backtop


Print 122 comment(s) - last by murphyslabrat.. on Dec 22 at 1:05 PM

AMD plans to keep "Brisbane" around, releases new chips based on it

Things at AMD may have gone from bad to worse with the lackluster Phenom launch in late November.  Not only did Phenom fail to appeal to professional reviewers, but the company ended up removing one third of its CPU lineup just after the big day.

Last week AMD CEO Hector Ruiz vowed that the company would stop hemorrhaging cash and return to profitability soon.  "That is our number one goal right now," Ruiz said in a conference in Bangalore

Making a profit at AMD apparently means refocusing on its older K8 architecture.  The company will introduce eleven 65nm K8 processors over the next two quarters.  By comparison: AMD launched two quad-core K10 Phenom processors in November with three more scheduled over the next two quarters.  Two tri-core Phenom processors will follow in March 2008.

Essentially, AMD will move any remaining Athlon 64 processors from the 90nm node to the 65nm node, with a few new frequency and TDP variations.

The AMD Athlon 64 X2 5600+ will be the first to jump on the new 65nm K8 bandwagon with a 65W TDP. The previous Windsor-based chip of the same featured an 89-Watt TDP. AMD will also add 100 MHz to the core frequency of the Athlon 64 X2 5600+, now rated at 2.9 GHz. Total L2 cache will be halved in the move to the Brisbane core, and the updated Athlon 64 X2 5600+ chips will feature only 1MB of L2 cache. Availability of these processors is scheduled for Q1 2008.

AMD's higher-end Athlon 64 X2 6400+ and Athlon 64 X2 6000+ will both be discontinued.

AMD will also update its "Energy Efficient" series and will release three new chips, the AMD Athlon 4850e, Athlon 4450e, and Athlon 4050e in Q2 2008. All of the new offerings will be based on AMD's Brisbane core and will feature a 45-Watt thermal envelope. AMD's current energy efficient "BE-2xxx" series will be phased out at that time. Respectively, the new chips will run at 2.5GHz, 2.3GHz and 2.1GHz.

All new Brisbane chips will be based on the Socket AM2 interface.  These processors are compatible with AMD's AM2+ socket designated for Phenom processors.


Comments     Threshold


This article is over a month old, voting and posting comments is disabled

RE: Goodbye AMD
By StevoLincolnite on 12/5/2007 10:55:27 AM , Rating: 3
You make it sound as if AMD is a Failure, the K6-2 and K6-3 were great value Chips in they're time, And the K8 was a great chip against the Pentium 4, And in-fact I still think the K8 is a still a great processor, As far as I last checked any AMD K8 Athlon 64 X2 is capable of still running anything demanding, and the K7 was just a solid chip with a very long life, Especially those Socket A boards...


RE: Goodbye AMD
By 16nm on 12/5/2007 11:00:51 AM , Rating: 4
The smartest thing for AMD to do now would be to keep working on K8 by refining it and getting more megahertz/efficiency out of it. I would not see anything wrong with them introducing a two-die quad-core K8 chip if it is reasonably feasable.


RE: Goodbye AMD
By cochy on 12/5/2007 11:03:31 AM , Rating: 3
There's only so far that you can push a CPU design.


RE: Goodbye AMD
By 16nm on 12/5/2007 11:16:48 AM , Rating: 2
Well, I think AMD are doing a pretty good job with K8 90nm, but they have a long way to go with 65nm. There is a lot of room for improvement there.

Conroe and K8, more or less, are the same design. Intel had great expertise with the 65nm process going into Conroe because of how hard they had to push it on the Netburst design. AMD have a lot more room for improvement with regards to their 65nm process. Fully expect them to introduce better, efficient steppings on 65nm. K8 is top technology and does not need to be retired. Updated - yes, retired - no. For example, SSE4 would not be a bad addition.


RE: Goodbye AMD
By Proteusza on 12/5/2007 12:04:43 PM , Rating: 3
K8 and Core 2 are similar but different. Like a Corvette compared to a Viper. Similar philosophies, different execution.

The Core 2 is superior in integer performance especially.


RE: Goodbye AMD
By MandrakeQ on 12/5/2007 1:32:56 PM , Rating: 2
Core 2 is optimized for high ILP, and it doesn't look like AMD is pursuing that route as aggressively anymore.

AMD appears to be spending their transistors on server features lately, and this means the desktop market gets neglected. I don't really blame them for this decision given the razor thin margins they have to deal with in desktop procs.

Sun has taken a similar path with their Niagara processors by focusing on thread level parallelism, and this appears to be working out well for them.


