Athlon 64 Debuts

Anandtech by Anonymous+Custard · 2003-09-23 05:39 · Score: 4, Informative

And Anandtech has a good article up, as well.

"Fast one"? by micromoog · 2003-09-23 05:39 · Score: 3, Insightful

Interestingly enough, Intel pulled a fast one (literally) and released a new breed of Pentium 4 chips with 2MB of on board L3 cache, just in time to boost their performance in the benchmarks for this launch.

So Intel cheated by, uh, making better hardware?

Re:"Fast one"? by NitroPye · 2003-09-23 05:56 · Score: 2, Interesting

A troll but I will bite. Did you happen to read that it was not running any applications optimized for it? The pentium 4 is all about just tossing in more cache and higher clock speeds where as the AMD chips are about efficiency. A 2.2ghz chip directly compeating with a 3.2ghz chip with a shat load of L3 cache shows that it is mearly just a smoke screen. Not only that but the 3.2ghz Pentium 4 costs a bundle more. What about the lower memory access times? I'd prefer lower memory access times.
Re:"Fast one"? by MisterFancypants · 2003-09-23 06:00 · Score: 2, Insightful

A troll but I will bite. Did you happen to read that it was not running any applications optimized for it?
Yeah because the applications don't exist. Maybe in a couple years time, AMD's processor will be the better choice, but if there are no apps there to support 64bit, its power is wasted.
Intel's chip is the better choice for the user who wants performance in apps that actually exist today.
Re:"Fast one"? by default+luser · 2003-09-23 06:14 · Score: 2, Insightful

The extra registers are not available in 32-bit compatibility mode. The gains in moving applications to x86-64 will largely come from these extra registers being made available.

--
Man is the animal that laughs.
And occasionally whores for Karma.
Re:"Fast one"? by Zathrus · 2003-09-23 06:41 · Score: 4, Insightful

Intel's chip is the better choice for the user who wants performance in apps that actually exist today.

Based on what?

The Athlon64 outperforms the P4 in virtually every benchmark, often by a hefty margin.

Ignore the results from the P4EE chip -- you said you wanted something that actually exists today. The P4EE will not be available until sometime in November.

On top of that the AMD Athlon64 3200+ costs ~$465, while the P4 3.2 costs ~$620. So you get slightly more power for considerably less cash. The motherboards appear to be roughly the same price for a Socket 754 board and a 865PE-based board (or you can drop more money for a 875-based board and get virtually no performance improvement -- your choice).

All of that said, if you want to buy the best x86 chip right now and money's no object -- buy Intel. Not because it's faster (it isn't), but because it's probably going to be more stable. We're talking about a 1st generation CPU on 1st generation motherboards. Not a great prescription for stability there. If money is an issue, but you still have to have the fastest thing out there, then buy an AthlonXP -- the 3200+ is about the same price as the Athlon64, but the MB's are cheaper and more stable (just as stable as P4 chips/mobo's).

If you want to be a realist and get the best bang for your buck then stop looking at the fastest chips available and start looking at the best price point -- a P42.6C or AthlonXP 2800+.
Re:"Fast one"? by sketerpot · 2003-09-23 06:50 · Score: 2, Insightful

With an OS where you can compile your own apps.
Re:"Fast one"? by ceswiedler · 2003-09-23 07:57 · Score: 2, Interesting

Modern processors already use massive register renaming tricks internally, so it's not as if these extra registers will be completely unused. There is an advantage to letting the compiler (as opposed to just the processor) pick registers, and in 32-bit mode, that will be lost.
Re:"Fast one"? by Ninja+Programmer · 2003-09-23 11:10 · Score: 2, Insightful

All of that said, if you want to buy the best x86 chip right now and money's no object -- buy Intel. Not because it's faster (it isn't), but because it's probably going to be more stable.
Remember that Athlon64 motherboard will be *MUCH SIMPLER* than previous generation 32-bit x86 motherboards since there is no seperate Northbridge that you need wired up. So there are fewer connections, and the mobo can be significantly smaller. Furthermore the power utilization of A64 has been much improved over the Athlon. I think that anything that AMD recommends or which is sold from the major vendors will be fine, and likely very comparable in stability versus anything from Intel.
Re:"Fast one"? by Hoser+McMoose · 2003-09-23 16:56 · Score: 2, Insightful

Just as a FWIW, the Athlon64 and Athlon64 FX are not first generation chips. They are currently on Revision C. Revision A (actually there were a few A versions, at least A0 and A1, possibly A2 as well) was the test silicon that was never meant for market. Revision B is what the first Opterons shipped as. The Athlon64 and Athlon64 FX are revision C of the exact same die.

Right now AMD is using the exact same silicon for all three of these processors. This is good in tha they've already had quite some time to get the bugs worked out (A0 silicon was out two years ago), though there's a slight downside in that the die is rather large. Even the regular Athlon64 has a 128-bit wide memory controller and 3 Hypertransport links on-die, it's just that half of the memory controller and two of the HT links are disabled.

All that being said though, the CPU is damn near the bottom of the list of components that cause system instability, regardless of what revision it is or who makes it. Things like RAM and the power supply are the big issues on the hardware front, though drivers are were things really cause problems. Here Intel and AMD are on equal footing except for motherboard drivers. Unfortunately no one has all that great of a reputation for motherboard drivers. Traditionally Intel has been seen as the best, but it was more of a "least bad" if you asked me. VIA had all kinds of troubles and ALi was worse. But things are a bit different now. First, Intel's current chipsets have very mature and stable drivers, and second, nVidia enter the market by making some chipsets for AMD processors, and they have totally raised the bar for everyone else, including Intel. Back in the days of previous crappy Intel drivers (ie the early i8xx drivers or the piss-poor first few releases of the PIIX4 drivers) the competition was mainly VIA and SiS, producing crappy drivers of their own. Intel's current drivers may be stable, but if they fall back on some of their old instability problems then nVidia (and by connection, AMD) could eat their lunch.

Yes, but... by vacaboca · 2003-09-23 05:39 · Score: 4, Funny

...how does it benchmark against the Commodore 64???

this is great but... by chef_raekwon · 2003-09-23 05:39 · Score: 2, Interesting

when will we see some serious APPS that utilize this technology?? why would joe user go buy one? to check their mail?

although not intended to be a troll, its looking that way....
if the OS, AND the Apps run 64bit - i'll buy one...till then, i'll stick with my original thunderbird, 1.4ghz.

--
We're like rats, in some experiment! -- George Costanza

Re:this is great but... by Anonymous Coward · 2003-09-23 05:41 · Score: 2, Insightful

Once Windows for 64bit AMD comes out (soon), I'm sure you'll start seeing quite a few more apps. We're in the transition phase now, where we'll buy 64-bit chips that can still work well in a 32-bit world. It's not an upgrade for Joe Homeuser quite yet, but give it a year.
Re:this is great but... by Trigun · 2003-09-23 05:42 · Score: 3, Insightful

No, a troll would be saying that you're going to buy some eggs to fry on this processor, or even better some grits to pour on Natalie Portman.

And the end user won't realize a big difference, and the bang-for-buck ratio won't be there either. But serious database apps, cad, and any other high-end market will most definitely benefit.
Re:this is great but... by cr_nucleus · 2003-09-23 05:54 · Score: 3, Funny

why would joe user go buy one? to check their mail?

