drives :-)

It's going to take a while for everyone to release x86-64 compiled versions of software anyway, so the real benefits of having an Opteron system won't show up for a while -- especially considering that the price for performance of any new processor is always sky-high. It may take 6 to 12 months after its realease for it to drop to a reasonable price for most consumers, and I don't want to wait nearly 2 years before buying a new system.

Once the price goes down on the Opterons, there is x86-64 code for them, and it makes sense to upgrade, I'll do that. (and maybe I'll give my old system to my nephew when I buy an Opteron, so everyone will be happy.. lol)

Re:Yummy

[akuma(x86)]

I wonder how many people are feeling the same way that you are. Is AMD "Osbourning" themselves?

Re:Yummy

[Anonymous Coward]

> Argh... do i wait for athlon64 or opteron, or do I get one of these bad boys?! Decisions, decisions...

Intel has had high speed, quality solutions available for quite some time now. But if you like to play catchup, check out AMD.

Aaargh!

[ancukiewiczd]

Just when I thought my KT400's 333 MHz bus was state-of-the-art and could handle any Athlon ever made, this comes along!

Re:Aaargh!

[ackthpt]

Just when I thought my KT400's 333 MHz bus was state-of-the-art and could handle any Athlon ever made, this comes along!

Just like mine... You realize were we erred, right? We jumped off the fence and actually bought somthing. At that moment the tech level of our aspirations froze and the obsolesence clock began ticking.

400MHz is all very wel and good, but what happened to Quad-Rate DRAM? Or .. (diabolical pause) .. is that to follow, just before the Athlon 64 is unleashed?

Re:Aaargh!

[GothicManSlut]

alas, Im in the same situation. bu there is a bright side, better AMD chips means the evil hord of intel will sweat.

Re:Aaargh!

[apdt]

That explains it. I was wondering what that smell was.

Re:Aaargh!

[twiztidlojik]

Errr...is it just my imagination, or is there actually no /6 divisor on the KT400? I'm running at 200 FSB, but it's got the PCI running at 40 MHz. Unfortunately, I had to go buy a chipset cooler for my northbridge and put the sink from the northbridge on the south =\

Re:Aaargh!

[Anonymous Coward]

Get over it. If you're going to "argh" every time computer hardware gets better, you'll have a heart attack by age 19.

Re:Aaargh!

[Anonymous Coward]

Come on, it's called KT400 for a reason. Nothing more than a BIOS upgrade will be required to support that bus speed once it becomes available.

Scoop?

[Max Romantschuk]

There have been rumors about AMD going for a 400MHz bus for quite some time now. Some chipsets even have experimental support for it. With the Athlon 64 being delayed until September I would say that is the only way for AMD to try and stay competitive with the Barton core.

Maybe I'm being a little arrogant, but I still feel this isn't really much to be that excited about.

Re:Scoop?

[Anonymous Coward]

yeah, [www.theinquirer.net] has had this news for at least several days now (1 web day =? 1 normal year)

Re:Scoop?

[Boone^]

NVidia's NForce2 comes with 400MHz (200MHz DDR) FSB support. The problem is that Intel's forthcoming 200MHz FSB isn't DDR, it's QDR (Quad Data Rate) so it's an effective 800MHz FSB.

Re:Scoop?

[cbreaker]

Yea but that's for use with Rambus memory. Rambus memory is clocked a lot higher then DDR, but it's performance gains (not much) don't justify the extra cost.

Re:Scoop?

[TheRaven64]

Maybe I'm being a little arrogant, but I still feel this isn't really much to be that excited about.

Indeed. I keep reading about new, fast CPUs, getting excited for a few seconds then notice that the CPU utelisation graph on my 1.33GHz athlon hasn't been above 20% for a while and wonder what the point is. I get more excited about separate co-processors for video, audio and encryption nowdays. Most tasks requiring a lot of processing power are not well suited to general purpose CPUs. When did anyone last use a software implementation of OpenGL for anything complex?

Re:Scoop?

[BlueArchon]

nForce 2 has a 333 MHz bus (166 MHz DDR)... But it can use up to 400 MHz memory (200 MHz DDR).
Memory speed != FSB speed. I'm running a old Athlon with a 266 FSB and using 333 MHz memory.
Read this

Re:Scoop?

[Anonymous Coward]

You need to start crunching SETI nits then...

Re:Scoop?

[Boone^]

If you were running a board with the NForce2 chipset, you'd realize that a 400MHz FSB was also an option (not just 400MHz DDR). Check out this press release.

Re:Scoop?

[xanadu-xtroot.com]

Rambus memory is clocked a lot higher then DDR, but it's performance gains (not much) don't justify the extra cost.

I didn't find it much more expensive. I recently built a new box and have RIMMs in it (with a PIV-2.4Ghz). It was about 80 bucks for a 256M stick (I bought two). That seemed to be a pretty decent price.

Re:Scoop?

[cbreaker]

Yea but it's about $115 for a 512MB stick of the fastest DDR (PC3700) memory available. And you can get a 1GB stick of PC3200 for $240.

Re:False post

[Eagle5596]

Ummm I didn't post this... *goes to change password*, could someone mod the parent down? Who the hell stole my password?

architecture

[qoncept]

AMD better forget these little incremental speed bumps and switch to a whole new architecture this year if they want to remain competetive. The current architecture is like milking a deadhorse and they are already running waay too hot. They need to make something big enough to give it a new name.

Re:architecture

[TopShelf]

"milking a deadhorse?"

Gee, good thing you know your metaphors, otherwise you'd be stirring a can of worms by leaving the wrong impression.

Re:architecture

[Sgs-Cruz]

Gee, you mean, like, Athlon 64?

It's not just a move to 64 bit. See Ars Technica's article (posted here, yesterday, I believe) for an explanation of some of the other advantages of x86-64... they've taken the opportunity to add some new features and remove some of the old ones that weren't being used anymore.

Re:architecture

[geniusj]

hahaha.. thanks for pointing that out.. I got a good laugh :)

Re:architecture

[10Ghz]

You DO realize that we are talking about the bus-speed, not the CPU-speed? You don't increase the bus-speed by huge amounts overnight. Move from 333MHz to 400Mhz, while not groundbreaking, is significant.

As to the "whole new architecture"... It's called Athlon64, and it has 800MHz bus (and loads of other improvements). Available in september in a store near you.

Re:architecture

[qoncept]

I'm not the brightest pencil in the box.

Re:architecture

[Anonymous Coward]

"You don't increase the bus-speed by huge amounts overnight."

Better tell that to Intel quick. They are launching QDR with an effective bandwidth of 800Mhz real soon now(tm) as a stop gap measure.

Re:architecture

[10Ghz]

Move from 533Mhz to 800Mhz is actually just move from 133Mhz to 200Mhz (533 and 800Mhz since it quad-bumped), just a bit bigger than AMD's move (166MHz ==> 200MHz double-bumped).

Re:architecture

[Anonymous Coward]

Move from 333MHz to 400Mhz, while not groundbreaking, is significant.

But not significant enough. Consider how many times AMD has changed the bus recently:

SlotA 200Mhz
SocketA 200Mhz
SocketA 266Mhz
SocketA 333MHz
SocketA 400MHz

Everytime they push the FSB higher people's motherboards become obsolete. 333 -> 400 is only 20% increase and it makes everyone's mobos obsolete AGAIN? What's the point of even having socketed CPUs anymore, the constant FSB hikes make upgrading totally impractical. Might as well move BGA. AMD keeps shitting on its customers, good thing for them Intel is no different.

Re:architecture

[Anonymous Coward]

Or a clothes horse.

Re:architecture

[Brian Stretch]

AMD better forget these little incremental speed bumps and switch to a whole new architecture this year if they want to remain competetive.

It's called x86-64. The Opteron ships next month.

The current architecture is like milking a deadhorse and they are already running waay too hot.

I did not need that mental image...

Current Thoroughbred and Barton core Athlons don't run all that hot. An Athlon 3000+ runs cooler than a 3GHz P4.

I reclocked my TBred core Athlon XP 1700+ to 8x202MHz (404MHz DDR) on my ASUS A7N8X Deluxe motherboard (Corsair PC3200C2 DIMM). I kept the default core voltage (1.5v). MemTest86 verified that it works reliably. Upping the FSB is mostly a matter of motherboard and memory support, not CPU support (outside of being able to adjust the clock multiplier). A few years ago I reclocked a 150MHz Pentium to 1.5x100MHz. Worked just fine.

Re:architecture

[Anonymous Coward]

The current architecture is like milking a deadhorse and they are already running waay too hot

(-1: Troll)
The P4 3.06HT runs hotter than the XP3000+.
83W vs 76W.

Re:architecture

[turgid]

Can't be done. Nope.

Re:architecture

[Courageous]

I just figured he had, you know, a THING for Stallions.

Hee haw! :)

Re:architecture

[Anonymous Coward]

yea...thats exactly my thought.

but i got modded down.

as overrated, nonetheless.
maybe if i had posted with extra karma i could see that...

what some of the mods are smoking...i'll never know.

but i want some ;oP

oh yea...and this is off-topic btw guys.

Re:architecture

[mabinogi]

Well, they havent changed it yet.

So in reality you're saying they've changed the FSB speed twice in about 3 years.

That's not 'constant FSB hikes'

Keep flogging that horse

[haplo21112]

It seems some what proper that so many of the revisions of the athlon have had horse names since they seem to keep beating it till they know its good and dead.

Is a 400MHZ bus really gonna help them all that much? How much more can this chip design take?

Re:Keep flogging that horse

[ancukiewiczd]

The Athlon can take a lot. People have overclocked the 1700+ and 2100+ Thoroughbreds to 3 GHz before.

Re:Keep flogging that horse

[gormanly]

Hmm. Yes, the K7 has gone from 500MHz to 2250MHz over its lifespan so far - but Intel's P6 core went from 150Mhz PPro to 1400MHz PIII.

Looks to me like they could still have plenty of room to play.

Re:Keep flogging that horse

[Anonymous Coward]

It seems some what proper that so many of the revisions of the athlon have had horse names since they seem to keep beating it till they know its good and dead.

Where can I get me a Horse XP? I bet it runs hot.

Re:Keep flogging that horse

[Andy Dodd]

I think it can take quite a bit more.

Even with half the stated bus throughput, the Athlon seems to do a good job keeping up with the P4, and at a lower price.

Even at a lower clock rate, the Athlon can beat a P4. The Athlon XP 3000+ has essentially the same performance as a P4 for almost all applications, even ones where you'd expect the P4 to excel at (Video encoding) for example.

It helps that throughput isn't everything - Latency is also important, and the P4 was designed around an extremely high-latency memory subsystem (RDRAM), while the Athlon was designed around a much lower-latency memory subsystem. All the throughput in the world isn't going to help you unless the turnaround between a data request and that data coming from memory is fast. The only exception is if you rearchitecture the whole system (and this includes changing the ISA, which means it can't practically be done for x86) around a high-throughput high-latency memory subsystem. (PS2 is the most valid example - That system is designed around throughput everywhere, and it's designed so that memory latency is a nonissue.)