RE: Goodbye AMD
By Andrwken on 12/5/2007 6:45:35 PM , Rating: 3
The margins are razor thin by not having a competitive product. Are you saying they shifted design philosophy 2-3 years ago when k10 design started because they thought the margins would be bad on the desktop? I thought they were doing quite well in the desktop space back then, and it would seem odd to neglect it on the current design, unless, you didn't expect your vastly larger rival to come back swinging after you started takin pop shots and marketshare.

It's called don't kick the 800 lb. gorilla until your are at least an 800 lb gorilla and can deal with the consequences of waking the sleeping giant. ;)

They just don't have the resources, or capital to play with the big boy at this point, and purchasing ati gave them enough marketshare for intel to pummel them even more without having to worry about the monopoly talk for a while.


RE: Goodbye AMD
By MandrakeQ on 12/5/2007 7:41:03 PM , Rating: 3
Whenever AMD and Intel have had roughly equal performance, margins are smaller for both sides. When one pulls ahead of the other though, the leader can charge more for the performance premium, and they make more money. When you're on the losing side of the battle, then your margins become 0 or negative (AMD is in that situation right now).

K8 was AMD's first processor to successfully compete in the server space (development started back in late 90's I believe). The design philosophy of the architecture is clear when you take a look at its two most defining features, x86-64 and the IMC (neither of which are particularly useful for desktop users). AMD took advantage that Intel wanted to force Itanium down everyones throat, as well as the fact that P4 was so inefficient that it took massive amounts of power just to get it competitive. Not to mention that Intel had at that point considered AMD a bottom feeder and left it completely unchecked.

K10 is an extension to K8 (still with a focus on servers) and it is going to do well in the 8+ core server space. Everywhere else it will lag behind, since there's no way a 4MB cache shared across 4 cores will perform as well as the 12MB cache the Penryn Xeons have (Unless of course you're bottleneck is bus traffic and memory latency, which is the case for the market previously mentioned).


RE: Goodbye AMD
By myocardia on 12/6/2007 1:38:39 PM , Rating: 2
quote:
K10 is an extension to K8 (still with a focus on servers) and it is going to do well in the 8+ core server space. Everywhere else it will lag behind, since there's no way a 4MB cache shared across 4 cores will perform as well as the 12MB cache the Penryn Xeons have (Unless of course you're bottleneck is bus traffic and memory latency, which is the case for the market previously mentioned).


Umm, two things to be noted here. First is that all Barcelona's (the K10 server chip) have been recalled. The second is that 100% of Intel's current quad-core CPU's have 2x4 MB of L2 cache, not 1x4 MB.


RE: Goodbye AMD
By MandrakeQ on 12/6/2007 3:34:42 PM , Rating: 2
I thought Penryn based Xeon 5400's had 12 MB L2 cache? Are these not available yet? As for the recall, its a bummer for AMD, but if they can get the problem fixed soon (hopefully by early next quarter), they should be able to sell a healthy number of them at least until Nehalem comes out.


RE: Goodbye AMD
By joemoedee on 12/6/2007 8:09:40 AM , Rating: 2
quote:
There's only so far that you can push a CPU design.


True, but then you have to consider that the current Intel CPUs that are wiping the floor with AMD performance wise trace their routes back to the P3 which was launched back in 1999.

There's a lot to be said for refinement and enhancement on CPUs. One great thing AMD has going for their K8 CPUS is Hypertransport, and I think they could still ring some good performance gains from their existing design.


RE: Goodbye AMD
By StevoLincolnite on 12/5/2007 11:05:31 AM , Rating: 4
I actually thought they would release a quad core K8, And if some of Quad's weren't functional could have sold them off as single and dual cores.
I Personally Have a Pentium M laptop, Pentium 3 Tualatin 1.4ghz machines and an Athlon 64 X2 4000+ - All solid machines in they're own right.

And considering the Price of the Athlon 64 X2's they are seriously great Performance for you're dollar in budget machines.


RE: Goodbye AMD
By murphyslabrat on 12/5/2007 11:07:47 AM , Rating: 3
Yeah, like imagine if Intel had've stuck with re-iterating the original Pentium instead of giving us the next generation of computing with Netburst...</sarcasm>

Not quite an acurate parallel, but it stands.


RE: Goodbye AMD
By iFX on 12/5/2007 11:52:20 AM , Rating: 1
No it doesn't stand. Netburth was a flaming POS, and K8 is a revolutionary, highly efficient CPU design that until mid 2006 was top dog. It has a long life left if AMD is smart.