You're right, it's useless. Any advance in processor technology should be stopped at once, we can already mails after all.
Re:this is great but... by Tumbleweed · 2003-09-23 05:57 · Score: 3, Insightful

Just because checking mail is all you use YOUR computer for, doesn't mean that's all everyone else uses their computers for. There are already plenty of apps around that will suck up this much processing power and still beg for more, and they're not as obscure as you'd think. I have a friend who does a lot of 3D animation, and rendering will ALWAYS leave you wanting more CPU power and RAM. Enough will never be enough for stuff like that. And when you want to rip your CD collection to mp3 (or ogg, or whatever), you're gonna want the fastest thing around, if you've got a decent collection. DIVX-encoding is pretty nasty on the CPU, too. These are all fairly common tasks now.
Re:this is great but... by NitroPye · 2003-09-23 06:00 · Score: 2, Interesting

A company I previoulsy worked for sells nothing but Opterons for workstations and servers. It is amazing what dual 64bit chips can do on a properly optimized workstation. It is quite impressive. We used suse 64bit version and gentoo linux on those machines
Re:this is great but... by snatchitup · 2003-09-23 06:01 · Score: 2, Funny

ignoramous! myopic! moron!

I write apps, that people use to read email, and I write these apps on the J2EE platform. Have you ever tried writing a 3 to 4 tier'd app? all running on one development machine running NT? Including the Oracle server? .... or even better...
I write apps, that programmers use to write apps for people to read their email. You can bet I need all the horsepower I can get!. .... or even better...
I write Virtual Machines that code jokey's use to develop tools to confuse, er uh, empower other developers to create tools for other people to write email readers so they can get their work done. I need more than just a little horsepower, I need at least a native 64bit processor and OS.
Re:this is great but... by Jeff+DeMaagd · 2003-09-23 06:04 · Score: 3, Interesting

I don't get this. Aren't you missing the point? The idea is that it is supposed to be forward compatible. Simply because there were no 32 bit apps at the time didn't mean that people didn't buy 386s and 486s.

If you are waiting for everything to catch up, then there is currently little advantage to x86-64.

Most apps don't need this power, the ones that do will be rewritten as the need arises, everything else can still run in 32 bit mode.
Re:this is great but... by Smallpond · 2003-09-23 06:14 · Score: 2, Insightful

I'm looking at a 1TB disk array. Since that's more than 2B blocks, I can't store the disk block address in a signed 32-bit number any more.

Barring that, what do I have that cries out for 64 bit arithmetic? Not much.

Also, try compiling the same app in 32-bit and 64-bit modes. The 64-bit app is a lot bigger and slower, since all the pointers doubled in size, so less code fits in cache, and I'm using more memory bandwidth.

The 16 to 32-bit conversion was forced, because it didn't take much of a problem to overflow a 16-bit number. The need for 64-bit hardware is a little less obvious.

Add in power and cost and I'm guessing that 32-bit hardware will continue to be sold for quite a while.
Re:this is great but... by secondsun · 2003-09-23 06:24 · Score: 2, Insightful

when will we see some serious APPS...
Add -m64 to you march flags then emerge -e world

--
There is nothing wrong with being gay. It's getting caught where the trouble lies.
Re:this is great but... by Tumbleweed · 2003-09-23 06:31 · Score: 2, Insightful

> 3D animation, sure - but are there 64 bit apps for it? or still 32bit apps? understandibly, i wouldn't give a rats ass if my mail was checking at 32bit, but, why would i spend the money if my 'money' app was still 32 bit aswell?

Apparently you haven't checked any of the benchmarks out there yet, or you'd know that 64-bit apps aren't even _remotely_ necessary to get great performance out of 32-bit apps on the Athlon64 (or PPC970 aka G5). Despite being '64-bit' chips, they still run 32-bit apps noticeably faster than their previous 32-bit counterparts. Yes, you'll undoubtedly see even more improvement once they're optimized for 64-bit OS & apps, but that's hardly necessary to see great improvements in performance right off the bat. I think that's quite an achievement, actually.

Also note the extreme price premium the FX & EE chips command puts them out of reach of the vast majority of people, and that noone's forcing you to upgrade. It's better to have the option and not need it, than need it and not have it.

The option _I'd_ like to have is enough money to get a G5 system. *sigh*

I've also seen people saying, "Yeah, but wait for Prescott." Uhm, 105W heat dissipation? NO THANKS.
Re:this is great but... by lightsaber1 · 2003-09-23 06:39 · Score: 2, Insightful

The 64-bit app is a lot bigger and slower, since all the pointers doubled in size, so less code fits in cache, and I'm using more memory bandwidth.
Of course, in a situation where you might want more than 4 GB of RAM (not me, but a good number of people), the 1 TERABYTE of RAM cap on the 64-bit machine sounds pretty enticing. I'm not really sure how allowing more RAM, and upping the on-chip cache, moving the DDR RAM controller to the cpu die, and optimizing the instruction set will cause things to move slower, even if you have a higher storage/instruction requirement.
Space is cheap. Time is not. Space gets cheaper and cheaper. Time is worth more and more every day.
I only wonder why they had to do this today...I look silly drooling on my keyboard at work.
Re:this is great but... by Slime-dogg · 2003-09-23 07:02 · Score: 3, Insightful

I'm not sure that you really understand the meaning of the word "Troll." Check here for a good definition of "Troll." The word came from "Trolling for n00bs," where people would purposely get something wrong in order to get all the new people to jump on them. A good troll is really quite amusing, and very difficult to pinpoint... such as this comment's parent's post, I would guess. ;-) Good trolls usually do not include Natalie Portman, Grits, or AYB. Those posts would generally be modded "Offtopic," which is a euphamism for "Useless."

What is interesting is that the grandparent thought they were posting a Troll by posting a valid thought held by many people. If there are no apps, then why would I buy the processor?

In answer to that question, I would propose the analogy: Would you wait until the flood hit to get sandbags? It's always good to have the sandbags on hand, they don't get in the way, nor do they cause you to live life in a different way. The Athlon64 is like that, a latent 64 bit platform that doesn't hurt to have, and even gives incredible performance gains in 32 bit apps.

--
You need to restart your computer. Hold down the Power button for several seconds or press the Restart button.

Will wait by Brahmastra · 2003-09-23 05:40 · Score: 2, Interesting

I'll wait until they actually come out with applications that use the 64-bit extensions to the x86 instruction set. Doesn't seem like its worth it to pay big bucks and buy this for 32-bit applications

Re:Will wait by Slack3r78 · 2003-09-23 06:08 · Score: 2, Interesting

But if you're in the market for a high end machine, this seems to be the way to go right now. The benchmarks I'm seeing show it outperforming the P4 3.2GHz chip, and from the pricing I've found so far, it's nearly $200 cheaper.

Mmm yesss by mesmartyoudumb · 2003-09-23 05:41 · Score: 3, Funny

Begun these processor wars have!

--
"Comedy's a dead art form. Now tragedy, that's funny."

Indeed a fast one up on AMD by tanveer1979 · 2003-09-23 05:41 · Score: 3, Interesting

Regardless, the performance levels for AMD's new flagship look very strong.

Though I am a die hard AMD supporter, i have to admit Intel has really pulled one up on AMD this time. The 64bit 3200+ is just about 15-20% faster than the stock and barrel Intel 32bit 3.2 GHz. Bad news for AMD this is, considering the retail price of these babies is 450 & 800$ (Normal and FX).

And BTW windows released XP 64bit Beta1 today.

--
My Aurora : http://www.youtube.com/watch?v=o91ZsGwJYyg
FB : https://www.facebook.com/TanveersPhotography

Re:Indeed a fast one up on AMD by digidave · 2003-09-23 05:50 · Score: 4, Insightful

Remember when the P4 came out and it was slower than the P3 at the same clock speed? Like P4 was then, the Athlon64 is designed with the future in mind, not blowing away everybody on day one and having no room to scale from then on.

--
The global economy is a great thing until you feel it locally.
Re:Indeed a fast one up on AMD by Anonymous Coward · 2003-09-23 05:52 · Score: 3, Funny

And BTW windows released XP 64bit Beta1 today.