And don't forget x86-64... That architecture is making me drool. (Forget the 64-bit registers - What's important in the short term is that AMD doubled the number of GPRs and vector registers.)

x86-64 >>>> IA-64

Re:Keep flogging that horse

[Kevin Stevens]

Call me crazy... but I thought that the whole point of Rambus was that it was much lower latency.... which gave the P4 higher framerates in some apps like Q3A when it was first released- due to its excellent memory subsystem.

Re:Keep flogging that horse

[yarbo]

Nope, higher bandwidth at the cost of latency, which will help in games but hurts almost everything else.

Re:Keep flogging that horse

[PCBman!]

I think the point of Rambus was equal latancy, much higher bandwidth. The problem was that Rambus' first word latancy sucked the life out of them for business apps, but it's great if you're streaming a whole array--donno how it performs traversing a binary tree or linked list though.

Re:Keep flogging that horse

[JCholewa]

> Even with half the stated bus throughput, the Athlon seems to
> do a good job keeping up with the P4, and at a lower price.

It is also valid to note that the Athlon is at this almost-equivalent level even though Intel's fabrication progress is at least half a year more advanced than AMD. Intel goes through process shrinks six to nine months earlier than AMD. This is a huge advantage, as it means that if AMD and Intel were using the exact same chip architecture, the Intel-fabbed chips would be nontrivially faster. So it says a lot for the AMD microarchitecture that they're getting similar performance versus chips coming out of Intel's factories, which are a half year ahead of them.

That's why I like the AMD microprocessor design teams, I think. They can make a non-inferior product using a comparably primitive manufacturing technology. It's as if some guy made a helicopter out of chalk that happened to be just as structurally stable as your Humvee.

-JC

Re:Keep flogging that horse

[ConsistentChaos]

If Intel's fabs are ahead of AMD's, wouldn't that mean it costs less for Intel to produce a P4 than AMD an Athlon? If so, that's *very* bad news for AMD, since they can't command the retail price an Intel-branded chip can, and thus recoup costs.

Re:Keep flogging that horse

[isoteareth]

New fabs aren't free...

Re:Keep flogging that horse

[workindev]

the Athlon seems to do a good job keeping up with the P4, and at a lower price...The Athlon XP 3000+ has essentially the same performance as a P4

If you did a little research, you would find that the XP 3000+ is actually $9 more expensive than the Pentium 4 3.06GHz. The XP 2800+ is $22 more expensive than the P4 2.8GHz.

Re:Keep flogging that horse

[Cynikal]

see in the real world, alot of real people cant afford to change their case, motherboard and ram everytime a chip maker decides to change their architecture. intel especially pissed me off when they switched from slot to socket right in the middle if the p3 series.

Basicly intel has rubbed me the wrong way everytime they changed architecture, which is the whole reason i went to AMD, they had a track record of offering higher clockspeeds in form that quite a few people were "stuck" in. i really liked the idea of upgrading with only having to buy a new chip instead of forking over 3-4 times as much for the whole new kit..

i mean their choice, beat the dead horse, or keep jumping horses every year or so and leave a good chunk of their customers in the dust. i would certainly like to see them pursue both avenues, a faster 400fsb chip, while still supporting their current architecture.

Re:Keep flogging that horse

[diablobynight]

But with the NForce2 boards, The Athlon's are blowing the Pentiums out of the water on multiple CPU systems, because their are actually two memory processors on the NForce2 boards. And when XP's new cpu comes out I would be highly surprised if their wasn't a huge server buy in. Because each 64 bit processor will have it's memory processor built into it. So that means you don't bottle neck the FSB with memory cache runs coming from multiple processors. Like in a quad processor XEON system, where you only get 25% of the 4th processor and 75% of the third processor, because of cache mis calls, where the data has to reference to insure it's the most recent.

Re:Keep flogging that horse

[haplo21112]

I've been seeing this equally clocked AMD's cost more than the intel part....AND in many checked if you compare the PR# of the AMD part to an intel running at that clock, the intel is cheaper...AMD is definatelky behind in this race, maybe permanently...They really need to push the chip harder if they want to even stay in the race, honestly if they could clock at the same speed as intel they might even have a superior chip, but they would never keep the price in range it seems.

Re:Keep flogging that horse

[Anonymous Coward]

Well, the thing is not so much that they doubled the GPRs, but they made them open to the programmer or compiler. I believe, and correct me if I'm wrong, but there are in the chip already quite a few so-called rename registers. The P4 I know has 128 registers, but only four can be explicitly named. The others are used by the processor itself to optimize the instructions. But this way, the registers can be optimized at compile time, instead of having more loads and stores to RAM.

Re:Keep flogging that horse

[mmol_6453]

It may depend a great deal on how their traces are layed out on the die. If the traces are two close together, then signals, thanks to electric fields, start jumping from trace to trace, causing interference.

If the traces are too far apart, then the whole circuit occupies more space than it needs to, which means the traces are too long. The longer the traces, the more like an antenna the trace becomes, possiby causing noise by RF interference.

They have to find the right size, and even then, their block layout for their different CPU portions may be the problem. A reorganization or the layout may be what's called for.

Re:Keep flogging that horse

[Keeper]

I think cost would be more related to die size and yields more than anything else...

Re:Keep flogging that horse

[Anonymous Coward]

Because each 64 bit processor will have it's memory processor built into it.

Yeah, thats great until DDR 533 comes around- then instead of just upgrading your chipset, you have to buy a new processor to upgrade your memory. That will probably happen just as DDR 666 is rolling around, and DDR 800. Whoops. Maybe integrating the memory controller into the processor wasn't such a great idea after all.

AMD is so ass backwards. At their current pace they will be out of business in a year, and good ridance.

Re:Keep flogging that horse

[Zathrus]

Clock rate means nothing - a 2.0 GHz Athlon will outperform a 2.0 GHz P4 by a considerable margin.

The PR is a much better indicator, although not always accurate. And when you compare Athlon XP PR's to Intel MHz ratings then the Athlon's beat out Intel in nearly every case - the top of the line is unusual in that the P4 2.8 and 3.06 is actually (slightly) cheaper than the Athlon 2800 and 3000. But for every other processor speed the Athlon is cheaper - often much cheaper. Hell, for the price of a P4 1.8 ($103) I can buy an Athlon 2200 ($97) (and this is the most favorable comparison at the low end -- I could also buy a P4 1.3 for $103, but that would be silly).

AMD is certainly behind, but not by much. And as far as $/performance is concerned, they're considerably ahead of Intel.

Re:Keep flogging that horse

[FueledByRamen]

Having a faster bus come out on the market won't help you much anyway - (using your example) you wouldn't overclock your DDR400 bus to DDR533 to take advantage of the faster speed, you'd buy a CPU and board that supported it. This works in the same manner as current Athlons: faster RAM (333 or 400) won't help when you have a bus that's still 266 (or 333).

Re:Keep flogging that horse

[diablobynight]

ummm...even without the memory controller integrated, you ussually wouldn't keep the same processor if you were upgrading your motherboard, Idiot. I mean seriously. If I build a PC133 1.2 ghz athlon system, and I upgrade, I am not going to buy a brand new motherboard, and brand new ram and then shove in my old as 1.2 ghz processor. AMD will not go out of business in a year.

Re:finally

[ancukiewiczd]

The 4GHz is only when heavily overclocked. The fastest retail P4 runs at 3.06 GHz, which actually is about the same speed as the 3000+. The 3200+ should give a nice speed boost. Actually, the fastest-overclocked Athlon runs at about the same speed as the fastest-overclocked P4. Not only that, but both processors are at about the same price.

What was with 333MHz?

[Anonymous Coward]

Makes you wonder why AMD even bothered to release 333MHz capable procs if they are planning on switching to 400 already.

Is PC3200 (400MHz) DDR SDRAM even a JEDEC standard yet?

Re:What was with 333MHz?

[Carewolf]

They dont need DDR400. Atleast not if NVidia releases a new 400MHz chipset. They already have 2 memory-controllers, and have so for not have much to use them for since bus and memory speed matched. If the bus starts to be significantly faster than memory, NVidia will have a real advantage.

Serial Attached SCSI

[jaavaaguru]

If they keep on getting faster at this rate, we'll soon be able to benefit from the speed of this ;-)

The downside is....

[irn_bru]

You computer will blow up if the processor speed-steps below 600Mhz...

Re:finally

[Eagle5596]

The big issue however is not clockspeed anymore, it's cache, and cache speed, so this bus increase will probably do more than a few extra clock cycles would anyway... depending on what the cache size is in the new chip set.

It's worth looking into, but I doubt it will hold a candle to Intel anyway. I used to have hope for AMD, but have never had a success personally, with them.

Re:so?

[Anonymous Coward]

No, Intel is not at 800MHz FSBs yet. They are about to move to 200MHz quad pumped, whereas AMD are about to move to 200MHz dual pumped buses.

Re:finally

[rgm3]

What the hell are you talking about? Clockspeed and bus speed aren't really related. Intels 3GHz chip runs on a 533MHz FSB. They also have chips that run on a 400MHz bus. Check out:

this page

Besides, AMD is trying to break the "megahertz myth," remember? :)

The computer clueless..

[Anonymous Coward]

Won't care. They'll react to Intel's crap the same way most of you react to nude pictures of CowboyNeal.

(Uh, with lust, that is.)

This in spite of the fact that unless you're compiling software, searching for extra-terrestrial life, or playing crack-the-encryption, you don't need 1+ GHz.

What you need is what AMD is offering - more speed on the bus. Depending on what you run, more ram. And everyone could use a faster hard drive. ..Even those who compile software will usually see better results from increasing bus speed + hard disk speed.

Do your part. Educate some of the clueless today!

Re:The computer clueless..

[julesh]

This in spite of the fact that unless you're compiling software, searching for extra-terrestrial life, or playing crack-the-encryption, you don't need 1+ GHz.

What you need is what AMD is offering - more speed on the bus. Depending on what you run, more ram. And everyone could use a faster hard drive. ..Even those who compile software will usually see better results from increasing bus speed + hard disk speed.


Actually, compiling software will probably benefit more than most because with a modern processor it is I/O bound (I rarely get to above 50% processor usage when compiling these days, although admittedly I do tend to compile files off a network filesystem).

There are two things the masses do with processor power these days:

- play 3d games which are getting ever more processor hungry...
- mess around with digital video, which is becoming a more & more common thing to do.

Both of these will probably benefit from higher CPU core speeds as much as they will from improved bus speeds. At least until memory speed catches up...

Re:The computer clueless..

[Zak3056]

Won't care. They'll react to Intel's crap the same way most of you react to nude pictures of CowboyNeal

And

What you need is what AMD is offering - more speed on the bus.

You do realize that Intel already offers a P IV that runs on a 533MHz FSB, or 33% faster than what AMD is introducing now... right?

Re:The computer clueless..

[master0ne]

no offence to julesh, but: ---begin rant--- your also forgetting the impact that faster processors has on os development (atleast winblows devel). the faster the processor the more spiffy 3d graphics and the worse unoptimized code to make the os run even worse on your existuing hardware so you look past its huge gaping security flaws and other numerious problems, while gawking at how buteaful it is and speding extra $$ on the hardware to make it run. All in the mean time Micro.. I mean the os manufacturer is still beta testing the os, and once you get your system running right. viola, theres a service pack to fix most major security problems! --end rant---

Re:The computer clueless..

