Slashdot Mirror
Inside AMD's Phenom Architecture
An anonymous reader writes "InformationWeek has uncovered some documentation which provides some details amid today's hype for AMD's announcement of its upcoming Phenom quad-core (previously code-named Agena). AMD's 10h architecture will be used in both the desktop Phenom and the Barcelona (Opteron) quads. The architecture supports wider floating-point units, can fully retire three long instructions per cycle, and has virtual machine optimizations. While the design is solid, Intel will still be first to market with 45nm quads (the first AMD's will be 65nm). Do you think this architecture will help AMD regain the lead in its multicore battle with Intel?"
intel released the core 2 duo in 1996????
Whoa, I'm more behind than I thought.
I had a 2P dual-core opteron 2.6GHz box as my workstation for several months. To be honest I couldn't really find a legitimate use for it. And I was running gentoo and doing a lot of my own OSS development [re: builds].
While I think quad-cores are important for the server rooms, I just don't see the business case for personal use. It'll just be more wasted energy. Now if you could fully shut off cores [not just gate off] when it's idle, then yeah, hey bring it on. But so long as they sit there wasting 20W per core or whatever at idle, it's just wasted power.
To get an idea of it, imagine turning on a CF lamp [in addition to the lighting you already have] and leave it on 24/7. Doesn't that seem just silly? Well that's what an idling core will look like. It's in addition to the existing processing power and just sits there wasting Watts.
Tom
Someday, I'll have a real sig.
Ultimately, it's performance that makes a successful product, not gigahertz or nanometers.
Sure, the 45nm process has great potential for better performance and higher efficiency, just like faster clock speeds had great potential - until AMD made a better architecture and achieved better performance at a lower clock speed than Intel's offerings at the time.
Let's wait and see how it really performs before passing judgement.
=Smidge=
In terms of market share, no. In terms of tech yes. See Opteron v. Intel P4 Xeon for example.
Tom
Someday, I'll have a real sig.
Quad core is all well and good, but are there really that many apps as of yet that can take advantage of it? TFA claims this is for servers and for desktops, and I'm not certain of its utility on the latter just yet...
P4 was never multi core... multi processor yes... but core no...
The problem is, analysts don't look at technology so much as they do market-share. The unfortunate battle between techies and beancounters wages on...
When it comes to multi-processing scalability, AMD's Barcelone/10h/Phenm single-die four core with hypertransport inter-chip interconnects will do far better than the two-die four core shared-bus Intel chips. Also, both the old and new AMD architecture will do relatively better on 64-bit code than the Intel Core 2 architecture: Intel's micro-op fusion does not work work in 64-bit, and their 64-bit extensions are a relatively recent add on to the old Core architecture. The FPU power of the new 10h architecture will be excellent as well. On the other hand, Intel chips will remain very competitive on integer code, cache-happy benchmarks, particularly when run in 32-bit mode. Also, the SSE4 extensions of the upcoming 45nm Intel ships will help for encoding/decoding and some rendering applications, provided that the software has been properly optimized to take advantage of them.
The Athlon X2 was superior to the Pentium D. It wasn't until Core 2 Duo that Intel took the lead in desktop CPUs.
OSSRocks, meet Pentium D - P4 dual-core with new marketing name.
I introduce to you the Pentium D.
The enemies of Democracy are
This article seems to be trying to rewrite history:
Amazing that, a 10 year lead on dual core parts!
todo - The developer's equivalent of confession: "Forgive me Father, for I have sinned..."
ahh yes i forgot about that one, i didnt realize they were sold i thought they were just engineering samples, i didnt bother biulding a new box until core 2 was out... probably why lol good catch!!
Take another look. He's making fun of the date they mentioned (1996).
Patrick Doyle
I mod down every jackass who puts his moderation policy in his sig. Oh, wait a sec....
I hope AMD neglects to advertise this one too.
Someone makes this same comment every time advances in CPU technology are mentioned.
Patrick Doyle
I mod down every jackass who puts his moderation policy in his sig. Oh, wait a sec....
HotHardware.com has more details on the AMD Phenom processor, including die map shots and other specifications as well.
1996? Wow, have *I* been misled. Mid 1996 is the vintage of my Dual Pentium Pro 200Mhz, and I *really* thought that it was state-of-the-art.
Colour me disappointed...
Just another "Cubible(sic) Joe" 2 17 3061
Actually, I just got a 65W Athlon X2 4600+ from Newegg which uses less power than my current 6 year old Athlon XP 1800+. The motherboard (ECS w/ ATI 690G) I ordered supposedly is also energy efficient. I guess I could save $60 by getting a single core, but almost all single core Athlons are rated at more than 65W. Why buy a single core when it costs more long term and is slower when multi-tasking?
I think that AMD beats intel, as a cheap underdog brand it kicked Intel of the top.