*shudder*

Don't speak like that again. :)
Re:Indeed a fast one up on AMD by Zathrus · 2003-09-23 06:17 · Score: 5, Insightful

Though I am a die hard AMD supporter, i have to admit Intel has really pulled one up on AMD this time. The 64bit 3200+ is just about 15-20% faster than the stock and barrel Intel 32bit 3.2 GHz. Bad news for AMD this is, considering the retail price of these babies is 450 & 800$ (Normal and FX).

Yeah. I can't believe that you can get a brand new CPU that's 15-20% faster than the previous champion for only 25% less! Oh, and that's in 32-bit mode. If 64-bit computing takes hold then Intel is SOL at the moment - despite the rumors flying around about Prescott and 64-bit instructions.

Yeah, the Athlon64 3200+ is $465 (note, look at the retail price w/ heatsink to be fair), but the P4 3.2GHz is $619.

The Athlon64 FX-51 is certainly overpriced, given how miniscule the performance differences are, but that's hardly a surprise.

What Intel did pull wasn't a price/performance coup (because it isn't, by any means) but a paper launch debacle. Every single review I've seen thus far includes benchmarks for the P4 3.2EE -- which isn't available until November, and at prices similar to the Ath64-FX (based on preliminary 1000 CPU lot prices). The P4EE is competitive with the Ath64, but it's a smokescreen. You can buy an Athlon64 right now, but the P4EE is non-existant.

And BTW windows released XP 64bit Beta1 today.

Unfortunately it appears that there's a severe lack in available drivers. All the sites are having to use nVidia 5900FX's and even then can't bench any DX9 apps in Win64 due to no 64-bit DX9 being available.
Re:Indeed a fast one up on AMD by Slime-dogg · 2003-09-23 07:10 · Score: 2, Interesting

Take a minute and look at the platform that the applications were compiled for. Only the OS was compiled for the 64 Bit chips, and even that was a pre-beta. Most modern apps released allow for updates that include specifically compiled code (ala Windows, as much as I hate to say it). Once this is taken advantage of, then you can be sure that you have a larger performance delta.

The other thing that comes to mind is that these chips are brand new, and are not running at the speed for which they were planned. Adding 2 stages to the pipeline should allow AMD to jack up the Mhz level, yet they are releasing something that is just a little past the matured silicon of the Athlon XP. I would expect these numbers to rise once production takes off, and the batches have higher yields.

--
You need to restart your computer. Hold down the Power button for several seconds or press the Restart button.
Re:Indeed a fast one up on AMD by NovaX · 2003-09-23 08:21 · Score: 2, Informative

Not really. The P4 had a longer pipeline for the future, and when SSE2 was used, it had similar performance clock for clock. The K8 doesn't have such a design. Instead its an Athlon with a longer pipeline so they can ramp it, but with improved algorithms so the IPC is better (meaning it does roughly the same amount of work per clock as an athlon, but can ramp higher). With the 64bit mode, they have extra registers so there is a performance boost, and the memory controller is a nice add-on too.

The only "future in mind" would be through compilers making use of the 64bit mode more efficently. However, Intel has the same with the P4 and IA-64, as well as they actually build them unlike AMD. So, your statement is simply wrong.

AMD was supposed to release this chip months ago, but they screwed up trying to go .13 and SOI at the same time, causing major delays. They're just lucky Intel only caught up and they're still quite competitive.

--

"Open Source?" - Press any key to continue

Benchmarks by I_am_Rambi · 2003-09-23 05:42 · Score: 4, Informative

Here are some more benchmarks

AMDzone
AnandTech
XbitLabs
Ace Hardware

There are even more at AMDZones main page.

Paper launch? by dzym · 2003-09-23 05:43 · Score: 3, Interesting

I'm seriously hoping that this isn't a paper launch. There are as yet many factors that could limit the potential for this powerful new product, and a paper launch would significantly weaken AMD's thrust.

First, we still don't have a mass market consumer OS for native x86-64, and even when Windows XP for x86-64 does come out (the word from the betas is that it's very very good and solid), we still need to wait for x86-64 compatible drivers to be developed and released by the various manufacturers, and it would be no small feat to even have a small core sampling of drivers available by say, summer 2004. Personally, I'm hoping that Nforce3 and ATI Catalyst drivers will be ready very early on.

Mad props to AMD, but they're not out of the woods yet on this release.

Re:Paper launch? by SixDimensionalArray · 2003-09-23 05:54 · Score: 2, Informative

With regards to the OS, here's just a little more interesting plot info. I just noticed that Internet.com has this article which mentions that Microsoft launched the official beta of Windows XP 64bit as a "precursor" to AMD's launch today.

AMD! Don't cater to the evil Redmond mole-men! ;)
Re:Paper launch? by Jungle+guy · 2003-09-23 05:56 · Score: 2, Informative

If this is a paper launch, Alienware is selling some interesting paper.
Re:Paper launch? by p3d0 · 2003-09-23 05:57 · Score: 2, Insightful

First, we still don't have a mass market consumer OS for native x86-64...
That's the beauty of it. You don't need one. Normal IA32 Windows runs as fast on AMD64 as on a Pentium.

--
Patrick Doyle
I mod down every jackass who puts his moderation policy in his sig. Oh, wait a sec....
Re:Paper launch? by randombit · 2003-09-23 08:48 · Score: 2, Informative

I'm seriously hoping that this isn't a paper launch.

Why don't you go go buy one from here or here and let us know?

First, we still don't have a mass market consumer OS for native x86-64

Very true. And some of us just don't care. I wanted benchmarks for GCC and OpenSSL and Postgres, and all they showed was 32-bit Windows apps, which nobody but nobody cares about. Well, life is tough, I guess.

Wow! by Anonymous Coward · 2003-09-23 05:43 · Score: 2, Insightful

Finally, a 64-bit desktop processor!

Oh wait it's been done!

But seriously, there are a number of reasons why this is a great chip.

Namely, if Intel actually thinks putting a cache that's too big on a chip is actually going to help, good luck.

A Toast to AMD for once again making the superior product.

Oh no! by gpinzone · 2003-09-23 05:44 · Score: 2, Funny

A site with the name "HOT" Hardware had to review an AMD chip. Why not just post a Slashdot article asking for AMD overheating jokes? Open the floodgates!

Spread arund the /.ing. by freidog · 2003-09-23 05:45 · Score: 2, Informative

HardOCP
Tom's hardware
Ace's Hardware
As you would expect, no chip is dominate. though the more interesting thing for me than the nip and tuck between $800 CPUs, is the Athlon64 3200+ performs right there with the 3.2C in mosts lets. Not bad considering it retails for more than $100 less.

A happy coincidence by RealAlaskan · 2003-09-23 05:47 · Score: 5, Informative

By a happy coincidence, I just today got my copies of the x86-64 programmer's manuals. There are five volumes:
1) Application Programming
2) Ssytems Programming
3) General Purpose and System Instructions
4) 128-Bit Media Instructions
5) 64-Bit Media and x87 Floating-Point Instructions

Get them here.

Then go make your favorite compiler or windowing system work better on this.

--
See what I've been reading.

What's up with AMD's model names lately? by Jugalator · 2003-09-23 05:48 · Score: 4, Interesting

We saw Athlon XP get released when Windows XP was still "hot" and aggressively marketed by Microsoft.

We now see Athlon FX get released when the GeForce FX graphics card series is the state-of-the-art among hardcore gamers, along with ATI's Radeon series. Hardcore gamers are also coincidentally a target group for AMD's processors since they're known to look for the latest and greatest processor-wise.

I wonder if this is just coincidental, or if AMD is actually using the popularity of other brands to market their own? Are they even using dirty tacticts to try to fool people into thinking "Ooh, this Athlon XP should work especially well together with this newly released Windows XP then, right?" and "Oooh, great, I must have the Athlon FX for my latest video card!"