[cel4145]

There are two things the masses do with processor power these days:

- play 3d games which are getting ever more processor hungry..

yep. nethack roars on the best processor power.

Re:The computer clueless..

[master0ne]

yeah, horriendously overclocked!

Re:The computer clueless..

[Anonymous Coward]

yes, it's called dual channel PC2100, or DDR266.

Spacka, read Intel's marketing buzz on why.

Re:The computer clueless..

[Anonymous Coward]

"play 3d games which are getting ever more processor hungry..."

Give me a GeForce 4 and a 600mhz processor, and I'll be fine.

"mess around with digital video, which is becoming a more & more common thing to do."

True enough. I'm sure the masses don't know enough to get a decent hardware card to increase video playback capabilities. The crazy bastards probably run software-only players.. Yeah, that would eat processor time.

(I do hope you're not talking about actual video creation - that's about as common as kernel hacking for the 'masses'.)

Re:The computer clueless..

[be-fan]

Actually, g++ will peg your CPU. Compiling C++ code is processor intensive enough that reading files from disk isn't the bottleneck. During a big C++ compile, my processor usage will however around 70-80% on a 2GHz P4.

Re:The computer clueless..

[pseudochaotic]

Technically, it's a 133mhz FSB that they quadruple for some reason, while AMD's is a 200mhz FSB that they double for some reason.

Re:The computer clueless..

[master0ne]

accualy i have tried winblows. versions 3.1, 95, 98, 98se,NT 4.0, ME, 2k, 2kpro, 2kadvanced server, and xp pro. all of which grow progressivly worse in the area of graphics, and with the exception of 2k and XP pro (with some tweaking) all have horrible memory leaks. not to mention all ms software code is horribly unoptimized. This is what allows them the time to release the security updates. The end user has to upgrade his hardware to use the new os, by the time he has useable hardware, viola the patches have been released. And i still do run XP pro on my laptop (be it only cuz of lack of 802.11a wireless nic drivers). I am stil trying to figure out why win XP requires appox 800Mb... how can an os be that big? there have been versions of linux trimed down to around 500 Mb which have all the useability and more of windows. btw when was the last time your system went below 1% processor useage, and below o say 20Mb ram useage? thats my useage during adverage computing.. ie mozilla, xmms (winamp port), and vi. My fancy gui bairly even registers a hit in performance (i run windowmaker). thankyou very much smeg head.

Re:finally

[FrostedWheat]

Not only that, but both processors are at about the same price.

Not to mention the AMD runs cooler than the Intel chip.

Whoa .. never thought I would say that!

Great news

[RalphBNumbers]

I was actaully a little worried that when a Macs switched to the PPC970, memory fast enough for it's initial 450MhzDDR bus would be prohibitively expensive. They might have been forced to increase the bus multiplier to maintain their target price point, or they might have just needed really expensive RAM.

With this 400mhz bus and a bit of upwards evolution, this shouldn't be a problem by the time 970 based macs are released. yay

Re:Great news

[Anonymous Coward]

Yep, going by Apple's past record I reckon you should see those PPC 970s in a mac by about 2010. There should be plenty of cheap 450 Mhz memory about by then.

Re:Great news

[Puu]

Why not dual channel 233 DDR memory? Should satisfy a single 450 DDR CPU bus nicely.

Don't know about the dual CPU Macs, though. IBM has said PPC970 relies on its good 2-way and 4-way scalability for top performance, and Apple sure will offer (at least) a duallie PowerMac as usual, so this has to be addressed.

Maybe PPC970 CPUs are good at talking to each other and sharing L2/L3 cache (maybe 2-way is meant to use only one CPU bus anyway). Or maybe quad channel is feasible on a higher end, more expensive Mac. Or maybe RAM speed will catch up by the time PPC970 Macs appear?

Re:Great news

[Puu]

Oh sheesh, didn't have my head screwed on.

Of course there are no 233 MHz DDR (466 MHz effective) DIMMs available. I meant 133 MHz DDR but somehow made a typo and managed to happily continue with it through the rest of the thought :-P

[To match a single 450 MHz DDR (900 MHz effective) CPU bus, you'd of course need *four* 133 MHz DDR (266 MHz effective) memory channels, 112.5 MHz DDR (225 MHz effective) not being available.]

Apologies to everyone for the waste of time. Fortunately most of that was my own.

Re:Great news

[immybaby]

Actually the 970 supports:

the 970 also sports a cache-coherent, 900-MHz processor bus capable of data rates up to 6.4 Gbytes/second

This is a double-pumped 32-bit wide bus running at up to 900MHz.

The memory subsystem does not have to keep up with this, it can be PC3200 or whatever. It does however mean the maximum speed supported is _fast_ and the bandwidth between the processors is very good.

Re:finally

[rogue409]

Usually, i read really good comments on slashdot from readers. This one was completely asinine.

Question!

[FrostedWheat]

Just how much influence does the bus speed have on the system as a whole?

My CPU is running at 266mhz now, what improvment would I see if I upgraded to a 333mhz bus chip with the same clock speed?

Just curious!

Re:Question!

[qoncept]

Just how much influence does the bus speed have on the system as a whole?

With the Athlon's architecture, not a whole lot. It wasn't designed with high speed memory in mind. Hell, they started with 100mhz SDRAM. Pentium 4 is better at taking advantage of higher memory bandwidth, but it does help Athlon. If you're building a new system, go for PC2700. If you've already got PC2100, the upgrade might be a waste.

Re:Question!

[julesh]

Well, first of all, you'd probably need to upgrade your memory to be faster as well, or you wouldn't notice a thing.

But, once you've done that, memory access times will drop substantially for those cache misses, which means about 5% of instructions will execute about 20% faster, so you'll see about a 4% improvement in speed, more or less, depending on how much memory access and IO your application performs.

I think. Somebody flame me if I'm wrong here...

Re:Question!

[Junior J. Junior III]

Depending on what you use the PC for, you might not notice anything at all. Current desktop PCs are more than adequate for web/email/office work, and have been since Intel first hit 300 MHz or so. I have a PII 400 running Windows 2000 at work that does not seem slow at all running all the basic, standard applications.

If you do stuff that involves digital video, compiling source code, or other types of activities that actually push the CPU, you might notice a difference between a 266MHz system bus and a 333MHz system bus.

The speed of the front side bus determines in part how fast information can get to the CPU from main memory. If you have fast memory + a fast FSB, you can get your CPU to work pretty darn fast. Your main performance bottlenecks are still going to be memory latency and hard drive access speed, though.

But once information gets from there to the main system memory, if you can keep that CPU at high utilization, you'll notice a pretty significant boost in performance.

Re:Question!

[SWPadnos]

... about 5% of instructions will execute about 20% faster, so you'll see about a 4% improvement ...

Not a flame, but your math is off. If 5% of instructions execute 20% faster, then you save 20% of 5%, or 1% of the time. (those 5% of instructions execute in 4% of the time, instead of 5%)

Re:Question!

[kauttapiste]

My CPU is running at 266mhz now, what improvment would I see if I upgraded to a 333mhz bus chip...

Hehe..If you're CPU is really running 266Mhz (I'm not gonna pick on your millihertz this time), you'll discover a whole new world with a BUS running faster than your old computer.. All I'm saying is put those old boxes with a nice FreeBSD installed to the closet and use them for screen/IRC-client! Imagine the uptime..

Re:Question!

[Kevin Stevens]

youve gotten some good replies already, but I just want to add that these are Front Side Bus speeds that we are talking about... so peripherals and such which are still stuck sharing the slow lane (and at this point its like comparing a congested single lane road to a 6 lane interstate). Increasing your FSB and your mem to a faster grade will give you performance increases, but nothing drastic, and nothing youll see in 'daily' usage- I think at this point the main bottleneck in opening word or loading windows are your hard disks. I think the really cool thing about this though is that chipmakers are finally trying to even the playing field between the busses and the processor speeds. Ok, well maybe 'even' is a strong word, but at least trying to get the busses to keep pace. It was only a few years ago that we were at a 66mhz bus- and that bus speed had lasted about 2 years (If I recall correctly, its been awhile... these are rough figures). Now bus speeds are increasing every few months, and soon we will have high speed busses to connect peripherals to.

Re:Question!

[FrostedWheat]

Hehe..If you're CPU is really running 266Mhz (I'm not gonna pick on your millihertz this time), you'll discover a whole new world with a BUS running faster than your old computer.. All I'm saying is put those old boxes with a nice FreeBSD installed to the closet and use them for screen/IRC-client! Imagine the uptime..

lol, dang it I didn't think anyone had noticed :p

I have an old 25mhz machine running in the attic, uptime so far is 658 days and counting. Not that it does much these days, I just can't stand to shut it down with that uptime ;)

The machine I was talking about however is a tiny bit faster Athlon XP 2000.

Re:Question!

[julesh]

Doh!

Re:Question!

[muckdog]

I was upset when checking up on one of my servers that up for over three years. Its uptime stated that it had only been up for 26 days. When I checked the logs it hadn't rebooted. The uptime counter had rolled over which ment 1122 days. We had to replace the batteries on the UPS. Thanksfully the UPS had a DC bypass switch so we didn't have to bring down the server. Its still running

Re:Question!

[JCholewa]

> My CPU is running at 266mhz now, what improvment would I see if
> I upgraded to a 333mhz bus chip with the same clock speed?

Do you mean a 266MHz chip with a 333MHz bus? That'd be odd, and such a creature does not exist. You'd probably get a substantial performance boost in data access intensive applications, ones that don't have a small code loop and don't exercise much redundancy. For small loops with lots of redundancy (which is not too uncommon), the caches on your processor offset some of the lower performance of the older memory bus.

But some apps benefit tremendously. A lot of games depend on loading massive amounts of textures continuously, so a jump from 66/66MHz (533MB/s) to 333/166MHz (2667MB/s) would alleviate a lot of low framerate problems. Anything with a lot of loading and saving to disk would probably benefit a lot, too.

But it's tough to say exactly how much you'd benefit. The reason why they decided to keep the bus speed ramp lighter than the cpu speed ramp is because caches could compensate for the performance loss, and putting caches into chips is much cheaper than the harsh work of getting the entire system to run at the cpu frequency.

-JC

Re:Question!

[zentigger]

One thing that nobody here seems to have mentioned is RAM. Again, assuming you are merely using typical internet/office applications, the best bang for your buck would be to install a plethora of RAM. You should be able to pick up a couple of DIMMS for your 266 for less than price of a tank of gas.

Re:Question!

[Anonymous Coward]

Maybe you would have better luck with the ladies if you worries less about your uptime and more about your personal hygene.

Re:False post

[His Nastiness]

Learn to speak english in either post and I'll believe it. Remember to click "Post Anonymously" next time. Dummy......

Damn

[Apreche]

Despite the bad economy these companies just keep making faster and faster computers. AMD, NVidia, Intel, ATI, they all just keep coming out with somethig newer and faster? I need a new PC really bad (running a P3450). I'm looking to get something that can play doom 3 well. When is going to be the right time to buy a computer? I've built a lot of computers, but non for me in many years. When is the time of year that's best? Right now I'm thinking April, since I'll have money then.

