AMD Dual-Core Performance Revealed

Timmus writes "In two separate articles, FiringSquad takes a look at the performance of AMD's dual-core Opteron CPU. The first article examines the performance of dual-core in scientific computing applications (MATLAB and LS-DYNA) as well as digital photography, while the second story focuses on the performance of dual-core Opteron paired against Intel's dual-core Pentium Extreme Edition in video encoding, Cinebench, and a few other applications. The performance improvements are pretty impressive in multi-threaded applications that take advantage of the technology."

Comments

YESSSS

[sehryan]

I am running one right now, which is why I got first post!

Re:YESSSS

[antifoidulus]

Unfortunately, it looks like the GNAA has a beowulf cluster of these and beat you to it.....

full article mirrors

[winkydink]

here and here.

OK then.

[millennial]

So we have:
scientific computing applications (MATLAB and LS-DYNA)
digital photography
video encoding
Cinebench and
"a few other applications".
So what about the average user? Will the college kid who just needs to type their papers, the parents who want to do their taxes, the gamers who want to play high-end stuff, etc. get any sort of boost from this?

Re:OK then.

[antifoidulus]

Why would they need a "boost"? These are expensive and obviously aimed at high end users. You can already get sub $1k laptops that really do all the stuff you described, so why would they buy a dual core desktop system?
If you are using a dual core system to run word either a) you have WAY too much money, or b) the code bloat at Microsoft has REALLY gotten out of hand......

Re:OK then.

[millermj]

The people who want to do all of that at once, maybe? Honestly, ever tried to MD5SUM your CD1 ISO at the same time as you were encoding your MP3s for CD2? Dual-core processors would make multitasking much smoother.

Re:OK then.

[Anonymous Coward]

So what about the average user? Will the college kid who just needs to type their papers, the parents who want to do their taxes, the gamers who want to play high-end stuff, etc. get any sort of boost from this?

Here's a clue: The Bottleneck Ain't The Processor.

Re:OK then.

[nmg196]

> the code bloat at Microsoft has REALLY gotten out of hand......

I wish people would stop talking about Microsoft code bloat when nobody else does any better.

Currently, 50 processes. The two highest (memory and VM wise) are Thunderbird which is using (60mb of main memory) and Firefox which is using 55mb of main memory. All the microsoft products I'm running like Visual Studio.NET 2003 are WAY down the list as none are using more than 10-15mb of main memory.

Nearly all popular linux distributions now come on more than one CD (even if you ignore the source code) and the default installations are WAY bigger than that of Windows XP.

Re:OK then.

[Davorama]

TFA goes on to mention in the conclusion that gamers may like the ability to do real work in the background while not interupting their game.

What the average user gets out of this? Hopefully a nice price cut on all those boring single core procs sometime soon.

Re:OK then.

[timster]

I don't know much about memory usage under Windows, but here on Linux I am running X with KDE 3.3.2, Firefox AND Konqueror, Citrix, Konsole and some other stuff, and my total actual RAM in use is about 165MB. Of course there's a bunch of RAM in use by the disk cache but it doesn't make sense to count that. Nothing is in swap.

On Linux at least the memory usage numbers reported by individual applications can be misleading, since it's hard to account for things like copy-on-write. It's easier to look at a system's overall numbers, and I think 165MB is pretty respectable.

As for the "more than one CD" complaint, Linux distributions these days come bundled with about all the software there is. Windows XP doesn't even include a C compiler as far as I know. But I always just install Debian from a net-install CD and download the packages I actually want. The resulting installed system is not particularly huge.

Re:OK then.

[NanoGator]

"So what about the average user?"

Windows is multi-threaded and behaves better in a multi-processor environment. Even the average user will notice this.

Re:OK then.

[masklinn]

Dude, Opterons are server-oriented processors, they're absolutely not geared towards regular boxes (much like Xeons). The people using this won't be playing and won't be typing text in their word processors either. The soon-to-be-released dual core A64 will more than likely be much much cheaper, but they won't come out right now. Please wait a few months

Re:OK then.

[Bwian_of_Nazareth]

Oh my, here it comes again. Comparing pears to apples. If I install MS Windows, what do I get? Operating system with a few (let me say - lousy) applications. If I install Linux distro of my choice, what do I get? Depending on my choice, it can be a full blown suite of application ranging from development to office apps to video processing.
And further more, e.g. KDE has been quite successfull at speeding up between 3.2 and 3.4. I am not so sure about the memory print, but that is no concern for me today (RAM is abundant).

Re:OK then.

[cthulhuology]

Face it, most people don't NEED more processing power. You can browse the web, send emails, and do your taxes on a Pentium 75, with enough ram, Firefox works just fine. Constantly upgrading is pure consumerist crap. You don't need this. Hell I develop games for a living, and I don't need this sort of processing power. Can a college kid use this stuff? Sure, they can run a P2P tool, download lots of pirated content, play their ripped DVDs/MP3s, all while pretending to do a paper. With a dual core machine, you could even keep your torrents downloading and play Doom3. Granted if you really feel you need to be doing this, you probably have a host of other problems and should probably consider going outside.

Re:OK then.

[2nd Post!]

So if I use Mac OS X, is it okay for me to talk about Microsoft's code bloat?

There are people who do it way better!

Re:OK then.

[pg110404]

So what about the average user?

The average user probably won't need it right now. Considering that they still sell very low end soldered-right-on-the-board athlon based processors at (last I heard) 1400MHz or 1600Mhz.

Those are the processors that are good for the 'joe averages'.

What makes the dual core processors useful are the higher end systems where 'more power' is not just the call of the techno-whore but for those who really do need it, such as the video editing types.

On an aside, I've also been playing around with spyware on a sandbox computer and it's actually very very very easy to pick that stuff up. It's also been recently that one key piece of spyware will not do anything on its own but will download and launch several dozen other spyware which actually do the damage. The main damage in this case is poorly written spyware that have unoptimized loops or infinite loops and bleed the CPU cycles dry, turning any speed into the clunkiest piece of shit you could possibly imagine. In this case, the second core will help mitigate that.

Also, as I've pointed out a while back, what about joe average's parents who think what they see is what there is? While it might be true that when longhorn comes out you can turn all that crap off, I seriously doubt microsoft will leave it off by default, which means when it does, you'll need that second processor to help take the load off the first just to maintain all them thar pretty graphics.

Re:OK then.

[toddestan]

Nearly all popular linux distributions now come on more than one CD (even if you ignore the source code) and the default installations are WAY bigger than that of Windows XP.

That's not really fair, as Windows XP doesn't really include very much with it. A media player, a web browser, a few simple utilities, and that's about it. Most Windows XP users have to install quite a bit of software to do what they want. On the other hand, most Linux distros have tons of packages included, for many users the default install has everything they need.

Re:OK then.

[Otter]

Honestly, ever tried to MD5SUM your CD1 ISO at the same time as you were encoding your MP3s for CD2?

If you're in that situation so often that it's worth buying a high-end workstation to expedite it, and if the bottleneck is really in the number of CPU cores and not elsewhere in the system, then, good!

As far as the OP's question about taxes, though -- unless you're really, really good at doing your taxes, the CPU does not even begin to approach being the limiting factor.

Re:OK then.

[LWATCDR]

Okay for the typing papers, doing taxes, sending email, and surfing the web you do not need even 1 Ghz. Frankly a 600 Mhz PIII will do all that at least using Linux and or W2k and yes I have done it.
Playing games it will depend on the game. Most games are single threaded so not yet. When multi core is common will that change? Possibly as soon as the game engines are recoded to take advantge of multi core. It may help some buy running other tasks on the second core like your anti virus, tcp/ip stack and so on.
Look at the list though.
Digital photography and video encoding ARE being used by average users. How many digital cameras, digital video cameras, and DVD RWs are sold today?
Digital imaging is no longer the realm of only the super high end user.

Re:OK then.

[Otter]

I don't know much about memory usage under Windows, but here on Linux I am running X with KDE 3.3.2, Firefox AND Konqueror, Citrix, Konsole and some other stuff, and my total actual RAM in use is about 165MB.

I've never used Citrix, so don't know how heavy that is, but -- 165 megs for the OS, desktop, file browser, web browser and a terminal isn't exactly what I'd call svelte. Open a GNOME app, dragging in a whole other suite of libraries, and you'll be pushing 200 megs without doing anything heavy.

Re:OK then.

[Chris Burke]

Currently, 50 processes. The two highest (memory and VM wise) are Thunderbird which is using (60mb of main memory) and Firefox which is using 55mb of main memory.

Point conceeded. Some OSS software chews up the memory, and FireFoo are major culprits.

Though I'd hope to hell Visual Studio is way down the list. It's just an IDE! It has a GUI and a text editor. All the memory-chewing hard work is done in the compiler back end. With that comparison, my Emacs session is 6MB.

Nearly all popular linux distributions now come on more than one CD (even if you ignore the source code) and the default installations are WAY bigger than that of Windows XP.

Of course they are -- they include reams of free software! Nobody would complain about the large size of Windows installations if that installation came with practically every piece of software you would ever need! Even a 'default' install that doesn't install everything still has vast swaths of software from compilers to office suites to web browsers to web servers to image manipulation to whatever.

Who could possibly complain about getting more free stuff, even if it takes another CD or two or three to fit it? Consuming disk space for useful things is fine. Windows installs are considered bloated because the size increases but the perception is that you're not actually getting more stuff. Honestly -- what comes with the XP install these days?

Re:OK then.

[bfizzle]

The main damage in this case is poorly written spyware that have unoptimized loops or infinite loops and bleed the CPU cycles dry, turning any speed into the clunkiest piece of shit you could possibly imagine. In this case, the second core will help mitigate that.

Until they get two resource hog spyware apps going at the same time. Then they are back to square one.

Re:OK then.

[huge colin]

Nearly all popular linux distributions now come on more than one CD (even if you ignore the source code) and the default installations are WAY bigger than that of Windows XP.

Also, they work properly.

Re:OK then.

[Anarke_Incarnate]

and most of those come with full suites of software. Microsoft's one CD is just the OS. Then you have to download the latest patches (Which you should do for any OS). I wonder how big Windows would be if it included several text editors, compilers, office suites, games, etc.

Re:OK then.

[alecks]

Meanwhile, everyone's complaining that MS is bundling a friggin media player and internet browser with THEIR OWN OS!!! I've always wondered, and i'm sure (err hope) there's a good explanation... Why doesn't Apple get sued for packaging everything with their OS?

Re:OK then.

[tehcrazybob]

I've never used Citrix, so don't know how heavy that is, but -- 165 megs for the OS, desktop, file browser, web browser and a terminal isn't exactly what I'd call svelte.

I wouldn't complain too much. Most Windows XP systems are using more than that at startup. I think mine uses about 180 right off the bat, and right now, running Firefox, Thunderbird, Gaim, the OpenOffice Quickstarter, and a few special drivers, it's using 241.

Re:OK then.

[Alan]

Regarding more than one CD, remember that to duplicate the Linux install software wise in windows you'd need the OS (1 cd), office (1, 2 or 5 cds depending on how you get it), sql server (1 cd), visual studio (1 dvd, though to be fair I don't think linux's offering has this yet), a whole lot of misc system and user apps (another 1-2 cds). Of course, you have the choice to install what you want remember, so you can only install the base os if you so choose.

Re:OK then.

[rathehun]

So we have: scientific computing applications (MATLAB and LS-DYNA) digital photography video encoding Cinebench and "a few other applications".

The college student who's typing out his college paper does NOT need an 64-bit dual-core processor.

Wait - I haven't used Office 2003.

R.

Re:OK then.

[Alan]

Yup. Of course, hardware always drops in price, so I figure in under 2 years you'll find that dualx64s are the standard for consumer machines, if not something better. By then the servers will have quad-core 5Ghz chips which people will say "those are for servers only", but they'll come down and a couple of years after that those chips will be in consumer machines....

And on, and on, and on :)

Re:OK then.

[pg110404]

Hence mitigate, not prevent.

Which means you could likely be infected with twice as much spyware for the same lack of performance.

