AMD Releases Image of Phenom/Barcelona Die

MojoKid writes "A few weeks ago, AMD released information on new branding for their desktop derivatives of the Barcelona core, now dubbed the Phenom FX, X4 and X2. If you're unfamiliar with Phenom, the processors will be based on AMD's K10 architecture. They've been tight lipped about specifics, but we know that it will feature a faster on-die memory controller, support 64-bit and 128-bit SSE operations, and they'll be outfitted with 2MB of on-chip L2 cache (512KB dedicated per core) in addition to 2MB of shared L3 cache. This week, instead of revealing some more of the juicy details regarding those enhancements, AMD just sent over a tasty photo of a Phenom die. At least it's something."

Comments

wow a photo

[jonathan DS]

can you see how fast it is? How about some specs we understand?

Re:wow a photo

[farkus888]

well marketing now tells us the number of cores is the only important factor in performance. this has 4, most desktop pc processors are 2 right now, that makes it exactly twice as fast as current processors.

Re:wow a photo

[luder]

Well, if I understood correctly, it still has overheating issues, as they're only capable of delivering a photo of a dead processor. Also, on a side note, isn't it funny that there's a website specially dedicated to hot hardware?

Re:wow a photo

[Jacer]

I understand your cynicism. Especially considering how a lot of processor intensive applications that most consumers use (games and other multimedia, and to a small margin running outlook, internet explorer, and MSN messenger together) get absolutely no benefit from SMP. However, as multi-core chips are rapidly becoming the defacto must have for everyone, I think developers are going to start coding to take advantage of this. We've already heard plenty of rumors about offloading most/all of the physics processing onto a chip that does it's computations in matrices rather than in any sort of linear fashion, streamlining the process both in method of computation and by freeing up your cpu cycles any number of other tasks (potentially an increase in game artificial intelligence, so it behaves less predictably, maybe do away with all of the nested tree structures and boolean choices) The only potential problem is increasing the complexity of development. Applications to take full advantage of all the new widgets will also take exceedingly more development time, support time, QA time, ect which will (alm0st) inevitably lead to a rise for consumers.

Re:wow a photo

[ceeam]

Frankly - who cares? It will be fast enough, no doubt. Speed is not the #1 characteristic of CPU anymore.

Re:wow a photo

[Lehk228]

can't fap to specs (well you can, but it's not very fun)

Re:wow a photo

[marcosdumay]

They also told us that it has a fast memory controller and large I/O words. Those matter too.

Re:wow a photo

[Chandon Seldon]

Especially considering how a lot of processor intensive applications that most consumers use (games and other multimedia, and to a small margin running outlook, internet explorer, and MSN messenger together) get absolutely no benefit from SMP.

Don't underestimate how well supported SMP is already. It's true that there aren't that many single applications that get a linear speedup to 4 cores, but dual core processors have been common for a while now. All of the new games support multiple cores - they have to, neither the PS3 and XBox360 has one simple processor. All of the media encoding applications support 4+ cores. Some video playback programs support multiple cores. And, as AMD and Intel love to remind us, if you run more than one program at a time *that* can use multiple cores.

Re:wow a photo

[servognome]

Don't worry the "Ocho" is due out soon

Re:wow a photo

[Anonymous Coward]

Anybody who encodes audio and video will love quad core. Xvid 1.2+ supports parallel encoding, and sweet jesus does it do it quickly on the proper machine. I can encode a 90 minute film to 1400MB Xvid w/multi-pass encoding in about 45 minutes on a stock E6600. On my stock AMD64 3000+, the same operation takes about 150 minutes.

Re:wow a photo

[Jacer]

Thanks for that. I wasn't aware of the next gen system architecture at all, and as such wasn't aware that most (all?) new games are being developed with SMP in mind. I did mention that one of the best uses would be people who run multiple applications at once though. Additionally, sorry for my formatting I don't really care enough to add boldface and fancy line breaks everywhere, so it may have been annoying to read everything in my original post.

Re:wow a photo

[mgblst]

This is a common misunderstanding, due to the English languages having many meanings for one word. In this example, hot is not meant to mean warm to the touch, rather it is meant to convey the sexiness of the CPU. Clearly by the pictures, this is a very sexy beast (my favourite kind)

Hype it up

[jhfry]

I know that this is just a ploy to build up hype for the new processors... I just hope that the processor performs up to expectations.

AMD really needs to respond to the Core 2 Duo's with something that tells the world that they are still in the race. I really don't want to see Intel become the unchallenged winner of the silicon wars... it would hurt us users in the long run.

I fear that it is a real possibility however. The cost of fabs, R&D, and marketing have grown so much in the last few years that it would be VERY difficult for any newcomer to compete with Intel unless they managed to develop a completely different and low cost way to manufacture their chips... or they are very heavily backed.

Re:Hype it up

[beheaderaswp]

"I fear that it is a real possibility however. The cost of fabs, R&D, and marketing have grown so much in the last few years that it would be VERY difficult for any newcomer to compete with Intel unless they managed to develop a completely different and low cost way to manufacture their chips... or they are very heavily backed."

AMD is not a newcomer. And the speed "crown" has passed between AMD and Intel a couple of times since the K6 and probably will again.

Maybe I'm missing something, but it appears to me that neither Intel nor AMD are ill equipped to compete, in the long term of course.

