Beta

Slashdot: News for Nerds

×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Quad Core Battle, Intel Yorkfield vs AMD Altair

CmdrTaco posted more than 7 years ago | from the four-is-way-more-than-three dept.

172

Joe writes "Yorkfield Extreme Edition based on the 45nm Penry core architecture will meet heads-on with AMD Altair based on the 65nm K8L core in Q3 2007 as reported by VR-Zone. Due to its advanced 45nm process technology, Yorkfield XE is able to pack a total of 12MB L2 cache (2 x 6MB L2) and still achieving a much smaller die size and higher clock speed of 3.43-3.73Ghz. Yorkfield will feature Penryn New Instructions (PNI) or more officially known as SSE4 with 50 more new instructions. Yorkfield XE will pair up nicely with the Bearlake-X chipset supporting DDR3 1333, PCI Express 2.0 and ICH9x coming in the Q3 '07 timeframe as well."

cancel ×

172 comments

I for one... (5, Funny)

Demanche (587815) | more than 7 years ago | (#16305413)

I for one... Will... wait for those 80 core CPU's intel said they will have in a 'few' years. I'll refuse to upgrade till I get one! :D

Re:I for one... (2, Interesting)

CastrTroy (595695) | more than 7 years ago | (#16305481)

Well, I recently (In June?) just bought a new computer after having my previous one for 7 years. If my current one lasts as long as my last one, I could very well be upgrading to the 80 core CPUs that Intel said would be ready in 5 years. If you buy something good with lots of room for expansion, and take good care of your computer, you shouldn't have to replace it every 2 years.

Re:I for one... (2, Insightful)

jimstapleton (999106) | more than 7 years ago | (#16305537)

I don't think I've ever replaced my computer.

Rolling upgrades for 10 years or so. Never more than half the computer has been replaced at any one time.

Re:I for one... (1)

CastrTroy (595695) | more than 7 years ago | (#16305635)

Well, when I say same computer, I mean same Motherboard and Chip. I had replaced the video card, added a hard drive or two, some RAM, and a CD Burner and DVD drive onto the original. I don't really consider it the same computer if you replace the CPU an Motherboard. I hope my current CPU and motherboard combination last another 7 years. BTW, it's AMD64 3200.

Re:I for one... (1)

jimstapleton (999106) | more than 7 years ago | (#16305731)

ok, two motherboard replacements in that time...

the first died because it had a disagreement with the PSU, and they both lost. Antec + Tyan = bad.

The second... Well, you buy cheap crap, you get what you pay for...

Re:I for one... (4, Interesting)

michrech (468134) | more than 7 years ago | (#16306211)

Antec + Tyan = bad

In my higher than average experience, Antec = bad. I couldn't believe how many of their P/S's (all above 400w) I've had to send back compared to the cheap-oh CoolerMaster 350w supplies we were using. Got to a point where Antec tech support number was being answered by a voicemail (we couldn't get a live person any longer).

When it got to a point where it was taking them *weeks* to get back to me (if they ever did at all), I got fed up and sent an email to the complaint email link they have on their support page (Yes! A complaint link! Only company I've ever seen that *needed* one due to such poor tech support!). In the email I stated my position, that I had a handfull of supplies I was going to dumpster because I could not get anyone to respond to me, and that I'd be reccomending *against* anyone using anything Antec again.

Long story short, someone actually replied fairly quickly, dragged me along for a couple weeks telling me how he'd get me help (he had me fill out an RMA form several times because he claimed there were errors, then refused to take the supplies back because I didn't have a recepit available (it got filed away and sent to storage) even though they were *well* within the manufacturer 3 year from the date of manufactuer instead of date of sale!)..

I've since switched to Enermax or Thermaltake and never looked back. Never had anywhere NEAR the quantity of P/S's to send back and of the FEW I have, it was taken care of right away.

In short, Antec can kiss my ass.

Re:I for one... (3, Interesting)

mrmag00 (200868) | more than 7 years ago | (#16306537)

I agree 100%. People praise Antec, but from my experience they consistantly fall short of their performance and have a comparable lifespan to, say, a $20 no-name PSU.

Re:I for one... (1)

michrech (468134) | more than 7 years ago | (#16307179)

I agree 100%. People praise Antec, but from my experience they consistantly fall short of their performance and have a comparable lifespan to, say, a $20 no-name PSU.

With your message (and it's positive moderation), and the moderation on my message, I'm glad to know I wasn't alone. I was half expecting a bunch of Antec "fanbois" to come in and mod me into the ground. :)

Re:I for one... (1)

El Torico (732160) | more than 7 years ago | (#16307559)

Antec has fanbois? I had a power supply die after 9 months and I haven't gotten a replacement for it from Antec yet.

Pretty cases and crappy power supplies. No, no Antec fanbois here.