Re:OK then.

[RupW]

Though I'd hope to hell Visual Studio is way down the list. It's just an IDE! It has a GUI and a text editor. All the memory-chewing hard work is done in the compiler back end.

When editing, auto-complete will require a version of the compiler's language parser and a some processed version of the project code. This will usually include a lot of system headers, so it's large. For many languages, the parser in the language UI plugin is also used for on-the-fly syntax checking and syntax highlighting. There's also the source browser (navigate to definitions, declarations and uses) which I think has a separate dataset from the auto-complete.

When debugging, you have all of the above plus the program symbols in memory. This includes symbols for all referenced system DLLs, so that's large too.

And there's probably plenty I've forgotten. So I've no problem with VS using lots of RAM.

Re:OK then.

[indifferent children]

My vi session = 1.8MB! Only old South Koreans use Emacs.

Re:OK then.

[anglozaxxon]

Isn't this technically already being done for games? I mean, there's the regular ol' CPU then there's the GPU. And then there's sound acceleration (not sure if that uses some sort of SPU or not...). I mean, aren't games basically the ultimate test in doing ten kajillion things at once?

Re:OK then.

[r_naked]

As for the "more than one CD" complaint, Linux distributions these days come bundled with about all the software there is. Windows XP doesn't even include a C compiler as far as I know.

I am sure Microsoft would love to include a stripped down version of VS.NET with XP. With all the grief they are given for including a web browser, what kinda headaches do you think they would get if they started shipping a compiler?
I hear compliants on here all the time about how you get so many more applications with a Linux distribution, and in the same breath how Microsoft should remove things from their OS.
I would LOVE to see Microsoft be allowed to ship more software with the OS. A compiler, an office suite (not just wordpad and Outlook Express), a decent DVD authoring package, something better than MSPaint etc...etc... but people would scream UNFAIR!
To me unfair is Microsoft trying to compete with free when the free side of things doesn't have the same restrictions. Hell even Apple ships most of the popular GNU software with OSX.
Please, someone tell me how that is fair. Did anyone stop and think that maybe if Microsoft didn't have to keep hacking their OS to make various groups / government bodies happy that the OS might be more secure? I mean they already use BSDs ftp (and probably several other BSD utils), it would be real nice if they could have thrown pf in instead of developing their own (potentially less secure / less configurable) firewall.

Re:OK then.

[Grayputer]

------------------
Nearly all popular linux distributions now come on more than one CD (even if you ignore the source code) and the default installations are WAY bigger than that of Windows XP.
------------------

Gee, don't do much do you. My windows install comes on a crate load of CDs, first one is Win 2000, then several for Visual Studio, then a couple more for Photoshop, then several more for MS Office and finally several more for MS SQL. I think I have about 10-15 CDs for my Windows install.

Oh yeah, my Linux desktop does all that (and more) from I think about 4 CDs (haven't counted recently).

Oh, and as far a space goes, I have a Linux server (old terminal server software) booting from a floppy. I think my total disk requirement is 1.3MB and it runs in 32MB of RAM (I had it so I used it, probably could use less). While I seem to remember Win 3.1 running is less than 32 MB of RAM, I don't think it could boot from floppy.

Methinks you missed the point, you can have it ALL or you can take almost NOTHING. How it is packaged is only marketing. Now admittedly Granny cares about marketing and ease of use more than I do (and Linux can learn from Windows there). But Linux CAN be both smaller and provide more functionality per dollar than Windows, depending on your needs and expertise.

Re:OK then.

[Mycroft_VIII]

The reason Microsoft gets so much more scrutiny and leagle flack on the bundling issue is because they have been found (leagly) to be a monopoly. This changes the rules for them so as to prevent them from locking out any future competition or taking over related/inlinked markets by virtue of thier having a near captive audience.
Apple with it's small slice of the market is very unlikely to say put opera out of bussiness by shipping thier own browser for free with thier operating system like Microsoft did to Netscape(I know that's a simplification of ie/netscape history, but it serves to illistrate my point I hope).
In short Microsoft is a victim of thier own success here.

Mycroft

Re:OK then.

[Waffle Iron]

My system becomes very sluggish if I do something like md5sum an ISO. However, my CPU utilization is not maxed, and lowering the process priority doesn't seem to help. If I let it run by itself and then come back to the machine when it's done, I see all of the apps and window manager have been totally swapped out and will take many seconds just to repaint the first time I clear the screensaver.

This tells me that the sluggishness is due to the interaction between virtual memory and heavy I/O activity, not CPU load. Another CPU probably isn't going to help at all in this situation.

Re:OK then.

[Jim_Maryland]

I have to admit, XP doesn't crash very often compared to previous MS releases. I have had it crash/lockup though due to applications and the only thing to do is power it down. I can't recall seeing a BSOD with it yet. On the other hand, my Solaris systems for the last 7 years and the DEC Ultrix systems for 3 years prior to that have had better reliability with zero OS freezes/crashes. I have had applications hang, but they haven't hung the OS in the least.

Microsoft has to overcome years poor stability reputation among IT professionals. Linux will likely face the same issue. The major difference I see though between the MS and Linux systems is that when I pay for the OS, I'd expect better quality, just because I'm paying for it. I'm probably a bit more forgiving to the Linux system crashing than a Microsoft OS.

As for the hardware comments, vendor support for the dominant desktop OS makes sense. Unless the hardware vendor sees a particular reason to release drivers simultaneously for multiple operating systems, I would expect MS to be the first driver released.

If you're looking for hardware to install into your PC, then how about showing me where I can get MS Win32 drivers for Sun GigaSwift PCI Ethernet Card. Likely this piece of hardware will only provide Solaris drivers. Note: I'm only showing this as an example to disprove your statement. Generally speaking though, you are correct about the hardware.

Re:OK then.

[masklinn]

ok, you're a moron, strange you weren't modded as such.

Here are some hindsights:

• AMD Opteron are server-oriented CPUs, they require the extremely expensive registered ECC RAM to work and special Motherboard, they use the Socket940 which has (surprise !) 940 pins. It's been designed to work in server environments and multiprocessor boxes
• AMD Athlon64 is the desktop version of the Opteron, the older generation ran on Socket754 and the current one, along with the AthlonFX CPUs (which are basically the high end A64) runs on Socket939. These chips are not compatible with S940, and run just fine with regular DDR Ram

As strange as that may seem when you know these facts, when going to AMD's website you'll be able to notice that Opteron are in the 'Server Processors' category while A64 and A64FX are in the 'Desktop Processors' one

Need more?
Opteron prices start at around 300 with the 1.6GHz Opteron 242 while A64 around 90 for S754 (2800+, 1.8GHz) and 110 for S939 (3000+, 1.8GHz)
Opteron start with 1MB L2 cache and go up to 2MB while lowest priced A64 have 512k and highest ones (including AthlonFX) are at 1MB
A64 have a single Hypertransport link (8Gb/s) while Opterons feature up to 3 coherent HT links (24Gb/s)
Opterons also features ECC correction for both L1 and L2 cache (on top of the RAM ECC), a lower core voltage (1.2V Vs 1.4 to 1.5 standard for A64)

Finally, it should be noted that Opteron boxes can get up to 16GB RAM per processor (8 DIMMs each), and while I don't know of the A64 capacity in this field I'm not sure it gets past 4 DIMMs.

Sorry mate, but the Opteron IS a server CPU, labelled and priced as such by AMD, and A64/AFX is the desktop one and will be (much) cheaper than the current dual core opterons when it'll be released as dual core in a few month.

Re:OK then.

[RicktheBrick]

I for one have two amd computers in front of me. I have one keyboard, mouse and monitor. All I have to do to go from one computer to the other is hit the scroll lock key twice in less that a half of second. I can do a computer intensive application and less than a second be on the second computer and do whatever I want. For the next year or so I will have probably have paid about half of what they will charge for the duo core sytems too.

Re:OK then.

[millennial]

No kidding. A 2gHz P4 should not cost $190 at Best Buy...

Re:OK then.

[SuperQ]

The average user is supposed to buy a Sempron.. DUH

The average user doesn't even need 1/4 the computation power of even athlon64 machines.

Hell.. I do most of my work on a 1.4ghz Pentium M, and a Duron 800.

Re:OK then.

[OwnedByTwoCats]

Writing papers and doing taxes have been low-challenge tasks for a while. The cheapest of the cheap computers will do this just fine.

For the gamers, look at the game benchmarks included in the article.

Re:OK then.

[OwnedByTwoCats]

The european anti-trust case against Microsoft is insisting that MS offer versions of their software without the add-ons. It will be interesting to see if MS reduces the price of the "stripped" version. MS would love to ship a version of their software where you pay for the add-ons, but don't get to use them.

The compelling anti-monopoly argument is that if MS only offers the fully-bundled version of their software, then consumers are forced to buy Microsoft's Office Suite, and Microsoft's DVD Authoring package, or Microsoft's Photo Manipulation program, whether they want to or not.

Re:OK then.

[OwnedByTwoCats]

Because Apple isn't violating the law. Microsoft was.

It is illegal to leverage a monopoly in one market (PC Operating Systems) to gain a monopoly in another market (Web Browsing applications).

Microsoft violated that law, and was convicted of violating the law.

Re:OK then.

[kesuki]

While I seem to remember Win 3.1 running is less than 32 MB of RAM, I don't think it could boot from floppy.
Actually it can boot from a floppy, but only if you hack it to ;) I remember running windows under 4MB of ram, and when they made web browsers that could actually browse the web with that little ram.

Re:OK then.

[juanescalante]

Yes, that is true. A more comparable open source product would be netBeans... and it takes up a LOT of RAM.

Re:OK then.

[dsci]

what kinda headaches do you think they would get if they started shipping a compiler?

I believe the complaints are not due to what they incluce as options, but what they integrate with the OS. There's a big difference in including a package on a CD for the user to install if they want it, and sharing so much code across 'diverse' applications that some apps can not be effectively removed.

Did anyone stop and think that maybe if Microsoft didn't have to keep hacking their OS to make various groups / government bodies happy that the OS might be more secure?

Huh? Who is making MS 'hack' their code? Microsoft actively markets their products to various groups/government bodies on various premises. I think you got cart before the horse.

Re:OK then.

[Chris Burke]

When editing, auto-complete will require a version of the compiler's language parser and a some processed version of the project code. This will usually include a lot of system headers, so it's large. For many languages, the parser in the language UI plugin is also used for on-the-fly syntax checking and syntax highlighting. There's also the source browser (navigate to definitions, declarations and uses) which I think has a separate dataset from the auto-complete.

Emacs does all that, so the comparison is still apt. Of course I didn't have to read the paragraph to say that; emacs does everything. ;)

When debugging, you have all of the above plus the program symbols in memory. This includes symbols for all referenced system DLLs, so that's large too.

Right... so Emacs + gdb session, which chews up a variable amount of memory.

And there's probably plenty I've forgotten. So I've no problem with VS using lots of RAM.

I have a problem with it using lots of RAM for basic editing (including syntax highlighting and source browsing functions). When it's doing actual data-intensive work such as debugging, sure, it will use more. I'm more concerned with the base footprint, because you don't want your debugging session to be cut short because base mem + debug was too much for your system. :)

Re:OK then.

[RzUpAnmsCwrds]

"Honestly -- what comes with the XP install these days?"

Internet Explorer, Windows Media Player, Windows Movie Maker, Outlook Express, MSN Explorer, IIS, Windows Messenger, Windows Netmeeting.

Oh, and a *lot* of drivers.

It's not as much as an Ubuntu desktop, but it is by no means "tiny".

" Even a 'default' install that doesn't install everything still has vast swaths of software from compilers to office suites to web browsers to web servers to image manipulation to whatever."

I don't know what distro you're using, but the default installs of Fedora and Ubuntu don't include Apache or GCC.

One more reason I like Ubuntu - the basics come on the CD, but you can add more with apt-get. Unlike Fedora, where invariably you will end up needing a few packages on the 2nd, 3rd, and 4th CDs.

Re:OK then.