Re:Damn

[EllF]

If you want to buy a computer that can run D3, wait for it to come out. About a month after its release, shop around for a new system -- you'll both be able to build a system that will be assured to work well for playing Doom, and reap the benefits of the latest hardware at the time.

The general rule of thumb for upgrading it to put it off for as long as you can, and then buy as close to the top of the line as you can afford.

Re:Damn

[DeathPenguin]

Good point. By the time Doom 3 makes it to retailers, everything will have changed in graphics and central processors. ATi and nVidia should have plenty of great new offerings available in video cards, plus this generation's products will be relatively cheap. S3's DeltaChrome also looks promising, though S3's historically been inept at writing drivers. Perhaps we'll even see 64-bit AMD CPUs by that time.

Re:Damn

[Junior J. Junior III]

I'd say the right time to buy a computer that will play Doom 3 well is about 6 months after Doom 3 gets released. You could also try buying one the day Doom 3 is released, if you've got a lot of money.

Seriously, it hasn't been released yet, so what's the point in speccing out hardware that will run a vapor app "well"?

Doom 3 may well run perfectly fine on a sub 2GHz machine for all we know.

Re:Damn

[qoncept]

First, there is no bad economy. There are high gas prices.

I usually try to help my friends build their computers. The key things to remember are: the best time to build a computer is as long as you can hold out, because the longer you wait, the faster and cheaper your hardware is going to be. Also, look at pricepoint. You know how Walmart has those labels on their price tags that say "17.8 cent/oz" ? Do that with your hard drives. Right now, 120gb drives are $100 on pricewatch, 160gb drives are $160. 120 has the best price point. Athlon prices increase about $10 each until you get to 2200, where they jump by like $40. Keep in mind that if you build a computer for a reasonable amount of money you can do it more often if you choose to and you'll never finish building one and be unhappy with the speed.

Re:Damn

[ryanvm]

The general rule of thumb for upgrading it to put it off for as long as you can, and then buy as close to the top of the line as you can afford.

No, that's the fool's way of upgrading. The smart way to do it is to upgrade whenever you want, but buy 6 month old hardware. That way you get a system 75% as fast as high-end for 50% of the cost.

Re:Damn

[Anonymous Coward]

That is the sensible approach, but it does have a cost: one quarter of your penis length.

Re:Damn

[ryanvm]

That is the sensible approach, but it does have a cost: one quarter of your penis length.

Yikes, that's one inch I can't afford to lose.

Re:Damn

[Reziac]

I recommend that if you need to shave costs, you still buy the best motherboard you can afford, then skimp on easily-upgradeable parts that will come down in price very rapidly after the first spasm of bleeding edge sales, such as CPU and video card.

By starting with a good motherboard, you also maximize the upgradeable lifespan of the system, because it is more likely to support newer components on down the line.

Re:Damn

[EllF]

I retract my earlier statement. If you can indeed buy a system for 50% of the cost that nets you 75% performance, you have a better general rule of thumb. I have a better exceptional-case rule. I'm taking your word for it -- I haven't bought a brand new system in quite a few years.

Your argument has merit if you're going to be upgrading your system regularly. Let's say it costs $1500 for a top of the line system. Let's also make two assumptions here: computer prices aren't changing, so any $750 system is 75% as fast as a $1500 cost system, and also that computer speed doubles in a year. You buy a system at some point in time for $750. So, you wait 9 months, and buy a second system at the same price, wanting to stay at 75% of current "top" speed. You repeat this cycle for 36 months -- 3 years. At the end of 3 years you've spent $3000.

Now, imagine that I buy a top of the line system at the same time that you bought your 75% system. I wait a full year -- 12 months -- and spend $1500 again, wanting to stay at 100% of the current "top" speed. In another year, I spend yet another $1500. In three years, I've spent $4500, or 150% more than you.

However, let's now imagine that each of us waits about 3 years to upgrade, as seems to be the case for this guy, who's running a 450mHz system. Computer speed has now tripled: you're 375% behind the current top speed, and I'm 300% behind. That's only 75% of difference, sure, but if I'm only upgrading every 3 years, I'd sure as heck want that extra 25% of leeway per year.

"Buy the most expensive system you can, and hold off for as long as you can" costs more in the end, but if you're only able to upgrade once in a great while, you'll make that system last longer. So what you propose makes more sense, if you can drop the $750 every 9 months. If you can't, though, and need to depend on a windfall of some sort coming in (which could reasonably occur every few years), buy the best possible when you do.

Re:Damn

[4of12]

The general rule of thumb for upgrading it to put it off for as long as you can, and then buy as close to the top of the line as you can afford.

A friend of mine gave me this good advice:

If you want the best value in a new computer, buy the top of the line system from 6 months ago.

Re:Damn

[Powercntrl]

I'm looking to get something that can play doom 3 well.

Karma be damned... Screw Doom 3.

I didn't like Doom 1, nor did I like Doom 2. My Athlon XP2000 is fast enough to convert DVDs to XviD in realtime and it creates a sauna effect in the room while doing so. In fact, I still keep my prior system (a PIII 850MHz) around for just about every other task since it gives off a LOT less heat. I'm not sure want to upgrade to anything that produces more heat just so I can play a game I won't even enjoy. I live in Florida and my electric bill is already insane.

Where was I? Oh yea, Doom 3... I don't want it. Now as long as Duke Nukem Forever doesn't use the Doom 3 engine, I'll be all set. THAT'S the game I'll install a bigger air conditioner for.

When is that game coming out anyway?

Re:Damn

[DevilJeff]

Damn is right! Why do you people keep wanting "a computer that can run Doom 3" It seems like EVERYONE is putting off doing anything until doom 3 comes out! Get on with your lives, get that new computer that you want, it's going to be able to play Doom 3 just fine, trust me.
I just got myself an Athlon XP 1800+ about five months ago, and I'll be damned if it won't run Doom 3 with the graphics turned to max as it is.

I find your ideas intriguing,

[Anonymous Coward]

And wish to subscribe to your newsletter.

Re:False post

[Jugalator]

ROFL! This was a funny thread :-D

haha

[ftvcs]

laugh, it's funny!

I'm already at a 400MHz FSB

[eamber]

I've got my Athlon XP 2100+ running on a 400 MHz FSB (of course, that's overclocked)... but it definitely does seem "snappier" than the 266 MHz FSB. Certain apps seem to benefit from the extra bandwidth, but not everything.

Finally a CPU that has THE REAL 400Mhz...

[Anonymous Coward]

...I wonder how long it will take until we have CPU's that are capable of THE REAL 3Ghz :)))

Re:False post

[Anonymous Coward]

Who the hell stole my password?

I don't know, but good thing you logged on a mere two minutes afterwards, so you could change your password.

No, really, I believe you...

Already got this luvin :)

[GweeDo]

I just recently bought an Abit based NForce2 Athlon Motherboard. I have my DDR3200 running at a pretty 200mhz (so 400mhz DDR) and my FSB is at 181mhz (so 362mhz DDR). I have made some changes so I need to try for a 200mhz (400mhz DDR) FSB again. I can tell you that just upping the FSB and your memory bandwidth can have great performance benefits for memory intensive apps (such as gaming). So this will be a great boost for the current XP line. Oh, and in case anyone is wonding, I have an XP2100+ (1.73ghz) running very nicely at 2.2ghz!

Re:Already got this luvin :)

[Anonymous Coward]

Almost all benchmarks show no benefit from running your memory/cpu out of synch. Either run 181/362 or 200/400. In fact, it can sometimes hurt performance and stability. Don't mix them.

Re:Already got this luvin :)

[Boone^]

The problem with the NForce2 is that performance gains with the dual-ddr setup is nullified when you FSB and memory bus become asynchronous. Benchmark it now, and with your memory set to 181/362 and see if it's true in your case as well.

Re:Already got this luvin :)

[Sloppy]

That NForce2 stuff looks pretty cool; I wish there was a dual-MP motherboard like that.

Re:Already got this luvin :)

[Anonymous Coward]

It's not a dual-DDR issue. It's simply a timing issue. The same lack of performance increase can be seen in other, single-ddr boards. It's simply not efficient to run them out of sync.

Re:Already got this luvin :)

[sawilson]

I have my thoroughbredB 1700+ running at exactly the
same speed and it only cost me 45 bucks. It DID
require a boost to 1.75 volts from the stock 1.5 and
a thermalright slk-800 + 120mm fan though :)

Re:Already got this luvin :)

[GweeDo]

Thanks for the advice, it actually caused me to look into some other ideas and now my box runs solid as can be at 200mhz FSB and Memory BOTH! (so 400mhz DDR). In 3Dmark2003 I saw a decent boost with the added FSB bandwidth and having them both in sync now :) Oh...and I can pull 2.3ghz it runs out ;) So to back up what AMD is doing, uping your FSB IS very important. They only officially support 333mhz right now, so going to 400mhz = a gain of 66mhz or about 20% over 333mhz, so that is good candy if you ask me :)

Um, No...

[RMH101]

...and order of magnitude would be 10x faster. Yours was a pointless post and wrong too.

It depends.

[Andy Dodd]

What sort of applications do you run.

Apps that operate on large datasets and are memory-intensive will probably see significant improvements. (Scientific computing, maybe video encoding/decoding)

Game engines are typically designed so that the core of the engine is relatively small. Back in the days of the older P2 and P3, Celerons were considered the kings of gaming because of the fact that while their cache size was only half that of their big Pentium brother, the Celerons had full-speed cache while the P2 or P3 only had half-speed cache. It happened that the main rendering pipeline of most games at that time fit comfortably within the 128k cache of the Celeron - As a result, Celerons were actually significantly FASTER for gaming than their big brothers.

So for games you might not see much improvement, although these days things might be different.

Re:It depends.

[dreamchaser]

Things are very different. Look at the Quake 3 engine. The higher the FSB goes, the faster it renders. Most modern (and not so modern, Q3 is OLD now) engines not only take up more space than the available cache, but they also rely on streaming large amounts of data from main memory. Higher FSBs, as long as memory speeds keep up with them, will accellerate most modern and not so modern gaming engines.

Re:finally

[muyuubyou]

Huh? they're talking about bus speed. Intel is about to reach 200Mhz quad-pumped (800Mhz) and AMD is about to reach 200Mhz double-pumped (400Mhz).

If you don't know what you're talking about prevent yourself from posting.

Re:Fed Up With AMD Systems

[solidox]

the only intel system (p4) i used (work machine) managed to kill every harddrive i put in it after about 2 days (4hdds in total). switched to amd, not a single problem since. i've been using amd since the k6 200mhz, and have never been disapointed with them.
my VIA KT400 chipset has no problems either btw.

Re:What was with 333MHz? - price cuts

[muyuubyou]

Hmm I guess 333Mhz will be cheaper when 400Mhz comes out.

Not everybody goes for the latest and greatest stuff all the time...

Not always

[Andy Dodd]

I can choose a "top of the line" system, or a system that has 75-80% of the performance for half the price.