Re:Hype it up

[kestasjk]

I think the Nintendo Wii will do better than the 360 and PS2, even though the specs are lower. Microsoft and Sony were both reaching for performance but they forgot all about the fun!

Re:Hype it up

[suv4x4]

Not everybody in slashdot knows how silicon industry works.
Some of us actually welcome people like the grandparent who puts the news into perspective.
Although for someone it may be obvious information, for other people it's not.

Orly :P Let's recap what he said in short: ..:: Chapter 1: Obvious statements ::..

The image is part of a press release.
Since AMD currently has worse chips, what AMD needs to do is have better chips.
If AMD dies, Intel becomes a monopolist. Monopoly is bad.
It's very expensive to start a chip business in your garage nowadays. ..:: Chapter 2: Hopes and Fears ::..

He hopes AMD's chip is very good.
He fears if it's not, AMD won't take the lead.
He fears if AMD keeps underperforming, it may go bankrupt (now, that's actually a bunch of bull, but I won't refute it since I'm lazy).
He fears all of this is "very real possibility".

It's not his fault really, it's the moderators. Anyone of us is free to come here and compile a list of crap and post it. Hell, even the GNAA do it, and they never had anything interesting to say so far. But if moderators keep modding pedestrian observations of this kind, people will just lean to posting more of it. I remember there used to be some interesting stuff in the comments before, people came here to read the comments.

Re:Hype it up

[billcopc]

Actually, this would be AMD's late response to the Core 2 Quad. Given that Intel is going to be slashing those prices again in July, AMD needs to be ready with something faster and/or cheaper to stay in the race.

Seriously, a super-clockable Q6600 for $266 bucks ? Hello!

AMD is playing catch-up right now, but Intel is doing what little it can to block the opposition by eliminating the price gap. AMD really needs to pull a rabbit out of a hat this time, or they will be left sitting on the bench until their next breakthrough.

Re:Hype it up

[tomstdenis]

Where is it $266? The Q6600 at the local shop here in Ottawa is $670 CAD [currently on sale for $639]. I'll wait until I can actually find one for around $266 USD.

Also keep in mind that the AMD design is a true quad-core. They didn't just hack two dual-cores together over an FSB. This is a true quad-core (e.g. the L3 is shared between all four cores) over a higher speed internal bus, attached with it's own memory controller, etc....

Will the average OpenOffice or Firefox user notice the difference between the Q6600 and Barcelona? Most likely not. But if you're doing number crunching [say media filtering, encoding, chemistry, etc] the AMD design will likely pay off better.

Tom

Re:Hype it up

[GweeDo]

Yup, that is why I buy AMD CPU's...they are so fun!

Re:Hype it up

[tomstdenis]

I guess I forgot that it's currently not next month. I'll hold my breath for the price cut.

Tom

Re:Hype it up

[ChrisMaple]

The only reason that AMD is still alive is that Intel made a series of blunders. Intel went exclusively with the expensive RAMBUS technology, kept the northbridge off-chip, chose clock speed over processing power. During the same period AMD integrated the memory controller, developed hypertransport, and emphasized processing power over clock speed. As a result, for several years AMD maintained a small performance advantage and slowly gained market share. Because Intel maintains a superior process technology, AMD's advantage was only a small one. Intel is much larger and can afford the huge expenses invloved in keeping the process advantage.

Now that Intel is mostly past its blunders, it still has the advantage of superior process and is likely to maintain that advantage. Unless AMD can pull more rabbits out of its hat, its goose is cooked. I want AMD to regain the performance lead, but I don't think it's going to happen.

Re:Hype it up

[fitten]

There have been four references to the performance of Barcelona lately. Three are pretty much in agreement that single threaded performance will be from 0%-12% faster than the equivalent clock speed K8. These can be found in the form of SPEC published by AMD, the 16 core 1.8GHz K10 demonstration using POVRay (where it wasn't 2x as fast as an 8 core K8 machine at the same clock frequency and the 8 core Intel Woodcrest, which you can already buy, was faster than it, as shown in an Intel response specifically to the AMD POVRay demonstration), and an NDA leak from one side that said they've been seeing up to 12% improvement. Another was the press expo where someone commented that "H.264 encoding was done in almost realtime" but that is too vague to be useful. If it were "close" to realtime, it'd be about 6x as fast as today's machines so there was something going on there (maybe the stack of the latest gen - to-be-released ATI video cards doing the work, who knows).

Re:Hype it up

[fitten]

Clarification to the above... 0%-12% faster than an equivalently clocked K8 in integer and x87 loads. There hasn't been anything really out yet for SSE that we've seen and a lot of work and resources were put into K10 to make SSE faster.

Re:Hype it up

[grammar fascist]

Unless AMD can pull more rabbits out of its hat, its goose is cooked.

INTEL [twirling evil black mustache, laughing]: Another! Another!

AMD [producing rabbit after rabbit]: I'm going as fast as I can!

INTEL: You'll pull more, or I'll cook... THIS GOOSE!

[Shot of lone GOOSE in a cage, legs in handcuffs.]

GOOSE [dejected]: Honk.

AMD: No-o-o-o-o-o-o!

(I had to. Your metaphors made me do it.)

