AMD confirms quad-core on AM2

Status
Not open for further replies.
But how long? Kentsfield will be out early next year...I think we'll be waiting until next summer for AMD's quad-core mainstream parts.
 
Did you happen to notice this statement?
And the fact that the AM2 socket allows us to have a seamless transition to quadcore tells me that I can buy a 4×4 platform, in the later part of this year, and then take it to eight cores sometime in 2007. And that's phenomenal.
Click to expand...
 
First of all AM2 is a socket that K8l will use so in 2008 when 4core K8l appears you will be correct. If you have a roadmap showing 4core K8L before 2008 please post it.

Amd has said mid 2007 for the 2core K8l . So AM2 supporting 4 cores has nothing todo with K8 . AM2 is a socket not a cpu. AM2 will also be used with K8L


Now, one thing, just to build on your question, is the notion that there's part of the desktop that's becoming workstation class, to a great degree. Our 4×4 enhancement is taking workstation-class performance to the high end of the gaming and prosumer space. I'm actually very excited about 4×4. As a gamer and enthusiast PC user myself, it's kind of a dream come true, and I think that once again, in that particular space, we are re-inventing the industry and pushing the limits. And the fact that the AM2 socket allows us to have a seamless transition to quadcore tells me that I can buy a 4×4 platform, in the later part of this year, and then take it to eight cores sometime in 2007. And that's phenomenal.

See the under lined part read it and comprehend. Intel will also have a 4x4 workstation M/B out soon on the green creek chipset and Intel sells it right now. So intel will have 8 cores on one board long before AMD by about 1 year.The cost will be about =




High performance board delivers the power of multi-core, dual-sockets to the demanding workstation environment

S5000XVN

The Intel® Workstation Board S5000XVN provides incredible performance and memory capacity for demanding workstation operations. This workstation board is ideal for customers needing server class performance and reliability, in addition to high-end graphics. The S5000XVN is acoustically optimized to minimize distractions in shared environments. I would wait for the ASUS model .




AMD down 26cents @ $24

INTC up 29cents @$18.34
 
$BangforThe$ said:
First of all AM2 is a socket that K8l will use so in 2008 when 4core K8l appears you will be correct. If you have a roadmap showing 4core K8L before 2008 please post it.

Amd has said mid 2007 for the 2core K8l . So AM2 supporting 4 cores has nothing todo with K8 . AM2 is a socket not a cpu. AM2 will also be used with K8L

AMD down 26cents @ $24

INTC up 29cents @$18.34
Click to expand...
I'm your huckleberry.



I was actually there for this event on June 1.
 
My input is this: Until developers start utilizing multiple cores, I'm not creaming my shorts over 2, 4, 8 or even 16 cores. As far as I'm concerned, the apps I run can still only use one at a time.
 
Rocco123 said:
My input is this: Until developers start utilizing multiple cores, I'm not creaming my shorts over 2, 4, 8 or even 16 cores. As far as I'm concerned, the apps I run can still only use one at a time.
Click to expand...

Well, I think the non-SMP-aware apps list is restricted to only games and basic office applications right now. Most graphics/video/audio applications are SMP/SMT-aware, I'm sure.
 
Does anyone actually know if the current am2 boards will support K8L and whatever else comes up down the road? Or are they just saying the socket?
 
Bona Fide said:
Well, I think the non-SMP-aware apps list is restricted to only games and basic office applications right now. Most graphics/video/audio applications are SMP/SMT-aware, I'm sure.
Click to expand...

Games and basic apps are all I use for home. As far as the commercial market, there are thousands (well maybe not thousands) of applications that are SMP aware. But again, none that most of us here use for recreation.

Make Oblivion and HL2 SMP aware, then I will, in fact, cream my shorts.
 
Chris_Morley said:
I'm your huckleberry.



I was actually there for this event on June 1.
Click to expand...

Why would you want to be my blueberry?

If you were referring to Huckleberry Finn than I am your Samuel Clemens(Mark Twain)
 
$BangforThe$ said:
Why would you want to be my blueberry?

If you were referring to Huckleberry Finn than I am your Samuel Clemens(Mark Twain)
Click to expand...

Um I think that might be a Tombstone reference..........
 
$BangforThe$ said:
Why would you want to be my blueberry?

If you were referring to Huckleberry Finn than I am your Samuel Clemens(Mark Twain)
Click to expand...
Well, what you should have taken away from my post is that you asked someone to post a roadmap showing K8-L quad core availability sooner than 2008. I did that.

As to "I'm your huckleberry", you can read this.
 
OK Chris but that roadmap link didn't work for me try another method. I can link to 2008 being first 4 cores more I am sure. Interesting statement you made about huckleberry. I think blue berry is a much more true to life compare. You know how it stains your hands after you touch them.
 
$BangforThe$ said:
OK Chris but that roadmap link didn't work for me try another method. I can link to 2008 being first 4 cores more I am sure. Interesting statement you made about huckleberry. I think blue berry is a much more true to life compare.
Click to expand...
I'd be happy to email it to you. PM me your email address. This came directly from AMD at their analyst day on June 1, 2006, at their headquarters in Sunnyvale, CA. I attended this event and saw the presentation that included this slide.
 
$BangforThe$ said:
SO you can't share with all of us . I would preferr an offical document . But OK no need for PM [email protected]
Click to expand...
Are you kidding me? I am telling you what I saw at Sunnyvale. I have posted a link that at least one other person has been able to view as they were able to "right click, save as".

Now I will email it to you.
 
link worked fine for me , thanks for the info

...and a huckleberry is a ..umm ..a huckleberry ..

not a blueberry

:)

(that was a great scene from Tombstone tho .. :p )

 
enelson125 said:
Does anyone actually know if the current am2 boards will support K8L and whatever else comes up down the road? Or are they just saying the socket?
Click to expand...



In early 2008, AMD's corporate roadmap claims a quad core desktop CPU will make an appearance, dubbed Greyhound. Greyhound is slated to become the first quad-core AMD chip to use the HyperTransport 3 bus, and the memory controller is slated to support DDR2 and DDR3. Unlike Deerhound, Greyhound will use the K8L architecture, and all the goodies that come with it, including the 5.2GT/s HyperTransport support. Unless AMD's plans change drastically between now and 2008, the processors will require a new socket.
 
I may have missed this but K8 may have a Quad core


Behind closed doors, insiders revealed to DailyTech a few tidbits of the long term quad core roadmap. AMD will introduce no less than four quad-core families over the next two years, with the first being Deerhound. Deerhound, we are told, will be a Socket F server processor expected to ship late next year on the K8 -- not K8L -- architecture. Deerhound did not appear to support FB-DIMM.
 
Here's a PICTURE I took of the slide, and a picture from my hotel window. Enough proof?




The slide says "next generation core" - I assume that means K8-L. Maybe it means something we haven't even heard of. Who knows.
 
ThreeDee said:
link worked fine for me , thanks for the info

...and a huckleberry is a ..umm ..a huckleberry ..

not a blueberry

:)

(that was a great scene from Tombstone tho .. :p )

Click to expand...

huckleberry a dark blue berry that looks like a blueberry. Blueberry was used to aid visual affect.
 
$BangforThe$ said:
Here is the link . Now I don't know who is rght who is wrong this is all speculation anyway . But if its proof we are looking for were all shit out of luck.

http://www.dailytech.com/article.aspx?newsid=2388
Click to expand...
Good lord. If flying to California and attending an analyst day hosted by AMD at AMD's headquarters is "speculation" that there is a roadmap showing quad-core in 2007, what is proof?
 
Is this: how many cores do you guys think they can cram into one processor before the memory bottleneck becomes noticeable? As I understand it (from reading up on NUMA) only one core can access main memory at a time, so the others have to wait. NUMA (non-unified memory access) is used on server boards, giving each cpu socket (or I believe seperate sets of sockets in 4+ socket systems) it's own RAM banks, hence eliminating that waiting period. My impression (can't clearly remember any sources atm) is that this does make a performance difference, but that it's a small percentage.

I imagine that with 4 cores it won't make much of a difference, I'm just wondering when and how this going to be addressed as the number of cores in mainstream systems increases. If a 4x4 system with 2x 4 core cpu's is all using the same memory it may become a slight issue.
 
Chris_Morley said:
Good lord. If flying to California and attending an analyst day hosted by AMD at AMD's headquarters is "speculation" that there is a roadmap showing quad-core in 2007, what is proof?
Click to expand...


Chris I don't doubt you at all. I believe you believe what you were told and shown thats doesn't make it the trueth . The whole trueth and nothing but the trueth.

Until they are actually out and in customers hands its all speculation . To pull this off AMD is going to have to have a perfect excution . Going by AMD's past record of bring new Arch. to the masses . I have every right to be cautious on any roadmap by AMD .

Socket F delayed . When was it suppose to be released?
 
Slartibartfast said:
Is this: how many cores do you guys think they can cram into one processor before the memory bottleneck becomes noticeable? As I understand it (from reading up on NUMA) only one core can access main memory at a time, so the others have to wait. NUMA (non-unified memory access) is used on server boards, giving each cpu socket (or I believe seperate sets of sockets in 4+ socket systems) it's own RAM banks, hence eliminating that waiting period. My impression (can't clearly remember any sources atm) is that this does make a performance difference, but that it's a small percentage.

I imagine that with 4 cores it won't make much of a difference, I'm just wondering when and how this going to be addressed as the number of cores in mainstream systems increases. If a 4x4 system with 2x 4 core cpu's is all using the same memory it may become a slight issue.
Click to expand...

Right now with AMD they are setting better because each cpu has its own IMC. On the Intel side we will have to wait for Tigerton which is said to be better than CSI. Due out in 2007.AMD's Opteron uses both point-to-point interconnects like Tigerton's and an integrated memory controller. But Intel has been reluctant to embrace integrated memory controllers, said Dean McCarron, principal analyst with Mercury Research in Cave Creek, Arizona. Leaving the memory controller on the chipset allows Intel to more easily accommodate changes in memory standards, because it's much easier to change a chipset design than a processor design, he said.

One reason for moving out the common architecture target date, and therefore the potential integrated memory controller, could be that Intel plans to increase the size of cache memory on its future processors, McCarron said. Cache memory stores frequently used data right next to the CPU where it can be accessed much more quickly than data stored in the main memory chips.

The combination of larger cache memory and the direct connections between Tigerton processors and chipsets could provide a significant performance boost for Intel-based servers in 2007, McCarron said. It's very early to know for sure, with even sample chips still far away, he said.

However, Tigerton will certainly be more efficient than Whitefield because of the new interconnect design, Brookwood said. "What happens on the chip matters a lot, obviously, but ultimately the performance of these chips is constrained by how fast you can feed them data," he said
 
$BangforThe$ said:
Right now with AMD they are setting better because each cpu has its own IMC. On the Intel side we will have to wait for Tigerton which is said to be better than CSI. Due out in 2007.
Click to expand...

I mean, even with each cpu/core having it's own imc, there's still only one pathway out to the memory, which is where the bottleneck is.
 
$BangforThe$ said:
Chris I don't doubt you at all. I believe you believe what you were told and shown thats doesn't make it the trueth . The whole trueth and nothing but the trueth.
Click to expand...
You asked for a roadmap. I delivered an official one straight from AMD. End of story.
 
Chris_Morley said:
Good lord. If flying to California and attending an analyst day hosted by AMD at AMD's headquarters is "speculation" that there is a roadmap showing quad-core in 2007, what is proof?
Click to expand...
I know, you could show him the video, of you asking a question to Henri Richard.. Obviously, that can't be faked. :p
 
Chris_Morley said:
You asked for a roadmap. I delivered an official one straight from AMD. End of story.
Click to expand...

Yes Chris you did but can you Link to another Web site that has the same Info . Thats all I ask for . Chris you know I like Intel better than AMD right. But I don't believe everthing Intel says. Its not like their lieing .Its just that Shit happens . Look at Whitefield complete failure. But when Intel sent it to India I was already exspecting it to fail. In a perfect world what you were told may be true . But the world is far from perfect.After all I gave a link that supports what I am saying and I can post more links supporting it also.
 
$BangforThe$ said:
Yes Chris you did but can you Link to another Web site that has the same Info . Thats all I ask for . Chris you know I like Intel better than AMD right. But I don't believe everthing Intel says. Its not like their lieing .Its just that Shit happens . Look at Whitefield complete failure. But when Intel sent it to India I was already exspecting it to fail. In a perfect world what you were told may be true . But the world is far from perfect.
Click to expand...

Who else would know, other than AMD? What other website/analyst/crackpot is going to know AMD's own roadmap better than AMD? Stop creating bickering for the sake of bickering. Your ass was handed to you on this matter on page one, and here on page two, you're still grasping to reasons to think that quad core K8L won't hit until '08. What is your trip?
 
robberbaron said:
Who else would know, other than AMD? What other website/analyst/crackpot is going to know AMD's own roadmap better than AMD? Stop creating bickering for the sake of bickering. Your ass was handed to you on this matter on page one, and here on page two, you're still grasping to reasons to think that quad core K8L won't hit until '08. What is your trip?
Click to expand...

I showed a roadmap that says quad core K8L in 2008. and maybe quad 4 K8 late in 2007 .

I would like to see more links. Thats my trip.
 
$BangforThe$ said:
I showed a roadmap that says quad core K8L in 2008. and maybe quad 4 K8 late in 2007 .

I would like to see more links. Thats my trip.
Click to expand...

you are such an AMD hater... all you do is start crap to trash AMD... I despise and HATE Intel, probably as much as you dislike AMD.. but I dont go into the Intel section and post daily over and over trashing Intel like you do here in the AMD section.

Fine we get it. You dont like AMD, you dont believe AMD... now STFU Punk!
 
Tetrahedron said:
you are such an AMD hater... all you do is start crap to trash AMD... I despise and HATE Intel, probably as much as you dislike AMD.. but I dont go into the Intel section and post daily over and over trashing Intel like you do here in the AMD section.

Fine we get it. You dont like AMD, you dont believe AMD... now STFU Punk!
Click to expand...

QFT!
 
Tetrahedron said:
you are such an AMD hater... all you do is start crap to trash AMD... I despise and HATE Intel, probably as much as you dislike AMD.. but I dont go into the Intel section and post daily over and over trashing Intel like you do here in the AMD section.

Fine we get it. You dont like AMD, you dont believe AMD... now STFU Punk!
Click to expand...

I wasn't trashing AMD in any way shape or form. I watched part of the web pubcast. I will watch it all later. All I have said is that Quad K8L Is 08 and the pubcast supports that.

Also as far as being accurate. The fab guy made this statement. Dresden Fab has never failed to meet its goals. If I remember it correctly Hammer was very late and the Dresden fab did not meet its goal. So now you want me to believe everything that the pubcast says to be absolutely accurate. They are merely laying out timeframes if they achieve said time frame great. But that is one big if.

I have in the past in this forum said and I mean it. K8l looks promising and I will look very hard at it. As possiable gaming PC's. But not till I look at Penryn.
 
$BangforThe$ said:
Chris I don't doubt you at all. I believe you believe what you were told and shown thats doesn't make it the trueth . The whole trueth and nothing but the trueth.

Until they are actually out and in customers hands its all speculation . To pull this off AMD is going to have to have a perfect excution . Going by AMD's past record of bring new Arch. to the masses . I have every right to be cautious on any roadmap by AMD .

Socket F delayed . When was it suppose to be released?
Click to expand...


Talk about calling the kettle black..... I once said the same thing, and you jumped all over me.
 
Status
Not open for further replies.
Back
Top