Re:I for one... (1)

Khyber (864651) | more than 7 years ago | (#16307395)

Anyone with a brain only buys Antec for the massive cases and cooling/expansion potential, not to mention they're fairly cheap. Fuck any PSU that comes with a case, I've blown ALL of them.

Re:I for one... (1)

DarkSarin (651985) | more than 7 years ago | (#16307533)

I don't know what is typical, but I've never had trouble with any Antec product, (I have an Antec case and PSU). I realize that I am dealing with a limited sample, but all the same, I've not had any trouble.

My experience with their customer service (my children busted the front USB ports on the case) was positive and reasonably responsive.

Of course, I wasn't dealing with them in volume, and I have only had the one incident, so I may have gotten the one good guy, but I don't know.

Just my experience.

Re:I for one... (1)

Aardpig (622459) | more than 7 years ago | (#16307545)

Funny you should say that, but I plugged an Antec PSU into a Tyan S2895, and the fucking mobo actually caught fire -- a little yellow flame, and lots of smoke.

Re:I for one... (1)

lymond01 (314120) | more than 7 years ago | (#16306403)

You say that "if you take care of it" a computer can last you 7 years. You could say that about a car, perhaps, but if you want to stay current with software updates (not even upgrades), you'll need something more than a Pentium II 400 MHz computer. If you play modern games, it won't cut it. If you run Windows, the only OS that is remotely secure is Windows XP SP 2 fully patched, and I won't say that's secure without a 3rd party firewall. You can't run XP SP2 without 512 MB of RAM (it's *possible* put hardly efficient).

So if your 7 year old computer wasn't on the internet, or was running a fully patched linux install and you mostly just did text coding, then sure...7 years is fine.

Crack smoking? (1)

Gr8Apes (679165) | more than 7 years ago | (#16306565)

You've got to be kidding - a secure OS?

The only way to have a secure computer is to have it separately firewalled from the net for worms, and to run with a lowest user priviledged account, using non-MS software.

Modern games are another ball of wax, and I've actually gotten to the point of creating a separate OS installed partition for any new games.

Re:Crack smoking? (1)

Neil Hodges (960909) | more than 7 years ago | (#16307481)

The only true way to have a secure computer is to not connect it to any network and not have any media connected to it or put in the drives.

Re:I for one... (1)

CastrTroy (595695) | more than 7 years ago | (#16306589)

Ah yes, but you can run VectorLinux, or Mandriva with Fluxbox/XFCE on a P2-266 with 512 MB of RAM. Even KDE wasn't that slow. Sure you can't play the latest games, but I never said I did. That's what I was running on for the last 7 years (RAM was upgraded, I think it was originally 64 MB running windows 98). Windows always ends up raising the processing requirement without any reason. Just look at the hardware requirements for Vista. Sure you can say look at the nice new Aero desktop, but then compare that to the 3D Compiz Window manager in Linux and look at the specs required for it.

Re:I for one... (1)

MrHanky (141717) | more than 7 years ago | (#16307283)

Games? Of course not. But browsing the web, sending email, writing documents in OpenOffice.org, etc., all work well on a ~300 MHz computer, as long as it has enough RAM. Yes, even OpenOffice.org 2.0. A fresh XP install doesn't need more than 256 MB to run efficiently for simple tasks like that (it will boot with 128 MB), and will stay good until you install loads of crap on it.

Re:I for one... (1)

Doctor Memory (6336) | more than 7 years ago | (#16307151)

when I say same computer, I mean same Motherboard and Chip
I can see your point, but I'll respectfully differ. I bought a motherboard with the "new!" AGP slot and a top-of-the-line AGP card. I eventually wound up replacing the motherboard (+CPU +memory) with one with a "legacy" AGP slot. I was able to re-use all my other components (except for the too-lame-to-die ISA Ethernet card, but the new MB had Ethernet built-in). Had to do a little device driver tweaking, but everything worked without too much hassle. And it was just like you'd expect: like some little gnomes had come in and turned the CPU up to 11.

Re:I for one... (1)

AceCaseOR (594637) | more than 7 years ago | (#16305653)

My next upgrade is the closest I'm getting to "replacing" the computer - I'm upgrading the MoBo, processer, video card, and probably getting an sound card instead of using the onboard one.

Re:I for one... (0, Redundant)

Hoi Polloi (522990) | more than 7 years ago | (#16307349)

I've been happy with my Creative X-Fi Music board. I use it with 5.1 headphones (great for gaming). Very good quality sound.

Re:I for one... (1)

Aladrin (926209) | more than 7 years ago | (#16305755)

I had the same experience up until a lightning strike totalled my system. Not even a hard drive or NIC left. -sigh-

Re:I for one... (1)

Demanche (587815) | more than 7 years ago | (#16305931)

Jokes aside, I also purchased a dual core computer back in Feburary. I can't say im too impressed with dual core because I am maybe a minimalist multitasker. I think my ideal cpu would be have to be customized really, perhaps that will be the big market once people get tired of cramming in cores.