Chip fly-over

[Asgerix]

I wish there was an application - sort of like Google Earth - where you could zoom in on the die and do a 3D fly-over.

Re:Chip fly-over

[WhodoVoodoo]

There is; it's called "Photoshop". Remember that at this point in the game, a die is flat =)

The advantages of four cores on a single die

[Eukariote]

On-chip connectivity can be much broader and lower-latency than off-chip connectivity. The two-dual-core in one package "quad cores" of Intel have to talk via the off-package north bridge. As you can see from the AMD Barcelona/K10/10h snapshot, the cores live together on a single piece of silicon.

The space between the the cores is a very broad crossbar, allowing fast inter-core synchronization/cache-coherency. The uniform block at the edge of the chip, outside the cores, is the L3 cache shared by all four cores. Each core has its own L1 and L2 cache. This design is nicely symmetric: each core has equivalent resources. It should do very well on heavy-duty symmetric multiprocessing applications.

Re:The advantages of four cores on a single die

[suv4x4]

On-chip connectivity can be much broader and lower-latency than off-chip connectivity. The two-dual-core in one package "quad cores" of Intel have to talk via the off-package north bridge. As you can see from the AMD Barcelona/K10/10h snapshot, the cores live together on a single piece of silicon.

According to Intel engineers though, communication between the chips was never a bottleneck, so the avantages of one vs the other design are questionable. I'm not a processor engineer, but that holds true everywhere: throwing resources in improving something that's not a bottleneck.. actually does NOT help performance. Logic 101.

And BTW:

AMD just sent over a tasty photo of a Phenom die.

All right! So I can print is and try it on a compatible motherboard, right? Right..

Re:The advantages of four cores on a single die

[Eukariote]

According to Intel engineers though, communication between the chips was never a bottleneck

What a load of crap. For quite a few applications, it definitely is a bottleneck. If you have single-threaded tasks that sit happily on their own processor and do not intercommunicate, then, yeah, it does not matter much what connectivity the cores and dies have. But in the real world, multi-threading and SMP tasks do need to intercommunicate, often heavily so. Also, processes will often migrate from one core to the next because the core it was running on before is in use. At that moment, fast inter-core synchronization of the caches is very helpful.

Re:The advantages of four cores on a single die

[slashthedot]

The single die - four cores processor architecture from AMD could be a result of their collaboration with Sun which already has 8 cores in a single die general purpose processor UltraSparc T1 for more than a year. It's surprising though that the two chip makers, Intel and AMD, still lag behind Sun in terms of cores per die.

Re:The advantages of four cores on a single die

[TheRaven64]

The number of cores per die is limited by two things:

1. Number of transistors per die.
2. Number of transistors per core.

Sun can put more cores on a die by having fewer transistors per core, it's as simple as that. Sun is bucking industry trends quite heavily at the moment (see here) by reducing the amount of die space take up by cache. Intel are right at the opposite extreme, with well over 50% of the Itanium die taken up with cache. Modern x86 chips are sitting at around the 50% mark. Intel could easily make a quad core chip with no cache for the same price as their dual-core chips, but performance would be much worse. They could make a single core chip with 50% more cache for the same price, but, again, performance would be worse.

Exactly what the best trade-off is depends on your workload. Sun are aiming at the web-app server market. It's a good business decision, since this is a rapidly growing area. It's also one of the easiest workloads to run, since it's inherently massively parallel; each web-app typically has a few tens to a few thousand users per server. If one thread in a T1 has a cache miss, then there are a huge number of others that are able to take advantage of the processing resources. Intel and AMD have to support a lot of legacy single-threaded code. A cache miss in one of these is expensive. Main memory accesses are of the order of 100-200 cycles, and so a cache miss every 100 cycles would cause a 50% performance reduction. For the T1, with its 8 contexts per core, it would cause a negligible performance reduction overall, as long as the other threads still have work to do.

Re:The advantages of four cores on a single die

[edxwelch]

Actually even if all cores are on the same silicon, inter-core communication is not as good as it could be. This artical has some interesting information on the topic:
http://www.xbitlabs.com/articles/cpu/display/dualc ore-dtr-analysis_12.html

Re:The advantages of four cores on a single die

[Jeff DeMaagd]

In an ideal world, yes. In this non-ideal world, AMD still has to surmount the fact that they are still a fab generation behind. If AMD can't get a clock that is close 3GHz rather than the 2.5GHz I seem to recall AMD reps throwing around, I don't think the same-die advantage will be enough help. Right now, there is no "true" four core x86 on a single die, and there are still a lot of uses for Intel's Core Quad chips, without an AMD competitor in this sector, the advantage still goes to Intel vs a product that's still in process. You should also keep in mind that Intel is not that far behind in getting four cores on one die.

Re:The advantages of four cores on a single die

[Jeff DeMaagd]

I also get the impression that Sun's dies are a lot larger too. Big-iron chips tend to be that way. That's part of why those systems are very expensive. So that's another reason why one shouldn't compare the technology between the consumer to small iron market and the big iron market, they are different markets.

Re:The advantages of four cores on a single die

[Wdomburg]

Sun put eight in-order single-issue integer only cores on a single die. AMD is putting eight full superscalar cores with branch prediction, virtualization extensions, vector units, blah, blah, etc, etc. Very different design philosophies producing chips with very different aims.