But perhaps they just happen to choose the same abbreviation as other popular brands at the time for the second time in a row. However, I still can't say AMD's Athlon marketing smell good to me at least.

--
Beware: In C++, your friends can see your privates!

Re:What's up with AMD's model names lately? by Tumbleweed · 2003-09-23 05:51 · Score: 2, Funny

Easy - the Athlon 64 FX 51 is 5.1GHz.

Right? :)
Re:What's up with AMD's model names lately? by Tumbleweed · 2003-09-23 06:02 · Score: 5, Insightful

I couldn't care less about their silly-ass naming schemes - I'm much more irritated with their ridiculous lack of planning on socket formats & memory controllers. Single channel on that kind of processor (the non-FX) - what were they smoking? Maybe if they want to call that the new Duron, but otherwise, that's just idiotic. They should also have increased (doubled!) the L1.

It's not like any of this _really_ concerns me, mind, as I'm now saving up for a G5. It's time to 'Switch'! :)

I'd say with CPU prices like that AMD & Intel are now charging, the whole 'PCs are cheaper than Macs' theory is gonna get a lot less credence.

I'm _really_ interested in seeing some real & comprehensive benchmarks between all the processors, now that they're all out. G5 vs P4 vs P4EE vs Xeon vs Athlon 64/FX.
Re:What's up with AMD's model names lately? by gl4ss · 2003-09-23 06:21 · Score: 2, Insightful

well, it's still better than intels EXTREME EDITION.

i
-

--
world was created 5 seconds before this post as it is.

Doesn't look like it. by eddy · 2003-09-23 05:49 · Score: 2, Informative

There's a limitied supply, but they're available. Here's one: Athlon64 3200+ bundled with a MSI K8T Neo-FIS2R, in stock and available, for "only" 6600SEK (~$835 /w tax)

I'm salivating, but the limited supply is going to keep the price way, way up for a time.

--
Belief is the currency of delusion.

Come on now by MalleusEBHC · 2003-09-23 05:50 · Score: 5, Interesting

I would hope that Tom's Hardware would at least try to keep the anti-Mac bias down to a tolerable level.

"At the same time, Apple laid claim that with the G5 model, it would offer the world's most powerful desktop system. Apparently there are users who will believe these kinds of claims. Whatever - at least the G5 also has 64-bit support with regard to the software. Nevertheless, there is still no final operating system available for it."

FUD, lack of evidence, and outright lies - they call this journalism? They dismiss the Apple's claims about the G5's performance without a) including it in their later benchmark or b) citing any references. Why am I suppose to believe someone who is trying to put down a group of users like they are petulent 2 year olds. Also, what is the crack about not having a "final operating system" out yet? 10.2.7 is a fine OS for the G5. If he means that there is no 64-bit OS, why not just come out and say it?

Pfff. I can't believe I took time away from constantly reloading my Fedex tracking page to read that drivel.

PS - FEDEX, BRING ME MY DAMN G5 ALREADY!!

Re:Come on now by TheNetAvenger · 2003-09-23 06:53 · Score: 5, Insightful

FUD, lack of evidence, and outright lies - they call this journalism? They dismiss the Apple's claims about the G5's performance without a) including it in their later benchmark or b) citing any references.

Because sane, savvy people in the technology industry should know better.

We have had 64bit AMD systems running here before the Apple G5 announcement, additionally, we have had Itanium with Windows XP 64 edition running here for over a year.

Considering we ACTUALLY have these 'shipping' systems already in our office and labs for quite some time, do we need a reference from the article's author to know that Apple was lying out their ass?

Where in the hell have you been?

Apple was NOT the first 64bit desktop PC, their performance numbers were 'admittedly' pulled from a comparison of slower 32bit Xeon CPUs, and only showed the specific few tests that the G5 actually outperformed even the older Xeon chips.

And the last nail in this Apple shenanigan is that Mac OSX (even the new release for the G5s) is NOT a 64bit OS, and has no plans to be a 64bit OS in the near future.

I don't dislike Apple or their products, but their marketing department needs to be whacked up side the head. Instead of billing the G5 for what it IS and its TRUE good points, they go out on this hyperbole that is false and make a fool of themselves.

I'm sorry you and other people buy into it. If you are so worried about facts and citing references, why haven't you checked the facts that Apple has been purporting? You would have found they are false which is what the rest of the industry already knew.
Re:Come on now by WiseWeasel · 2003-09-23 07:45 · Score: 5, Insightful

You claim to have been using 64 bit x86 CPUs for over a year (Itanium) as well as from AMD (Opteron), but what you fail to realize is that Apple is claiming first 64 bit DESKTOP. The Itanium and Opteron chips are not desktop CPUs, but server and workstation chips. This is just silly semantics, and anyone who believes marketing drivel should have their head examined. But to simply dismiss a whole platform for these stupid marketing claims is ridiculous. The 1.6 and 1.8 GHz G5s have been in the marketplace with a final version of MacOS X 10.2.7 for over a month, making them the first viable 64 bit DESKTOP solution available in my book, since they run a final, stable version of MacOS X which is as tuned as it needs to be for 64 bit operation (mostly 42 bit memory address space support and a few 64 bit math libs), and a good consumer user experience. Someone trying to use Windows XP 64 right now had better be an expert user, to deal with all the problems with missing drivers. If you want to argue that the opteron was a desktop CPU or had desktop solutions (since it can run Win32), that might be a valid argument, but you can't fault Apple's marketing department for trying to cash in on the 64bit buzz, especially since most people who watch those commercials have never heard of an opteron.
As for benchmarks, it's silly to omit the G5s from the comparison for religious reasons, and people want to see how they stack up. The only way to counter Apple's marketing drivel is to do actual real-world benchmarking using cross platform apps and benches. The 'slow' Xeon chips Apple compared the G5 to were the fastest available at the time from Intel. I don't know why they didn't compare themselves to the Opterons, but them's the breaks. What we need now is for all these PC tech sites to get some dual G5s and compare them to what they think is the top of the line on the x86 platform. FUD and hostility will only make people see these sources as biased and unprofessional. Objective comparisons between G5s, Athlon64 FX and P4 EE will drive lots of page hits, since people are starved for this info right now (due in part to Apple's excessive marketing claims). So, either put up or shut up, and be happy that Apple's marketing claims drove so much interest in G5 comparisons w/ x86 offerings.

--
"I like systems, their application excepted", George Sand (French)

Re:What about supporting hardware? by CausticWindow · 2003-09-23 05:52 · Score: 5, Funny

Have they even created one yet?

Actually, no.

If you take a closer look at the pictures in the article, you will see that they've attached a wire to each and every pin of the processor. It's all a loose mess of wires and duct tape. If you want an Athlon 64, you'll probably have to do the same yourself, because there aren't scheduled any motherboards for it before sometime after christmas.

--
How small a thought it takes to fill a whole life

Implementation or spec? by Realistic_Dragon · 2003-09-23 05:53 · Score: 2, Interesting

A physical address space that can support systems with up to one terabyte of installed RAM, shattering the 4 gigabyte RAM barrier present on all current x86 implementations.

If you implement x86 with PAE then you get 8gb.

It's only 4gb if you exclude dirty hacks - but if you did that then you would have to rule out x86 in it's entirety.

--
Beep beep.

Re:Implementation or spec? by Zathrus · 2003-09-23 06:25 · Score: 2, Informative

If you implement x86 with PAE then you get 8gb.

You also get horrible performance and require different compile flags for the application.

Compile something for x86-64 and you get 48-bit addressing out of the box with no performance hits and no special flags.
Re:Implementation or spec? by afidel · 2003-09-23 06:59 · Score: 2, Informative

No, PAE does not allow a single program to access more than 4GB, only the OS. So while they were stupid to add the physical portion to the address space tag they were basically correct. How many people that run applications that need 4GB or more of ram run more than one at a time? Basically anyone who needs more than 4GB needs it available to just one app so PAE is worthless in most cases.