RE: Goodbye AMD
By TomZ on 12/5/2007 12:33:25 PM , Rating: 2
Nah, the only future K8 has is if AMD sells 'em cheap. It doesn't compare to Core 2 either performance wise or in terms of energy efficiency. The only other possible angle is to make them cheap enough so that performance/dollar competes with Intel at this point. And in the meantime, Intel is not standing still either...


RE: Goodbye AMD
By mindless1 on 12/5/2007 11:16:56 PM , Rating: 2
They already were, beating Intel in by far the most popular price segment, the under $140 and more recently under $85 category.

Their future depends on lowering manufacturing costs and securing more OEM contracts, because AMD was in 2nd place, is in 2nd place, and will continue to be in 2nd place to Intel. It's not a competition with Intel for performance or efficiency, it's a competition to provide a cost effective solution to OEMs.


RE: Goodbye AMD
By murphyslabrat on 12/6/2007 1:22:13 PM , Rating: 2
The only problem with that is that a consumer will read stuff like, "Intel CPU's overclock like crazy", and "AMD just can't beat Intel when it comes to ultra-high performance"; and they'll go with an Intel CPU, assuming that these reports have a bearing on SOHO.

So, in other words, the "more educated" consumers that we have nowadays, as a result of Intel's Netburst, will go with an Intel CPU. Short of that, we have the domnodies that have absolutely no clue, and will probably end up buying a Netburst Celeron/Pentium D.


RE: Goodbye AMD
By 16nm on 12/5/2007 1:06:24 PM , Rating: 1
K8 and Conroe are just evoluntionary Pentium Pro/P2/P3 designs. There's nothing Revolutionary about K8. Netburst was just a weird design to wow the buyer with great megahertz numbers. Intel thought they could design the long pipeline to be very intelligent wrt instruction ordering and achieve great performance but that never happened. The ultra long pipelined Prescott was supposed to be the pinnacle of this effort but it was a total flop, but boy did Intel ever learn a lot about manufacturing processes with it. It catapolted them to where they are today. Their manufacturing processes are the absolute best. Wait till they fine tune their high k metal gate 45nm process (something AMD do not even have in the works). It will be something to behold, I'm sure.

AMD need to get some of the Intel Mojo in their 65nm process. I suspect it's mostly a question of dinero standing in the way of their success right now. And that's too bad, because it doesn't seem as though they'll be earning a whole lot any time soon.


RE: Goodbye AMD
By weskurtz0081 on 12/5/2007 1:29:54 PM , Rating: 2
AMD and IBM are working on HKMG's actually, and we MIGHT see them by the end of next year.


RE: Goodbye AMD
By 16nm on 12/5/2007 2:14:55 PM , Rating: 2
That's good news. I've only heard that IBM had high k 45nm but that AMD were not interested (in paying for it, probably). I do know with certainty that AMD's current 45nm effort is of the low k gate variety.

It's very good to hear that IBM and AMD are working on high k together. I hope AMD lessen their focus on a single die quad and concentrate their efforts on dual core and dual die quad core chips. From a manufacturing standpoint, this seems to make sense, and performance does not appear to be a huge issue (Kentsfield > Phenom).


RE: Goodbye AMD
By weskurtz0081 on 12/5/2007 4:50:10 PM , Rating: 2
good post. The first round of AMD 45nm will indeed be SOI, and they are supposed to switch over to HK after they release and get the 45nm process working fairly well. It probably will not be until 09 until we see it, but they are working on it.


RE: Goodbye AMD
By murphyslabrat on 12/5/2007 2:17:32 PM , Rating: 2
The comparison I am making is to the negative step away from a very capable little brother. Netburst was an all-new design, while Barcelona is an evolutionary step; however, neither were any better than their precursor. Netburst, as you said, was a flaming POS, and Barcelona is so much more costly (money, R&D, and transister/silicon-wise) for similar performance compared to K-8. Amd would probably have done better by sticking to pathway optimization, silicon and transistor innovation, and, possibly, channeling it directly into die-shrink R&D. As is, we have something, that at present, was a catastrophic waste.

Maybe we just aren't seeing the full fruition of the technologies that debuted with Barcelona, but the suspicious absence is akin to the massive performance increases promised with Netburst.

BTW, few things make me as angry as Netburst. Do you realize that, barring available platform-specific considerations--memory-speed, FSB-bandwidth, etc--the Tualatin revision of the Pentium III not only performed much better than the Pentium 4, but had a die-size of approximately half of a same-node Pentium 4. The only shortcoming is in FLOPS, where the Pentium III and Pentium 4 were about even, clock-for-clock.