Sun's foray into more traditional processor designs - the Rock - isn't expected to ship until 2008 and will feature only four cores.

The only designs actually on the market with eight traditional cores would be the IBM POWER4 and POWER5 lines, but those are dual core dies in a multi-chip module, reminiscent of the Intel scheme.

Re:The advantages of four cores on a single die

[LWATCDR]

They lie. I write multi-threaded code. My threads pass a lot of data back and forth. I try to keep it to a minimum but minimum is still a lot. Not only that but when threads do pass data between code it is often a critical path for the program. One thread is waiting on the other so it will effect the performance of the system.
Here is logic 101.
Your latest product has a weakness.
do you.
1. Admit the weakness and loose sales?
or
2. Downplay that weakness and say it is never a problem.

Intel might be right. The inter chip communication is never a bottle neck because memory access bottle neck is the real limiting factor. Any data heavy tasks that would be bottlenecked by the inter chip communications are already bottlenecked by memory access so it is never a problem.

The real truth is that very few programs are multi-treaded heck even FlightSim X had to be patched to take advantage of multiple cores. So yea it is a freaking lie.

Re:The advantages of four cores on a single die

[Wdomburg]

AMD has been a fab generation behind for years, and still spanked Intel prior to the Core 2 release. And there's far more factors than just clock speed, number of cores and number of dies. Intel's replacement for their aging frontside bus is years behind; the first chip they announced would use it (Whitefield) was canceled back in 2005, and now they're promising it will debut in the later half of 2008. An integrated memory controller, which AMD has had since 2001, will debut at the same time. Using a dedicated L2 cache and adding an L3 cache has the potential to reduce cache contention. More advanced hardware virtualization (including I/O virtualization) could improve performance in consolidation (one of the areas that heavily parallel designs make the most sense).

Of course it's all speculation one way or another until real hardware leaks out. And of course subsequent releases from Intel may well erase some of AMDs advantages. But the converse is also true.

Re:The advantages of four cores on a single die

[TheSunborn]

Currently both Amd and Intil produce using 65nm, so how can AMD be behind now?

Re:The advantages of four cores on a single die

[yabos]

Intel is already moving to 45nm by the end of 2007 or earlier http://www.intel.com/technology/silicon/45nm_techn ology.htm

Re:The advantages of four cores on a single die

[slashthedot]

Sun's foray into more traditional processor designs - the Rock - isn't expected to ship until 2008 and will feature only four cores. Wrong! Rock has 16 cores. Read http://blogs.sun.com/jonathan/date/20070410

Re:The advantages of four cores on a single die

[thanasakis]

Correction, it is 4 execution threads per core. The next gen model will be 8 per core (e.g. 64 exec threads), but I don't think you can buy that one yet, maybe later this year.

Also, the reason that they are ahead of Intel and AMD at this point is probably the fact that they decided to go multicore several years ago when others were trying to squeeze insane MHz out of their chips. If I remember correctly, a lot of folks in slashdot were laughing at this so-called throughput computing strategy, but it turns out that Sun was correct.

Re:The advantages of four cores on a single die

[Wdomburg]

It seems to depend on how you define a core. Released details are sparse, but what's out there suggests there will be four sets of four "processing engines" sharing essential chip resources, including L1 cache. Is that really sixteen cores as the term is commonly used in the industry today, or is it stretching the definition? If you count them seperately, is it unfair not to count the discrete execution units that make up a "core" in Intel or AMD's offerings seperately too?

Cores are the new megahertz - a near meaningless way of comparing performance between chip makers, or even between product lines (especially with Sun, considering the considerable design differences between the Rock and the Niagara). As chip makers diverge even further in their designs, the nomenclature is getting muddied, with each vendor pushing whatever definition gives them big numbers to advertise with; hence the disagreement as to whether multi-chip modules are truly "multi-core"; hence Sun advertising the T1 as "32 thread" when it can only execute 8; hence Apple advertising dual processor machines as "quad Xeon".

Sun does a lot of very clever things in their designs, and I'm very interested in seeing what kind of performance Rock pushes, or Niagara 2 for that matter. I expect, as usual, that performance will vary highly depending on workload. In the case of Rock, if the reports of only 64K L1 cache (32K/32K split for data and instruction) per group of four processing engines is true it may scale poorly as data set grows. It's hard to predict though, since they may have a particularly large L2 cache with low latency access or some other scheme for minimizing the impact of sharing such a basic resource.

It's hard to make any prediction without detail, though, and the design significantly diverges from their only other recent new design in a number of areas. [Note: I'm not counting the UltraSparc IV+ as a new design; it's mostly a die shrunk IV with more cache, which itself was just a pair of IIIs on a single die.]

and socket type?

[Danathar]

And it will probably require ANOTHER slot type and force me to upgrade my motherboard yet AGAIN!

Geeze...please let me keep my motherboard for 6 months!

Re:and socket type?

[Anonymous Coward]

No it doesn't require a new socket. The socket is still AM2, so you can keep your mainboard. The boards, that do come out now are AM2+ boards, they offer a new power saving tech for the Phenom's, which will save you about 10% power consumption.

Re:and socket type?