--
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.

the benchmark I want to see by asv108 · 2003-09-23 05:54 · Score: 4, Interesting

Tom, if your listening this is want we want to see:

A benchmark of Linux on the following systems:

2 GHZ Dual G5
2 GHZ Dual Opteron
Athlon 64 FX
Pentium 4 3.2 Spicy Edition

GCC Settings for each system should be optimized for the best possible stable performance. I'm so sick of seeing 32 bit windows benchmarks for testing 64bit processors.

It's out. by BoomerSooner · 2003-09-23 05:59 · Score: 2, Informative

You can get Windows 2003 Server and Windows XP 64bit 2003 Edition from the MSDN Website (assuming you have an account).

Re:It's out. by BoomerSooner · 2003-09-23 06:26 · Score: 2, Informative

Windows XP 64-Bit Edition For 64-bit Extended Systems (English)
Locate in Contents

Date Posted
9/18/2003 1:38:00 PM
File
srv03sp1_usa_1069_amd64fre_pro.iso, ISO-9660 CD Image
Size
419.32 MB
Minimum Estimated Download Times
T1 37 minutes
128 KB 7 hours, 38 minutes
64 KB 17 hours, 11 minutes
28.8 KB 38 hours, 12 minutes

Description
Product key: XXXXX-XXXXX-XXXXX-XXXXX-XXXXX

Instructions
An ISO-9660 image file is an exact representation of a CD, including the content and the logical format. The most common use of an image file is to write it to a blank CD-R resulting in an identical copy of the original CD including file name and volume label information. ISO Images can also be extracted directly to a file location by many CD-R utilities. For more information about image files, please visit the MSDN FAQ.

Competition by Stalyx · 2003-09-23 05:59 · Score: 2, Interesting

Although the numbers look impressive, I personally would wait till their are more 64 bit applications in the market. If you are a gamer there would be modest increases in performance but really waiting a few more months will not hurt if you are more interested in value.

I love the fact that AMD is competing, this bodes well for all computer users. A technology war will definitely keep prices down. In the same sense, I am hoping that Intel comes out with a "decent" 64 bit chip. Not that I am fond of Intel, I prefer AMD but still the fact remains that whenever their is conflict we see the most marked increases in technology.

Another interesting read from Tom's Hardware by deltagreen · 2003-09-23 06:01 · Score: 3, Interesting

Tom's hardware has got another article, called The Intel v. AMD Performance War: You Lose , about the more cynical, money-making sides of the launch. Perhaps it's a bit conspiratory, but certainly worth a read, as it raises many valid points.

The biggest benefit... by chill · 2003-09-23 06:01 · Score: 5, Insightful

...of both the new AMD-64 and the Pentium 4 Extreme is that the prices of the older chips should start dropping like a stone.

--
Learning HOW to think is more important than learning WHAT to think.

Threat to Athlon64: Prescott (not Pentium 4) by reporter · 2003-09-23 06:02 · Score: 4, Insightful

Unfortunately for AMD, the upcoming Prescott by Intel could kill the Athlon64 -- and the UltraSPARC III. Please read "Prescott has 64-bit compatibility built in".

AMD created the 3DNow! extensions to the 80x86 instruction set architecture (ISA), also known as IA-32. They were a significant improvement over the original set of MMX extensions. However, later, Intel created the SSE (and SSE2) extensions. Guess what? AMD was forced to incorporate them into its future chips in addition to the 3DNow! extensions. Ignoring the SSE extensions would have cost AMD dearly in terms of marketshare. The fact of the matter is that Intel sets the global standard for the IA-32 ISA.

Now, AMD has created its own x86-64 extensions to the IA-32. You can be sure that Intel has created a different set of 64-bit extensions (which we shall call "INTEL-64") to the IA-32. After all, why would Intel support AMD in any way? Once Intel activates the INTEL-64 extensions in the upcoming Prescott, AMD will be forced to go back to the drawing board to incorporate the INTEL-64 into all future chips. The current Athlon64 will be like the K-5 -- interesting but without a future.

AMD will probably take an additional 2 years to produce an INTEL-64-compatiable chip. By that time, Intel would have locked 90% of the 64-bit desktop market with Prescott.

The worst news is for Sun. With Prescott, Intel has a 64 bit chip that will be significantly faster than the UltraSPARC III/IV. Right now, the Pentium 4 crushes the UltraSPARC III in performance. Please review the performance characteristics of the Pentium 4 at the SPEC web site. Since Prescott (successor to the Pentium 4) will be faster than its predecessor, Prescott will clean UltraSPARC's clock. Moreover, the number of applications that will run on Prescott -- the heir to the software empire of the x86 -- far exceeds the number of applications that run on UltraSPARC III/IV. On the key TPC-C benchmark, Prescott will clearly deliver outstanding performance, compared to the UltraSPARC III/IV.

In short, when Intel activates the INTEL-64 extensions in Prescott, Intel will force (1) AMD back to its usual state of borderline bankruptcy and (2) Sun into being a software company.

... from the desk of the reporter

Re:Threat to Athlon64: Prescott (not Pentium 4) by Bo+Diddly+Squat · 2003-09-23 06:32 · Score: 5, Insightful

While it's likely that Intel has some form of 64 bit instructions built into Prescott, it is highly unlikely that they will enable it. They would have to lose much ground to the Athlon64 and Opteron systems to do that.
Remember that they are betting on the Itanium as their 64 bit processor. Enabling the 64 bit instructions on Prescott will almost instantly kill the Itanium and I doubt Intel wants to do that.

As for 64 bit Prescott killing Sun. I don't see any reason for that. It seems to be fashionable to predict the death of Sun these days, but they are not in that much problems. The Itanium already crushes UltraSparc for performance, but Sun is still doing fine. A 64 bit Prescott will not change that. The software base difference doesn't really matter here either as there is simply stuff that runs on UltraSparc but not on x86 systems. Also, a processor is not the only part of a computer. While UltraSparc's performance may be lagging at the moment, Sun's systems are very well designed and no x86 system can compete with their high end systems.

Hell, Sgi is in much more trouble than Sun and they still design their own processors (MIPS).
Re:Threat to Athlon64: Prescott (not Pentium 4) by afidel · 2003-09-23 06:33 · Score: 5, Insightful

Actually the 64 bit instructions from Prescott ARE x86-64 which is AMD's extensions. Intel has a liscense for x86-64 and has been working on project Yamhill for some time to integrate the x86-64 instructions into the P4 core architecture. They did this when they Microsoft agreed to make Windows for x86-64 as well as IA64, MS will NOT keep three forks of their codebase so Intel's option was so ceed the low end to AMD or use their liscense, they of course chose the latter since almost noone has installed Itanium systems. Btw Intel chips have been faster than Sparc for some time, people don't buy Sun machines for number crunching they buy them for stability and scalability. Until there is an Intel machine that can scale like the 6800 and is as stable as Sun hardware there will be a niche for Sun. Of course Oracle with it's RAC initiative may reduce the scalability argument.