As a result, it's cheaper to buy a "lower end" system at a lower price and just replace it with a "lower end" system a year later. I'll get two systems, one of which is better than today's "top of the line", for the same price as one "top of the line" machine today.

Make sure you get something upgradable, of course.

Just look at CPU prices: Athlon XP 2500+ CPUs run around 2x the price of a 2000+. 3000+ CPUs are double that again. That's 4x the price for 1.5x the performance. Same for RAM, and to some degree hard drives. (With hard drives, you often get more "bang for the buck" by getting something close to top of the line. 120 gigs or so is currently the sweet spot as far as price per gig, and that's close to top of the line these days. But as soon as you jump to 160 or 200 gigs the price skyrockets. If you go down in size, you're spending not much less and getting significantly les.)

Re:Not always

[emotionus]

Those CPUs you quote are also running DIFFERENT cores. One being knewing, with a larger L2 Cache and BUS speeds

Re:Not always

[emotionus]

dammit. One being NEWER

Re:Not always

[Anonymous Coward]

Actually, Devsdeals.com regularly has 180GB and 200GB drives for less than $1/GB. Usually requires a rebate to get that price.

Re:Not always

[Anonymous Coward]

Athlon XP 2500+ CPUs

Ummm... there is no such thing.

http://www.amd.com/us-en/Processors/ProductInfor ma tion/0,,30_118_1274_3734^609,00.html

Re:Not always

[Zathrus]

Well, if you look purely at CPU speed/price then the best buy is the Athlon 1800 for $58. The 1900 costs $63, which is a 5.5% performance improvement for a 8.6% price increase.

Except that that's rather silly. I'd recommend buying an Athlon 2200 ($97) or 2400 ($123) - the performance improvements are marginal, but so are the price increases. It's not like it used to be where CPU speed steps were several hundred dollars apart. I certainly wouldn't recommend anything higher unless you had a good reason for it - and good reasons for faster CPU are few and far between when the bottleneck is more likely to be RAM, HD, or (most likely) the user.

Other than that, I very much agree with your point, and it's how I've been building systems for sometime now. The product cycles are too short and speed improvements too large to buy top of the line constantly - you'll be purchasing a new $1500-2000 system every year for a 20% speed increase. That or buy a $750-1000 system every year for the same increase and minimal realistic differences in performance.

Re:Not always

[DevilJeff]

Yeah, funny that it doesn't show up there, huh? Maybe AMD's official store lists it? Oh yeah, it does!
https://aac.avnet.com/AMDstore/index.jsp

Re:Not always

[Turbyne]

I used to use lower-end hardware, thinking that soon I'd upgrade. I've noticed that with lower end stuff there's a lot of time to spend with maintinence of the machine. This can break, that can break, etc. etc. emachines and wal-mart PCs may be cheap but they're certainly no more than a 2-year computer. I like building higher than mid-range equipment. Stuff that was top of the line at one point, but has been discounted down, but from good manufacturers. These computers usually end up being at least 5 year machines in my hands, and are inheritable/sellable. If you don't want to deal with this stuff, buy a Mac. [ducks]

Re:Not always

[SmittyTheBold]

120 gigs or so is currently the sweet spot as far as price per gig, and that's close to top of the line these days. ..provided you don't care about speed, which is something about which I care very much. The speed of my Atlas 10K3 drive makes it I barely care that it's only 18 GB. I wish I had all my media online sometimes, but then I launch a game and remember why I don't have the extra space...and the launch speed makes things alright.

Re:Not always

[Andy Dodd]

With the exception of the 3000+, all of the speedgrades listed in the original page are ones that the non-Barton cores were only available in.

The only 2500+ that exists is a new Barton core. (i.e. that page is probably out of date. NewEgg seems perfectly willing to sell me a 2500+ Barton.)

Re:architecture...troll?

[caino59]

hmmm sounds trollish, but I'll bite.

like one poster already said, bus speeds don't ramp up overnight.

but what really gets me....

way hot?

maybe you are thinking of your 400mhz k6-2/3 ;oP

intel is just as hot, if not hottor (i know someone will post the numbers)

and move to a new architecture...why?

they committed to the socket a, which I see as a Good Thing, as it provides a decent upgrade path for users. Unlike Intel...who knows whats coming next...maybe we'll see a hex-shaped socket ;oP

Not to say they *should* stick with that socket architecture forever, gotta move on when it's slowing you down.

Go look at some benchmarks, I don't think it's really slowing AMD down.

As an aside...how many over-clockers are out there already running at 400+ mhz FSB?

just curious.

one word says it all

[athlon02]

WOOHOO!

Makes me glad I didn't spend money on an Asus A7V8X yet.

Re:one word says it all

[Anonymous Coward]

All nForce2 motherboards already have built in support for a 400 MHz FSB. I don't have the link on hand, but nVidia stated this on their web site quite some time ago and well before Sis tried to claim all the glory for supporting it.

Re:one word says it all

[athlon02]

really? i thought that all nForce2 supported was DDR400, didn't know it supported actual 400MHz FSB for the processor too. neat! Welp if I can find an nForce2 with all the same features as the A7V8X that may be what i go with next.

Albeit, my next (and probably last for at least a while) computer purchase is going to be a 12" powerbook with 640MB, 40GB, DVD/CD-RW, as soon as I can scrounge up the $$.

Re:Fed Up With AMD Systems

[ergo98]

Do you Intel employees get a bonus for posting this sort of FUD? I have a KT333, previously using a KT266, and have had absolutely zero chipset related problems. Indeed the machine has been rock solid. Of course I have two Intel machines that are rock solid as well.

Intel FUDmeisters have always relied upon the "It ain't Intel so it's unreliable" FUD as their primary defense: If Windows 95 crashes on a Intel box, it's Microsoft's fault, but if it crashes on a AMD box, it's AMD/VIAs fault. Man does it get tiring.

Now I can

[Anonymous Coward]

Run KDE!

Re:finally

[Pike65]

Yeah, the guys over at HardOCP managed a stable overclock of a 3GHz P4 to 4.4Ghz, but it took them two tries. The first one they blew up at 4.2GHz, and the second one they managed to get to 4.4GHz stable but only with help of two Vapochill units (think refridgerator and you're getting there) as well as watercooling for the GPU.

That's the best clock I know of and I don't think anyone else has come close. Can anyone correct me? I'd be interested to see if that really was the fastest . . .

Re:64-bit

[emotionus]

64 Bit basicially allows more RAM (theoretically 312 TeraBytes I think) and larger instructions to be passed through the CPU. Your Bus speed has nothing to do with Windows, but to answer your question is yes Windows will take advantage because it's processes will be ran faster. Again, bus is independent of BITs as it's just a speed as to how fast something is going. For the sake of the question, yes a 64 bit CPU could use a 400MHZ.

AMD Codename Schemes

[Puu]

What's next for codenames? (Leaving aside brand names such as Athlon, Duron, Opteron.)

K6 was AFAIK nothing but numerals: K6, K6-2, K6-3...

K7 has been horses: Thunderbird, Morgan, Thoroughbred, Palomino, Barton...

K8 has been tools: Hammer, Clawhammer, Sledgehammer...

What's in for K9?

Dogs? ;-)

Re:AMD Codename Schemes

[be-fan]

I vote: Porn Stars!

'Sylvia' is the initial version with a 400 MHz DDR bus and 512K of cache.
'Jenna' is the second version, with extended SSE instructions and a bus speed jump to 500 MHz DDR.
'Traci' is the big jump to a 800 Mhz DDR bus and a 1MB L2 cache. It makes its debut with the 'Peter' Northbridge chipset along with a Southbridge that has no name, but a very wide bus...

Re:AMD Codename Schemes

[Anonymous Coward]

I second that.

ObSpeed

[ciscoeng]

"Whoa!"

Um.... well......

[walker2030]

400mhz DDR does not mean 400Mhz FSB plus the VIA KT-400 runs DDR 3200 (400 Mhz) and AGP 8x bus and its been out for a least 5 months

Re:Fed Up With AMD Systems

[one_line_enough]

Check out the CRC problems on MSI's (VIA) KT400 mbs. If winxp/via drivers happen to drop ide speeds from UDMA to PIO due to CRCs (still looking into it), other KT400 mbs might be experiencing the same problem, except that the users don't know about it - maybe this is the case where ignorance is not bliss. In the case of KT4V, MSI's idiotic solution to CRC problem is to slow down the ide speeds. The latest trend in the AMD world is to blame manufacturers of different components on the motherboard or the "other" guy, when something is wrong. Good examples are AMD and Tyan, Tyan and Crucial.

MOD PARENT UP +5 INSIGHTFUL

[Anonymous Coward]

Indeed, the audience takes it right up the whoop.

Death of the upgrade.

[Wino]

Anyone remember just a couple years ago when you could actually plan out a simple upgrade to your computer that would make it perform better for a modest price?

Toss in some extra RAM, wow no swapping!
Replace that CPU, doesn't Quake run good now!

The furious pace of bus speed changes have pretty much killed these types of upgrades for home/desktop users. Adding more PC2100 ram to their system when they know they're getting a DDR400 mobo is highly annoying. And forget about popping a new P4 or Athlons into your 1 year old mobo. Gotta buy $300 of new RAM and a $200 new DDR666-PC31337 AsusBitDragonMSI Ultra Deluxe to go with it!

Bleh.

Re:Death of the upgrade.

[Anonymous Coward]

It isn't really that expensive... I'm upgrading right now, and am looking at ~$30 for new RAM, $120 for a motherboard, and $100 for the new CPU.

Re:Death of the upgrade.

[Anonymous Coward]

(to clarify, I'm upgrading from a 700mhz SDR setup)

Re:Death of the upgrade.

[Wino]

I'll conceed that RAM prices have dropped dramatically over the last year. However, my main point is still valid. It'd be nice if the motherboard (the backbone of your computer) had a longer useful life than the things you plug into it. Otherwise we start heading down the road we've been on lately... every upgrade is replacing your ram/cpu/mobo all at once.

Re:Death of the upgrade.

[Jeff DeMaagd]

I really don't think it is much different than it was in the past. If you wanted the latest CPU, you often needed to get a newer board. You could still get the fastest CPU the existing board takes, which is often not nearly as expensive and get a significant boost.

With my P4 MB, I bought the best MB I could get at the time, and the slowest CPU still sold new at the time (1.7GHZ CPU), after the 3.0something P4 was released, a BIOS was released to take advantage of it.

It would be nice if CPUs weren't clock multiplier locked, you youd get a bit more flexibility, i.e. run a higher multiplier to work with a slower bus board.

I think it may improve with the Hammer line of chips, as the memory controller is part of the chip. Upgrade the chip, you might get a faster memory controller, thus you can upgrade the RAM without replacing the main board.

Re:Death of the upgrade.

[bogie]

Actually its not that bad it you buy AMD. They unlike Intel have stuck to basically one socket for a while now. Besides the newest Barton chips there are plently of people with mobo's that are well over a year old and are still able to upgrade to a pretty fast chip.

Re:Death of the upgrade.

[fbg111]

I think you overstate the case. The Nforce 2, for example, provides a superb upgrade path. It will take everything AMD spits out the fab in the foreseeable future, except Hammer.