[Chris Burke]

Internet Explorer, Windows Media Player, Windows Movie Maker, Outlook Express, MSN Explorer, IIS, Windows Messenger, Windows Netmeeting.

Even IIS? Well, that's more than I expected. Then again, the last time I installed an MS operating system it was '95 and it came with Wordpad and a calculator. :)

I don't know what distro you're using, but the default installs of Fedora and Ubuntu don't include Apache or GCC.

I just installed Hoary, and now that you mention it I don't think it installed gcc. I'm used to plain Debian, which does (though the old Debian installer, there really wasn't a 'default', just usage models and i always checked 'developer' or whatever it was called). Frankly, I think not getting gcc is crap. :)

One more reason I like Ubuntu - the basics come on the CD, but you can add more with apt-get. Unlike Fedora, where invariably you will end up needing a few packages on the 2nd, 3rd, and 4th CDs.

I guess... apt-get is great, at least if you have broadband, but I liked having Debian come on 3-6 CDs and installing every god damned thing under the sun straight out of the box. I wouldn't mind if Ubuntu did the same, giving additional CDs with extra packages if you wanted.

Re:OK then.

[Sawopox]

>Nearly all popular linux distributions now come on >ore than one CD (even if you ignore the source >code) and the default installations are WAY bigger >than that of Windows XP.

This is an apples and oranges argument. A "linux distribution" SHOULD come on multiple CDs, WindowsXP is "solely" an OS. A distribution is designed as a complete software solution: office, web, security, games (for some), hardware support, printing, graphics, programming, database, etc, etc. The "OS" should, to someone like me with no official OS-design theory (and opposable thumbs) be solely for running the above applications.

With regards to your VS.NET 2K3 not using that much RAM, were you actually compiling something? Firefox perhaps? Also, what are you viewing in Firefox? Do you have twelve tabs open to amateur porn like me? Is thunderbird your gmail forward, with 350MB sitting in the inbox? The amount of RAM usage varies, but most people would agree that the same program (Firefox for instance) will run more efficiently on a Linux based OS than a Windows based OS given the same hardware.

Re:OK then.

[Jim_Maryland]

Sun has been very good to me for stability and performance (mostly running Oracle, WebLogic, Tomcat, Apache, and Java). XP Pro has also been pretty good, but between the two, XP Pro has had rare lockups that I haven't encountered on Sun systems. My in-laws actually have XP Home Edition and they seem to have a few problems with it but they disregard most safe computer practices (when I visit, I have to regularly clean up their system from spyware and the occasional virus - they do at least leave the AntiVirus running but often install anything they can download). I'd imagine XP Home to be pretty similar in performance to XP Pro.

I haven't looked, but Dome used to make high end medical use graphics cards that were Sun compatible. I don't know that I've ever seen them mentioned for MS Win32 use. That might be an example of a non-Sun part that may not have MS Win32 drivers first.

Re:OK then.

[toddestan]

Is it really bloat? Besides, every distro installer I have seen lets you pick and choose what you want to install. You can install as little as you want with Linux. Windows XP may not include a lot, but generally you are stuck installing what it does come with, whether you like it or not.

Re:OK then.

[budgenator]

Constantly upgrading is pure consumerist crap.
Lighten up, some people Hot-Rod cars, some People Hot-Rod computers; niether is realy practical, but both are a lot of fun. I have some old software that came bundled on my 8MHz 286, which on a whim I loaded onto a Pentium 75, those old window managers, and word processors are wicked fast on a pentium.

If everybody still played DOOM II, you'd be saying Hell I used to develop games for a living..., enough people write their own levels to keep the game interesting for quite a while.

Re:OK then.

[catprog]

Probaly because you can remove it from Apple (I not sure if you can but I think you can) but with windows you can not remove it without breaking the computer.

Re:OK then.

[millennial]

I know this as well as you, but what really bothers me is that there are people out there who will spend $ridiculous for a computer that they'll do basically nothing with.

Example: My dad bought a $3000 laptop. THREE THOUSAND DOLLARS. What is he doing with it? Listening to music, using iTunes with his iPod shuffle, playing DVDs, surfing the web and OCCASIONALLY writing something. As a geek, it physically pains me to see such a beautiful machine (64-bit processor, DVD+/-R/RW, Harmon-Kardon speakers...) being wasted. But there it sits.

Re:OK then.

[budgenator]

blender, and video encoding is pretty slow on my 700 MHz athlon, but everything else runs OK

Re:OK then.

[StikyPad]

Look, just because you don't find Character Map, Onscreen Magnifier, and Paint useful...

Re:OK then.

[AvitarX]

I don't know what it is doing, but I want my web-browser to eat memmory if it is there.

I want to hit back and have it come from RAM, and it would be real nice if it was for a while back.

Thunderbird has no excuse.

Re:OK then.

[Deliveranc3]

Because they could better be served by an internet distribution system?

They should definitly stick to 4.7 gigs for the immediate future.

Re:OK then.

[nmg196]

You're arguing my point for me. The linux distributions are bloated and install many applications that most users will never use. Primarily, this is probably because installing these applications yourself as a new user would be a nightmare and would put you off using that distribution.

Re:OK then.

[nmg196]

It's an Operating System! Why *should* it come with more than that? Microsoft gets enough flack for simply including IE. Imagine the ensuing row if they also included things like Office and "everything they need" and users never needed to buy *any* third party applications. Yet linux does this all the time. The only reason it doesn't get slated for it is that it's not the market leader.

You can't have it both ways.

Re:OK then.

[nmg196]

> I'd hope to hell Visual Studio is way down the list. It's just an IDE!
> It has a GUI and a text editor. All the memory-chewing hard work
> is done in the compiler back end.

You've obviously never used Visual Studio. Notepad and Wordpad are text editors. Visual Studio is NOT just a text editor. To get the same features under linux you'd need about 9 different programs. It has a very powerful help system, a debugger, build tools, resource editors, intellisense code editors for many languages, many different GUIs for designing web and windows applications etc etc etc. It even includes a web browser on top of all of those things and has a memory footprint which is still a *fraction* of the size of firefox's which is *only* a webbrowser.

Re:OK then.

[Bwian_of_Nazareth]

:-) You're given the choice - not much is installed unless you agree to install it. We could argue about what "bloated" means - does it mean offering a lot (and maybe installing by default)? And does it even matter with disk space being cheap?
Anyway, you can't compare size of Linux distro with size of Windows installation disk. Those are two different beasts.

Re:OK then.

[Chris Burke]

You've obviously never used Visual Studio.

If something can be both obvious and wrong, then you're correct.

Notepad and Wordpad are text editors. Visual Studio is NOT just a text editor. To get the same features under linux you'd need about 9 different programs. It has a very powerful help system, a debugger, build tools, resource editors, intellisense code editors for many languages, many different GUIs for designing web and windows applications etc etc etc.

Understand that I'm using the word "text editor" in a way that allows Emacs to be called a "text editor". Notepad is a text editor, but a crap one. Visual Studio is a text editor, but with actual features. Emacs can do all that too, except the GUI design, I'll grant. But my text editor has a command shell and a psychoanalyst. ;)

I have used VS. It's a fine IDE, but it isn't that special, and rattling off features as though it's a miracle each one doesn't use 100MB each doesn't impress me.

It even includes a web browser

Heh. What a useful feature! So, did they write a browser from scratch, or did they re-use the IE libraries that are loaded automatically on bootup and thus probably aren't being counted in the memory footprint?

Emacs has a web browser, too, for what that's worth. ;)

Chance for someone to karma whore...

[Amiga Lover]

OK. Anyone have a quick simple explanation of why Dual Core over Dual CPU motherboard? are there inherent advantages to dual CPUs so close together?

Re:Chance for someone to karma whore...

[BlueCodeWarrior]

Imagine two dual core CPUs plugged into a dual CPU mobo.

Pop an erection yet?

Re:Chance for someone to karma whore...

[Anonymous Coward]

Presumably the one major factor is cost of the motherboard.

Dual-cores can run on existing motherboards, with a bios update.

Re:Chance for someone to karma whore...

[Anonymous Coward]

It cuts down the wait time of communication between the CPUs as the dual core chips don't need to travel on any sort of MOBO bus to communicate thus effectively giving a clock speed bus of inter CPU communication.

Re:Chance for someone to karma whore...

[Anonymous Coward]

Less heat, less space, less energy requirements, eventually less money because there is only one chip.

Re:Chance for someone to karma whore...

[Anonymous Coward]

With Opteron, you get extra HyperTransport interlinks without the extra pins that would be required.

With either architecture you get shorter wire runs, making timing easier to control.

Re:Chance for someone to karma whore...

[Moraelin]

Probably the biggest advantage is that it's cheaper. (Although if by much, that remains to be seen.)

Plus, AMD's promise was something like being able to double the number of CPUs without having to buy a new motherboard. Though how much saving that will be (I expect AMD to price these pretty high), and whether it will mean that you're stuck with much slower cores to keep the TDP limits, that remains to be seen.

There are other possibilities for improvement, such as using a shared cache and IMC instead of just throwing two cores together and going over HT like on a dual CPU system. But AMD hasn't yet done that.

Re:Chance for someone to karma whore...

[Jjeff1]

Cost.

With a BIOS update, I turn my single socket board into a dual CPU rig. Now your 799$ low end server gets almost 2x the CPU horsepower. This is killer for a cluster or similar. 2x the CPU in the same rack space.

For the bigger boxes, it turns a 4-way high end box into a 8-way. Think database servers, virtualization servers or any other multi-threaded app that uses a lot of CPU. If you need an 8-way box, your cost just went down by 40% or more.

For the board makers, they no longer have to build a big board with all the Hypertransport traces between sockets. There are 1/2 the traces to memory because there is only 1 memory controller.

Re:Chance for someone to karma whore...

[Anonymous Coward]

OK. Anyone have a quick simple explanation of why Dual Core over Dual CPU motherboard? are there inherent advantages to dual CPUs so close together?

Double your cpus without double the license fees, not to speak of the lower footprint.

Re:Chance for someone to karma whore...

[shawnce]

Yes.

One example... dual core (true dual core) CPU have the ability to exchange data between the cores at faster rates and more importantly with less latency then when having to exchange data between CPUs on a dual CPU system. This can improve SMP flow.

Another example... good dual core implementations will utilize some form of cache unification to allow better bulk sharing of data between cores while still allow high-levels of independent cache activity (the IBM's Power5 is a good example of this).

Re:Chance for someone to karma whore...

[AFairlyNormalPerson]

Your dual processor board now supports 4 processors with a bios update. I guess 4 is better than 2.

Re:Chance for someone to karma whore...

[uujjj]

Advantage 1: A dual-core motherboard is cheaper than a dual-CPU motherboard, since it only needs one CPU socket. This price advantage is even bigger for a 2x dual-core mobo vs. 4-CPU mobo.

Advantage 2: Communication between the processors no longer requires you to send a signal off-chip, which is much faster. The dual-core Opteron takes advantage of this, though the way the dual-core P4 is designed still requires sending communication off-chip.

Advantage 3: Although the current gen dual-cores don't do this, you could, in theory, share the L2 cache between the processors. This further speeds communication, and it also provides a larger cache per thread.

Re:Chance for someone to karma whore...

[hobuddy]

1) Cost.
Since there need only be half as many sockets, the motherboard can be smaller, less complicated, and therefore less expensive. This is especially true in the case of single-socket motherboards, which are usually 50-60% as expensive as their dual-socket brethren. AMD has sweetened the cost savings even further by arranging it so that most single-socket motherboards already in use with a single-core CPU can accomodate a dual-core CPU after just a BIOS flash.

2) More efficient interconnection between the cores.
This advantage currently applies to AMD's design but not Intel's. As explained here, "As you can see, AMD didn't simply glue a pair of K8 cores together on a single piece of silicon. They've actually done some integration work at a very basic level, so that the two CPU cores can act together more effectively. Each of the K8 cores has its own, independent L2 cache onboard, but the two cores share a common system request queue. They also share a dual-channel DDR memory controller and a set of HyperTransport links to the outside world."