--
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
Re:Threat to Athlon64: Prescott (not Pentium 4) by dAzED1 · 2003-09-23 06:37 · Score: 4, Interesting

fortunately for SUNW shareholders, Sun released some rather good news yesterday about a breakthrough they had in CPU interconnects. So, the ultrasparc3/4 might be devalued, but Sun was about to devalue them with their own stuff anyway. In addition, the ultraspac3/4 are still very solid chips, and its easier to have an E15k with zillions of them than to try to figure out how to get zillions of p4's to operate in the same space.
Check out the new info posted about sun's interconnect tech here (free reg, blah blah). With the new tech, processors and/or memory can be directly connected to each other.
I still wish my wife had let me buy AMD a couple months ago when it was in the 5's. They're losing money fast, but...since the stock is up 140% from its price in late June, I guess investors still have hopes for them...
Re:Threat to Athlon64: Prescott (not Pentium 4) by Algan · 2003-09-23 06:38 · Score: 2, Insightful

I seriously doubt that. Even if Intel already has those 64bit extensions ready in Prescott, I believe they are compatible with AMD's. I think it is too late for Intel to come out with another incompatible set of 64 bit instructions, since the major operating systems on Intel platform (Windows and Linux) are already supporting AMD's 64 bit instructions. A better move on their part would be to come up with a compatible processor and then rely on brand name, market share and better production efficiency (0.9 microns, larger waffers) to outcompete AMD.

--
If con is the opposite of pro, is Congress the opposite of progress?
Re:Threat to Athlon64: Prescott (not Pentium 4) by pmz · 2003-09-23 06:41 · Score: 4, Insightful

"Right now, the Pentium 4 crushes the UltraSPARC III in performance."

I honestly don't care, because there's more to a computer than SPEC drivel. Quite honestly, I'm suprised you forgot to mention Itanic in your anti-Sun rant, because INTEL-64 will make the multi-billion dollar IA-64 go bye-bye. Have you ever seen several billion dollars seem as if it never existed? Intel may.

BTW, here are the actual competitors for your entertainment:

Itanic 2/Power4/UltraSPARC III
Opteron/PowerPC 970/UltraSPARC IIIi
Pentium 4/Athlon

You see, the people considering the first trio aren't necessarily even considering the second trio, because their needs are different. Fancy that, Mr. Troll.

--
Healthcare article at Kuro5hin
Re:Threat to Athlon64: Prescott (not Pentium 4) by jacksonyee · 2003-09-23 06:44 · Score: 2, Interesting

Your argument that AMD is not the market leader and has to follow Intel's lead has been true in the past, but this particular situation is a bit different. Intel has sunk all of their resources into the Itanium project for 64-bit computing, and although the rumored Yamhill project may still yet come to pass now that Athlon 64's performance leadership is unquestionable, I haven't heard a single mention of anything other than the EPIC architecture for Intel, while AMD has been push x86-64 for years now and letting compiler writers and programmers become used to the instruction set. It takes a good amount of time for programmers to utilize a new instruction set effectively after it comes out (look at MMX when it was first introduced), and even if Intel reveals Yamhill extensions, if they're not x86-64 compatible, it's going to take a while for non-Intel programmers to catch up.

Personally, if Intel does reveal Yamhill, it'll be all the better for the industry. More competition will just mean lower prices for us - the consumers.
Re:Threat to Athlon64: Prescott (not Pentium 4) by ball-lightning · 2003-09-23 08:27 · Score: 2, Insightful

While it's likely that Intel has some form of 64 bit instructions built into Prescott, it is highly unlikely that they will enable it.

Actually, I would dissagree. Remmember way back (at the very least, I haven't encountered such a message in a while) when a program would say, "Requires a 386" and you would try it on a 286 and it would complain? Most of the time today, if you are trying to install a program and your computer does not meet the minimum requirements, you can usually squeek by and still have it run (just slowly). If x86-64 becomes popular and is actually used, anyone with an Intel processor will not be able to use a program taking advantage of the x86-64 registers. While in the beggining there will probably be more than one version, if it becomes popular Intel CPUs will start to have a "compatibility problem" if they don't have the x86-64 extension. Look for them in the Pentium 5.
Re:Threat to Athlon64: Prescott (not Pentium 4) by Short+Circuit · 2003-09-23 08:52 · Score: 2, Interesting

I agree with you in your description of Intel's business practices, and while I see several refuting posts against INTEL-64, I'd like to point out a couple things:

First, if Intel was going to launch a 32-bit compatible processor with their own 64-bit extensions, they really ought to have announced it by now, to get people to put off buying AMD's Opteron and Athlon64.

On the other hand, if Prescott was going to support AMD-64, then they really ought to have announced it by now, to bring back attention over to them. The less press AMD gets, the better. (for them.)

Their strategy could make sense, if they wer planning some sort of massive surprise. If you read the Inquirer's article, you'll notice that Intel has been denying that Prescott is 64-bit.

--
tasks(723) drafts(105) languages(484) examples(29106)
Re:Threat to Athlon64: Prescott (not Pentium 4) by Short+Circuit · 2003-09-23 08:55 · Score: 2, Funny

"Pentium 5"?

No, they'll just call it Manganese, the 25th element.

--
tasks(723) drafts(105) languages(484) examples(29106)
Re:Threat to Athlon64: Prescott (not Pentium 4) by akuma(x86) · 2003-09-23 09:41 · Score: 2, Interesting

The Power4/Itanium/UltraSPARC advantage has always been in their scalability. They have point to point links with gobs of bandwidth to handle many parallel CPUs which are capable of handling large transaction-oriented workloads.

Now that Opteron has Hypertransport -- a highly scalable big bandwidth point to point link protocol, what exactly is the advantage of the other RISC contenders in the high end enterprise space?

A nice instruction set? Sorry, that doesn't cut it anymore. The only thing holding back x86 in the enterprise is the lack of an enterprise class OS. Linux will be dominant here in a few years thanks to large investments made by IBM.

As much as people don't like to admit it. x86 will become the dominant instruction set across all computing sectors with the exception of embedded.

Too big? by BoomerSooner · 2003-09-23 06:05 · Score: 2, Insightful

Did you see the performace gain by the cache you say is too big (Normal 512K P4 versus Extreme 2MB P4)?

The cache is the main reason the G5 Dual 2.0 isn't that much better than a G4 Dual 1.42. If the G5 had 2MB Cache as the G4's do it would perform even better (L2 would be significantly better than L3 as well, but costs increase significantly).

Next time you're trying to run some 3D modeling or any other intensive CPU based software, just realize the cache is significant.

Re:What about supporting hardware? by NitroPye · 2003-09-23 06:07 · Score: 2, Informative

Drivers are not to hard to get on the linux side. There is a x86-64 nvidia driver. As far as other drivers.. all in the kernel baby. Doom3 will run on the processor but it most likely will be a 32bit build. I would not be suprised at all if there was a x86-64 build and even a smp x86-64 build.

Re:What about supporting hardware? by paitre · 2003-09-23 06:07 · Score: 2, Informative

There's boards out from MSI, Gigabyte, Shuttle...
*shakes head*

Chicken and egg by siskbc · 2003-09-23 06:08 · Score: 4, Insightful

Unless they come from the same company, either the 64-bit OS had to come before it had a 64-bit chip to run on, or the 64-bit processor had to come before it had anything to run. Which makes more sense - releasing a chip so the OS makers can finalize their designs, or releasing an os that uses mythical 64-bit opcodes?

I think we can all see the wisdom of releasing the new processor before the new OS.

--

-Looking for a job as a materials chemist or multivariat

Re:Chicken and egg by Wonda · 2003-09-23 06:33 · Score: 2, Insightful

not really, linux for instance was ported to the emulated x86-64 long before the first x86-64 cpu existed :)

Re:What about supporting hardware? by tuffy · 2003-09-23 06:10 · Score: 4, Informative

If you take a closer look at the pictures in the article, you will see that they've attached a wire to each and every pin of the processor. It's all a loose mess of wires and duct tape. If you want an Athlon 64, you'll probably have to do the same yourself, because there aren't scheduled any motherboards for it before sometime after christmas.

This sure looks like a motherboard that supports the Athlon64, shipping same business day.

--

Ita erat quando hic adveni.