Imagine this senerio:

You pay X dollers for X amount of on die cache You pay X dollers for each additional core over 2 you pay X dollers for a custom cpu (ie $100-$300)

I know this is expensive now, but intel could probably find a way it can make the initial process for cpu production streamlined for this market.

Imagine paying $800 for a cpu with 6 cores, 2 cores with 2mb of memory, 4 cores with 512kb. (just hypotetical)

Windows could allow you to map the 2 cores with 2mb of memory for your intensive applitions and use the other 4 cores for things like web browsing and email.

The problem with 4 cores is each core gets the equal amount of cache and not all processes are as intensive, so its not as efficent as it can be, more cores will solve this probably, but thats millions down the road, and sooner or later we will be tired of hearing about cores as much as we are about clock speed by now.

Just my opinion imo.

Re:I for one... (1)

Dun Malg (230075) | more than 7 years ago | (#16306167)

You pay X dollers for X amount of on die cache You pay X dollers for each additional core over 2 you pay X dollers for a custom cpu (ie $100-$300) I know this is expensive now, but intel could probably find a way it can make the initial process for cpu production streamlined for this market.
Errrr..... no, not likely. Creating the mask for the photolithography is not a simple task. Such a custom chip would require scores of photomasks, one for each possible combination. Then the finished chip would require testing, to see if it it actually works. If it doesn't (due to any of a number of possible issues, from mask errors to unexpected leakage or capacitance) then the mask has to be redesigned. Then, even if it works, you'd need to make several extra of each ordered core design in order to be sure to get a deliverable yield. The "initial process for cpu production", as you call it, is basically 99% of the hard work in CPU manufacturing. There's no magic automation wand available to make it any easier-- not beyond the incremental improvement we've been seeing for the last 30 years, anyway.

Re:I for one... (1)

Demanche (587815) | more than 7 years ago | (#16306275)

I know its beyond our times....
But it would be damn neat if you could customize your cpu you must agree ;)

Re:I for one... (1)

Amouth (879122) | more than 7 years ago | (#16306819)

if customization is what you want then look at FPGA's you know the ones that can be reprogramed.. you could just get some with diffrent cache amounts and put them in cards in your back plane and be set.. they are wonderful things except they suck at FPU.. other than that they are nice... other problem is people don't want to deal with backplane computing.. i never understood why.. it offers some of the best posiable expansion and upgradability... you will never see single chip customization like you are talking about for the genral public.. not until we can figure out a diffrent way of building chips..

Re:I for one... (1)

Doctor Memory (6336) | more than 7 years ago | (#16307223)

ISTR that Freescale (nee Motorola) or one of the FPGA suppliers (Xylinx?) had something like this. Basically, you would pick electronic "assemblies" (PLLs, caches, ALUs, etc) that were all guaranteed to work together. I don't remember if they would then build and ship the chips to you, or if it was up to you to program one of their devices, but I thought it was a pretty neat idea.

Re:I for one... (1)

Khyber (864651) | more than 7 years ago | (#16307249)

I still have my Packard Bell 8088 with an ungodly 2 megs of RAM, running DR-DOS. BTW, I've had that computer for well over 18 years now. I don't think any other geek could say that, at least not with that kind of computer. Hell, the 300 baud modem STILL WORKS, which just goes to say that today's manufactured hardware is pure and simple CRAP. Antec PSU, anyone?

Major rebuild (1)

Hoi Polloi (522990) | more than 7 years ago | (#16307251)

I built myself a new system last year to replace my ancient Micron 486. It was so old that the CPU didn't even use a fan (had plenty of dust in the heatsink though), the VLB graphics board had 2 MB vram, 48 MB RAM, the monitor was a 14" CRT that had ghosting probs, and the hard drive space was less than most high-end MP3 players. Even the mouse and keyboard barely worked anymore. I pretty much milked it for every penny I paid.

I did a 100% rebuild. Now I've got a AMD 64 X2 3800+, Lian-Li case, UPS, 19" LCD, 2 GB RAM, 500 GB total SATA HD space, NVidia 7800GT, etc. I have no idea how many generations I must've jumped. I felt like a hermit walking out of his cave and blinking at the sun. I've even gone from dial-in to high-end DSL. I'll replace the CPU/MB when apps start grinding on it. 5 years prob?

Re:I for one... (1)

genrader (563784) | more than 7 years ago | (#16305857)

I agree. I ran a 1.4ghz Athlon with 768 DDR and a RAdeon 9500 for 4 years. Just now started to annoy me from being too slow.

Re:I for one... (1)

thePowerOfGrayskull (905905) | more than 7 years ago | (#16307341)

If you buy something good with lots of room for expansion, and take good care of your computer, you shouldn't have to replace it every 2 years.