After reading the TechReport article I linked to above, it looks to me like AMD is way ahead in the dual core market in all of the areas that count: better backward-compatibility, better cache coherency, and lower heat.

Re:Chance for someone to karma whore...

[interiot]

The biggest advantage is it's the only way to get a faster processor. You won't see a 4GHz processor in production soon. It's a matter of physics. So instead, you'll wait for applications to better support parallel processing, and then get a dual-core CPU at that point.

Re:Chance for someone to karma whore...

[__aazofn1209]

Well, for example, quad-cpu mobos are so expensive you'd think they were made of platinum. They're also slightly larger than Rhode Island. So dual-core + dual-socket is a good alternative if you want to bump your server performance up a bit. This will make sense economically once AMD rakes in some product-launch cash and lowers the prices on their dual-core server chips.

Re:Chance for someone to karma whore...

[Senor_Programmer]

Two dual cores on two socket MB.
Or better yet, 4 ;-)

Intel's dual core is like a dual processor MB using one socket.

AMD's dual core has another bus, built in.

Ultimately it's

propagation time is dictated by wire length leading to a limit on how high one can go in frequency before adding wait states.

increased wire length for a given wire size requires more power to get a signal from one end to the other (AMD vs Intel thermal specs say a lot between the lines).

Re:Chance for someone to karma whore...

[Daravon]

"Pop and erectino yet?" No. It's broken. -Bob Dole

Re:Chance for someone to karma whore...

[telecsan]

If you're turning a 4-way box into 8-way, your costs aren't anywhere near related to the hardware. You're going to be running per CPU licensed software, and likely dual-core will not help in this regard. Sorry about your luck.

Re:Chance for someone to karma whore...

[drinkypoo]

One example... dual core (true dual core) CPU have the ability to exchange data between the cores at faster rates and more importantly with less latency then when having to exchange data between CPUs on a dual CPU system. This can improve SMP flow.

Is this really true in the case of the Hammer core? CPU to CPU and Core to Core links both use HT.

and, as you fail to mention but allude to by mentioning the technology, there is no cache unification in hammer (yet?)

Re:Chance for someone to karma whore...

[masklinn]

Doubling the calculation power without changing the encumbering? (aka layout and space optimization)

Re:Chance for someone to karma whore...

[obender]

There are two kinds of dual Opteron motherboards:

1. the cheaper ones have the memory controlled by one processor and the other processor gets to the memory via the first one over hypertransfer e.g Tyan Tomcat, Tyan Tiger
2. the more expensive ones that give each processor its own memory (NUMA), eg Tyan Thunder

The cheap dual mobo will be easily beaten by the dual core as the communication between the two cores has been optimized by AMD.

With the expensive mobo things are not so simple. In theory the dual mobo should be better but this depends on how NUMA aware is the OS. AFAIK only the latest XP has this and some Linux kernels. If the OS is not NUMA aware the performance of the NUMA setup is not significantly better than the one with shared memory access. More information here

OTOH I would expect the dual cores to fit inside a dual processor motherboard and thus offer even more power.

Re:Chance for someone to karma whore...

[shawnce]

Well I am not an expert on the latest AMD stuff (of course they had the two core design in mind from the beginning of the Opteron) but from what I have seen the two dies in the Opteron interface with a common crossbar switch (the each have an independent L2 cache so no real unification at that level).

So they are not interconnected via HT on die but by some cross bar interconnect that presumably allows some level of concurrent point-to-point (core to core , core to HT link, core to memory, etc.) transfers and at high clock/data rates.

see this diagram ... or in a nice pdf

Re:Chance for someone to karma whore...

[CompVisGuy]

Not sure if anyone answered this properly yet...

The main advantage of dual core over dual processor (where the processors are not in the same CPU package) is that it should be possible to allow the two CPUs to communicate at very high speed.

Inside a single CPU, data is moved around at, or very close to, the clock speed of the CPU (e.g. 2.7GHz). Outside of the chip, the longer distances signals need to travel mean that it is more difficult to run data busses at high speeds. So talking to a hard disk or other peripheral component is slower (e.g. 166Mhz).

If you had two separate CPUs (i.e. not in the same package), then they would be limited to the 166MHz buz (or whatever). However, if you have two CPUs in the same package (and ideally on the same piece of silicon), they can communicate at or close to the clock speed of the 'package' (i.e. 2.7GHz). The CPUs can then work co-operatively more efficiently.

The main drawback to putting one or more CPUs on a single piece of silicon is the cost of doing so. CPUs are etched onto silicon wafers. These CPUs are then physically cut out (e.g. with a diamond circular saw) and then packaged. If the wafer has a local imperfection, any chip that intersects with that imperfection is likely to be faulty, so you have to throw it away. If the cutting process damages a chip, it must be discarded. The smaller the CPUs, the less likely a particular one will be damaged. The larger they are, the higher the probability. As the chips get bigger (e.g. if you bolt two CPUs together into a single chip), the yield goes down, and the cost of an individual chip increases.

Re:Chance for someone to karma whore...

[karnal]

Man, you can't even copy and paste correctly.

Maybe you need a dual core athlon to do that...?

Re:Chance for someone to karma whore...

[taskforce]

Under AMD's solution the on die memory controller also gives lower latency. (Although this is a moot point for Intel, becuase their's is still controlled on the Northbridge.)

Re:Chance for someone to karma whore...

[Deliveranc3]

The same logic which would allow this should also allow sharing memory acces, which would double the amount of memory available to each CPU (less HD access) double their memory bandwidth (Considering the latency to the cpu is trivial) and double system memory trivially.

Re:Chance for someone to karma whore...

[archen]

Which brings up the point of us hitting a limit now. So what do we do? Add another processor. And after a year when that is old hat? Add 2 more? Are we just going to keep adding processors now? If that is the case (which doesn't seem far fetched anymore) then the Cell Processor technology looks even more interesting.

Re:Chance for someone to karma whore...

[fitten]

No. You won't see a 4GHz Pentium 4 in production because it produces too much heat to be cooled reasonably well (read: cheap/reliable/easy enough for mass consumers). There is nothing magical about the number 4GHz. Portions of Pentium 4 chips run at 2X core speed (which is more than 4GHz on any part over 2GHz core speed), for example.

Re:Chance for someone to karma whore...

[fitten]

It's not cheaper initially, that's for sure for AMD parts :( Check out AnandTech's review. Dual core Opterons cost 3.5X as much as single cores at the same speed. Dual core A64s are going to cost 2X as much as the single core speed equivalents. Intel's parts costs 1.5X as much (at the entry level) as the single core speed equivalents.

Re:Chance for someone to karma whore...

[flaming-opus]

imagine. I've had an rp2600 under my desk for a couple years now.

Yeah it's fast; that does not intrinsically make it exciting, or sexy.

Re:Chance for someone to karma whore...

[shawnce]

Can you better state things? I am not sure I follow you.

HD access? What do you mean by CPU? (core, dual core package, etc.)

For example in the case of the dual core Opteron the two cores share the same memory controller and buses to main memory. The amount of memory attachable to those buses is determined by the capabilities of the memory subsystem outside of the CPU package and the single onboard memory controller. In other words it being dual core hasn't affected, in a positive, fashion the amount of memory supported or memory bandwidth.

What it likely does help with is cache synchronization since that can take place on die via the crossbar switch (at least ideally).

In the case of the Power5 (another dual core CPU) it also shares a single memory controller but it has an integrated cache hierarchy on die (not to mention a huge amount of L3 and L2 cache) that sits in front of the system interconnect. This allows the cores to share a much larger pool of cache, avoiding aspects of cache synchronization (at least at some levels).

Re:Chance for someone to karma whore...

[Tore S B]

You're going to be running per CPU licensed software, and likely dual-core will not help in this regard. Sorry about your luck.

Yeah, my company avoided upgrading our Apache-running webserver because the license costs would've killed us.

Re:Chance for someone to karma whore...

[PitaBred]

Only if you are using the AMD dual-core processors. Intel's still use the FSB to communicate between the cores. It's literally just two cores just shoved onto one die.

Re:Chance for someone to karma whore...

[mikis]

That is not true, unless you compare single-core 1xx serie with dual-core 8xx. Check the prices here:

Entry level dual-core Opterons to be priced same as high-end single-core parts

For example, if you compare best CPU's from 1xx series, dual-core Opteron 175 costs 57% more than single-core Opteron 152 ($999 vs. $637)

Re:Chance for someone to karma whore...

[fitten]

Those aren't clock speed equivalent. Look at the prices for single-core 2.2GHz vs. dual-core 2.2GHz, both with 512K L2.

one step closer to the day

[Prince Vegeta SSJ4]

where I can encode mpeg2 DVD (maybe it will be HD-DVD by then), rip & copy a DVD, Download a huge torrent, and Play UT with a respectable framerate.

Re:one step closer to the day

[superpulpsicle]

Well te UT framerate part falls along the lines of graphics card issues more than the processor. Blame nvidia/ATI for that.

Re:one step closer to the day

[drinkypoo]

Uh, no. This has never been true. Unreal engine games have always been CPU-bound.

Re:one step closer to the day

[Anonymous Coward]

Bah, there's no use bragging about your multitasking abilities here on Slashdot, where 98% of the user population has learned to surf the web with only one hand!

Re:one step closer to the day

[advocate_one]

Pah... I can do that already... I've got 4 linux boxes networked and three of them are being driven using X over ethernet... so basically, yes, I am doing all four tasks at once on the same display... however, they are being done on four separate computers...

Re:one step closer to the day

[jensen404]

CPU time per frame is about the same at any graphics setting. GPU time per frame scales greatly based on the settings. So there is no way to say a game is CPU-bound or GPU-bound unless you state the settings you are using.

1 Dual Core vs Dual CPUs?

[uofitorn]

Wouldn't a better benchmark be to compare a dual core setup to a similarly configured dual processor workstation?

Re:1 Dual Core vs Dual CPUs?

[Dun Malg]

Wouldn't a better benchmark be to compare a dual core setup to a similarly configured dual processor workstation?

If you're not going to RTFA, why do you think you're qualified to discuss what they did and did not benchmark? They did compare dual core and dual processor setups as well. They also discussed the relative advantages of both.

Re:1 Dual Core vs Dual CPUs?

[timeOday]

I think the benchmark for comparison should be whatever else you can get at a similar price.

Dual processor motherboards and CPUs were never priced to make them attractive for widespread use, whereas dual core chips supposedly will be. We shall see.

The simple future

[caryw]

Why stop at dual core?
Once a way to link multiple cores of a CPU is firmly implemented scaling the chip to 4, 8, or even 32768 cores should be relatively easy.
With chip dies getting smaller and smaller the only real reasons not to continue this multi-core scaling would be physical space and power usage.
Perhaps they could scale multiple cores vertically instead of just making the chip wider and longer.
And perhaps the cores could only be "turned on" when called for instead of using up juice all the time.

Interesting look at the future of chips.
Sony's Playstation 3 is using a "cell processor" or similar multi-core design that has already been covered here in the past.

Arstechnica article on the cell processor here.
--
NoVA Underground: Fairfax County, Loudoun County, Arlington, Price William chat and local forums

Re:The simple future

[Detritus]

It wouldn't work. Why do you think we have processors with two or three levels of cache? There is a serious speed/bandwidth mismatch between the processor and the main memory system. There are ways of increasing main memory bandwidth, but they are very expensive. There's no point in adding more processors if they are going to spend 95% of their time stalled, waiting for cache lines to be filled.

Re:The simple future

[danheskett]

You are straight on. One of the major problems with large pipeline chips is that they fill the various cache slots with what is predicted to be needed. If you have a huge cache pipeline, and the CPU thinks its going to need certain commands in that pipeline, and it in fact doesn't, all those prefetched bits are wasted, and the CPU is suddenly memory bound for a huge number of cycles. On systems with tons of CPUs, this is a very large problem. If you have 32 CPUs, and 15 of them incorrectly predict which branch will be taken in a single cycle, suddenly all those CPUs are going to be choked.

Re:The simple future

[Ignignot]

Why stop at dual core?
Once a way to link multiple cores of a CPU is firmly implemented scaling the chip to 4, 8, or even 32768 cores should be relatively easy.