Reviews of Mobo and Chipsets by carbona · 2003-09-23 06:13 · Score: 5, Informative

For those of you lucky enough to be already considering what mainboard your new Athlon 64 will be running on, OCWorkBench has been posting reviews in the past month on three motherboards/chipsets:

K8VNXP VIA K8T800
K8NNXP nForce3 150
M1687 K8.

Looking for a Linux/Windows/Mac admin/support tech monkey in the Los Angeles area? Please see my resume.

Re:Linux by TheNetAvenger · 2003-09-23 06:18 · Score: 3, Informative

Microsoft 64 2008?

XP 64bit Edition has been out for a couple of years now for the Itanium processors. Just waking up or just not into the 'computer' thing?

For the AMD (64bit extended) processors, SP1 beta, which includes the AMD 64bit version is already out in most tech labs.

We are running it here even and it is even available to MSDN subscribers.

Considering we haven't even posted a bug for it yet, I would say, that not only is it working well, it will hit the release schedule later this year.

Contiki ported to Athlon by yerricde · 2003-09-23 06:19 · Score: 2, Informative

Athlon 64 can run 99 percent of Commodore 64 software in emulation, including the C=64 version of the Contiki OS. In addition, it can probably run almost all these other ports.

You want native? If the Athlon 64 processor and chipset can boot up as a normal x86, it can run Contiki for PC DOS.

--
Will I retire or break 10K?

Re:Intel is dying.. by Phigs · 2003-09-23 06:20 · Score: 2, Informative

Check Pricegrabber, PCSuperDeals.com is showing 83 units in stock at a price of $454.28 for the Athlon64 3200+

available at newegg by uiucryan · 2003-09-23 06:21 · Score: 3, Informative

It is already available at newegg.
here

HotHardware.com is pretty stupid by Paladine97 · 2003-09-23 06:32 · Score: 2, Insightful

They state on one of the early pages:

Many applications improve performance due to the removal of the 32-bit limitations.

Nice try HotHardware.com. Anybody who is anybody knows that performance does not increase just by increasing register bit size!

tiger has a system for sale now by squarefish · 2003-09-23 06:32 · Score: 2, Informative

here

it looks pretty interesting and it's pretty reasonably for what it is at $1650 including a dvd burner;
AMD Athlon(TM) 64 3200+ PC Processor
3 IEEE 1394 Firewire Ports
512 MB DDR 333 main memory (2 GB Max)
10/100 NIC Port
160 GB Ultra ATA/100 Hard Drive
128 MB 8X AGP ATI Radeon 900 SE Video Card
17" Monitor (1280 x 1024 Res.)
1-Year Limited 24Hr/365 Day Tech Support click here to upgrade the monitor
30-Day Money-Back Guarantee Details
56K Max v.92 Send/Receive Fax Modem
DVD-RW with Ulead VideoStudio 6
DVD Bonus Software: DVD X Copy Xpress
Microsoft Windows XP Home Edition
4 USB 2.0 Ports

--
Creationists are a lot like zombies. Slow, but powerful and numerous. And they all want to eat our brains.

Threat to INTEL-64: Microsoft (not Athlon64) by tugrul · 2003-09-23 06:34 · Score: 4, Insightful

I would take your post seriously if you had mentioned what part you believed Microsoft would play in the world of competing 64-bit extensions.

WXP for the Althon 64 is well on its way, as seen in the linked HotHardware review. Will Microsoft and the driver writing departments at hardware firms put up with a stealth announcement of another set of 64-bit extensions?

Re:this is great but...you're not reading the resu by j-turkey · 2003-09-23 06:34 · Score: 4, Insightful

if the OS, AND the Apps run 64bit - i'll buy one...till then, i'll stick with my original thunderbird, 1.4ghz.

If you actually want the industry (and 64-bit computing as a whole) to move forward -- then go out and buy a chip...if nobody buys these because no OS/Software manufacturer writes 64-bit code, then there will be big trouble for little AMD. The idea was to build a chip that runs 32 and 64 bit code very quickly. What it sounds like you're saying is that you're waiting for Intel to release a 64-bit chip so that everyone optimizes their code for that. I, for one would rather back the company who was innovative and was first to market with a pretty cool product than the biggest guy on the block (er...industry...whatever).

From the benchmarks I've seen -- it does a pretty damn good job at both 32 and 64 bit code (especially for a first release without a good 64-bit clean codebase). It manages to beat the existing 3.2 Ghz P4 for cheaper.

I hear what you're saying about waiting on at least one level though -- I can't remember ever spending $400 on a CPU, but I'm confident that in 6 months, those things will be selling for quite a bit less than they are now...they'll be a good bargain.

Finally, I have a 1.4Ghz Thunderbird as my primary desktop. At this point, it should be called a 1.4Ghz Shitbird. I intend to replace it -- the only questions are: when, and replace with what? I know that if I decide to get an Athlon64-ish chip, I won't be making that decision based on whether or not there are 64-bit apps/OS'es for it. For what you pay, it works great...I'll make my decision based on price:performance ratio on what's available now...and from what I can see, it already shines against a much more mature platform.

--Turkey

--

-Turkey

Re: what about the free games they advertised? by Svet-Am · 2003-09-23 06:37 · Score: 2, Interesting

I watched the live webcast and they said that the intial wave of adopters would be able to register their processor and receive free games.

The microsoft rep also said that there is a site where folks can download the WinXP 64bit BETA.

Where are these things? If I could download WinXP64 (Even in BETA) and get free 64-bit games, I'd be willing to give it a whirl since I need an excuse to upgrade my PC anyway.

--
[move .sig! for great justice, take off every .sig!]

Re:are all the reviews by idiots? by aSiTiC · 2003-09-23 06:38 · Score: 3, Informative

Um...NO. If its a true 64-bit processor, OS, and Application...you'll see FAR more than a 5-15% increase. You'll see an exponential increase. No, not just 100%...true 64 bit everything is more than twice as fast as true 32 bit.

32 bit = 4294967296 possibilities.

64 bit = 18446744073709551616 possibilities.

Your simplification of 64-bit processing is quite astounding. If your simple logic was indeed correct I'm sure Intel, Motorola, AMD, would be creating 128,256,512-bit systems by now and we would be living in a grand age of no disease, no war, and no death...

In fact the arguement for 64-bit processing is much more complicated. Simply put, 64-bit processing allows you to do two main things:

1. Move more data in one clock cycle

2. Perform more complicated instructions in one clock cycle

The first of which is widely accepted to be a good thing and the second is not so widely accepted (due to heretics like Hennessy...jk).

--

Where the Music Matters

I want to see 64bit software compared by bluGill · 2003-09-23 06:39 · Score: 4, Insightful

I read several of the reviews, and all stuck with 32 bit code for the comparition between the Intel P4 and the AMD Athalon 64. Linux runs on the Atahlon in 64 bit mode, wouldn't be hard to compile PovRay and Doom on a 64 bit compiler and see if anything changes. Thats just for an easy test.

Many real world (science?) applications benifit from 64 bit processors, find some (presumably running on UltraSparc, PPC, Alpha, or such) and port them over to see how the 64bit abilities of this chip compares to the other existing chips.

I run open source OSes, and open Source applications. I don't care about 32 bit performance because I'm fairly sure that if I did have an Athalon 64 I wouldn't run 32 bit code very often. I can choose between many chips, compatable instruction sets to me means gcc (or other compiler) has an output for them. 32 bit x86 compatiabily is nice for the few times I have to run something 32 bit (normally in Wine) and that doesn't happen very often.

Athlon 64 Laptop by Experiment+626 · 2003-09-23 07:00 · Score: 4, Interesting

I ran across this article today... apparently someone is already trying to put together a mobile system around AMD's new 64-bit offering.

Re:Athlon 64 Laptop by puppetman · 2003-09-23 09:08 · Score: 2, Informative

There are mobile Athlon-64 CPUs.