Let's not deceive ourselves with that "have to". Need has nothing to do with this.

Re:I for one... (1)

mendaliv (898932) | more than 7 years ago | (#16305921)

Yeah, and when AMD comes out with some they'll be called the S-Tair Master. (props to whoever gets this reference)

Re:I for one... (1)

cerberusss (660701) | more than 7 years ago | (#16306619)

wait for those 80 core CPU's intel said they will have in a 'few' years
Good. Then you probably have the right specs for Vista.

Re:I for one... (1)

rthille (8526) | more than 7 years ago | (#16306971)

Yeah, I'll wait too, especially since it seems every time I upgrade I spend all my time fighting with driver issues for the new hardware...

Which of 'My Rights Online'? (-1, Offtopic)

FreakinHippie (636913) | more than 7 years ago | (#16305501)

...does this story have to do with?

Altair? (2, Funny)

gEvil (beta) (945888) | more than 7 years ago | (#16305503)

Ooooh. Blinkenlights on a processor!

Re:Altair? (1)

einhverfr (238914) | more than 7 years ago | (#16305533)

Yeah, who would recommend an Altair for modern computing tasks?

Re:Altair? (1)

Doctor Memory (6336) | more than 7 years ago | (#16307761)

Hmmm, I wonder if an Altair with sufficient memory (8K might be enough) and some kind of network interface could run a text-based IM ("IRC Classic")? Assuming it had an appropriate terminal connected, of course. I mean, 99% of its time would be spent waiting for the user to type something, or waiting for input on the network interface.

Re:Altair? (1)

in2mind (988476) | more than 7 years ago | (#16305691)

" Ooooh. Blinkenlights on a processor!" Ah ! Blinkenlights. Is it still on anywhere?( I think the last I saw it was in Germany..) May be quad cores can run Blinkenlights.Guess someone will try it.

Isn't that going a bit far? (0)

Anonymous Coward | more than 7 years ago | (#16305517)

I mean, frankly... isn't 12MB L2 overkill? We're barely putting today's 2-4MB to good use.

Re:Isn't that going a bit far? (1)

Yvan256 (722131) | more than 7 years ago | (#16305573)

12MB cache divided by four processors = 3MB cache.

No, I didn't RTFA.

Re:Isn't that going a bit far? (0)

Anonymous Coward | more than 7 years ago | (#16305669)

It says 2x6mb. It'd make much more sense if it were 4x3mb, but nope.

Re:Isn't that going a bit far? (4, Informative)

Wdomburg (141264) | more than 7 years ago | (#16306075)

That's because Intel is cheating. They don't have a quad-core die, they have two dual core dies shoved onto a multi-chip package. Each die has a shared 6MB cache.

Cheating? Yep... (0)

Svartalf (2997) | more than 7 years ago | (#16306395)

Just like the Duos are "faster" on the benchmarks that have been ran- they've got 2-4 times the L2
and as long as you sit in the L2 for the large part, you're going to be "faster". Take the L2
advantage away from the Duo and it's a slower system overall to the Athlons, still.

I'll consider an Intel right now, mostly because the Jury's still out on the GMA X3000 display
GPU (All the reviews are using the old drivers which run it in the older mode of operation with
no T&L, etc.- I've not seen ANY benchmarks using the enhanced drivers or the Tungsten Graphics
developed Linux drivers for the GPU yet...); but otherwise, it's not really as compelling as
the review sites make them out to be.

Re:Cheating? Yep... (1)

Wdomburg (141264) | more than 7 years ago | (#16307801)

True, though they're to be lauded for creating an architecture that can take advantage of a ginormous cache as well as it does. They really did learn to "work smarter, not harder" in their current chip. And of course, in the end, the consumer doesn't care how performance is achieved. :)

In the case of Intel's quad core solution, they seem to be achieving higher overall performance, as expected, but at the expense of pushing their thermal envelope back up to 130W even after the die shrink. AMD, on the other hand, has commited to shipping a 68W quad part and that's including the integrated memory controller.

Still, it's not only a few months of bragging rights that Intel is buying - Woodcrest only supports dual socket configs which means four cores at most unless you fall back to Dempsey, which simply isn't price or performance competitive with the Opteron. This at least scales them to eight core configurations. Of course on the AMD side of the house you can pick up an eight-way board, which gets you to sixteen cores today, and will scale up to a whopping thirty-two cores next year. Between that and Intel's lackluster bus technology, they've got serious problems at the high end of the server market.

Re:Isn't that going a bit far? (1)

shawnce (146129) | more than 7 years ago | (#16307609)

Actually based on current information it looks like Yorkfield is two dual core blocks (with related cache) on a single die NOT an MCM. This is much the same as current AMD packages and the coming K8L.

The K8L looks likely to have 512 MiB of cache (L2) per core with 2 MiB of cache (L3) shared among all four cores while Yorkfield will have two independent 6 MiB cache (L2) blocks shared between two cores and on die glue between the independent dual core blocks and the FSB.