How about because the benefit from adding a 2nd core is not a 2x speed increase. You get diminishing returns as you rapidly run out of scalable instructions. Besides the architectural nightmare of a multiple core setup, there is the economic factor. Will you pay 4, 8, or 32768 times as much for a 10%, 13%, or 15% speed increase over a dual core? The problem isn't the hardware, the problem is that software is often inherently singlethreaded and cannot be changed from that.

Re:The simple future

[mobiux]

I am waiting for intel's 32768 core processor.

iirc, the p4ee dual core puts out 225 watts at full power.
That would make the chip putting out roughly 3686400 watts, or 3.686 megawatts.

It's too cold on this planet anyway.

Re:The simple future

[jd]

IIRC, the latest generation of Sun UltraSPARCs has 6 CPU cores. An alternative approach is to have "virtual" cores - have a stack of registers and pools of computational elements. This does require some extra element of sophistication, to share out resources, but if you have two programs with very different CPU needs, both programs should run faster. Also, if you have fewer programs than there are virtual cores, but instruction parallelization can be performed, you still get a speedup.

The idea of turning off parts of the CPU would work, if you have a large enough cache. What you would need to do is prefetch all possible paths far enough ahead that you could turn on any deactivated part of the CPU before the instruction needed to be executed. You then have an independent "monitor" processor (an MPU?) which purely scans the cache and turns off all elements on the CPU that aren't needed within the lifetime of any of the contents of the cache.

Another poster noted the bandwidth issue between processor and main memory. That is certainly a problem, but one that may be fixable. One way is to sped up memory (and the bus). The other is to look at ways of reducing the amount that needs to be transferred, by putting some of the CPU in memory. (The technique is called "Processor-In-Memory", and has been around for about 10-15 years.)

Re:The simple future

[shawnce]

Look at the Power5 which is a combination of dual core dies integrated in a multichip package. It provides for L3 to L3 sharing as well as ring style interconnect I believe.

Amazingly the following is the best info I could find that isn't private...
IBM's POWER5 Chip with 8 cores and 144MB cache showcased

Not exactly what you are talking about but close... of course the cell processor is closer to what you would likely get on a single die at this time given feature sizes and heat issues.

Re:The simple future

[masklinn]

An alternative approach is to have "virtual" cores

You know, the word "Hyperthreading" comes to my mind when i read that...

Re:The simple future

[LWATCDR]

That would not be too bad. Just use the heat to boil water and run a steam turbine to help power the cpu. Let's not let all the heat go to waste after all. Many gas turbine systems already do this:) Of course the CPU will have to run at 100C plus so something besides water might really be better. Like.. NH4... gives the whole liquid cooled cpu idea a whole new meaning.

Re:The simple future

[Deliveranc3]

Each chip has multiple calculation units.

So really the concept has already been implemented, but managing these cores in applications which can't be optimized for them can yield a net loss in speed. So... But the 875 totally smokes even on old benchmarks.

Re:The simple future

[beowulf2003]

Your RCed wrong. The latest version (ultrasparc IV) is a 2 core processor (the two cores being the same as the ultrasparc III) and can run 2 threads (one per core) from http://www.sun.com/processors/. But they are talking about their next generation (which is not ultrasparc) called Rock and Niagara which are supposed to have more than two cores per processor. Some of what you are talking about like having "virtual registers" are already being implemented by techniques like register renaming where the registers visible to a program are less than the actual set of registers on the CPU. This is useful for things like function calls which involve saving the register states, but could also be useful for multiple threads on the same core.

Re:The simple future

[flaming-opus]

Because this is really only useful for a few, very specific tasks. The cell works because all those cores churn over a very small set of data, and do a lot of math without needing to get a whole lot of stuff from main memory. That doesn't work for most tasks.

Server tasks, unlike multimedia tasks, are generally branch-limited or IO-limited. They require large amounts of data, and program code. You can't easily improve memory bandwidth or I/O bandwidth without adding more pins to the back of the chip. If you start doubling the number of pins, well, you start to ask why you bothered putting two cores on a chip.

2 cores, well because it's easy and the chip makers can't really come up with other good uses for the transistors. 4 cores - I'm sure it'll happen, trends don't die quickly. 8-cores, you're gonna need a lot more memory buses, I/O connections, clock pins, ground pins, etc. I'm sure they'll find clever ways to put 2000 pins on a chip, maybe 4000, but you scale this up too high, and the chips will be too fragile for users to install themselves. Sooner or later the benefits start diminishing, and the costs jump above the point where a more complex motherboards ends up being easier.

Re:The simple future

[TeknoHog]

That would make the chip putting out roughly 3686400 watts, or 3.686 megawatts.

Such power radiating from a square centimetre requires a die temperature of about 28000 Kelvin (using Stefan-Boltzmann law :)

As for gamers (from TFA)

[ahsile]

Notice the lack of an Athlon 64 FX version of AMDs dual-core strategy. For the time being, its recognized that games are exclusively written for single-threaded operation and as such run better on single-threaded processors at elevated frequencies. Thus, the FX series marches on at 2.6GHz for now. ... so for games, keep to your single core CPU.

Re:As for gamers (from TFA)

[millennial]

I suppose that makes sense. The question this raises, though, is whether there are any games designed to work better on hyperthreaded/multiprocessor systems.

Re:As for gamers (from TFA)

[phoenix.bam!]

Definitely not on the hyperthreaded system. Hyperthreading is only useful is you have 2 or more low demand threads. The benefit of hyperthreading disappears when 1 process needs 99% of the cpu, like many games. This can even be deterimental as your game will never be able to use the entire processor.

Re:As for gamers (from TFA)

[99BottlesOfBeerInMyF]

I suppose that makes sense. The question this raises, though, is whether there are any games designed to work better on hyperthreaded/multiprocessor systems.

I very much doubt it. I've always thought of Blizzard as being one of the better companies when it came to "doing it right" with regard to coding their games. I know playing Warcraft III it always consumed 100% of one processor and did not put a dent in the other. I have not noticed any games that do a better job.

Re:As for gamers (from TFA)

[DigitumDei]

Depends I guess. I know I don't have the luxury of keeping my gaming machine seperate from all other applications I use, so my gaming machine is also my work machine and it tends to have a lot of stuff running at any given time. Now when playing shooter games I often notice a sudden drop in fps when some service or other decides it needs to do something. A dual core machine would be a lot less prone to this I guess.

Also, from the article. "And although the company says dual-core isn't for gamers quite yet, perhaps it is, only in a different usage model. Alan Dang and I were discussing processor benchmarking moving forward and he came up with the idea that we don't run compute-intensive tasks in the background today because we think they can't be done. However, if a dual-core processor enables a DVD encode while you're playing Half-Life 2: Deathmatch, there's a good chance that the way we think about demanding tasks may change. Even though games aren't currently threaded, the background processes a dual-core processor enables may very well catapult the technology into favor with game enthusiasts."

Re:As for gamers (from TFA)

[John Courtland]

And then detrimental again because both processes share the L1 cache... I don't know if Intel fixed that problem yet, but the cache sharing actually decreased performance compared to a processor with HT disabled while running high-demand single-threaded applications (games).

Re:As for gamers (from TFA)

[Mr. Ghost]

Galactic Civilizations from Stardock has a mode that can take advantage of hyperthreading. Of course it is a turn base strategy game and is able (I assume) to offload a lot of background processing to take advantage of it.

Re:As for gamers (from TFA)

[TheRaven64]

Hyperthreading is only useful is you have 2 or more low demand threads.

Not quite. SMT is useful if you have two threads which make mutually exclusive demands on the processor's execution units, for example one doing a lot of integer arithmetic and another doing a lot of floating point calculations. Additionally, SMT is marginally useful if you have two processor intensive threads, since the cost of a context switch between two threads is less on an SMT system than on a single context system (although the addition of a third thread can significantly reduce this benefit).

Re:As for gamers (from TFA)

[i.r.id10t]

IIRC Quake3 used separate threads for sound and video.

In my own experience with both a dual P2-450 system and a dual AMD 1.2ghz system is that the game will run on one cpu and the OS will use the other... simply balancing the load between both processors. Quake2/3/UT/ET ran on the dual 450s with comparable oomph to a single 1ghz system.

Re:As for gamers (from TFA)

[Malc]

So have there been any multi-threaded games since Quake 2? That was awesome on a dual proc machine. When things in the game got hectic, the framerates didn't drop in the same way they would on a single proc machine.

Re:As for gamers (from TFA)

[Mancat]

That just means that WC3 is not multithreaded.

Re:As for gamers (from TFA)

[adamruck]

I believe that ut2004 is completely multithreaded

Re:As for gamers (from TFA)

[Anonymous Coward]

If you can afford a "gaming machine" that can actually run a modern game you can certainly afford to offload some services onto a another box. Complete PIII class dells sell on ebay for $50-80 everyday.

Re:As for gamers (from TFA)

[tokabola]

Not yet, but there will be. The Playstation 3 and the XBox 2 are both designed to rely heavily on parallel processing. Even the PS 2 did somewhat. The result of this will be that game developers will start thinking in terms of multithreading when writing for the platforms, which of course means multithreaded ports for PC.

This was all mentioned in TFA, perhaps you should have read it before asking silly questions?

Tommy

Re:As for gamers (from TFA)

[tokabola]

The reason games are written without multithreading is because most games are written with consoles in mind. While the PS 2 has a type of parallel processing, neither XBox nor Nintendo support any kind of SMP.

The next generation of consoles from Sony and MS are supposedly going to be fully SMP capable, so game developers will start taking advantage. That will make multithreaded ports to PC a no-brainer.

There's nothing about gaming that makes multithreading less usefull - in fact the need to run a real time physics engine as well as an AI make most games excellent candidates for multithreading. The only reason it hasn't been done is because most hardware (console and PC) doesn't support it and it doesn't make sense (financially) to write a second version of the game that can, just for the 30 or so people who could actually use it. As dual core cpu's, along with the PS 3 and XBox 2, become more common, the software will begin taking advantage.

Re:As for gamers (from TFA)

[niko9]

Or, for games and doing all those things in the background that you would rather leave uninterupted (streamripping, news feeds, dvd/mp3 encoding, folding@home), get a dual core. Check this link. There's only a few fps difference when gaming and having more than a few background apps running on a dual core setup.

Re:As for gamers (from TFA)

[NanoGator]

" The question this raises, though, is whether there are any games designed to work better on hyperthreaded/multiprocessor systems."

Why worry about the past? If these processors take off, new games will support them.

Re:As for gamers (from TFA)

[dcw3]

Now when playing shooter games I often notice a sudden drop in fps when some service or other decides it needs to do something.

Interesting point. I play World of Warcraft on a Athlon 1800+, 512M machine, over cable modem. Whenever any background task launchs (virus scan, skim backup, etc.) I notice a significant lag. I bought a nice new $300 video card, but that was no help...I'm going to try more memory next since the game uses nearly 300M. Other than that, I figure I'll have have to upgrade to a faster machine, but not until I try it on my daughters 3300+ first.

Another poster suggested offloading tasks to another machine, but obviously, you can't do that with virus scanners, backup, and other essentials.

Re:As for gamers (from TFA)

[ZorbaTHut]

City of Heroes is multithreaded.

One of the reasons more games aren't multithreaded is that people generally don't buy dualproc systems for games. As dualproc systems get cheaper and more reliable, this will change.

Even on a game system, dualproc is useful just so you can do multiple things at the same time. Especially at LAN parties - you can run the server *and* play, and each of those runs at full speed. (Which I've done many times - amusingly, people with a single faster CPU still can't do that because the game keeps trying to devour an entire CPU.)

Re:As for gamers (from TFA)

[ageoffri]

Galatic Civ is multi-threaded and shows a good improvement even with a HyperThreading P4.

Re:As for gamers (from TFA)

[Anonymous Coward]

UT2004 is *not* multithreaded. However, the Unreal 3.0 engine they are working on now will be.