I just upgraded to an Abit NF7-S Nforce2 mobo, with an AXP 1800+ and 512MB of PC2700 for less than $300 (from Newegg). My next immediate upgrade was to OC the CPU bus to 166 to synchronize CPU/RAM at 333. That provided a decent improvement, according to Futuremark. And my next upgrade will be to replace the CPU and RAM with a Barton 3200+ and PC3200/3500, running at 200/400 FSB (which Nforce 2 is capable of), once the price comes down a bit. I also plan to add dual SATA drives in RAID 0 config to boost that hdd performance. Dual SATA is included on the mobo.

In fact, I can't remember a time since I started building my own PCs in '97 that one mobo has provided such outstanding upgrade potential. Of course, if you're rich and can buy that NF2, Barton, PC3500, and dual SATA all in one fell swoop, more power to ya. But for those of us with only a few hundred to spend at a time, Nforce2 provides a very nice upgrade path with plenty of longetivity. At least until we start lusting after K8 mobos...

Re:Death of the upgrade.

[joedavis123]

Remember when every few years there's a big jump required to upgrade because of new technology?

Not any worse now than it ever was. Infact I see it as better because with fast processors and such, there isn't much of a requirement to upgrade, unless you feel you have to run the new Doom3 Alpha at more than 25 FPS

Re:Death of the upgrade.

[Anonymous Coward]

dunno what you're talking about... I recently bought 512MB of PC133 memory for my P3 600 for something like $60.

If I go buy the latest and greatest Althon, why would I care about $60 considering how much that'd cost?

Re:Death of the upgrade.

[Ramze]

I've been frustrated with this as well. I remember when PC133 ram was always backwards compatable with PC100 and PC66, but now memory makers are making PC133(only) chips... making it difficult to upgrade older systems at a reasonable cost (the PC133 chips are on sale for about 1/3 the price (with rebates)of the PC100's b/c there is a limited supply of PC100's)

BUT, as long as motherboard dimensions continue to shrink, chips become more integrated, and prices continue to fall, I don't see this as being a bad thing for the computer industry. Sooner or later, I think we'll have a small motherboard with a processor and connections for some standard I/O ports (possibly fiber optics) and a power supply connection. I think eventually, RAM, networking hardware, and video will all be on the processor die. (to me, this makes sense as connections need to become faster in order to increase performance of the system)It'll be simple to pull the motherboard and processor out and replace them with newer ones for a relatively low price b/c they are one unit. (processors could become integrated into the motherboard -- no need for a socket if they'll never be swapped out)

Granted, I'm thinking 10 years into the future... but, think. We're already putting 8 megs of cache on processors (I remember when I had only 8 megs of ram in my computer), Intel is putting wireless technology into CPU's, Modem and Networking hardware has been built into CPU's in the past, and now AMD is putting memory controllers into the CPU. I don't think it will be long before Nvidia partners with one of the CPU makers and integrates their GPU with someone's CPU. Graphics cards require ever-faster connections to the CPU. Changing BUS speeds and graphics card slot designs are great, but the graphics cards need increasing amounts of power and are suffering from overheating -- my solution would be to put the GPU and CPU together, let one heat sink work for both & let them share RAM and possibly even each other's registers as needed.

I think the upgrade of motherboards, cpus, and ram as seperate entities will go away, but the upgrade of a machine will become cheaper and simpler. Just pop out the motherboard with integrated cpu,gpu,ram,network card, and chipset and pop in a new one and connect to peripherals by some standard serial I/O ports.

Re:Death of the upgrade.

[MarcQuadra]

The problem there is that such deep integration costs more than the benefits it produces. While it makes sense to stick the memory controller and maybe a few I/O processors on the CPU core it's very diffucult to cost-justify including a GPU and audio. It would cost millions to integrate the GPU on to the CPU, and any changes in the design of either part would nescessitate both the GPU and CPU teams to hash it out. There's little advantage to integrating the CPU and GPU, AGP is a very fast and cheap way to get tons of data back and forth between those components, and I don't know of anything that pushes AGP to it's limits, the bottleneck is more often than not the GPU and the GPU-VRAM pipe.

What I do see in the future is more modular designs. Eventually this serial craze will boil down to the chipset level and there will be packet/serial interfaces between key components. Since the different components will 'speak' to each-other instead of signalling over customized paralell interfaces there will be much greater flexibility and reduced R&D costs. This type of system would be much slower with today's technology, but advances in serial signalling, superconduction, and fiber optics will bring us much closer to a modular future.

Re:Death of the upgrade.

[Ramze]

I see things as going the other way. You're talking about today's technology and today's costs, where as I'm talking about 10 years down the road. I see things as going towards integration instead of modularization. 10 years from now, the die size for processors will be smaller than 1/16th the size of todays -- making it easy for today's processor core to sit on the same die as a GPU and many other devices.

Once key components are defined and standardized, they don't need to be seperate from the CPU other than an interface for devices. Every time the die size of a processor shrinks, more circuits can be laid on the same surface area -- it would make sense to put standard components onto the die size when that happens. I would bet that Nvidia would merge with either intel or AMD to create system-on-a-chip CPU's. Such systems would be low power, standard, and cheap compared to todays systems & fiber optics could connect the CPU to any storage devices or peripherals that aren't wireless by that time. I wouldn't be surprised to see Intel merging with ATI and AMD merging with Nvidia.

You're correct that changing either the cpu or gpu would require consultation with the other design team, but not if you only have one design team to begin with. As for design costs, Intel spent big bucks designing the Itanium line of processors that is likely to tank. I think it'd be a worthwhile investment to work with companies that create chipsets, video cards, and other standard components to improve the processor any way we can. We've added MMX instructions and other multimedia extensions to the processor... it only makes sense to put the whole gpu on there with it once it's economically viable. AGP may be fast by today's standards, but when we have processors running at 48 Gigahertz, I think we'll have a noticable benefit having a GPU of the future talking directly to the CPU. We added math coprocessors to processors a long time ago, I think we'll add graphics coprocessors as well (we are headed that way w/ all the multimedia extensions as it is). In order to keep up with competition, the graphics cards of the future will either have to have a much wider interface to support a huge cooling system anyway -- may as well be on the processor and share the heat sink. A large reason the GPU's aren't running at 3 Ghz is due to heat and power constraints which wouldn't be as much of a problem if they were integrated with the CPU b/c they could use the same power source and cooling system. Also, the main bottleneck for graphics cards is RAM these days... they're the main people pushing for faster RAM so they can keep pumping out ever-faster cards with cheap memory, but when we can fit more on ever-smaller silicon wafers, they'll be using an onboard cache for their GPUs as well. So... you'll have a CPU with a large onboard cache... and a GPU with a smaller, yet still adequate onboard cache... why not combine the two and share the same cache? Why have a CPU with 3 gigs of onboard cache sitting 6 inches away from a GPU with only 1 gig of cache... instead of a CPU/GPU chip with 3 gigs shared (assuming that the cpu never uses the full 3 gigs and the GPU can steal at least 1 gig when it needs it... or hell, put more cache on the combined chip). Sure, larger chips give lower yeilds, but I think that makes some sense... lol... I'm rambling and it's 1:00 AM -- got class in the mornin'.

I think your idea is interesting, but your main argument against my idea is cost, which goes down exponentially over time in the computer industry. I don't think there will be very many players in the computer industry either as it moves towards maturity... so, companies will buy each other out and combine technologies, things will become more integrated and very very tiny. I wouldn't be surprised 10 years from now to buy a computer the size of my alarm click that plugged into the back of my plasma screen monitor and wirelessly connected to a my HDTV, stereo, a keyboard, mouse, RAID drive in the basement, and a microphone

Re:Death of the upgrade.

[MarcQuadra]

The number one cost in making these will be the engineering labor to integrate the components. If there are standardized ways of connecting these parts then we are in the same place we are today and we're very close in our thinking. I agree that more will be integrated onto a single big chip, I think that's a great idea, I think the CPU, memory controller, several I/O DSPs, and a huge L1 cache should be packed onto the CPU core or in the same IC. On low-cost systems integrating the GPU might make sense too, but the implementation I find most likely for this is a whole mini-AGP bus INSIDE the Main-IC, custom-designing a CPU-GPU bridge for each processor will be way too costly and time-consuming. The devices ought to have seperate logic but in some cases be inside the same package for size/heat/cost issues.

I think we're on the same page but in different paragraphs. :-)

Re:Death of the upgrade.

[Ramze]

hmm... you're probably right on that. It would allow for a wider diversity of GPU and CPU's to communicate without serious redesigns -- great for marketing different products for different consumer bases. It would mean that you wouldn't have to redo a whole fab plant for CPU's if you came out with a better GPU to work with it, or vice-versa -- cutting the cost of shifting product lines. Also, if it's found that a CPU and a GPU don't need the same clock speed or can ramp up or lower their clock speed with usage(highly likely), it'd be easier to impliment the speed-stepping individually if the two were on seperate chips. Of course, that's all implying that there will be lots of innovation in the processor and graphics chip areas 10 years from now... but, I think that's a safe bet to take.

Too cool...

[Anonymous Coward]

...now if we only have Serial ATA disks with sub-millisecond access times!

Re:Fed Up With AMD Systems

[bigbadwlf]

You're jumping to quite a conclusion by insinuating that AC is an Intel employee.

Now I've never had an AMD machine, but I used to have a Pentium MMX on a non Intel chipset and never really realized how unstable it was until I upgraded to a PIII with a BX chipset. That system's been rock-solid.

So my conclusion based on my personal experience is this:
You're better off with a chipset made by the same manufacturer as the CPU (regardless of who it is).

Re:finally

[Anonymous Coward]

Well, it was a sad day when CPU clock speed stopped being integer multiples of bus speed.

400mhz bus and 3200+ XP?

[Anonymous Coward]

Uh...uH... UHGGAg.. ... I just wet my pants. Thanks slashdot!

[OT] Re:Already got this luvin :)

[Psiren]

Are you running Linux on this machine (or anyone else reading)? I'm looking at getting an NForce2 board (probably an MSI with the Geforce4 and SATA) myself sometime soon, and would be interested to know if you had any difficulties with it. Any problems with DMA and that sort of low level stuff? Cheers.

Re:[OT] Re:Already got this luvin :)

[Quickening]

Linux on NForce2 getting there.

Did you have any OS install issues?

[Robber Baron]

I just picked up the same board with a XP 2500 (Barton), a Gig of DDR 2700 (2x 512) and an 80 Gig ATA 133...but I'm having a bit of bother getting W2k to install (kept getting stop errors :( ). I only just put it together last night and probably should've waited until this morning when my head was clearer, but you know...geeks and their toys! Anyway, with my latest attempt (started this morning and formatting as I type this) I first disabled the Serial ATA as I don't yet have anything to go there and it seems to be working so far a little better. It was probably looking for a SATA driver and freaking out when I didn't give it what it wanted. Once I get W2k running I'm going to try loading Mandrake and BeOS.

Re:Did you have any OS install issues?

[Moloch666]

You should load gentoo or something on that baby.

Re:Did you have any OS install issues?

[Robber Baron]

Yes I've used it loads of times and never had a problem. I have access to some more recent ones however (it has SP3 already included)...might try one of those later if the one I have doesn't work.