Re:Isn't that going a bit far? (1)

RingDev (879105) | more than 7 years ago | (#16305765)

It's shared though, not dedicated (IIRC). So it's a 12MB cache, so one proc could in theory use 5 megs of it while another only used 1, could it not?

-Rick

Re:Isn't that going a bit far? (1, Interesting)

Anonymous Coward | more than 7 years ago | (#16305909)

Its not exactly 3 MB a core, and you cannot count it in the same method as 12MB for one core either. The reason for this is that future multicore processors will share L2 cache. As the number of cores per cpu increases, it is less likely that a single thread will resume on the same core it was once running on. As a result, instead of refetching data from memory, a shared L2 cache allows recently fetched memory to stay in the cache and be accessed by multiple cpus. The problem that arises is that multiple threads can be using this cache simultaneously and cause thrashing of the data. One way to solve this is to increase the cache size enough that memory thrashing is kept to a minimum. As a result, the Cache hit ratio on a single core cpu with 12MB would be very high, and the quad core version not quite as high, although, probably still substantially higher than giving each core only a 3MB l2 cache.

Re:Isn't that going a bit far? (3, Insightful)

DrDitto (962751) | more than 7 years ago | (#16306793)

I mean, frankly... isn't 12MB L2 overkill? We're barely putting today's 2-4MB to good use.

Are you kidding me? With a 4-way superscalar processor running at 3GHz, any cache miss can result in the processor being completely idle for 50-100ns. At an aggressive 50ns memory latency, this is up to 600 wasted opportunities to retire instructions.

Re:Isn't that going a bit far? (1)

TheRaven64 (641858) | more than 7 years ago | (#16307607)

This is ameliorated to a significant amount by SMT. If you have a cache miss in one context, you switch the execution units over to the next one and let it run for a bit. Of course, this requires developers to realise that they're not writing code for a PDP-11 anymore...

y halo thar (-1, Offtopic)

Anonymous Coward | more than 7 years ago | (#16305519)

o hey guys so wuts up with like pentiums n stuf? lol

Competition rocks (4, Funny)

mattnuzum (839319) | more than 7 years ago | (#16305557)

I've said it before, I'll say it again: This is exactly why competition rocks. Soon, we'll say Moore was no prophet, he was a pessimist!

Re:Competition rocks (0)

Braino420 (896819) | more than 7 years ago | (#16306007)

WTF says he's a prophet? He merely stated his observation, nothing more. Sure, it may have set some sort of goal that was to be achieved, but he is becoming waaaay overrated.

Moore's law is nothing about this (1)

Flying pig (925874) | more than 7 years ago | (#16306077)

It specifies an approximate period in which transistor packing density doubles. Getting to 80 processors on a die is about size and yield as well as packing density. The cost of die size depends on economies of scale among other things, and the increased demand for silicon (especially large consumers like solar panels) must help drive the price down.

Like many laws, people mention Moore's without actually knowing what it says.

Re:Moore's law is nothing about this (1)

mattnuzum (839319) | more than 7 years ago | (#16306717)

quote:
"Due to its advanced 45nm process technology, Yorkfield XE is able to pack a total of 12MB L2 cache (2 x 6MB L2) and still achieving a much smaller die size and higher clock speed of 3.43-3.73Ghz"
I know what the law says. Wouldn't you say the above statement puts this discussion into the realm of Moore's law?

One sided (4, Interesting)

Daemonstar (84116) | more than 7 years ago | (#16305565)

Ok, so we have all this neat info about the Intel chip; what about the AMD processor (it gets a whole sentence and a half)? If this is supposed to be a "battle", it seems that most of the comparison has already been done in favor of Intel before the event even takes place, if this article is any reference. :P

Re:One sided (4, Funny)

LarsWestergren (9033) | more than 7 years ago | (#16305681)

Ok, so we have all this neat info about the Intel chip; what about the AMD processor (it gets a whole sentence and a half)? If this is supposed to be a "battle", it seems that most of the comparison has already been done in favor of Intel before the event even takes place, if this article is any reference. :P

Ooh, but dont' count out AMD yet! According to the nifty diagram from TFA, the Windsor has a "HT1.0", and the Altair a "HT3.0", and I can't see anything like that for the Intel processors. I don't know what a HT1.0 is, but I'm TERRIBLY excited about it, let me tell you.

More bullet points or higher numbers in a press release indicates a superior system much more clearly than any real life performance tests.

Re:One sided (1)

prefect42 (141309) | more than 7 years ago | (#16305747)

HyperTransport. I believe 3.0 is twice as fast as the current incarnation, but could be wrong.