Re:As for gamers (from TFA)

[goates]

Unreal 3 will use multiple threads.

http://www.anandtech.com/cpuchipsets/showdoc.asp x? i=2377&p=3

goates

Re:As for gamers (from TFA)

[millennial]

Isn't there an option in the Task Manager for Windows XP on dual proc systems that lets you pick which processor a process should run on?

Re:As for gamers (from TFA)

[wernercd]

Beyond that, people like me who upgrade to near-top-of-the-line computers every year or two can simply keep the OLD gaming computer for 'normal' tasks - and still only need One monitor/keyboard/mouse thanks to KVM switch.

My two year old gaming computer now runs very nicely as a media server.

Go Figure eh?

Re:As for gamers (from TFA)

[ender-]

Isn't there an option in the Task Manager for Windows XP on dual proc systems that lets you pick which processor a process should run on?

Yes, if you right-click on a process, you can set the "affinity" to tell it which processor to prefer.

Ender-

Re:As for gamers (from TFA)

[rbarreira]

Actually this is not true. I was involved in the optimization of the clients of a distributed computing effort (ecc2-109, some of you may remember it), and hyper-threading made the clients about 30% faster.

When I optimized them, the speedup given by HT was smaller, but it was still something like 18% at least...

The reason for this is that all the units in the CPU are more heavily used with HT, which is also why those cpu's heat more than normal P4's...

Re:As for gamers (from TFA)

[rbarreira]

And oh, I should mention that no effort to optimize the clients for HT was done - the speedup given by HT was calculated just by running the same program twice, so both programs were doing integer operations only...

Re:As for gamers (from TFA)

[styrotech]

The question this raises, though, is whether there are any games designed to work better on hyperthreaded/multiprocessor systems.

Falcon 4 (flight sim)

A bit long in the tooth now though.

Re:As for gamers (from TFA)

[2ms]

I thought Id has always made a particular effort to multi-thread. I remember for a brief time dual G4 Macs were the fastest platform for Quake 3.

Re:As for gamers (from TFA)

[dwater]

Games that use OpenGL Performer can take advantage of multiple CPUs very easily. Some of the ones run by Disney do, for example.

full article coral cache

[Anonymous Coward]

article 1
article 2

For the lazy...

[Anonymous Coward]

...and for those who don't want to flip through pages and pages of flash banner ads:

Scientific Computing

MATLAB: Though the script includes a moderate amount of matrix math, it doesn't seem like much of it is parallelized. Our recommendation from two years ago still stands - for most Matlab users, the fastest performance will come with a single Athlon64 line.

LS-DYNA: I will bench the CPUs using two classic tests, a 3-vehicle collision and a single front-collision. The 3-vehicle collision takes more than 24 hours to complete - we do not have these numbers ready for this round of articles.

Digital Imaging

Capture One: With Capture One only supporting two CPU threads, the dual-core Opteron's lower clockspeed is a disadvantage.

Bibble: It took only 4 minutes to complete with the 2x Dual Core Opteron 275. 4 minutes! That's 4.2MB/sec of processing time - a 2x Dual-Core Opteron 275 can process RAW images about as fast as it takes to copy them from to your computer using a standard-grade USB 2.0 CF card reader!

Noise Ninja: On the slower Opteron 246, the fastest results were had with 4 threads, but on the faster CPUs, 8 threads was better.

Video

After Effects: Since the decoding of WMV-HD does not seem to take advantage of both CPUs, the performance gain from the Dual-Core AMD Opterons is virtually absent.

Re:For the lazy...

[UnknowingFool]

So in summary: This would be good in number crunching applications that are optimized for multiple cores, but not good to run multiple programs simultaneously. (i.e. play Half-Life 2 and run a webserver at the same time)

Anandtech article on same subject

[bersl2]

http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2397

All I can say is...

[FlyByPC]

...LONG LIVE COMPETITION!

I wish both AMD and Intel well. All the better for us. Lower prices and better performance.

All windows, all the way.

[eddy]

Yeah, but it's all windows, windows, windows. Sadly, the same is true for the Tech-Report.

Re:All windows, all the way.

[bersl2]

Well, at the very least, Anand put some Cygwin/gcc tests in there.

Most Windows users don't know what it means to multitask. It's much harder to do when you don't have multiple desktops, virtual or otherwise.

Re:All windows, all the way.

[Politburo]

Most Windows users don't know what it means to multitask. It's much harder to do when you don't have multiple desktops, virtual or otherwise.

Speaking as someone who has used a wide variety of systems with multiple desktops and/or monitors, I do believe you are talking out of your ass.

Re:All windows, all the way.

[bersl2]

As you are currently wording it, you are agreeing with me. Did you mean to say, "Speaking as someone who has used a wide variety of system, some with multiple desktops and/or monitors [and some without]..."?

Re:All windows, all the way.

[Politburo]

Maybe you missed this part:

"I do believe you are talking out of your ass."

The number of monitors and/or desktops has nothing to do with multitasking knowledge or ability. They do aid in organization of multiple applications and windows, but are not at all required.

Re:All windows, all the way.

[bersl2]

How many Linux sessions have multiple desktops, whether these be the virtual ones that most any window manager provides for, or actual ones hosted on a different monitor?

Now how many Windows sessions fit the same criteria? Do there exist any statistics on multi-monitor vs. single monitor setups?

The point is that a taskbar and minimization only go so far. If I have 20 different windows open, don't you think that a flat organizational hierarchy isn't very optimal? This is why I like browser tabs, and this is why I like having multiple desktops: everything isn't crammed in the same place; the organization is distributed and contextualized, making it easier to handle.

our chips have dual cores

[yincrash]

all we need is some dip, and we've got some heavy duty crunching!

Re:our chips have dual cores

[corngrower]

Correct grammar would be:

All we need are some DIPS and we've got some heavy duty crunching!

Although most ICs use higher density packaging these days (LQFP, SOIC, PLCC, etc.), and you'ld probably also need a mobo, capacitors, resistors, inductors, and a good power supply.

Wow!

[truesaer]

This is pretty disasterous for Intel. The game benchmarks show significant performance penalties for dual core chips, as expected. Intel launched its dual core specifically as an Extreme Edition for games.

On other benchmarks the AMD dual core gets 10-20% better performance! SiSoft Sandra is an exception, where there is a mixed bag between the two processors.

This pretty much verifies for me that Intel did a seriously rushed cludge to get this thing out the door. The only reason I can think of to target this to gamers is that no OEMs would want to buy them for server or desktop use, so you have to target people who like the latest technology even if it isn't that great.

AMD on the other hand seems to have a pretty good product here. I can't wait until the desktop versions come out.

Re:Wow!

[i41Overlord]

On other benchmarks the AMD dual core gets 10-20% better performance! SiSoft Sandra is an exception, where there is a mixed bag between the two processors.

In the article on Anandtech, they do a pretty good job of explaining this.

Basically, in the past, programs that had multiple threads heavily favored Intel's Hyperthreading chips since they could handle multiple threads at once. AMD's chips lacked this capability. Intel's dual core chips did pick up a performance boost, but they didn't pick up the large boost that AMD's dual core chips did. AMD picked up such a large performance boost because their chips lacked hyperthreading, and now with two cores they're able to handle multiple threads at once whereas they couldn't before.

Re:Wow!

[t35t0r]

AMD single processors can't handle multiple threads at once? What planet do you live on? So you're saying I can't run multiple threads of the same application on an athlon and do it effectively? The CPU will automatically split prioritization and CPU processing power evenly between the two. While this may not be as effective as Intel's hyperthreading technology, I'd take an athlon64 3200+ 939 pin or athlon xp 3200+ over a pentium4 3.2ghz any day, simply because of the fact that I haven't noticed any difference with hyperthreading on or off on the intel systems. All it does is make a fake virtual cpu out of a single cpu and reduce the processing time given to a single thread.

Re:Wow!

[i41Overlord]

AMD single processors can't handle multiple threads at once?

No, they cannot. They quickly switch between threads, but cannot process them simultaneously. Intel chips with Hyperthreading process the threads simulataneously.

What planet do you live on?

Earth. Welcome, visitor.

So you're saying I can't run multiple threads of the same application on an athlon and do it effectively?

And he sets up the strawman, and he misses. No, I didn't say that and nowhere from reading my post could a reasonable person draw that conclusion. I only said that the single core Athlons couldn't process multiple threads simulataneously, but now with dual cores they can, giving AMD's dual core chips a bigger boost in benchmarks than Intel's dual core chips received.

While this may not be as effective as Intel's hyperthreading technology, I'd take an athlon64 3200+ 939 pin or athlon xp 3200+ over a pentium4 3.2ghz any day, simply because of the fact that I haven't noticed any difference with hyperthreading on or off on the intel systems.

Me too, and I did choose an Athlon 64 3200 over a P4.

All it does is make a fake virtual cpu out of a single cpu and reduce the processing time given to a single thread.

It does a lot more than that. It actually improves performance quite a bit, and anyone who has analysed benchmarks would see its effect in multi-threaded tasks.

Educate yourself on the subject before making a post flaming me.

http://www.aceshardware.com/read.jsp?id=50000319

Re:Wow!

[avandesande]

This is a paper launch.. Nobody is going to buy this white elephant so Intel doesn't have to supply them.

Re:Wow!

[pjrc]

This is pretty disasterous for Intel.

What else would you expect from articles paid for by AMD?

Re:Wow!

[ChrisMaple]

Intel chips with Hyperthreading process the threads simulataneously.

My understanding of Intel's hyperthreading is that it is essentially an intelligent fast context switcher running in hardware. It still has to swap out registers and empty or suspend instruction queues. This means that threads are not actually executed simultaneously. AMD's single core threading has to do all that in software or just wait around until stalls are resolved.

Re:Wow!

[Anonymous Coward]

My understanding of Intel's hyperthreading is that it is essentially an intelligent fast context switcher running in hardware. It still has to swap out registers and empty or suspend instruction queues. This means that threads are not actually executed simultaneously.

Nope. Hyperthreading can handle instructions from two different threads at the same time. If one thread stalls on a memory access, that leaves more execution units available for the other thread. It's especially important for Intel because they put their memory controller way off in a separate chip, so their memory latencies are horrible.

Re:Wow!

[truesaer]

I think Hyperthreading is important for Intel more because of their deep pipeline, although the increased memory latency makes it help too. The main issue is that on a branch mispredict or other stall you can end up with a much larger bubble on an Intel pipeline as compared to an AMD pipeline. So, by keeping some extra work hanging around they can keep execution units busy while the bubble passes.

This is one reason why AMD hasn't implemented this yet...they wouldn't see as much benefit, so the effort of implementation is not worth it.

Anyone realize that suddenly all P4's disappear?

[denominateur]

Now, it's struck me as very peculiar that the benchmarks where the dual dual core setup from AMD really shines leave out any comparison whatsoever to the Intel dual-core offering. This begs the question whether the person doing this review is a journalist or a marketing represenatative of AMD.

"We did not have time to evaluate the Intel platform with the Intel MKL, the P4 3.0GHz is an older reference measurement." is a very cheap excuse and indicates either lazyness or bribes on the side of AMD... I hate hardware review sites!

Dr. Dobbs and Threading/Multi-Core

[DanielMarkham]

Dr. Dobbs last month had an item regarding threading in real-world environments. The authur said that while multi-threaded applications run a lot faster than single-threaded applications, that always isn't so. In addition, there are some significant issues in running in a multi-tasking, multi-threaded environment, not solved with the use of mutexes and semaphores.

Multi-threading and mult-cores are definitely the way the industry needs to go, but the current development methodologies and application architectures (as well as computing theory) may need to catch up a bit.

Re:Dr. Dobbs and Threading/Multi-Core

[zx75]

The computing theory and architectures are already there. Now that Java has finally jumped on the bandwagon of reliable multi-threading with v1.5 (or v5.0, or whatever the hell they're calling it today), chances are unless you're using really legacy code the language will have the appropriate system calls available to it.