Re:Did you have any OS install issues?

[Brian Stretch]

I used my existing Red Hat 8 install, which worked fine. I suspect that Win2K is too old to know how to deal with your state-of-the-art nForce2 board completely OK, especially if your install disc doesn't have any of the service packs integrated (heck, Win2K really wasn't stable until SP2). Yeah, I disabled SATA too. WinXP would be a better choice if you have to run Windows, but once you get Win2K loaded, SP3 installed, and the current nVidia drivers, you should be OK.

I'll assume you have a proper power supply. If it's P4 capable (has the extra 4-pin square power plug that uniprocessor Athlons don't use) you'll be OK.

Be sure to flash in BIOS 1002 from ASUS's website if your board didn't come with it. There were issues with many of the prior BIOS versions. I'm still running 1002 Beta 1.

Re:Fed Up With AMD Systems

[Anonymous Coward]

I'm not sure Intel is immune from these types of problems.

The Intel 82801AA (ICH South Bridge) used in many i810 (and i8xx series) boards had data corruption issues.

Re:64-bit

[Rcknight]

were there as many typos in that program as there were in your post?, must have taken u forever to get it to compile!

Re:finally

[Anonymous Coward]

The fastest retail P4 runs at 3.06 GHz, which actually is about the same speed as the 3000+.
Actually claim that is disputed by a lot of people

http://www.aceshardware.com/read.jsp?id=50000372
"

Comment removed

[account_deleted]

Comment removed based on user account deletion

Comment removed

[account_deleted]

Comment removed based on user account deletion

What about the Intel 800MHz bus

[workindev]

Why is this a big deal when Intel is moving to 2x faster (800MHz) bus later this year?

Re:What about the Intel 800MHz bus

[Anonymous Coward]

because intel is hyperpipelined so they do fewer IPC than AMD does, therefore a dualpumped amd200 bus is far more valuable than a quad pumped 200 p4 bus...

Re:What about the Intel 800MHz bus

[Anonymous Coward]

IPC and bus speed are completley unrelated, but you AMD fanboys love ignoring common sense.

Go take a look at some memory benchmarks that compare an Athlon with a P4 and try your argument again.

Re:What about the Intel 800MHz bus

[ProtonMotiveForce]

Wow. You're stupid.

Performance is all that matters in the end, not IPC.

Re:What about the Intel 800MHz bus

[NedTheNerd]

exept the fact that it doesnt really matter all that much you get nominal performance gains with increases in buss speed. Im not a rockt scientist but I "think" the reason this is news is because x86-64 seems forgin to some people even though its a hacked up version of x86.

Re:What about the Intel 800MHz bus

[soulsteal]

Right now, both chip manufacturers are running the same base bus speed: 133MHz. While AMD is running it's chips on a dual-pumped bus (266 MHz), Intel is running on a quad pumped bus (533 MHz).

With AMD and Intel moving to 400 and 800 MHz respectively, they're both moving their FSB base speeds up to 200 MHz. That's still a 50% increase in base bus speed for both parties.

Re:What about the Intel 800MHz bus

[workindev]

Exactly. My point is, why make a big deal about AMD announcing it when Intel has had this in the works for almost 6 months?

some please explain quad/double pumped BUS speed

[zaqattack911]

I am well aware that the 400mhz bus on p4s is quad pumped, and is truely running at 100mhz fsb (or a bit more for 533).

I am aware that AMD 400mhz bus is a double pumped 200mhz fsb.

Could someone explain what "double pumped" actually means? if I think back I remember hearing something about how in doulbe pumped.. the cpu grabs data off the bus at the beginning, and the end of a single clock cycle. is there a downside to doing things this way?
Or perhaps.. this is the best way things should be done, and cpu designers should concentrate on LOWERING mhz (for heat/energy reasons), and UPPING the amount of data/instructions it can do in a single clock cycle?

So eventually we could move toward a computer that can run on a single clock cycle, which would be a mhzless computer? I know there is theory somewhere in there :)

Would it not help voltage/heat greatly if the bus was 33mhz and (12x) pumped?

--Zuchini.
(I keep writing my name, erasing and using an alias instead :) bad habbits)

Re:finally

[Anonymous Coward]

umm they still are multiples of the fsb. are you all high?

So what?

[Andy Dodd]

Note that the "2000+" etc. speed ratings are designed to compare the CPU to a P4.

So an Athlon XP 2500+ will have approximately the same performance regardless of the core. Despite the newer core, the 2500+ Barton (clocked at 1.83) is cheaper than the 2600+ non-Barton.

Different cores or not, the 2500+ gives approximately 25% more performance than the 2000+ at twice the price. The 3000+ gives approximately 50% more performance at over 4-5 times the price.

Re:Fed Up With AMD Systems

[ergo98]

I was being facetious in claiming that the AC was an Intel employee, though it isn't beyond consideration that there could be Intel employees doing just that: One of the primary apprehension causers for people considering AMD processors is the perfect example of FUD at work -- Oooh, maybe it'll burn up (in actual practice the heat of an AMD is not a concern. Bogus claims about the greatness of Intel processors because of their heat throttling is like claiming that a model of car is great because it has a parachute, and then showing a demonstration of a car being pushed out of a cargo jet), and maybe it'll make all of my apps corrupt, etc. Of course he could just be an Intel fanboy, and there are lots of those (just as there are lots of AMD fanboys).

Regarding chipsets: Intel has put out their share of fragile or completely non-working CPUs, chipsets, etc. The BX is one Intel chipset that hung around so long specifically because it proved itself, but in amongst were a serious of blunders and mistakes. Again I run a KT333, previously a KT266, and it is rock solid: I have never had a single problem and run the gamut on my machines. Of course at the same time I have an Intel system that runs rock solid.

My experience

[ComputarMastar]

I built my first Athlon system after seeing how well a friend made out with his Slot-A Athlon system. Socket-A 1.2GHz was the bleeding edge at the time so I got the 1.0GHz model w/ pc133.

After I'd pulled out half of my hair trying to figure out why it refused to run stable (or as stable as my Celeron 300A w/ Win98), I ditched the mobo for one of the new KT266A boards. It ran sort of OK for a month, then stopped running for more than 1 or 2 minutes at a time. After more hair-pulling I noticed that the power supply was putting out 3.9v on the 3.3v line and the power regulators on the board were too hot to touch. A new power supply, motherboard, and CPU later I was back in business... or so I thought.

It still had weird unexplainable crashes at odd times, so I installed Win2K and that helped a little, but it still wasn't stable. I was determined to have an Athlon system that was stable, so instead of going Intel, I started scouring message boards for anything that would help me get through this.

Many posts pointed at a buggy implementation of ACPI in the first revision of Soundblaster Live's, which I had. Got a Game Theater XP, but same problems. Finally found some info about how VIA wasn't even implementing PCI to spec, so it was time to look for a non-VIA board.

At the time it was mostly all-VIA or an AMD northbridge + VIA 686B southbridge combo. nForce had just come out, so I was hesitant to try an untested chipset, so the only other alternative was AMD's new SMP chipset. SMP was something I'd always wanted to try out and the Asus board could use my old RAM and CPU until I could afford a pair of the pricey Athlon MPs, so thats what I got.

About a year later, I'm still running it with a single XP 1800+ and it has been very stable. The only time it ever crashes is when I let it get too dirty and the geForce3 overheats.

Oh yeah, review sites are absolutely worthless if you're looking for stability. "Its very stable!" really means "It didn't explode while we overclocked the hell out of it for the 3 days we had it."

So anyway, my point is that some of us get burned bad by Athlon systems. Its only because I'm so stubborn (or maybe insane) that I stuck with it until I got it stable instead of switching to an Intel system and preaching the evils of AMD.

Re:My experience

[Anonymous Coward]

You are more likely to experience these problems with an AMD system, although it was probably the SB Live! at fault.

I had an SB Live! in a P-II 233MHz system. There were occasional bursts of static in the sound output and Win95 was rather unstable. I naively blaimed Microsoft.

Then I put in a VIA motherboard and Athlon 1.0GHz (Thunderbird A). Now both Windows and Linux were horrendously unstable. By reading on the internet I learned that the SB Live! was basically a piece of crap, and that it worked particularly poorly with VIA chipsets. I ditched it in favor of a Turtle Beach Santa Cruz, and things have been fairly good ever since.

(Well, until I upgraded my linux kernel and started suffering from the athlon extended paging bug... which was annoying but at least there was a simple and effective workaround.)

Sir, there not shitting on me

[diablobynight]

I beg to differ, I think the front side bus upgrades are the most important thing they do. You see, FSB is how fast you can get data to and from that super fast processor. You don't want a bottle neck between your memory and your L2 cache, or your Cache and your processor. I like that their upgrading, but I read an press release about it, over the summer, So I think it's retarded that their putting it in slashdot now.

Re:Sir, there not shitting on me

[Anonymous Coward]

I know full well what FSB is and its importance. Yes raising the speed will bring benefit, but there is also the downside I mentioned. If you are going to obsolete a huge pile of hardware you should make it worthwile, not for a 20% jump that will only gain you 5% of real world perfomance at most. Bigger jumps in the order of 200 -> 333 or 333 -> 500 would be a lot better.

Besides AMD just released Barton, the extra L2 should make it somewhat less constrained by FSB.

Re:Sir, there not shitting on me

[Zider]

I assume these new 400MHz-bussed cpu's still can run at lower speeds, right? So the "old" mainboards can still run the new cpu, just not att it's peak performance.

Correct me if I'm wrong, I'm just guessing here. I'm still on a Celeron 300MHz.. (CPU speed, not FSB)

your upgrading argument ....

[diablobynight]

needs to stop. It has been made before. I personally use Athlon chips myself, but I also believe that it's worthless to upgrade processors and not motherboards, in most cases. Like I have my old system that is PC133 1.2ghz and when I upgraded to my NFORCE2 system, I just built a whole new box for like 580$, I transfered my soundblaster live card and my Geforce 4 card to my new system and bought a new case, keyboard, mouse, processor, ram, motherboard. I put an old PCI card into my 1.2 ghz box and now I use it, as a webserver,router, file server, and occasionally just to sit their and encode divx when I don't want to take up processor cycles on my new box.

The REAL question...

[NerveGas]

The real question, at least in my mind, is whether they will make AthlonMP's with the 400-MHz bus. While it's not a wrap-up, indications seem to say that they won't, because it would compete with the hammers.

Seeing as how the AthlonMP motherboards have seperate busses for each processer, imagine if Nvidia made an "nforce" chipset with dual-channel memory for dual Athlons - each processer could get full memory bandwidth at the same time. That would be truly impressive, especially for RDBMS servers where you live and die on bandwidth.

But, of course, such a monster would be a direct competitor with the Hammers - and AMD's got too much at stake to let the Hammers fail.

steve

Always Behind

[Remlik]

Whooo hooo! Huge upgrade to 400Mhz still leaves them behind Intel by 133Mhz! Makes me wanna rush right out and upgrade my system to below competition specs!