Re:One sided (4, Informative)

Wdomburg (141264) | more than 7 years ago | (#16306903)

The specifications list bandwidth for the 1.0, 2.0, and 3.0 specs as 6.4, 11.2, and now 20.8GB/s respectively. AMD is jumping from 1.0 to 3.0. They're actually pushing a bit more than the original spec on the current processors though, since the spec originally only included bus speeds of up to 800MHz and they've got it running at 1000MHz which bumps throughput to 8.0GB/s. So, assuming they max bus speed, it'll be about two and a half times faster.

This is where I think AMD gets themselves a big win. Intel's FSB, even clocked at 1333MHz (actually it's 333MHz QDR, but we'll not quibble) pushes only 10.6GB/s. And that's not accounting for the off-die memory controller. Even with dual buses (like the 5000 series chipsets tout) they only just barely have enough aggregate throughput to handle memory transfers.

Re:One sided (1)

Joe The Dragon (967727) | more than 7 years ago | (#16305779)

HT = HyperTransport
HT 3.0 is HyperTransport v3

Re:One sided (0)

Anonymous Coward | more than 7 years ago | (#16306293)

Actually, you should blaim AMD for that, since they have been and continue to be very scarce with information about future products.

Re:One sided (0)

Anonymous Coward | more than 7 years ago | (#16306545)

Yeah! I mean, come on! All of us here at Slashdot have worked long and hard at a history of fair, honest reporting showing both sides of the story. I would hate for the world to get the wrong idea about us because of this one biased article.

Ah Numbers... (3, Funny)

glowingsnowball (973747) | more than 7 years ago | (#16305593)

Processer speed as well as cores are just numbers to me. The only thing high processer speed means to me is that I am able to write unefficient code and get away with it. For(int i = 0; i9000;i++){For(int j = 0; j9000;j++){For(int l = 0; l9000;l++){System.out.println("More Cores")}}}

Re:Ah Numbers... (2, Funny)

bobtheinsanecow (789624) | more than 7 years ago | (#16305995)

// May as well use all the cores...
int i,j,l;
fork();
fork();
for(i = 0; i<9000;i++)
  for(j = 0; j<9000;j++)
    for(l = 0; l<9000;l++)
      printf("More Cores");

Re:Ah Numbers... (1)

Yetihehe (971185) | more than 7 years ago | (#16306247)

WTF? Why not k?

Although it's good example for multiprocessor hello world.

Re:Ah Numbers... (0)

Anonymous Coward | more than 7 years ago | (#16307225)

WTF? Why not k?

Because he's already used "fork", and doesn't want to be confused!

Re:Ah Numbers... (0)

Anonymous Coward | more than 7 years ago | (#16306633)

Well, you _are_ programming in Java....

I've mocked intel before... (1)

gentimjs (930934) | more than 7 years ago | (#16305613)

I've mocked intel in previous threads for not beating AMD by a much larger margin with core2 than they actually did. This stuff (mentioned in post) is the kind of performance jump I was expecting to see. Bravo! If they get this stuff out the door ontime, Intel just might make it back onto my vendor list.

Just another qualitity advertisement... (-1, Offtopic)

mmell (832646) | more than 7 years ago | (#16305615)

brought to you by CmdrTaco.

What, this surprises you?

Bulrathi or Sakkra pwn the Altair (1)

BadAnalogyGuy (945258) | more than 7 years ago | (#16305637)

Ground invasion is where it's at. Space battles can be equalized with sufficient technology taken from captured planets.

Re:Bulrathi or Sakkra pwn the Altair (1)

julesh (229690) | more than 7 years ago | (#16306309)

As long as I can have lots of big ships stocked to the gills with x2 missile launchers, I don't care.

Re:Bulrathi or Sakkra pwn the Altair (1)

NDPTAL85 (260093) | more than 7 years ago | (#16307231)

Its all about the Psilons son. Their homeworld of Mentat will rule all of creation someday.

Re:Bulrathi or Sakkra pwn the Altair (1)

BadAnalogyGuy (945258) | more than 7 years ago | (#16307631)

Not when they have to keep offering gifts of technology to their lizard and bear overlords.

Interesting. (3, Interesting)

LWATCDR (28044) | more than 7 years ago | (#16305667)

Intel is going to need that HUGE cache because of it's limited FSB. It will be interesting to see how they do side by side.
The AMD with it's Hyper-transport could have an advantage over the Intel chip but right now it is all pie in the sky.
I wish that AMD had access to the Intel Fab tech. Just how fast and low power would their chips be if they where 65nm right now like Intel's?

Re:Interesting. (1)

Joe The Dragon (967727) | more than 7 years ago | (#16305985)

There chip sets also far behind as well Bearlake-X with is a year a way is there first with 2x16 pci-e. right now they only have x8 x8 with not much left over.

Right now nforce 590 has that plus lanes left over for x4 and x1 slots. There is even a nforce 590 board with 2x16 2x1 and pci-x slots.