The difficulty is that in order for multi-threading to be worthwhile, a developer really needs to know their stuff. It is not easy, there are a number of things that must be taken into consideration that simply do not occur in single-threaded programming. A programmer who just picked up a 'C++ in 24 hours' book is most likely not going to have the tools available to them in order to handle or understand the complexities of multi-threaded programming.

That being said, there are many situations where multi-threading is not appropriate, but if you think the theory needs to play catch-up, you might be surprised at how common it is in professional development.

Re:Dr. Dobbs and Threading/Multi-Core

[ldspartan]

How has Java changed in regards to threading in 1.5? Unless it's on the Sun JVM side, I'm unaware of any language changes in 1.5 that effect threading.

Thanks!

--
Phil

Re:Dr. Dobbs and Threading/Multi-Core

[DanielMarkham]

I believe the point was that there are transactional qualities associated with complex object heirarchies that are not adequately captured by current language and programming paradigms.

Certainly, language constructs and techniques handle the simple cases. The issue is both scalability and sharing. "Really knowing your stuff" may or may not be adequate. For instance, OOP does a great job of separating common pieces of data and functionality, but various pieces of data and functionality can be executed in an ad-hoc manner by outside callers, creating various race or lock conditions that are not evident to either programmer. This is not due to poor programming, rather the need to look at depolyed functionality in a new way.

I hope that help explain it. I could write the 3-thousand-word version of this if you like. But only if I get paid for it! (grin)

Re:Dr. Dobbs and Threading/Multi-Core

[zx75]

It does to some extent, but I guess I keep looking at it from a different perspective.

OOP as a design method has tremendous advantages, but you're right, it doesn't address many multi-threading issues. Its appropriate design and understanding accessibility issues that allow for multiple thread access. It must be in part due to the fact that I've spent the last couple years doing real-time device level programming, kernel level programming, and a number of complex client-server projects that make multi-threading natural for me.

For kernel level stuff, if you need specific functionality or specialized resources for something important you just write a process administrator to manage it and spawn worker threads as necessary. Once that sort of thing becomes second nature, multi-threading simply becomes a matter of syntax. But without my now 5 years of university, I doubt I ever would have reached the same comfort level with it.

Re:Dr. Dobbs and Threading/Multi-Core

[zx75]

AC answered your question probably better than I could, but I'll put it very simply.

Java 1.5 added a Monitor that works. The major flaw in all previous version of Java was that its multi-threading was non-priority blocking (most of the time) and to get things to work as expected you had to be very careful with conditional loops, and even then it was easy to starve a thread if things weren't perfect.

Monitor that work mean that we can have reliable priority blocking multi-threading constructs that require minial work to implement correctly, and thus make complex multi-threaded Java applications practical and easy to implement (without having to worry about a lot of niggling details).

In other news ...

[Luscious868]

Dell still won't sell servers with them ....

Re:In other news ...

[Silverlancer]

Can't you put these components together yourself and save the $5K surcharge?

Yes.

Re:In other news ...

[orionware]

What you are often paying for is someone's ass to ride when it breaks and actually get them out there that day to fix it. Idiots in the corner offices also think that since it's more expensive, is MUST be better quality and more reliable.

Freaking Duh! :)

Re:In other news ...

[barc0001]

But IBM will...

So who cares? Dell is synonymous with low-end crap to a lot of people anyway. It's hard to maintain an image of your servers being the upper end of Tier One quality when you're selling Dimension 6whateverthousand desktops with Genuine Intel leftovers on TV spots for $449...

I mean, nobody would ever dream that KIA would make a dependable, bulletproof car after watching one of their ads where they practically beg you to find one of their cars worthy of spitting on.

Dell seems to be content to shoot for the lowest denominator while sucking up to Intel. If that keeps them happy, great. I just won't be buying their crap.

Re:Anyone realize that suddenly all P4's disappear

[Martin Blank]

Check out the article at Anandtech linked a little bit above, which does have comparisons with the Pentium lines. The results are still clearly in AMD's favor.

Anandtech has some cost comparisons/benchmarks

[IronChefMorimoto]

Anandtech has an AMD dual core Opteron and Athlon64 X2 article that might compliment the original poster's story pretty well. It has a sh*tload of benchmarks:

http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2397

I really wish they wouldn't do gaming benchmarks with an Opteron in stories like these. Just because the Opteron used has similar specs to the dekstop processor that hasn't been released doesn't necessarily mean that the gaming benchmarks are all that useful. Just my 2 cents.

It'll be interesting to see how soon prices fall for these AMD processors (server and desktop) when they go mainstream. Read the cost comparisons for these badboys in the article.

Finally, I'm glad that Anand decided to demonstrate that the new AMDs will be backwards compatible with Socket 939 motherboards WITH BIOS revisions. Intel's dual core processors don't offer that luxury, from what I read in the article.

IronChefMorimoto

why not more than two core?

[kris_lang]

Okay, let's think about two things:
the limited three dimensions of space, and the limited number of layers you can put on the chip wafer as it is fabricated. The limited number of layers on the wafer is a simple concept to get. The limited 3-d space to work in also limits interconnects between multiple processors in clusters by limiting the topology which the cluster can form.

The uber-cluster concept was in the Thinking Machine (TMI) something-or-other which had 1024 processors linked together in (effectively) a ten-dimensional-hypercube. When you've got clusters currently, they've got to be networked and have an efficient and rapid method for passing messages and data. If you go mebbe to four processors, I can see the interconnects fitting on the wafers tightly. Maybe even to eight...

But beyond that, you start having difficulties maintaining direct interconnects between processors. The Thinking MAchines supercomputer effectively implemented 10 interconnects for each processor by having each processor (defined / labelled as a 10-bit address) connect to each processor whose address differed from its own address at exactly one bit. So a message could pass from any of the 1024 processors to any other of the 1024 processors in at most ten steps. And like the internet fabric, there were redundant pathways for the message to take. (fairly cute bit flipping algoritms for it too).

But I don't see compressing the network connections as easily for a hypercube on a wafer.
But maybe a simple linked line of processors.

hmmm...

Re:why not more than two core?

[ChrisMaple]

the limited number of layers you can put on the chip wafer as it is fabricated

Over the years, the number of layers possible has grown. If crystaline silicon can be grown on top of a few layers of interconnect (new technology, AFAIK) then true 3D construction would be possible. Flaw density would be likely to rise (reducing yield), but shorter interconnects would help with power dissipation.

there's no conspiracy.

[i41Overlord]

Now, it's struck me as very peculiar that the benchmarks where the dual dual core setup from AMD really shines leave out any comparison whatsoever to the Intel dual-core offering.

They couldn't test a dual core multiprocessor chip from Intel because one doesn't exist yet. They've only released single processor dual core chips so far.

AMD introduced dual core on their multiprocessor server chips first, with desktop chips coming later on. Intel introduced dual core on their single CPU desktop chips first, with server chips coming later on this year, or in early 2006.

The problem is that you can easily run a single multiprocessor-capable CPU in a system, while you cannot run two single-processor-only chips in a system since they lack that capability.

I'll believe it when I see the BIOS!

[Anonymous Coward]

Given ASUS has not updated the socket 940 SK8V BIOS in over 6 months, and hasn't even had a new beta BIOS (yuck) since December, what are the odds of any motherboard really supporting these CPUs? How many companies rush to support the older boards?

In fact, the SK8V is suddenly gone from both the motherboard page and the retired products page on the asus website. Hmmm...

Re:I'll believe it when I see the BIOS!

[Renraku]

I wish compotent people worked on BIOS revisions.

The newest BIOS revision for the K7MCD board increases the temperature by 15-30F for no reason. It doesn't really do anything more than the last release of the BIOS, either.

Is this article worth a darn?

[GweeDo]

With comments like:
"Even grandmothers own 8-megapixel consumer digital cameras now"

I really have to question the intellegence of this poor guy. I don't know many grandma's that drop $700-$1000 on digital camera's.

Re:Is this article worth a darn?

[necronom426]

Does the average person really need more than a 3 or 4MP camera anyway? I have a 3MP camera and I almost never have it on the maximum size setting.

---
Do a good deed for today - Click here to help me get a free Opera licence

Re:Is this article worth a darn?

[hawk]

Actually, they buy them for their kids--you get more 8 megapixel pictures of the grandkids that way :)

hawk

Re:Is this article worth a darn?

[necronom426]

A few times. I sometimes use the maximum size as a zoom, then crop it afterwards. I've tried printing lower res pics and they look almost the same as the higher res pics when printed on my Epson Stylus 680 at 4x3 inches. Maybe my printer isn't very good.

For the screen I obviously don't need anything like 3MP.

Re:Is this article worth a darn?

[necronom426]

See my other reply, but another reason is that the big pics eat up memory cards like crazy. I can happily snap away for hours with a smaller size. On maxumum I run out quickly.

I should probably get a bigger memory card!

Re:Is this article worth a darn?

[Vicente Gonzlez]

My Mother is a Grandma, and she does. In her mind there is nothing wrong with a good ol' $900 camera, of course she also has a $800 normal camera which is far superior.

Linux/GNU/Gnome memory usage

[Zan Lynx]

On my Gentoo AMD64 -O3 compiled system, running Gnome, Rhythmbox, Evolution, Epiphany, Liferea and GAIM, it is using 365 MB of RAM, not including buffers or disk cache.

In contrast, Windows XP running a similar set of applications was only using 230 MB.

Re:Linux/GNU/Gnome memory usage

[GreyWolf3000]

1) 64-bit binaries are usually between 20% and 50% larger than their 32-bit counterparts. Windows XP doesn't use 64-bit binaries.

2) Don't use -O3, use -Os.

3) Memory reporting isn't always accurate (as others have stated).

Re:Linux/GNU/Gnome memory usage

[Zan Lynx]

I do know what I'm doing.

I like -O3. -Os is for people with a tiny cache.

Memory reported by "free -m" is accurate. Adding up the memory reported by ps or top is not.

Re:Linux/GNU/Gnome memory usage

[GreyWolf3000]

Well, as far as optimizations go, compiling Gnome with -O3 really won't give a significant increase. Most of the applications just sit around waiting for input, and when they do something, the difference between -O2 and -O3 for them is negligible (the -f options tend to do more, but you have to expirement with them to get them right for each application). -Os will help startup times, and decrease your total footprint.

opterons and gaming?

[ahmetaa]

this enthusiat hardware web sites are simply ridiculous. how can you make a decent test with desktop applications and server processors and systems and reach conclusions? evet their DB tests suck.

Re:HEAT

[shawnce]

Yes and no... it depends basically.

Dual core (multi-core) dies are often larger then their single core siblings (not always depending on how the do single core version) and this gives you a larger surface area to thermally bond with as well as spreads out heat production. This can make it easier to cool even if it puts out more heat.

Emacs in 6MB???

[hawk]

>With that comparison, my Emacs session is 6MB.

Oh, come on. It took more than that 20 years ago :)

More seriously, I just launched one, and it immediately was using 9.5Mb on FreeBSD. It then did a couple of things on its own (???), displaying a notes buffer, and hopped up to 9.5Mb.

I shudder to think of what it will do if I actually type anything in its window . . .

hawk

Re:Emacs in 6MB???

[ksheff]

I launched Emacs when I got into work today and while I haven't used it as much as I normally do, it's only taking up 6.9M. Galeon is using about 64M.

Re:Emacs in 6MB???

[Chris Burke]

Dude, I'm not just talking out my ass. I checked what emacs was using right now that I've been using to develop code for several weeks. Okay, I did kill the buffer containing the 75MB of debug output, but other than that 6MB was a legitimate snapshot of memory usage.

Re:Emacs in 6MB???

[hawk]

Now you've got my curiosity up. Is this in X or a consle/xterm? And what version. Mine is 21.3 on FreeBSD (the default). When I've noticed a difference, it tended be that FreeBSD has used slightly less, rather than significantly more, memory than Linux.

hawk

Re:Anyone realize that suddenly all P4's disappear

[Deliveranc3]

I'm an AMD fan boy I was thinking it would smoke the Intel too hard.

The dual core is mostly tested in encoding an area where Intel has traditionally dominated. And it still kicks intel's ass.

uh-huh, keep going.....