At least AMD hasn't pulled that one on us...yet. And I sure as hell hope we never see anything like the Pentium III-to-Netburst again.


RE: Goodbye AMD
By StevoLincolnite on 12/6/2007 10:49:03 AM , Rating: 2
The Tualatin was actually very poor at clock speed ramping, it was more of a test of the new process node that would eventually be used for the Pentium 4 Northwood, the Willamatte still used the same fabrication nodes as the Pentium 3 coppermine, the other notable thing is, that the Celeron Tualatin actually performed about the same as the Coppermine clock for clock (Even while it had a 33% advantage in bandwidth).

Besides the Pentim 4 wasn't all that bad, think of it like this.
Intel Used the Pentium 3 Tualatin, Retrofitted it with the Pentium 4's Front side bus, Advanced tree predictor, Doubled/Quadrupled the cache and made it more energy efficient, and extended/improved the pipeline and called it a Pentium M, which is what our new beloved Core 2 Duo's are all based on, which clock for clock slaughtered the Pentium 4.


RE: Goodbye AMD
By murphyslabrat on 12/22/2007 1:05:03 PM , Rating: 2
quote:
the Celeron Tualatin actually performed about the same as the Coppermine clock for clock

Yes, while it had one half to one quarter of the L2 cache. If anything, that's a testament to the versatility of the P6 core.

And why couldn't Intel have done all of that whithout the Pentium 4?


RE: Goodbye AMD
By gusc3669 on 12/5/2007 12:13:15 PM , Rating: 2
I am not an IT person, but I agree that while playing "one-ups" by producing a chip that is faster (clockwise) than the competitor's chip is not the way to go (eventually), it did sell chips back in the Pentium era, along with bragging rights, until AMD took a next step and produced their Athlon (Socket A) chips. I still own a Socket A computer and remember that it was not the fastest when I purchased it, but it was more efficient and could perform operations on par or better than the faster, more expensive, Pentiums.

AMD needs this "next step" in innovation if it is to remain competitive and get back to its desired profitability. Yes, you could compete in the "budget" arena but not if the competition is providing equal or better offerings at relatively the same price point. AMD would also need to ramp up the volume sold if it wants to go this route which does not seem to be occurring as the majority of new the Dell/HP/Corporate mainstream purchases are Intel based (at least in the various companies I've been associated with). Making their products on the 65nm die, increasing the clock rate, and possibly lowering the price is fine if they want to remain competitive and sell some chips, but don't cut the other items which add to the chips overall performance.


RE: Goodbye AMD
By 91TTZ on 12/6/2007 7:58:08 AM , Rating: 2
quote:
Yeah, like imagine if Intel had've stuck with re-iterating the original Pentium instead of giving us the next generation of computing with Netburst...</sarcasm>


That's sort of what they ended up doing. They scrapped the Netburst idea and went back to reworking the Pentium Pro design.


RE: Goodbye AMD
By caqde on 12/5/2007 11:26:22 AM , Rating: 3
Unfortunately that is impossible because of the design of the K8. The on-board memory controller makes doing such a task impossible without major changes to the core one way or another. Hence the Barcelona core's design... A design like Bulldozer/Fusion will get rid of such issues though.


RE: Goodbye AMD
By Amiga500 on 12/5/07, Rating: 0
RE: Goodbye AMD
By Martimus on 12/5/2007 12:36:17 PM , Rating: 2
I don't know if they could introduce two-die quad-cores with the integrated memory controller. But maybe they can, considering they had two chip solutions with the Quad-FX setup.


RE: Goodbye AMD
By 16nm on 12/5/2007 12:45:06 PM , Rating: 2
I'm sure it is possible, but is it reasonable? They've surely thought about it, and I hope the reason we don't see any discussion of it is NOT because they have been making fun of the Intel dual-die quad. Ego should not get in the way of rational decision making...


RE: Goodbye AMD
By Master Kenobi (blog) on 12/5/2007 12:50:19 PM , Rating: 2
With Quad-FX, each controller had explicit control of one bank of memory, not both. So basically it was "separate but equal". This is where Intel is getting benefit by having the memory controller on the north bridge. They can glue two dice together and get away with it, just tweak the controller to compensate. AMD has to stick with the unified design because it's integrated, at least until they can do something very creative so that the two don't fight each other for fetches.


RE: Goodbye AMD
By Martimus on 12/5/2007 1:07:12 PM , Rating: 2
That is kind of what I had thought, thank you for answering. I am glad that I posted that here, as I have been wondering about that issue for a while and it would have been difficult to look it up an my own.