Also when will there work station chips sets come any close the nforce pro chip sets in number of pci-e lanes?

Re:Interesting. (1)

sgt scrub (869860) | more than 7 years ago | (#16306527)

Just how fast and low power would their chips be if they where 65nm

That is where I was going with my post so I'll just add this. If Intel can retool to .45 before AMD gets to .65 then I can't forsee any kind of battle. AMD has the edge on everything else, IMHO. Unfortunately, the die size is what allows for the best performance/heat and that is what is important to customers these days.

As far as the cache expectations go, I don't think Intel will go that high. Your right. They will need it to keep instructions flowing. Intel was supposed to have 4M cache on the core 2 duo and it is only 2. So all things considered, I expect 4.

Re:Interesting. (1)

shawnce (146129) | more than 7 years ago | (#16307743)

Intel was supposed to have 4M cache on the core 2 duo and it is only 2.


You may want to visit Intel's site... several of the Core 2 Duo based processors have 4 MiB shared L2 caches (X6800, E6700, E6600, Xeon 51xx).

Re:Interesting. (1)

Dr. Spork (142693) | more than 7 years ago | (#16306857)

AMD may not have access to Intel fab tech, but they do have access to IBM fab tech. This may be moving forward more slowly than Intel, but maybe on a more solid foundation with Silicon-on-Insulator. It's true that Intel made more progress in fab tech than people realized. The progress was largely invisible until they started fabbing something better than the dreadful Netburst cores. Now they seem to be way ahead, but for the first time next year, AMD will be able to crank out some serious volume thanks to Fab36.

I wouldn't count AMD out just because they're being beaten this quarter - for the same reason we didn't count Intel out when they were being beaten for years!

Re:Interesting. (1)

grumpyman (849537) | more than 7 years ago | (#16307195)

I wish that AMD had access to the Intel Fab tech. Just how fast and low power would their chips be if they where 65nm right now like Intel's?


Aie... Can you imagine Intel/AMD merge? When we have that we'll all yell and holla about the monopoly.

Re:Interesting. (1)

Jeff DeMaagd (2015) | more than 7 years ago | (#16307209)

I don't think the FSB is really as much of a limitation as you think. For one, the FSB bandwidth matches the memory bus bandwidth. Sure, there is other I/O, but that's much lower order of magnitude, and that memory bus bandwidth is what would be the limiting factor on AMD's chip as well. Hypertransport isn't necessarily what makes AMD's system better, what helps them more is the on-die memory bus to get better latency.

Intel 45nm is late 2007 / early 2008 (0)

Anonymous Coward | more than 7 years ago | (#16305711)

So I strongly doubt that it will come out at the same time as K8L unless AMD mess up and their schedule slips. Sadly that is a likely possibility for AMD at the moment, their fault for sitting back on their asses (well, they didn't but Intel have performed extremely well recently after a few years adrift and caught AMD by surprise).

K8L's core is meant to be quite a bit faster at some tasks than the K8 core, over 40% apparently, although I'm sure that is not in general.

No more instructions!! (1)

miceliux (654192) | more than 7 years ago | (#16305745)

Please, no more!

Re: No more instructions!! (3, Funny)

dp_wiz (954921) | more than 7 years ago | (#16305989)

That 80-cored chips definitely would require a desperate bureaucracy to operate.

Re: No more instructions!! (1)

wild_berry (448019) | more than 7 years ago | (#16306327)

Vote CowboyNeal for Governor of Core 0x4F!

Re:No more instructions!! (1)

pkulak (815640) | more than 7 years ago | (#16306137)

I say, if they fit, throw 'em on there; it can't hurt. And if someone writes a compiler for it, we win. RISC is dead. :D

10x (1)

sumi-manga (948999) | more than 7 years ago | (#16305763)

Intel has MORE than 10 times the market capital as AMD, and they're still going "head on"? You gotta love the cache size and clock speed on Intel's desktop though, just so... big. I mean, that's better right?

what new instructions? (3, Interesting)

backwardMechanic (959818) | more than 7 years ago | (#16305835)

I've often wondered, what are these new instruction Intel keep thinking up? Are they some sort of fancy array processing, new addressing modes? I'm curious. Whatever happened to RISC?

Re:what new instructions? (5, Informative)

Anonymous Coward | more than 7 years ago | (#16306023)

Here [intel.com] are details about the new instructions.

RISC is dead. I miss it too.

Sort out those names! (1)

MrNemesis (587188) | more than 7 years ago | (#16306053)

Penryn new instructions = PNI = SSE4

Prescott new instructions = PNI = SSE3

Therefore SSE3 = SSE4...?

Strikes me that Intel is running out of buzzwords! Was the marketing dept. severely depleted in the last round of purges?