Re:Always Behind

[NerveGas]

Remember that an Athlon on a *measly* 266-MHz bus can often out-outperform a P4 on a 400-MHz bus, and sometimes out-perform the P4 on a 533-MHz bus. Assuming they actually increase clock speed (and not another PR game), I'll bet that this will make the Athlons VERY competitive again.

steve

Re:some please explain quad/double pumped BUS spee

[zaqattack911]

Troll? Sometimes I just don't get this place.

Reply to me if yer gonna troll a perfectly valid idea, you little slut.

Re:some please explain quad/double pumped BUS spee

[ultor]

The problem with such an approach is that it would drastically increase latency between the memory and the processor. While theoretically you could transfer an enormous amount of data in a short amount of time, random access to data would be very slow. The actual bandwidth comes into effect when a large amount of data is being fed to the processor in a steady stream, but when you need only a small bit of data, it would travel at the actual bus speed across the pipe, in this case 200mhz for the Athlon and 133mhz for the Pentium 4. This is why the Pentium 4 excels using Rambus at 32ns while the Athlon prefers 8ns DDR.

Re:False post

[Eagle5596]

Actually I was logged on already... time for morning slashdot

Hilarious!

[Anonymous Coward]

Of the replies, the two that claimed that AMD suxxors got moderated "insightful", while the two that questioned the bogus anecdotal evidence got moderated down. Someone has an "Intel Inside his ass" dildo on superhigh.

Re:Hilarious!

[Anonymous Coward]

That reminds me of the ibrator.

Actually....

[Dawn Keyhotie]

Boromir was Denethor's eldest son, Faramir was Boromir's younger brother. None of them were ever king of anything, Denethor was the last Steward of Gondor, i.e. standing in for the real king, Isildur.

Isildur went off to war, whacked off Sauron's finger(s), took the One Ring of Power, decided not to destroy it in the Crack of Mt. Doom (btw that sounds kinda nasty nowadays), and subsequently got himself killed in an ambush up north.

And ever since then, the Stewards of Gondor faithfully waited for the real king's return. And of course the real king was Isildur's heir, after several (many?) generations, the Ranger of the North Frodo and his faithful sidekick, Samwise the Great.

Or something like that.

Just underclock!

[MarcQuadra]

Not true! I have my KT266a motherboard here running a barton, it's just got the FSB underclocked, it runs cool and faster than my old tbird. And this system has PC3200 DDR RAM=, it just is running at PC2100 speeds right now. My next purchase wil be a new mobo that can take FULL advantage of the CPU an RAM. Look at the Intel side, they change the PHYSICAL pinout so you CAN'T do this. The athlon has been on one single pinout while intel has done FC-PGA, FC-PGA2, 427(?), 472(?).

DOn't underestimate the power and value you can get from underclocking.

Here's what double/quad pumped means

[StandardCell]

All digital data is synchronized to a clock, be it source-synchronous (i.e. clock comes with data), which is the case with DDR, or recovered clock (i.e. clock information is based on rate of change of incoming data). Whatever scheme you get, you will still have a clock inside at some point.

Traditionally, the memory elements or registers on a chip will ignore incoming data until the clock signal undergoes a positive transition, i.e. logic low to logic high. At that point, assuming the data has been stable for a long enough period of time before and after the clock edge, it will be captured. However, since there is only one positive edge per clock cycle, data can only be captured on that edge.

In a double-pumped scheme, what you have is a set of 2:1 multiplexors that go to two different sets of registers. One is sensitive to positive edges, the other is sensitive to negative edges, i.e. logic high to logic low transitions. If you simply wiggle the data out faster, and you have a double-pumped scheme with a small FIFO buffer, you can recover data twice as fast as a single edged scheme. On the interface itself, there are special low skew low insertion delay clock distribution schemes that enable this to happen without too many problems.

In a quad-pumped scheme, you actually have two separate clocks that are 90 degrees out of phase with each other. In effect, you have two positive and then two negative edges to work with internally now. You wiggle data out at 4x the single data rate, and have 4:1 multiplexers to the registers, plus (again) a careful layout of the internal clocks.

The area overhead in such schemes is minimal (~10% for DDR) and really takes advantage of the speed of on-chip devices. It does take some special consideration, but from the perspective of increased die size, it's not a problem. Power, however, is significantly increased for both I/O (SSTL-2 type stuff) and for core devices because of the data rates, and that is also a consideration during design of not only the power distribution, but also the package/module design and the board design.

And, FYI, Rambus uses multiple serial/deserialization (SERDES) that wiggles data between a pair of signals (positive and negative) whose voltage differential is recovered, not for individual levels, which (supposedly but not actually) simplifies matters. Transmitting data via this differential is actually much faster than a single-ended scheme like DDR currently is (single ended meaning all I/O refer to a common ground (and voltage reference)). Then they even IIRC get into exotic schemes like multi-level differential (i.e. steppings between 0 millivolts differential and full swing). I could be wrong about the latter though...

Get Em While They are HOT!!

[atarione]

Just take the bathroom fan out and mount it on these bad boys .... ya'll will be good to go @ about 60c.

not quite..

[jwdeff]

A new chipset... supports a 400MHz front-side bus ... which appears to clear up questions about whether AMD would include that feature in the forthcoming Athlon XP 3200+ processor.

A chipset supporting a 400MHz FSB does not mean the Athlon XP 3200+ will have a 400MHz FSB. In fact, according to AMD at the very same trade show, it will not support a 400MHz FSB.
(http://www.tomshardware.com/business/20030314/ceb it2003_2-03.html)

Also, every time AMD adds more cache or increases the FSB speed, the processor gets a lower clock rate to product number ratio. The 2700+ with 256Kb of L2 Cache is clocked the same as the 3000+ with 512Kb. So, even if they shipped 3200+'s with a 400 MHz FSB, it would probably be clocked about the same as a 3000+ (at like 2166 MHz). All in all this isn't a bad thing, but you wouldn't be getting an extra 200+'s AND the increase in speed from the faster FSB, the FSB performance bump is figured in to the model number.

Who is hottest?

[steveha]

they are already running waay too hot.

Actually, AMD processors are cooler than the equivalently-performing Pentium 4 chips.

Athlon XP 3000+ max heat: 74.3W
Athlon XP 3000+ typical: 58.4W
Athlon XP 3000+ temperature limit: 85C

Pentium 4 3.06 GHz theoretical max heat: 109.0W
Pentium 4 3.06 GHz thermal design power: 81.8W
Pentium 4 3.06 GHz temperature limit: 69C

What Intel calls "thermal design power" is sort of similar to what AMD calls the "typical" number. It's 75% of the theoretical max temp, so the theoretical max temp for the Pentium 4 would be 109.0W. But the P4's clock throttling would keep it from hitting that theoretical max temp.

My source for all this:

http://users.erols.com/chare/elec.htm

Note also that since your power supply isn't 100% efficient, and since the power supply has to produce one Watt for each Watt your system dissipates, that a complete system with a Pentium 4 will dissipate over twice the difference of just the CPUs. In other words, for our example, the Pentium 4 dissipates about 23W more, so the Pentium 4 complete system will dissipate even more than 46W compared to the Athlon XP system. I'm not sure how efficient a typical power supply is, but if we assume 66% efficiency, the total for the Pentium 4 complete system would be about 58W more than the Athlon XP system.

steveha

Re:Who is hottest?

[steveha]

Aaack. I cannot do math today.

For the complete system numbers, you don't double the difference and add some for the power supply inefficiency; you just add some. The power supply doesn't dissipate 1W for each 1W it provides the system!

So, better numbers: if the Pentium 4 dissipates 23W more than the Athlon XP, and the power supply is 66% efficient, the Pentium 4 system will dissipate 35W more (the extra 12W come from the 66% efficiency of the power supply).

Sorry about that.

steveha

SMP box...

[Anonymous Coward]

should i go for 2 athlon MP 2200's now for 210 bucks www.newegg.com or wait for a 64 bit and/or 400 mhz fsb MP chip???

NO Opteron For YOU!

[Nazmun]

Sorry to rain on your parade and all (it's true I really am sorry) but the opteron will be more expensive then what you would spend on an entire PC. It may be cheaper then the itanium but it seems the pricing for this chip will be in the thousands of US Dollars. Opteron's (Sledgehammer) will be mostly for servers and high-end workstations (targeting Itaniums market).

However the Athlon 64 (*hammer - tasty but not nearly as Juicy as the Opteron) will be sold for people like you and me. These will start out expensive like the Barton 3000+ chip did when it came out.

Thats Not A chipset problem

[Nazmun]

Incorrect voltages and overheating related issues of mobo chips sounds like a mobo problem but not the fault of the chipset manufacturer (Via) or the chipset.

Voltage problems can be caused by a bad power supply or crappy voltage regulators. Unfortunately just like before there is a greater selection of crappy AMD boards then there are Intel boards. You just have to do a bit of research and be sure to buy a board from one of the better known manufacturers like Asus, Abit, MSI, and Epox (they have made crap but are doing an awesome job with their latest nforce2 line).

So how fast will it be?

[hendridm]

Is that 400 MHz in Intel or AMD numbers? Are they going to release it as the 533+ bus operating at 400MHz? You know, so consumers won't get confused...

Best Buy rep: Based on what you described to me, I would recommend this Compaq with an Athlon 2100+ processor.
Average customer: Is that a Pentium? How fast is it?
Best Buy rep: Actually, it's roughly equivelent to the Pentium 4 architecure, and runs at about 1.8GHz.
Average customer: Oh, give me whatever's cheaper.
Best Buy: *sigh* Have you taken a look at our eMachines yet?

Low-end != cheap

[Andy Dodd]

eMachines/Wal-Mart PCs are low-quality.

I was talking about a high-quality but not super-top-of-the-line PC, like a lower end Dell, or in the case of homebuilt, using an Athlon XP 2000-2200+ rather than a 3000+ and a GeForce4 Ti4200 or 4400 rather than a 4600 or an FX (Although these days 4600s aren't too expensive.), and only getting 512M or RAM in 256M sticks rather than 1 gig in 512M sticks. No quality difference, but a big price difference.

Although if there's any part of a system I would not skimp on, it's the video card. I reccommend budgeting at least $200 for the video card alone if you want good performance - The video card is the bottleneck in most games these days, and is the one place you shouldn't skimp. I have a Ti4600 and only a 1.1 GHz Athlon - My video card is still the limiting factor most of the time. The only reason I want to upgrade now to a 2000-2200+ myself is because I've been doing a lot of video encoding, which is one of the few apps where a fast CPU is REALLY useful. (22 minutes of 1080i HDTV video takes approx. 7-8 hours to do a two-pass decoding/deinterlace/resize/encode to DivX run.)

Re:Low-end != cheap

[Turbyne]

Interestingly you've pretty much spec'd out my Athlon. The Force is strong in this one.

Re:A polite but urgent warning to Americans

[Anonymous Coward]

You're a stupid bitch.
Grow up.
Tommy, is that you? My 4-year-old neighbor with a chip on his shoulder? Is it past your bedtime?

Fuckkkkkkking la-HOOOOOOOOO-zer.

Last Post!

[alpg]

Who wants to remember that escape-x-alt-control-left shift-b puts you into
super-edit-debug-compile mode?
-- Discussion on the intuitiveness of commands, especially Emacs

- this post brought to you by the Automated Last Post Generator...