From The Inquirer:

"First, off AMD has a mobile Athlon 64 reference design which includes 256MB of PC 2000 memory, the K8T400 chipset, and an ATI M9 graphics card. The mobile chip will be launched in September."

Re:Cost of manuals? by darkstar949 · 2003-09-23 07:16 · Score: 2, Informative

After alittle poking around on Intel's site I found the manuals avalible:
IA-32 Intel Architecture

There is also a good deal of pdf manuals, and I'm sure that more hardcopy manuals would be avalible if I actualy cared to look.

Costco sells it. by semenzato · 2003-09-23 07:26 · Score: 3, Interesting

My friend Boris informed me that Costco is selling one of these.

Great for GC and dynamically typed languages by PissingInTheWind · 2003-09-23 07:27 · Score: 4, Interesting

64 bits words have an advantage I would not have realized before programming compilers for dynamically-typed languages:

In a DT language , you need some way to 'mark' information to say wether it is a number, a pointer, etc. The usual technique is to mark the bit field with a 1 or 2 bit tag at the end.

Also, for garbage collection (ie Mark-And-Sweep) you need to be able to 'mark' the object that needs to stay alive so they are not reclaimed by the gc.

That being said, taking 2 or 3 bits on a 32 bit field is a lot, but it is very interesting to realize that that cost go away on a 64 bit machine.

So, 64 bits = more memory (larger address space) but it also means some techniques becomes much more viable in terms of feasability of implementation, which is a very exciting (for some loose definition of exciting) prospect!

--

A message from the system administrator: 'I've upped my priority. Now up yours.'

Re:are all the reviews by idiots? by Zathrus · 2003-09-23 07:29 · Score: 4, Informative

I don't want to bother going over the implications here...those that know them, do...

Fortunately, you're not one of those.

64-bit computing is not a clear win in and of iteself. If you're merely doing 32-bit calculations then odds are a true 64-bit CPU will be slower, because to do the same work you have to move twice as much data to do it -- you have to fetch a 64-bit operation instead of a 32-bit one, and ditto for data. Presuming that the actual operation takes the exact same amount of time per bit then your 64-bit CPU would be half the speed of the 32-bit CPU because it has to do twice as much work for the exact same result.

Of course, that's the simplistic case and ignores the actual case at hand. AMD's x86-64 is not a pure 64-bit CPU because it doesn't need to be. The x86 ISA is ungodly ugly (which is why nobody implements it anymore -- it may look like x86 ISA on the surface, but there isn't a modern CPU that actually implements the x86 ISA in the CPU core) and x86-64 operators are only ~10-15% longer than traditional x86 operators on average. So your operator length didn't double, and you don't need to double your instruction or combined cache just to maintain parity with old CPUs. Additionally, the current x86-64 spec only implements 48-bit addressing, which should be more than adequate for about a decade or so, so address fetches only increase in size by 50%, not 100%. Still an increase, but one that's fairly acceptable and it's a vast improvement for anytime you have to address anything >4GB in size (which is increasinly common, particularly for databases). On top of that, x86-64 adds 4 general purpose registers, which doubles or more the numper of GPRs available (the "or more" bit depends on the operation you're performing, since some of the GPR's in x86 aren't entirely GP). Yes, all modern systems have register renaming. Doesn't help much when you still only have a handful of registers available to a single process and it has to deal with only those -- the assembly code cannot know about the virtual registers, otherwise they wouldn't be virtual!

As far as the data goes, you don't have to perform 64-bit operations on all the data, nor does all of the data have to be 64-bits wide. Clearly how much impact this has will vary from program to program, but unless you're doing a bunch of operations on long long's (64-bit integers) then you probably won't see much performance difference in raw computing speed. In fact, because of the slightly larger operator size and the increased address size you're more likely to see a performance decrease, albeit a slight one.

Apple's marketing response? by mnemonic_ · 2003-09-23 08:09 · Score: 4, Interesting

I wonder if or how Apple Computer, Inc. will modify its marketing efforts. Will it continue to ignore AMD like they did with G4 ads, neglecting the fact that the "megahertz myth" had existed for years in the PC world in the competition between the Athlon and Pentium processors? Which led to numerous Mac users enlightening us lowly PC peons about how processor clock alone does not determine processor performance, a fact which many of us had been aware of long before the release of the Motorola G4.

It is true that the G5 was the first 64bit desktop computer processor. Now there is a second. Apple should show some G5 vs. Athlon 64 benchmarks, which should be a much more competitive comparison price-performance wise than one dealing with G5's and Xeons. And much more realistic, with both catering to (roughly) the same market.

AT and THG compared ! by pyrros · 2003-09-23 08:39 · Score: 5, Funny

>And Anandtech has a good article up, as well.

Ha! good article my ass.

The anandtech article runs a measly 18 pages, while tom's runs 53. So it is clear that the THG article is 194% better than the anandtech one (see fig.1).

55| __ 50| || | || 40| || | || 30| || | || __ 20| || || \--------- . THG. A/T

What's more, in our second test, "pretty pictures on the first page of the article" the beleaguered news site falls even further behind. While THG has 4 pictures on the first page, including one of the athlon XP (oh shiny!) anandtech has none. This could be due to a browser incompatibility or a hyper-active web-filter but we couldn't be bothered to check. (see fig.2)

A/T | | THG |================] \----------------- 0 1 .. 2 .. 3 .. 4

As we can see, the THG article has !DIVISION_BY_ZERO! times the images of the anandech one, and so it must be much better.

[I am quite surprised that this post passes the lameness filter, considering the amount of ugly ASCII art. The fact that ./ kills whitespace in ecode is annoying though ]

Re:AT and THG compared ! by placeclicker · 2003-09-23 09:15 · Score: 3, Funny

./
Talk about one of the worst spelling mistakes possible... :)

--

Browse at -1, because trolls are often the most creative part of /.
Re:AT and THG compared ! by pyrros · 2003-09-23 09:45 · Score: 2, Funny

And to think that I actually googled for beleaguered to get the spelling right ;-)

You're forgetting the 8 new GPRs by kylef · 2003-09-23 11:01 · Score: 2, Informative

Also, try compiling the same app in 32-bit and 64-bit modes. The 64-bit app is a lot bigger and slower, since all the pointers doubled in size, so less code fits in cache, and I'm using more memory bandwidth.

Actually, this is only partly true. Because the new AMD64 instruction set includes 8 more general purpose registers, compilers now generate far less load/store code during periods of register contention. This alone ALMOST makes up for the extra byte per instruction that 64-bit instructions require.

In other words, full 64-bit AMD64 instructions require slightly more bits per instruction than their IA32 equivalents to encode. But at the same time the TOTAL number of instructions in a given program has decreased because of less register pressure (which produces load/store code). These two competing factors tend to offset each other, although the presentation I saw indicated that the total code size has increased slightly (5% or so, IIRC).

You are correct in pointing out that the instruction cache will now be less effective because bytes per instruction have increased slightly. I assume instruction cache sizes will need to increase to compensate. But instruction memory bandwidth isn't really a huge issue anymore because instruction caches do such a good job of prefetching (and code accesses tend to be highly local, unlike data accesses).

Stupid comment by THG was Re:AT and THG compared ! by subsolar2 · 2003-09-23 13:00 · Score: 4, Informative

On
this page you will see the following stupid comment...

Companies such as ID Software, who are responsible for titles like Quake3 and Unreal Tournament 2003, are not ready to jump onto the 64-bit bandwagon.

Two dumb things about this...

1. UT2003 is made by Epic

2. UT2003 was a premier product that AMD was showing off as a 64bit application running under SUSE linux on the Opteron when it was released.

Two big mistakes ... but I guess if it's not running on Microsoft(tm) Windows(tm) it does not exist, even if 64bit Windows(tm) is not available yet.

102 of 481 comments (clear)