THe next 12 months or so are going to be a very interesting time for the CPU world. All Intel needs to do it get their chips' idling power down into the same ballpark as AMD, and AMD need that 65nm process in volume *now*! I've actually been finding myself forcing myself not to look at computer stores and upgrade my workstation because I know that six months down the line there'll be something orders of magnitudes better on the scene...

Re:Sort out those names! (1)

Afrosheen (42464) | more than 7 years ago | (#16306389)

Yeah it seems like the processor wars are really heating up. The days when every PC had PC100 or PC133 ram are way over, but those days lasted *years*. It was 'good enough' for a very long time. PC3200 had a good run but DDR2 is just beginning to replace it, and next year we'll be dealing with DDR3.

  I guess there's not alot to complain about. At least it's not more-valuable-than-platinum-per-ounce Rambus.

Re:Sort out those names! (1)

Gr8Apes (679165) | more than 7 years ago | (#16306705)

I couldn't wait - bought a AMD 3800+ X2 and motherboard for a whole $160. (Frys) works allright, and allowed me to not buy anything else to get a working PC. However, I'm looking forward 6 months and going wow - I can buy today's $500 setup probably for $100 in 6 months, if the current price drops are any indication, and I'd still be 2 cycles behind. The CPUs are definitely rolling out faster this year.

Re:Sort out those names! (1)

MrNemesis (587188) | more than 7 years ago | (#16307643)

Heh, I already have a 3800... and a 4200... I can't believe I was contemplating replacing at least one of them for one of the new AM2 energy efficent varieties (new mobo, new memory, new coolers... ARGH!).

I'm just really tempted by the "fake" Kentsfield quad cores since H264 encoding takes an age, even with both cores flat out, but I'm still not sold on Core2 wholsesale since it's only energy efficient when it's running full tilt. Maybe I'll get a Kentsfield for my workstation and keep my servers (idle 95% of the time) on AMD64 for now... or maybe I'll just die of silicon-induced indecision before then.

Aren't we past this already? (1)

nebulous_afterthough (943262) | more than 7 years ago | (#16306453)

From the quote of the article "and higher clock speed of 3.43-3.73Ghz"

/me shakes head disapprovingly

Let the Ghz versus instructions per cycle war begin, yet again...

AMD: Sell --- Intel: BUY ! BUY ! BUY ! (0)

Anonymous Coward | more than 7 years ago | (#16306655)



No more AMD for me !! AMD is going back to the "follow the leader" mold. The leader is INTeL !! and I only follow the leader !!

Re:AMD: Sell --- Intel: BUY ! BUY ! BUY ! (1)

raddan (519638) | more than 7 years ago | (#16307639)

Hey, did you notice? You were drooling. Stop.

Does anyone really care? (1)

argStyopa (232550) | more than 7 years ago | (#16306807)

I used to upgrade systems about every 3 years when CPU speed typically tripled or more.
So my first system was a 486-25.
Second system was a P-90.
Third was a 300MHz AMD.
Fourth was 1.2 GHz AMD.
Current system is a P4 2.7 GHz and it's at least 3 years old. And I don't feel any urgency to upgrade my basic system, perhaps a video card and some more RAM instead.

I simply don't see that CPU horsepower increasing in the steps like it used to. Yes, I understand multicore, more-cache, hyperthreaded CPUs are going to offer performance not indicated by something as simple as CPU speed, but is it THAT much?

Re:Does anyone really care? (1)

teeleton (148923) | more than 7 years ago | (#16307139)

You're buying into the intel clock speed = processor speed propoganda. There's a lot more to how fast your computer runs than just how fast the clock spins. Clock speed comparisons are reall yonly valid within a particular chip package (p4 to p4, opteron to opteron). Even trying to compare different generations of processor from the same manufacturer as you've done is sketchy at best.

Re:Does anyone really care? (1)

FuturePastNow (836765) | more than 7 years ago | (#16307705)

You'll be happy to know that CPU speed has, in fact, tripled in the least three years. I've got a 2.8GHz P4, and the high-end Core 2 Duo and Athlon FX chips fit your criteria.

I might celebrate my computer's third birthday by replacing it. Then again, I can't imagine what I would need to do three times faster.

Penhryn New Instructions (PNI) (1)

SETIGuy (33768) | more than 7 years ago | (#16306879)

Yorkfield will feature Penryn New Instructions (PNI) or more officially known as SSE4 with 50 more new instructions.

I hope they come up with a new acronym. PNI is already used for Prescott New Instructions.

I like the Irony. (1)

UnixRevolution (597440) | more than 7 years ago | (#16306927)

The Altair, AMD's Quad-Core CPU, being named for the first widely available home computer, the Altair 8800, is just too fun.

Let's hope AMD's altair is more useful.

Re:I like the Irony. (1)

rcamera (517595) | more than 7 years ago | (#16307425)

my altair is extremely usefull. without it, i'd have to stand downstairs and hold the door to my office open all day.
Load More Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Create a Slashdot Account

Loading...