Overall i think however that single or complex duo / quad cores or more are a dead end.
To get performance a complete redesign is required, not only the main chips.
Perhaps a tandam of ARM chips using 3 types of mem compactflash besides fast ram and a disk.
And a diffirent aprouch to a memory bush, perhaps rather clusters of small chips with some memory who together devide tasks simultane. spreading out load is often cheaper then tuning up load on a single (critical) part.
it's just a mater of time then we will see this.
Also i think a cheaper base material should be found, well perhaps the next chips are made by small nano fabrics
I know you're out there. I can feel you now. I know that you're afraid. You're afraid of us. You're afraid of change.
Intel, a giant corp with continued antitrust oversight, was quite happy to allow AMD to appear to be a genuine competitor to the US authorities, as long as AMD's market share remained a relative sliver. Once AMD's CPUs achieved a much better price/performance ratio than Intel, Intel moved to squash them like a deer tick in Olde Lyme, CT.
Does AMD stand a chance against Intel? Not unless they can make a profit out of having less than five per cent market share for the rest of eternity. Perhaps they can.
In any event, any competition is better than none. If it weren't for AMD's great processor architecture and 64-bit extensions, we doubtless would not have Core-2 duos and quads at an affordable price point now.
But Intel's R+D PIZZA BUDGET must be larger than AMD's total worldwide cash flow.
(Full disclosure: I build my own boxen, and have since 1991. I have never used an Intel chip in any computer I've assembled. So I have no grudge against AMD; I used them because they performed very well for a low price. They still do.)
Well it's not like anyone cared about the Pentium D, since it was pretty craptacular. Lessee, take a bandwidth-starved core design, slap two down in one package, downclock the system bus since the MCM has some of the same signal integrity issues as multi-socket, and you get... two cores that are starved for bandwidth even more. Yay.
So you were much better off waiting until Core 2 regardless, if you wanted an Intel dual core anyway.
The enemies of Democracy are
The original Core Duo led most benchmarks as well (I have a T2600 before me.) Core Duo is really an excellent design (made a nice change.)
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
It seems that AMD's research department is only concerned with beating Intel at its own game. This is foolish, IMO. AMD is doomed to always be a follower unless its engineers can come up with a revolutionary new CPU architecture based on a revolutionary software model. The new architecture must address the two biggest problems in the computer industry today: reliability and productivity. Unreliability puts an upper limit to how complex our software systems can be. As an example, we could conceivably be riding in self-driving vehicles right now but safety and reliability concerns will not allow it. Why? Because there is something fundamental wrong with software. Fortunately, a software model that solves these problems already exists. It is called the "non-algorithmic, synchronous, reactive software model. That's what Project COSA is about.
What's new in the article? All this was announced at the conference in Germany in January!!! Why is Slashdot even posting this? The only interesting thing is that the hype has made AMD's share go up 10 % in the last 2 sessions:)
**Life is too short to be serious**
This is, again, US style journalism: just parotting press releases. No depth, no investigation, no knowledge involved. These lazy writing parrots become so f***ing annoying and a waste of my time.
Please post links to articles when the contents includes some orginality and knowledge, otherwise just link to the press releases themselves!
At one point, the Opteron was single-core. And it still beat the P4 Xeon.
tasks(723) drafts(105) languages(484) examples(29106)
You might even be right, but my instinct says no.
Intel should have realized, from how U.S. govt. treated Microsoft and others, that they weren't in NEED of someone like AMD to cut deeply into their sales for a while with truly competitive products.
I'd say your scenario would hold much more merit if govt. had already broken up Microsoft into separate divisions or something....
The fact is, AMD has occasionally built a very comparable, yet cheaper alternative to Intel's offerings. (Remember the success of AMD's response to Intel's 486DX CPUs?) Whenever they pull this off, they do quite well, UNTIL Intel ups the ante with another huge R&D effort, and produces something "next gen" that AMD has to counter.
I don't think Intel is afraid of AMD "taking over the marketplace" in processors.... but I don't think they'd be wise to ignore them as irrelevant either (or only existing as a "straw man" for the sake of legal arguments). They certainly proved themselves much more of a competitor than, say, Cyrix.
Only in benchmarks and anecdotal evidence. Apart from that, who's to say?
Craptacular indeed (great new word) - the only thing craptacularer was the Celeron D they had out at the same time, which despite the name was not dual-core. Very amusing though, watching the 'tards with enough knowledge to be dangerous and who wanted a cheap PC,
"That one's a 'D', that's got 2 processors, that makes the internet faster"
Your technical analysis is correct, but the pentium D is so cheap on newegg that it replaced the celeron line out in october. Its now a good entry level dual core chip. My Pentium D 805 outperforms a Dell Precision 650 with 2 2Ghz Xeon chips. As bad as the Pentium D is, there as still a little progress. Cooling it is another issue all together.
MidnightBSD: The BSD for Everyone
They seem a bit slippery there. When will the Barcelona Opterons ship? Anyone know?
AMD's cool & quiet tech will shut down individual cores when you are not using them. I believe this is all new for the Barcelona. It idles down cores when you are not using them fully. It shuts off parts of cores that you aren't using (eg the FPU if you are only using integer instructions).
According to the last picture [imageID=9] in the Image Gallery, different cores on the same chipset can run at different voltages and different MHz's:
http://www.informationweek.com/galleries/showImag
He used the phrase "begging the question" properly!
Well, the Athlon X2 was probably a better desktop chip, but the Core Duo was a better laptop chip than anything put out by AMD.
Ewige Blumenkraft.
I've been using craptacular for years, don't make be break out the DMCA!
Funny thing is, everyone knew Core2 will beat X2, even before X2 came out. AMD must have crapped their pants when they saw the samples. (no, I'm no fanboy, but please search the archive from bitnet for example, and see for yourself). Core2 was the hammer for AMD like K8 was the "sledgehammer" for intel back in the day.
Uh, doesn't "make -j 3" gives you a good speedup? I'd imagine multi-core being great for development, at least for compiled languages.
Well, the Athlon X2 and Core Duo (Yonah) ran pretty even at the same clock, but it was only for mobile. We never got to see how fast the Yonah could run at the top end. The fastest one was 2.3GHz, but given the overclocking headroom of mobile processors, it could have beaten Athlon FX's with the right desktop platform.
The Core Duo had two big downfalls compared to the X2s: one was that it was a 32-bit chip and not a 64-bit chip. The second was that the Core Duo had the Pentium III/M's relatively weak FPU compared to the much more powerful FPU units in the X2. This made the Core Duo roughly as fast per clock as the X2s in 32-bit applications that did not stress the FPU much, but FPU-straining apps had the X2s pull away mightily from the Core Duo. I'd never seen a direct comparison between an X2 running in 64-bit mode versus a Core Duo, but other tests with X2s in 32- and 64-bit mode show about a 10% increase in performance by running in long mode. Thus, the X2s were potentially a bit faster than the Core Duo was if each was allowed to run optimal setups.
Just "gittin-r-done," day after day.
Hard to say, actually. The P4, for example, had some novel technology in it (advanced branch prediction, cache, etc.) but had other parts that caused it to have problems (pathological retry, for example). The Opteron also had some novel technology (IMC, HT, etc.) as well but a lot of its core wasn't necessarily that technologically advanced. It's hard to say which one was more technologically advanced... brute force can go a long ways for pure performance.
Troll? It's an honest question and was answered quite well by people that responded. Well done abusive moderator.
It's getting really boring to see all these comments in threads about multi-core CPU with clueless people ranting that: "besides multimedia editing, nobody will need multiple-cores".
In a near future I'll benefit from having, say, an 8 cores @ 2 Ghz CPU because, well, 16 Ghz CPU simply won't exist at the time that 8 cores @ 2Ghz comes out. This is, for a start, an undeniable fact that a great many people will benefit from a single core.
Then usually anyway you're using a multi-threaded OS running, say, multi-threaded applications (big surprise, Java applications suddenly became a lot faster while a great many C-written mono-threaded app are lamely using only one core ["phenomenon" that can be seen very easily in a CPU monitoring facility]).
Or maybe that some desktop users are running OS X + Windows under Parallels? Mutliple cores shines here.
So, please, stop whining in the next thread about 16-cores CPU that you'd rather have an hypothetical 32 Ghz CPU running mono-threaded apps on a single-tasking OS.
AMD was doing great, and for years they had the cost/performance ratio to justify their success. But now Intel has a better chip cost/performance wise (in the 200-300$ range), and they are on top. I've used AMD for years but when my buddy asked me how he should go last weeks I told him Intel. Nothing is better for the market than to have two teams who are almost equal - it's fun to watch and their hard work is our benefit.
It's hardly a battle. Each group pays attention to what it cares about.
The Athlon X2 was superior to the Pentium D. It wasn't until Core 2 Duo that Intel took the lead in desktop CPUs.
I think you mean "regained". Computers did exist before 2003.
But I does know now that AMD is losing, and very hard.
ghostbar page.
You want to back up those claims with some data or at least links?
AMD's lowest power usage CPU's are currently 35W and made on 90nm SOI for AM2 socket with F2 Stepping. They range from a 1.6 Ghz Sempron on the low end up to a 2.0 Ghz Athlon X2
You can get info on all available models at these urls:
Compare Desktop Processor Specifications
Compare Server/Workstation Processor Specifications
AMD Processor Pricing
The question is of course can you buy one retail?
Woopty Doo Basil, what does it all mean?!