RE: Goodbye AMD
By 16nm on 12/5/2007 1:15:21 PM , Rating: 2
Yes, they could try to do something very creative, or, simply copy Intel's design. It doesn't seem to matter much where the memory controller is placed judging from the performance of Conroe. At least, we don't hear much about the superiority of the IMC since Conroe debuted. That bullet has pretty much been shutdown. AMD must do something to get back in the game. K10 is not going well at all. That L3 cache design flaw is really absurd. They totally dropped the ball on that one. It seems everything hinges on K10 right now. If they can keep K8 going then they can relax and work on getting K10 right.


RE: Goodbye AMD
By Master Kenobi (blog) on 12/5/2007 5:50:47 PM , Rating: 2
Yes and No. Conroe gets away with kicking butt and having the Controller on the North Bridge because Intel's engineers worked some magic. AMD's IMC and HT bus can feed their processor faster than Intel can (Theoretically). Intel compensates for this by increasing the L2 cache into the 2-12MB range and using extremely slick algorithms to do intelligent pre-fetching. Allowing it to feed data into the chip at a rate the chip can use without a performance hit where the CPU is idling on cycles because it can't get information quickly enough.

To go with Intel's north bridge method, AMD would not only have to come up with a better prefetch method, but also increase the L2 and possibly L3 (Jury is still out on how effective L3 cache really is, Barcelona/Phenom was not the best of implementations based on the theories I've read about) into the megabyte range much as Intel has.

With Intel's move to Nehalem, they will be sporting the Native Quad Core design much like Barceona/Phenom, they will also be sporting the Integraged Memory Controller, and the new Intel QuickPath (Faster and more bandwidth than Hyper-Transport 3.0 supposedly). Now Intel is also adding Hyper-Threading back in so it will show up as 8 Processors to Windows (Should be pretty slick). If Intel can find a way to glue 2 of those together and not have the memory controllers fighting, AMD could copy that (If they haven't figured out how to do it by then). Right now they would have to re-engineer their chips and chipsets to a point that would make them both incompatable with current boards and chipsets, as well as force them to redesign the CPU's and a lot of the logic built into it that is based around being linked to the IMC. Sounds like too much cost and headache to get around it. Best bet would be to develop a way to not have the IMC's fight with each other.

In Summary.... Intel played their cards right and played their strengths and milked every last bit out of the North Bridge, Legacy BUS, and their processes. They were able to take a cheap route based on existing architecture and glue two dice together and make out like bandits.

AMD Ran with new tech like IMC, and Hyper-Transport and while it has worked well for them (K7/K8) it seems to have closed that door for them, up until such time as they can figure out how to get 2 IMC's to play nicely. That's really the only thing stopping them from doing exactly what Intel is doing.


RE: Goodbye AMD
By JumpingJack on 12/7/2007 2:37:30 AM , Rating: 1
Well done.... the argument boils down to cache really.

Fast bus + small cache or slow bus + large cache, the design choice of either combination is balanced to hit certain performance goals.

Cache hit/miss rate goes as a power law in terms of overall capability, the larger the cache the lower the miss rate (a miss being a situation where the processor needs data but does not find it in cache).

Small caches have high miss rates, to mitigate the penalty for going to DRAM, simply design a bus to get it there faster (high BW, low latency), so if you miss say 5 times out of 20 attempts, you may pay only 8 cycles penalty waiting.

Large caches have low miss rates, so you may only miss 2 times out of 20, but you must pay 12 cycles penalty waiting on the slow bus.... just an example.

It is completely possible with the 'archaic' FSB to out perform the novel and slick IMC implementation so long as the caching technology is par excellent (which C2D has), consequently high quality memory is not nearly as impactive on a well cached CPU as opposed to a cache lean CPU.

Where the FSB becomes more problematic are in large throughput, large working set situations where even a larger cache becomes overwhelmed ... hence SPECFP_rates (large data sets, high through put) really shine on the high BW busses.


RE: Goodbye AMD
By Omega215D on 12/5/2007 2:16:36 PM , Rating: 1
Refining K8 and bringing some of that to the Socket 939 iteration... ok I know it's a pipe dream but hey, I'm sticking with my 939 and DDR until it's absolutely necessary (gaming wise) to ditch my current setup.


RE: Goodbye AMD
By qwertyz on 12/5/07, Rating: 0
"It's okay. The scenarios aren't that clear. But it's good looking. [Steve Jobs] does good design, and [the iPad] is absolutely a good example of that." -- Bill Gates on the Apple iPad

Related Articles













botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki