backtop


Print 57 comment(s) - last by JonB.. on Nov 23 at 12:39 AM


AMD guidance details the backwards and forwards compatibility of the AM2, AM2+ and AM3 sockets.  (Source: AMD)

AMD details its platform progression identically to Intel's "tick-tock" architecture roadmap.  (Source: AMD)
AMD announces its Spider Platform in what could be its biggest launch of all time

Tomorrow marks the launch of AMD's platform launch, codenamed Spider. The launch consists of three components: AMD Phenom processors, the ATI Radeon HD 3800 graphics processor and AMD 7-Series desktop chipsets.

The first new AMD desktop architecture in four years will debut with the Phenom 9500 and Phenom 9600. Both chips feature a 95-Watt thermal envelope and a shared 2MB L3 cache.

The AMD Phenom 9600 ships with a 2.3GHz operating frequency, while the Phenom 9500 features a slightly lower 2.2GHz clock. Both processors run on HyperTransport 3.0 and feature a total 2MB L2 cache; 512KB per core. The chips also come with an integrated DDR2 memory controller and support speeds up to the DDR2-1066 specification, which is still pending JEDEC approval.

Directron.com carries the Phenom 9600 at a retail price of $322.00 and the Phenom 9500 at $286.00.

AMD guidance originally stated that a 2.4GHz Phenom, dubbed the Phenom 9700, would also launch on November 19, however last minute roadmap updates indicate this chip will come in December instead.

These two Phenom processors are part of AMD's first-generation, 65nm Stars family. In the second-half of 2008 AMD will announce its second-generation Stars family, which will be a migration of the current K10 core to the 45nm node.

The second-gen Stars lineup consists of five variants, including quad-core Deneb, dual-core Propus, dual-core Regor and single-core Sargas. The second-generation Stars CPUs will also include an integrated DDR3 memory controller, a first for AMD.

AMD's Spider Platform launch also includes the official announcement of its ATI Radeon HD 3800 Series. The launch of the Spider platform will add two new video cards to AMD's lineup, the ATI Radeon HD 3870 and Radeon HD 3850. 

AMD's new Radeon series received overwhelmingly positive feedback during AMD's November 15 media event.  Newegg representatives tell DailyTech the company sold out all of its stock on the first day, but is quickly replenishing inventory.

To round off the launch of the platform, AMD also announced its 7-Series desktop chipsets. The AMD 790FX targets the ultra-enthusiast market segment, and corporate guidance sets the retail price of 790FX boards between $150-250. Several partners already announced boards supporting this new chipset earlier this year.

AMD designates the AMD 790X as its performance chipset, slotted just below the 790FX.  AMD guidance states this chipset will run the consumer between $99-150.  The mainstream AMD 770 chipset will not see store shelves for several months, but AMD claims this chip will round off the mainstream segment for AMD chipsets. 


Comments     Threshold


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

Any 'experts' out there?
By wordsworm on 11/18/2007 7:33:58 PM , Rating: 2
Can anyone tell me what the bandwidth would be like between a GPU and CPU on the same die? Isn't the bandwidth supposed to considerably faster on the die as opposed to on the motherboard? I can't help but think that the motherboard would suddenly be freed of one of its most intensive bandwidth hogs.




RE: Any 'experts' out there?
By Zurtex on 11/18/2007 7:40:34 PM , Rating: 2
As it stands that kind of bandwidth simply isn't needed.

If it was on the same die, but not 'onboard the chip', it would probably communicate with HyperTransport instead of PCIe2.0. There's no real need for a graphics card to do this, otherwise AMD would probably be pushing for it.

However thinking about integrating stream processors (what powers modern GPUs) onboard to a modern CPU, opens up all sorts of really interesting possibilities. Having a quasi-sequential quasi-parallel chip would be awesome to play around with.

If only there was a company who built both modern x86 CPUs and modern GPUs, if only there was another company with billions and billions for R&D who were terrified of being beaten by the other company again in terms of performance. ^_^


RE: Any 'experts' out there?
By Ringold on 11/18/2007 8:34:03 PM , Rating: 2
quote:
with billions and billions for R&D who were terrified of being beaten by the other company again in terms of performance. ^_^


That's whats really sad about this whole scenario already. I heard some people talking on CNBC about how Intel is holding back out a real fear of completely destroying AMD; it's much easier to be a duopoly than be a nearly undisputed oligopolist. Intel's already had a small taste of what monopoly power feels like, and it needs to look no further than Microsoft to see how Europe treats dominating success. We're probably already missing out on big performance because AMD's fallen so badly behind.

Phenom doesn't look to be a C2D killer, and we don't even get a die shrink apparently until the tail end of next year. Expect nothing dominating from Intel, then, until when whatever is after Stars comes out; they'd simply be afraid of the consequences.

I also wouldn't be shocked to see prices rise over the next year from where they are now. Not because of a weak dollar, no, but because Intel can afford to without hurting volume much and because Intel knows it might throw AMD a bone if they can do likewise. Meanwhile, of course, they'd roll in the money as well.


RE: Any 'experts' out there?
By Zurtex on 11/18/2007 9:01:48 PM , Rating: 4
Well I live in Europe (the U.K), so for me the weak dollar drives prices down not up ^_^.

To be honest though, if Intel really are holding back 'as not to destroy AMD' rather than just milking the market for all its worth, good I guess. In the long run having a duopoly would far more help the market that a monopoly. Sure in the short term we'd get faster more power chips, but what would the incentive be to think of new ideas and not sure slowly ramp up clock speed? AMD and the ATi investment was all about the long term when it comes to R&D, so lets just hope it pays off.


RE: Any 'experts' out there?
By Amiga500 on 11/19/2007 5:31:13 AM , Rating: 2
A free market does not guarantee the best deals for the consumer - the sooner you stop tripping over every word the PR divisions of the multinationals come out with the better.

I also wouldn't be shocked to see prices rise over the next year from where they are now. Not because of a weak dollar, no, but because Intel can afford to without hurting volume much and because Intel knows it might throw AMD a bone if they can do likewise. Meanwhile, of course, they'd roll in the money as well.

If AMD were not around, Intel would already be charging us $1000 for a C2D E6600.

If AMD were not around, Intel would have virtually no need to push R&D. The only reason we got Conroe was K8.

With AMD so weak, Intel can relax and not push themselves - do you think they would be pushing harder with no AMD around? How naive are you?


RE: Any 'experts' out there?
By ZmaxDP on 11/20/2007 8:21:26 AM , Rating: 2
"A free market does not guarantee the best deals for the consumer"

Then how about enlightening us with what does provide the best deals for the consumer???


RE: Any 'experts' out there?
By Visual on 11/20/2007 10:56:53 AM , Rating: 2
a "for free" market does that.
but that's communism, not capitalism


RE: Any 'experts' out there?
By wordsworm on 11/21/2007 11:33:48 AM , Rating: 2
That's right... the Soviets didn't have money. China doesn't have money, because everything is free.

Are you a nuclear plant inspector working for someone named Mr. Burns?


RE: Any 'experts' out there?
By JonB on 11/23/2007 12:39:41 AM , Rating: 2
It was only two years ago that Intel was still saying that nobody needed 64bit processors except for high end servers. We'd still be running overheated, frequency locked P4's if not for AMD.


RE: Any 'experts' out there?
By SilthDraeth on 11/18/2007 11:09:54 PM , Rating: 2
quote:
If it was on the same die, but not 'onboard the chip', it would probably communicate with HyperTransport instead of PCIe2.0. There's no real need for a graphics card to do this, otherwise AMD would probably be pushing for it.


Actually AMD is pushing for, and designing that. It is called AMD Fusion.
http://arstechnica.com/news.ars/post/20061119-8250...
Is one article talking about it.


RE: Any 'experts' out there?
By Zurtex on 11/18/2007 11:31:08 PM , Rating: 2
My understanding was the main aim of fusion was really to explore all possible co-processors by allowing intercommunicating through HyperTransport.

The on-die graphics wouldn't be full graphic card power, put something strong enough to at least run a full 3D GUI / HD Acceleration. Sort of a stepping stone to a future I was describing. But I could be wrong, I've not looking in to it too much recently.


RE: Any 'experts' out there?
By Targon on 11/19/2007 8:25:10 AM , Rating: 2
There was a slot type for this called HTX which would allow for add-in cards that would communicate via HyperTransport, but it was under-hyped and seems to have died.

AMD has promoted the idea of having different processors that are socket compatible with the CPU, so if you take a 4 socket system, you could put two CPUs in, a PPU(Physics Processing), and a GPU chip into the system, or you could go with 3 CPUs and one GPU, or add other co-processors to provide whatever abilities you might want to see.

A problem with putting a GPU on a straight chip that goes into the motherboard is that video standards are changing too rapidly for this to be a good idea. You have DisplayPort, HDMI 1.3, Dual-link DVI, plus there will be new versions of these connection types out by the time R&D had the chance to release a new motherboard that supported the old standards for video.

If you look at how quickly GPU technology is improving, it seems almost foolish to go with integrated video these days because in the time it takes from the chipset release to motherboard release, a new low-end GPU comes out that allows for $50 video cards that makes the old ones look like garbage.

A HyperTransport connected video card makes the most sense, but would require video card manufacturers to support a new standard that may or may not survive. Look at what happened to Microchannel(old IBM PS/2 days) and VLB(Vesa Local Bus). There would also need to be a real benefit to the technology since something good on paper means nothing if real-world performance doesn't help. Then you have the support end from AMD where the new slot type would need to be properly supported(game players MUST be able to run their games, and problems with video cards not being recognized because they are not PCI-Express are a valid concern.


RE: Any 'experts' out there?
By MGSsancho on 11/19/2007 8:22:01 PM , Rating: 2
there are FPGAs that can fit. well its really a FGPA on a tiny board that has pins that fits in the slot.

There are infiniband as well as 10gbit Ethernet chips that use the HT bus. I think myranet but I am probably wrong on that. either way its mostly HPC and nothing us mortals really ever see.


RE: Any 'experts' out there?
By murphyslabrat on 11/20/2007 9:31:56 AM , Rating: 3
quote:
My understanding was the main aim of fusion was really to explore all possible co-processors by allowing intercommunicating through HyperTransport.

No, you're thinking of the Torrenza Initiative. Fusion is the idea of integrating a GPU-like co-processor onto the die, itself.

And, as AMD has been feign to let go, it would be far from an exclusively graphical utility. This kind of parallel co-processor could potentially alleviate some bottlenecks in certain applications.


RE: Any 'experts' out there?
By Gul Westfale on 11/19/2007 12:12:06 AM , Rating: 2
on an AMD64 system the RAM is already connected to the CPU through its own dedicated bus (the onboard memory controller of the CPU), thus th eonly really bandwidth intensive thing for the mobo chipset is handling traffic between CPU and GPU.

the amount of bandwidth is determined by the slowest part in the chain, meaning that if the GPU can accept data at 10GB/sec but the mobo chipset can only send it at 5, then the bandwidth is 5GB/sec.

if the CPU and GPU are on teh same die than this all still applies; but it is much easier connecting things on-die than through the motherboard, as there is no need for long traces in the mobo that can become hot and may cause interference. all modern mobos are actually 4 or more layers of board sandwiched together because there are just so many traces (and the larger traces need to be cut up into smaller ones to avoid heat issues). putting it all on the same die means the manufacturer is not limited by the mobo anymore, and thus bandwidth would be whatever the manufacturer would want it to be. latency would also be reduced.

at the moment there is no need for such a device yet, but cost also factors in: if you want to build a low-cost PC then the fewer separate components=the lower price.


RE: Any 'experts' out there?
By 3kliksphilip on 11/19/2007 4:03:02 PM , Rating: 1
All this is too complicated for me. All I know is that I liked the 'Confidential' 'DO NOT DISTRIBUTE' writing at the bottom of the page.


"We basically took a look at this situation and said, this is bullshit." -- Newegg Chief Legal Officer Lee Cheng's take on patent troll Soverain

Related Articles



Latest Headlines
4/21/2014 Hardware Reviews
April 21, 2014, 12:46 PM
4/16/2014 Hardware Reviews
April 16, 2014, 9:01 AM
4/15/2014 Hardware Reviews
April 15, 2014, 11:30 AM
4/11/2014 Hardware Reviews
April 11, 2014, 11:03 AM










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