[Josef Meixner]

In every press release AMD stated it will run in AM2 sockets. If I remember correctly it will not be able to use the new hypertransport links, support for the new power saving functions (it can switch off complete cores if they aren't needed) in AM2 sockets, it will need AM2+ for that. Sorry, I am far too lazy to search for a reference for those last bits of information, it is something I read in a magazine (paper version).

Re:and socket type?

[Eukariote]

Indeed so. Anyone having bought or buying an AM2/AM+2 desktop motherboard can drop in Phenom processors. When you have a performance AMD 4x4 (1,207-pin Socket F) board with FX processors, you can drop in the new quad core FX chips as well. Similarly, when you have a DDR2 Opteron server/big-iron, you can also upgrade.

That makes the current AMD platforms attractive: you can buy a cheap Athlon X2 chip to get good performance now, and later upgrade to a Phenom chip and get excellent performance and four-way multiprocessing. I plan to wait with my upgrade until the price comes down a bit.

Re:and socket type?

[minorproblem]

Who upgrades just their processor anymore anyways, most of the time its just not worth it. Ussually easier to just upgrade everything at once. If your going to be running a bleeding edge processor why would you want to run it with an old video card and only 1 gig of ram that probably isn't the right speed for it.

Re:and socket type?

[Anonymous Coward]

Yeah whatever. Remember socket 940?
That was the high-end socket for K8, for Opteron/FX chips, while Athlon64 took the cheaper socket 754.

Then AMD marketing wonks decided to invent socket 939 to differentiate the market further and isolate desktop and server platforms. (And don't fall for the marketing BS. For the last time, no, 939 doesn't have anything to do with unbuffered RAM. Sockets have nothing to do with that. Unbuffered support is purely a function of the new CPUs' fixed memory controller. Older K8 revs had horribly unstable first-generation memory controllers that couldn't drive multiple DIMMs without registered memory.)

The net result was anyone with a desktop socket 940, such as the ASUS SK8V, got the shaft. You had to start buying more expensive Opterons instead of newer FX CPUs. And then, quite strangely, Opteron 1xx migrated to socket 939 exclusively so you had to buy Opteron 2xx even for a single-CPU system.
And with socket 939 getting all the attention, companies like ASUS abandoned BIOS updating for the socket 940 desktop boards. You can't even use a dual-core CPU on them, let alone any newer K8 rev.

I predict the same fate for AM2. DDR3 is on its way, after all. Then the AMD marketing wonks will force a new socket on us and you AM2 users will be out in the cold. I'm waiting this time. Sweet revenge.

Re:and socket type?

[ceeam]

I think AMD would've stick with 939/940 sockets if memory manufacturers would not switch to cheaper (but otherwise the same performance) DDR2 memory.

Re:and socket type?

[Chandon Seldon]

For the last time, no, 939 doesn't have anything to do with unbuffered RAM. Sockets have nothing to do with that. Unbuffered support is purely a function of the new CPUs' fixed memory controller.

That's great and all, but it doesn't actually help you in this case. The registered / unregistered thing still meant that there was no chip / board combination that would have worked had they not introduced a new socket. A lot of people complain about the numerous AMD socket changes, but new versions of the same socket that were incompatible with old chips (or vice-versa) wouldn't have been any better.

Re:and socket type?

[Gr8Apes]

It's only been in the past 6 months or so that DDR2 prices fell below DDR prices.

a mobile version?

[Vo1t]

Any info on a mobile version of Barcelona ? I think that AMD should follow Intel in the sense of making an uber cool mobile processor first (that motivates squeezing the most from one Watt) and then give it full throttle for a desktop version. Just like it was done with the Core.

Re:a mobile version?

[Eukariote]

For mobile, AMD has gone a different route for now, they have reworked the K8 for extremely low power: http://www.theinquirer.net/default.aspx?article=39 894. The two cores and memory controller get independent voltage planes. And the cores can clock up and down independently. It makes good sense: for mobile, low power is crucial.

Many of the high-end features (double FPU units, hypertransport interconnects, and so on) of the Barcelona design are not required for a laptop, and add power draw caused by static leakage, even when not in use. In due time, though, AMD will no doubt rework the K10/Barcelona core into a mobile design. Probably they will release a moderately power mobile Barcelona version before that, for high-end workstation type laptops.

WTF?

[GreatBunzinni]

Photos of processor dies? WTF is this? Some kind of porn for uber-geeks?

Re:WTF?

[ScriptedReplay]

You must have missed Intel's marketing campaign after the C2D launch - CPU die pics were in every other subway in Manhattan. At least AMD only sticks them in press releases for now.

Today's processor marketing explained to geek

[DrYak]

sed " s/number of GHz/number of cores/ " marketing.txt

Re:Today's processor marketing explained to geek

[pakar]

So soon we will have 10^9 cores..

Don't even wanna think about the overhead for checking locks :)

Shhhhh!

[dreamchaser]

Some of them are busy fapping to the pic right now, so hush. You'll spoil the mood.

Re:Shhhhh!

[default luser]

Those cores aren't real.

Seriously, that "pic" is a render.

Danger! Danger!

[eebra82]

Now that AMD released high-res pictures of this core, Epson can use their transistor printers we have read about and start selling this CPU ahead of AMD. Good job, AMD!