[gosand]

Currently, 50 processes. The two highest (memory and VM wise) are Thunderbird which is using (60mb of main memory) and Firefox which is using 55mb of main memory. All the microsoft products I'm running like Visual Studio.NET 2003 are WAY down the list as none are using more than 10-15mb of main memory.

Did you just count those processes, or all of the shared DLLs that are loaded, explorer, etc that is necessary to run them?

Re:HEAT

[greenreaper]

AMD ones appear to have the same Thermal Design Power (about 90W) as their single-core chips.

Amateurs

[Jah-Wren Ryel]

How can anyone take an article seriously when the very first sentence just screams, "AMATEUR!!" like this one does:

Intel may very well go down in history as the first processor manufacturer with a dual-core solution, if only by three days.

IBM Power4, Power5
HP PA-8800
Sun Sparc IV

All full-fledged dual-core processors shipping long before Intel -- HP's been shipping for over a year and IBM's already well in to their 2nd generation of dual core processors with Power5.

Sure, you can excuse the author with some hand-waving about x86 context only or whatever. But if they really knew what they were talking about, they would have said it that way - or at least a competent editor would have corrected it. If these guys can't even get the trivial stuff right, how can anyone trust them to get the real technical details right?

Re:Amateurs

[leoc]

Another thing that pisses me off is that he tests these 64 bit CPU's with 32 bit Windows, claiming that Linux is "hardly mainstream".

What a load of crap.

These dual core chips are PERFECT for high performance NON-GAMER Linux systems, and yet these guys disregard the most mature and stable 64 bit platform to run game benchmarks on 32 bit windows.

Re:Amateurs

[Loki_1929]

"These dual core chips are PERFECT for high performance NON-GAMER Linux systems, and yet these guys disregard the most mature and stable 64 bit platform to run game benchmarks on 32 bit windows."

That could, perhaps, be due to the fact that 90%+ of their readers are only interested in benchmarks performed on a platform they run with applications they run.

If you're convinced there's a market out there for a site that does benchmarking with non-gamer linux systems, feel free to start your own site and do your own benchmarking. Until then, it does you no good to complain about the articles and reviews of successful sites not covering your small corner of the world.

Re:ok - heres the answer -

[sehryan]

Actually, no it isn't. The moron kid's friend is the father, so its his DNA.

Why is parent modded funny?

[Vaystrem]

Seriously. This is the level of usage that I'm trying to achieve, not UT (World of Warcraft). But I do want to be able to queue up a few torrents, rip and encode my movies (for a media centre setup) and play a game at the same time while running Itunes (Decoding lossless).

Back in the day, and I have posted this before, when I had a friend running an Abit BP6 (Dual Celeron board) he could burn a CD, encode MP3s, host a dedicated UT server and then join it. Dual Cores / Dual CPU setups have always been for us hardcore users and this is a remarkable step forward.

Here is the Anandtech AMD Dual-Core Review (I prefer it slightly to the FiringSquad articles)
http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2397Anandtech

Here is Part I of the Intel Dual Core Article
http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2388
Here is Part II of the Intel Dual Core Article
http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2389

That should provide some useful information.

Pricing

[nns6561]

It strikes me odd that AMD is pricing their Athlon x2's between $500-$1000. This seems particularly high when compared to the prices expected for Intel's offerings. The Pentium D's are supposed to be between $250-$500. I cannot see AMD getting that much higher of a price for similar processors.

Re:Pricing

[Nasarius]

They're Opterons, which work in SMP systems (4 dual core CPUs = much goodness). AMD should have the "eXtReMe g4m3r" [Athlon 64 FX] editions soon enough, followed by the usual consumer CPUs [Athlon 64], which are much less expensive than the Opterons.

Re:Pricing

[nns6561]

I wasn't referring to the Opterons. I can understand why they might be priced at $1000-$3000. They are competing against Xeons. I can't understand why the Athlon x2 64's are priced so high. They will soon be competing against Pentium D's for which the prices are suggested to be about half of the Athlon's current prices. AMD does not have the ability to charge that kind of premium for equivalent chips. One of the reasons AMD has done so well is that their chips generally outperform Intel's at similar prices.

Re:Pricing

[Loki_1929]

"It strikes me odd that AMD is pricing their Athlon x2's between $500-$1000."

Why? This is typical of the high-end processor line.

"This seems particularly high when compared to the prices expected for Intel's offerings."

You're judging AMD's pricing scheme against processors Intel hasn't even delivered yet? Aside from that, you're comparing apples and paint. AMD didn't go el-cheapo by strapping a second core on their CPUs using aluminum foil and a 'bungie' cord. While it may seem 'cool' to have McGuyver rigging processors for your multi-billion dollar a year company, the massive thermal envelope on those space-heaters is anything but.

"The Pentium D's are supposed to be between $250-$500."

Sorry, what's the pricing on those "dual core" EEs again, please?

"I cannot see AMD getting that much higher of a price for similar processors."

Similar in what way? I mean, I suppose if you boil it down to the subatomic level, they're virtually identical. That said, AMD's dual-core solution is a ground-up design intended to bring the best possible performance using existing boards within a reasonable thermal envelope. Intel's design is a rigged, last-minute rush job that spews heat and chews power like nothing you've ever seen. Coupled with the fact that the performance we've seen from their "dual core" "solution" has been less than stellar thus far, it's pretty obvious that AMD's planning (from 1998) is really starting to pay off.

It's looking like Intel's going to play another year of catch-up to AMD. If this trend continues, Intel may one day continue to exist only to serv as a warning to others.

Re:Pricing

[Chris Burke]

Intel is offering lower-end dual-core chips with relatively low clock frequencies. AMDs dual-core desktop chips will be a couple speed grades behind the single core, but are otherwise top of the line. What I imagine is happening here is that AMD wishes for their parts to be "boutique" sort of like the FX line at first, and then as those parts move down the line to being lower end chips (as faster ones are released) the prices will drop.

Remember that AMD, unlike Intel, probably can't afford to sell gobs of the higher-cost lower-yield dual-core chips, so selling fewer of them at a premium makes sense.

Depends...

[phorm]

even when I game I tend to have background tasks going (downloads, background services, etc)... so in such a case wouldn't it balance the load of the game on one and the background apps on another?

A better analysis on Anandtech

[Hack Jandy]

http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2397

As usual.

Um, think again...

[Svartalf]

Just because it'll priortize, etc. doesn't mean that it's running the threads simultaneously which is what "at once" actually means. The only way is to have Hyperthreading or SMP for that. In the case of the SMP machine, it'll priortize the threads and divvy them up across the CPUs/Cores on the machine, to be executed as in-parallel as is possible.

On a non-Hyperthreading, non-SMP machine, it's going to execute only as fast as the one-legged man is able to get to kicking asses...

Re:Uhh....RTFA

[fitten]

Has nothing to do with affording parts... maybe labor. Review sites are shipped units to test on loan and the site has to ship it back. If Anand only got shipped two chips, then, that's all they can review.

I imagine most folks are like me and want to know how two cores stack against each other when in the two single core vs. the single dual-core flavors. This lets me know which has more performance and gives a little insight into how well the integration of the dual cores was done. Basically, these numbers would tell whether I should buy a dual socket motherboard with a single core in each or a single socket motherboard with a dual core in it. Comparing the performance of 8 cores (four dual cores) vs. a quad single-core machine isn't too interesting other than to just see bigger/smaller numbers.

And yet the GHz are not optimized...

[gatkinso]

...for a Java virtual machine, let alone an x86 segmented architecture. Only time will tell if superscalar chip fabs are O(log n).... but I seriously doubt, SERIOUSLY doubt! that **any** pointer arthimetic will be performed faster on such a set up.

Re:And yet the GHz are not optimized...

[Slashcrap]

Ha ha!

10 points for including meaningless technical buzzwords with no context.

10 points just for the phrase "Only time will tell if superscalar chip fabs are O(log n)..."

And 50 points for getting some fucking retard to mod you up.

Congratulations!

Re:And yet the GHz are not optimized...

[tekcsound]

And (log n) points to you for being quite possibly the most sarcastic person on /.

My hat's off to ya!

We have this...

[Inoshiro]

"An alternative approach is to have "virtual" cores - have a stack of registers and pools of computational elements. This does require some extra element of sophistication, to share out resources, but if you have two programs with very different CPU needs, both programs should run faster. "

It's called hyperthreading. It sucks because multiple threads doing different things isn't the common case (unless you run BeOS). It also sucks because the L1 cache gets thrashed a lot, which is why it's actually bad to have it on for some activities.

Re:HEAT

[default luser]

No. The TDP for the .09 micron dual cores is the same as the TDP for the .13 micron single cores.

The TDP for the Winchester cores (.09 micron single core) is 63w.

So, take into effect that is is two Winchester cores with half the memory controllers (less power usage), plus the efficiency gains of a mature .09 micron process, and you've got around 90w TDP rating for the higest-speed chip they plan to introduce.

I seriously doubt 2.2GHz will suck 90w. My 2.0GHz Winchester uses less than 40w max. I fully expect them to hit as high as 2.6GHz in later releases of this core revision.

AMD desktops outsold Intel in 2004.

[JackAxe]

Is this real enough for you?

http://www.google.com/search?client=safari&rls=e n& q=AMD+outsells+Intel&ie=UTF-8&oe=UTF-8

Over here: Apples compared to: Oranges

[Anonymous Coward]

What is this supposed to tell me?

How does a dual core opteron stack up against a dual core pentium? Seems a simple question. But I guess they thought a better comparison would be to compare two dual core opterons against a single dual core pentium...brilliant.

This just re-inforces my belief that review sites never tell you anything of any use. Horrendous graphs (like showing scores of 1660 vs: 1640, but starting the graph at 1600 so the 1660 looks 150% times better than 1640, instead of 2%).

The hyperbole also gets to me....chips/systems are always 'blowing away the competition' when they are like, 5% faster. Hey...get real. When I upgraded my P166 to an Athlon 2400+, did I really care that it might have been 5% faster/slower than it's competition? It was over 10x faster than the system it replaced! Let's get some reality in here....

But of course, we know it's all about scoops and page views for advertising anyway...integrity is gone, if it ever existed in the first place.

the
canonical
anonymous
coward

That's blotted!

[budgenator]

my ed session only uses 16KB!

YAR - yet another review...

[cwg_at_opc]

found at The Tech Report:

http://www.tech-report.com/reviews/2005q2/opteron- x75/index.x?pg=1/

why benchmark a $850 processor?

[iamhassi]

That gaming article is currently completely worthless to average PC builders, who wants to buy a system with CPU that costs more than the rest of the entire system anymore? Why couldn't they benchmark a less expensive Opteron processor on games, like the 242 or 244, both under $200 right now.

I'd also love to see how well these overclock, since I have a athlon64 3000 (1.8ghz) that does 2.5ghz, which would put it at about 4100 AMD's PR scale.

If a ~$150 1.6ghz 242 can do the same speeds as the ~$700 2.4ghz 250, a 50% overclock, then we might have another Celeron 300A on our hands :)

Also at Hexus.net...

[cwg_at_opc]

hope this link works:

http://www.hexus.net/content/reviews/review.php?dX JsX3Jldmlld19JRD0xMTI5/

Common misconception in the article...

[Shanep]

There used to be talk that the V. 34+ speeds of 33.6 kbps represented the very fastest that US phone lines could ever handle. It just wasn't scientifically possible to go any faster.

The 33.6 limit was NOT due to the phone lines, it was due to the equipment at the end of the phone lines at the telephone exchange.

At the telco end, telephone audio would be converted to and from analog and digital so that an analog signal could be presented to your phone, but on the other side a digital signal could become part of a digital packet switched network to efficiently be routed to where your phone call needed to go.

Upgrade that equipment or allow the elimination of conversion between analog and digital and you upgrade the limits. The limits of the copper lines themselves are much higher than 33.6kbit/s.

Re:Anyone realize that suddenly all P4's disappear

[Deliveranc3]

Because there are no motherboards for dual 860's!

Intel has no need to produce such as the chip is available in such low quantities :(