Re:Direct links to JPEGs

[instanto]

I am thinking Sim City when I see this

http://www.amd.com/us-en/assets/content_type/Digit alMedia/43263A_hi_res.jpg

Looks like the industry areas are quite big, wonder how the pollution in that city is.

No fires though, so that is a good thing.

MMMM... Breakfast is computing

[failedlogic]

These multi-core CPUs are a great direction for the industry. The real question is, when is the 10 CPU processor coming out?

I think this will be a great option for people who get in early at the office. The original Pentium is able to cook an egg on top of the CPU. With 4 cores comes complete breakfast for one person: 2 eggs and 2 toast. I suppose the real key is a workgroup CPU with 10 cores would be useful each is used to cook in total 4 eggs, 4 toast and 2 cups of coffee (you do have to feed your co-worker ... I think).

Re:MMMM... Breakfast is computing

[stonecypher]

The Sun Niagara, which has 80 cores, has been on the market for several years now.

Re:MMMM... Breakfast is computing

[Mr Z]

Err... I think you mean 8, with four threads per core. To the OS it looks like a 32-way machine on one chip. It's Intel that has been in the news again recently about its 80-core research chip. And then there's those GPUs with 128 cores on them that you can program in C.

Re:MMMM... Breakfast is computing

[stonecypher]

You are correct. My apologies.

Nooo... Not a new hype word!!!

[LWATCDR]

Okay I really like my AMD system but they need to be slapped hard for inventing a new goofy marketing term.
MEGATASKING.
Dude if you have over a 1024 tasks running at once you need to run some malware clean up software.

Re:Nooo... Not a new hype word!!!

[forkazoo]

Okay I really like my AMD system but they need to be slapped hard for inventing a new goofy marketing term.
MEGATASKING.
Dude if you have over a 1024 tasks running at once you need to run some malware clean up software.

My friend, you fail to appreciate the lunacy of the intricacies of marketing. That which you have described would, in fact, be merely kilotasking.

Re:Nooo... Not a new hype word!!!

[dodobh]

Dude, that was my webserver being hit by Slashdot.

Re:Nooo... Not a new hype word!!!

[LWATCDR]

Of course it was just kilotasking. That is why the term megatasking is so dumb...
How's that for covering up a silly math mistake?

Of course Intel will kill AMD. They will jump right to Teratasking!

whip

[Jeek Elemental]

put the photo in L1 cache to send a not-so-subtle message to your cpu

AMD slideware

[kemo_by_the_kilo]

if you RTFA....
aw great now i have to protect my computer from AMD slideware....

It isn't?

[Xocet_00]

In all honesty then, what is?

Neg, that would be kilotasking

[Xocet_00]

Megatasking would be 1,000,000+ processes at once.

Finally

[Anonymous Coward]

they'll be outfitted with 2MB of on-chip L2 cache (512KB dedicated per core)

It's about effing time... maybe chip manufacturers have finally clued in that cache is the single biggest characteristic of a processor that affects (NOT impacts) performance. I have seen far too many 2-3GHz chips crippled by insufficient cache over the years, but hey, it was $20 bucks cheaper and the same speed so it must be a better deal right? Too bad that this will probably not make the market and the cache will be cut back to 64KB per core to shave a few dollars off the price and suck more people in to buying crippled gear...

Re:Finally

[Just Some Guy]

It's about effing time... maybe chip manufacturers have finally clued in that cache is the single biggest characteristic of a processor that affects (NOT impacts) performance.

Tell me about it. Those jackass chip hackers at Intel and AMD have been ignoring my advice for years in favor of their own cost/benefit analysis and engineering tradeoffs. If only they'd listen to us expects on Slashdot, there's no telling what they could accomplish!

Obligatory...

[bberens]

$670 CAD is $266 USD

Re:Obligatory...

[murray_420]

Actually by end of year they expect the Canadian Dollar to reach parity with the US Dollar. (Currently US dollar is $1.0604 Canadian)

Re:Obligatory...

[bberens]

If there ever comes a time when $1CAD == $1,000,000US I will still make that joke.

Yeah, this qualifies as second worst Slashdot news

[aliquis]

... item ever, after the big reset button which was posted this very day aswell.

Maybe it's time to shut down Slashdot?

Well then,

[TheCreeep]

Why don't they just release the CPU? I mean they have it working, they tested it and stuff.
I'm not trolling, I'm just curious to find out what changes a processor goes through in it's last months before being launched.

Re:Well then,

[ChrisMaple]

Two things come to mind. One is, AMD wants to make ultra-sure that there isn't some as-yet unnoticed mistake that will cause a product recall. The second is that there's a delay between the time somebody says "go" and packaged chips can be put into computers, most of which is the several weeks it takes to process wafers.

Memory! Lots of Memory.

[triso]

In all honesty then, what is? Memory capacity, greater than 4 GB.

speed

[itzdandy]

though AMD has lost a bit of thunder recently with the core2 line kicking butt, this multiple core phase the industry is moving towards is going to help AMD. They made a decision to change the bus for multi-core systems a while ago and this chip is the first one to really show off that bus. I think that Intel will have faster individual cores for a while and barcellona wont challenge that but AMD has a much better multi-core multi-processor bus so multi-core scaling will be much much better for AMD. an 8 core AMD will likely stomp an 8 core Intel chip though the Intel chips will be better per individual core.

I just hope that AMD can make a big enough improvement on their cores before Intel can get thier cpu to cpu speeds up to par. I have an Intel dual CPU/dual core Xeon server at work that is fast but i cannot tell the difference in most things between a single 2core and the dual 2core setup because of the week bus.

to clarify for those not up to speed on the current platforms, AMD has a high speed dedication point to point bus for each CPU and each CPU has a direct dedicated link to its own memory while intel has a fast shared bus that can get saturate especially when only running at 533 or 800 mhz giving AMD the advantage in that regard.

Coming soon....

[DrYak]

Featuring next year, an AMD sponsored article called the "The multi-core myth", AMD processor beging being branded with a N-code in their name (letter "N" + a number) which they pretend is only unit-less number to tell apart models of processors, but which incidentely is very close to the perfomance of an Intel with that number of cores.

Then, as part of their Torenza initiative and GPU onboard of CPU, AMD introduces processors with a huge amount of vector stream-processing units. It is supported by Linux even before hitting the market (thanks to previous work on Cell) and is immediately adopted by the scientific community.

Intel announces that their separate line of GPUs on PCIe is no more while simultaneously scraping their "Intel Core Treisdeka-duo 32" cooking-ware... oops.. processors and droping the "Core" name, and promise to release newer CPU based on an older technology and featuring massive amount of stream processing units, called "Intel Hub Trio".

Marketing for next decade :

" s/number of cores/number of steam processing vector units/ "

Wash, rince, repeat.

What is with the rinky dinky cache?

[tek_heretik]

I wound up buying a CD2 E6700 cuz it has the shared 4MB cache and it smoked the top-o-the-line AMDs too. I am sorry but I have found that in AMD's quest to be fast, their CPUs are not as stable as Intel's. My C2D compresses, encodes, encrypts, etc. like wildfire (in combination with a 3 drive Raid 0), runs cool, uses little power and doesn't crash. I can't say I have had the same luck with AMDs in the past. >:-/ Also, more cores does not translate into more performance unless they are actually utilized, for example, the new quad C2s have little performance increase (for J. Sixpack running everday OS and apps) over their dual core predecessors. >:-/

Re:What is with the rinky dinky cache?

[Chandon Seldon]

Is this a question, or is this Intel fanboy gibbering? Whatever, I'll just answer the question.

What is with the rinky dinky cache?

The Core 2 Duo E6700 has 4 megs of shared L2 cache and 2x64k of L1 cachee. All of the upcoming AMD chips will have 2 megs of shared L3 cache, 4x512k of non-shared L2 cache, and 4x128k of L1 cache. Nobody has rinky-dinky cache - it looks like everyone agrees about 4 megs of cache is the right answer at 65nm.

Historically, the reason that AMD has lagged behind Intel on cache is a combination of two factors: 1.) They've been a process generation behind a lot of the time and 2.) Their onboard memory controller lets them get away with cache misses easier than Intel's FSB plan. If they're lucky, this stuff (plus extra L1 cache, which is a really big deal) will let AMD regain the performance crown until Penryn happens.

As for "your computer doesn't crash", I have to say: Good job. You've managed to get on par with everyone else who doesn't have a defective computer. Can't say I've ever had a computer crash because of a defective CPU - defective ram? sure. Hard disk? all the time. CPU? Never. Not Intel, AMD, Cyrix, VIA, IBM, or any other CPU.

Re:What is with the rinky dinky cache?

[tek_heretik]

First of all, I am not a fanboy of anything (sept maybe Linux), I go with what works and is the best VALUE for my money, not an overheating (AMD are notorious for hot running CPUs, good fan/heatsink or not), poorly designed fan (the stock AMD fans are pathetic) that loves to get clogged costing many people money (I have a registered business and work out of my home fixing PCs so next time don't be so quick to throw around names). I had a 1733MHz Athlon once, ONCE, it would kahk under real heavy loads AND LOCK UP, so don't give me the "never had a bad CPU" crap, it was bad by design (when I was shopping to build this machine, I remembered that experience). Maybe it wasn't defective but is sure was a piece of sh*t. You are right, they are ALWAYS BEHIND INTEL (most of the time) when it comes to products and development. It's a free marketplace so I go with what will work AND LAST. I just had to replace a woman's computer because her AMD partially fried (it was fine until you get into windows, when it had some 'real' work to do, it would kahk), I tried everything to eliminate all other possibilities except change the motherboard (pretty hard to find an older socket board these days-at least one that works good) and it was all because of poor heatsink/fan design, they are certainly not built to last. See, you think you are some kind of hardware guru, well I have some news for ya pal, 95% OF THE POPULATION AREN'T, THAT IS WHO AMD SELLS TO TOO, SO STFU A$$WIPE. ,,l,,

Re:What is with the rinky dinky cache?

[Chandon Seldon]

Amazing... you had a bad experience almost 10 years ago, and then you had an issue with a heat sink, and now you're 100% sure that AMD products are "crap". All I have to suggest is this: Be a little less rabid about spreading that anti-AMD FUD. I may not currently be fixing computers for people as my primary job, but I frequently help a couple of friends who do - and I see no evidence of a significant reliability difference between Intel, AMD, or even VIA on consumer-level processors today.

As for historic processors, all I have to say is this: Make sure you get the heat sink mounted right on the original Athlons and the older Pentium IVs.

Re:What is with the rinky dinky cache?

[tek_heretik]

10 years ago? Boy, you sure are NOT a hardware guru (read the date of the review... http://www.sharkyextreme.com/hardware/cpu/article. php/991011 ). I don't care how many friends you have helped, except for a brief period just before Intel came out with its new line (C2Ds), AMD HAVE ALWAYS BEEN BEHIND, ADMIT IT, WHAT'S WRONG WITH YOU? Also, I had an old IBM mini-tower that came with a stock K6-II 350, I dropped a brand new K6-II 500 in it and abused the hell out of that little machine but of course it had a decent heatsink and fan on it, so it's not like I have never had a good AMD, you are just not listening, IN THEIR QUEST TO KEEP UP WITH INTEL, THEY PRODUCED HOT RUNNING, UNSTABLE CPUs. I can remember building many junkers to sell in my yardsales and NEVER had a problem with Intel CPUs, NEVER. They were on ocasion, way slower than there AMD equivalents but they always finished their tasks, RIGHT THROUGH TO THE END, where as an AMD would lock up. Look, I hate monopolies as much as the next person, in an ideal world I would own a quad AMD running some shiny Linux distro but unfortunately, that world isn't here yet. >:-/ I know it's hard to admit but Intel leap-frogged over AMD with the C2D line, lets just thank all the Gods of all the religions that AMD exists and Intel has some formidable competition to keep it in check. Whatever happened to AMD's anti-competition lawsuit against Intel anyway? BTW, my prediction is Europe will be mostly made up of AMD machines running Linux in 10 or 20 years, which is not such a bad thing. >:-/

Re:What is with the rinky dinky cache?

[Chandon Seldon]

Sorry, I'll admit I speed-read over "1733 MHz" and assumed that "Athlon" didn't mean "Athlon XP". So it was five years ago.

I'm still unconvinced by your vague anecdotal evidence that AMD processors are innately unstable. Like I said, I've personally seen a statistically significant number of computers built, deployed, and supported - and I've never seen processor stability issues that could be legitimately attributed to manufacturing quality. Are you 100% sure that your "unstable" Athlon XP didn't have bad RAM, or improper cooling, or a virus infestation?

Seriously, if there were some chronic stability problem with AMD processors I'm pretty sure I would have noticed it by now. I guess the last heat dependent crash I saw was an AMD... the processor was running at 90 deg C because the heat sink was clogged with cat hair. Surprisingly, the owner of that system hasn't complained about any residual stability issues since I vaccummed it out for him - I would have expected some lasting damage.

Re:What is with the rinky dinky cache?

[rrhal]

One other note about older AMD CPU's. Well realy it's a not about thier motherboards. If you take the northbridge heatsink off you will find (usually) 1) it's warped 2) The etruded surface has big gouges in it 3) it was mounted with some kind of TIM tape that is ineffective. If you sand down the northbridge heatsink until it is flat, sand it out to 600 grit so it is smooth, and remount it with a decent thremal paste, the memory controller will actually be cooled and you wont get memory errors. Same is true of the bus controller.

If you want before and after confirmataion just put your finger on the heatsink. Before you will feel almost no warmth because not much heat transfer is going on. After it will be warmish to the touch. Its actually conducting heat.

Re:What is with the rinky dinky cache?

[Chandon Seldon]

If you take the northbridge heatsink off you will find (usually) 1) it's warped 2) The etruded surface has big gouges in it 3) it was mounted with some kind of TIM tape that is ineffective.

Who made the motherboards you're talking about? That's definitely a motherboard manufacturer problem, not an AMD problem - you want to pay a bit of attention to who's making your motherboard, since it's an important component in the functionality & stability of the PC.

Re:What is with the rinky dinky cache?

[rrhal]

I've seen this on ASUS, Gigabyte, EPoX, and Shuttle motherboards all at various price points. It seems they all cut corners a little bit with the north bridge heatsink. And yes - thanks for clarifying the point - this is not an AMD CPU problem. But it does cause random crashes to the desktop that people frequently blame on the CPU.

Re:What is with the rinky dinky cache?

[tek_heretik]

Boy, you don't give up, you must work for AMD to be so adamant. Look, I could care less who makes my CPU as long as it is fast, stable, runs cool (there are residual costs to hot running CPUs like ambient cooling >:-P) and doesn't break the bank when I get my electric bill. Right now, Intel happens to fit that bill with the C2D. Here's another thing, how come AMD based machines are always the 'budget' line (you get what you pay for remember) and they are no different when it comes to price gouging either, when they were 'on top', they were charging outragious money for their 'top-o-the-line' CPUs. So one more time, I DON'T WANT TO HEAR HOW GREAT AMD IS, DAMN IT (Intel is certainly no angel but right now they make a better CPU). Let's agree to disagree cuz I have better things to do than go back and forth with you on this. >:-/

Re:What is with the rinky dinky cache?

[tek_heretik]

Here you guys, go look at this vid... http://www.youtube.com/watch?v=9AQmuUEDxGI&mode=re lated&search=

Re:What is with the rinky dinky cache?

[tek_heretik]

And yes, I know it's old but nothing changes sept the CPU speed. >:-P

sounds powerful

[webmonkey44]

This sounds powerful!