• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

AMD Multi core CPU's

Sadly that is servers in Mid 2005 and desktops in the later half of 2005. Sound more like fall to me.
 
probably in october with the launch of the next fx.. though i'm guessing they'll release two, one single core with a speed bump and the first dual core fx.. which may have a difference monikor (cross your fingers for this.. and hope for a64-dx, hehe)
 
I see that they will work with 940 sockets, is that all? will they work with 939?
 
think about it. current games can't take advantage of both cores, so the game would essentially see one core which will be clocked slower than the 'equivilent' single core chip. thus it'll be slower. once things are written for multithreading, dual core should be set.

and foz.. 940 is all they've announced, but there's no doubt in my mind that 939 based dual cores will be launched.. and they should be compatible with current boards just as the 940's are ;)
 
The definition of "dual-core processor," however, is broad and ranges from a single chip in which the two integrated cores share resources to something that consists of two functionally and physically separate pieces of silicon that happen to be in the same package. Packaging, which protects the silicon and contains the metal bumps for carrying signals from the chip to the rest of the computer, is an integral part of a processor.
Click to expand...
taken from: http://arstechnica.com/news.ars/post/20050228-4658.html

looks like amd may be the former depending on how the memory controller is doled out.
 
Most likely in the first generation that will be true. Still, it beats off-die access to a shared memory controller, though.
 
ThomasE66 said:
They might be jumping to a 6x number to play marketing with Intel's new 6xx P4's.
Click to expand...


6xx will be old news by the time dual core FX's hit the market.
If they were do something that pointless, it would have to be to an FX-81, not 61.
 
even though an fx-61 should be a 3.2ghz part, which seems very doable at the moment?
i'm thinking a a64-DX for dual core to compete with the pentium D if you're gonna speculate like that :cool:
 
A dual core Athlon 64 at 65 nm would compare favorably with Intel's Yonah, both in terms of computational and thermal performance.
 
I'm a whole lot more intrigued by this article...

http://www.theinquirer.net/?article=21531

2.4GHz Dual Core to start with??? Maybe faster?? Thats great news (I had been expecting 2GHz or lower, due to thermal and power charecteristics). The biggest question is whether or not AMD has the fab capacity to have dual cores (since for every dual core chip they make, its one less single core chip they make, and one less chip to make revenue on).
 
doormat said:
I'm a whole lot more intrigued by this article...

http://www.theinquirer.net/?article=21531

2.4GHz Dual Core to start with??? Maybe faster?? Thats great news (I had been expecting 2GHz or lower, due to thermal and power charecteristics). The biggest question is whether or not AMD has the fab capacity to have dual cores (since for every dual core chip they make, its one less single core chip they make, and one less chip to make revenue on).
Click to expand...
it's not AMD fabing most of them in the end anyway it's IBM so yea
 
East Fishkill? Hmm... wow, IBM must have a lot of fab capacity there... they're doing the Cell chip for PS3, the next Nintendo chip, G5s and AMD chips. If fab capacity isnt an issue, well, thats quite nice. I cant wait... dual core opterons in a dual socket 1U chassis. Mmm... 460GHz in a 48U rack. :D
 
doormat said:
East Fishkill? Hmm... wow, IBM must have a lot of fab capacity there... they're doing the Cell chip for PS3, the next Nintendo chip, G5s and AMD chips. If fab capacity isnt an issue, well, thats quite nice. I cant wait... dual core opterons in a dual socket 1U chassis. Mmm... 460GHz in a 48U rack. :D
Click to expand...

They just built another 370,000 square foot annex in their East Fishkill facility last year. That's quite a bit of room to make wafers in :)
 
incomudro said:
it's not AMD fabing most of them in the end anyway it's IBM so yea
Click to expand...
erm... dresden? fab 30?
austin.. fab 25 (now only puts out flash memory though..)
and the one they're building, fab 36?

ibm doesn't actually make amd's chips, they just help with the tech.
 
A dual core chip isn't double the size of a single core chip. Isn't the memory controller part going to remain the same (dual channel) but with an added crossbar into the caches for each of the cores? Presumably this will receive a speed bump to keep the caches full, but memory bandwidth on current AMDs is not the bottleneck it is on the P4s.
 
Quote:
Originally Posted by incomudro
it's not AMD fabing most of them in the end anyway it's IBM so yea
erm... dresden? fab 30?
austin.. fab 25 (now only puts out flash memory though..)
and the one they're building, fab 36?

ibm doesn't actually make amd's chips, they just help with the tech.

Same difference
 
How can you be so sure? Last I checked, the details of the AMD-IBM alliance were classified.
 
this was something i found at toms hard ware "It's sad, but true: with the Smithfield dual-core processor set to increase maximum power dissipation from 115 to 130 watts early this summer, Intel had to make small modifications to the pin setup of the Socket 775. It is thus beyond any doubt that all earlier boards on the market will not work with the dual-core processors. (For their part, AMD has made a clear statement to the opposite effect: the dual-core Athlons running the Toledo core will to be able to do their job in all Socket 939 motherboards. Thus, nothing stands in the way of an upgrade other than a BIOS update.) "

so it looks like there will be 939DC's with no new board buying

incomudro
 
incomudro said:
this was something i found at toms hard ware "It's sad, but true: with the Smithfield dual-core processor set to increase maximum power dissipation from 115 to 130 watts early this summer, Intel had to make small modifications to the pin setup of the Socket 775. It is thus beyond any doubt that all earlier boards on the market will not work with the dual-core processors. (For their part, AMD has made a clear statement to the opposite effect: the dual-core Athlons running the Toledo core will to be able to do their job in all Socket 939 motherboards. Thus, nothing stands in the way of an upgrade other than a BIOS update.) "

so it looks like there will be 939DC's with no new board buying

incomudro
Click to expand...

That has been my understanding as well. I certainly hope it holds true, because that's one of the reasons I went with s939 this time around.
 
incomudro said:
this was something i found at toms hard ware "It's sad, but true: with the Smithfield dual-core processor set to increase maximum power dissipation from 115 to 130 watts early this summer, Intel had to make small modifications to the pin setup of the Socket 775. It is thus beyond any doubt that all earlier boards on the market will not work with the dual-core processors
incomudro
Click to expand...

A 130W is the top end.
The low end will around 95W, for the first 8xx offerings. (But they're all going to require the upcoming 945 and 955 chipsets).

and Intel is alteast promising to move desktop chips back down to around 60W. Which may be an indication of a move towards a Penitum M on the desktop.
URL=Article from Tom's]http://www.tomshardware.com/hardnews/20050302_204846.html[/URL].
 
remember that amd rates tdp differently than intel.
intel effectively uses the power disspation at idle for their figures...
and amd rates one tdp for the fastest chip in the line (note that all 130nm a64's are rated for 89w with the exception of fx-55 at 104)
 
Yes, they do use different models for TDP. (Though Intel doesn't measure it at idle).

The processor’s power is specified as Thermal Design Power (TDP) for thermal solution design.
TDP is defined as the worst-case power dissipated by the processor while executing publicly
available software under normal operating conditions, at nominal voltages that meet the load line
specifications. The TDP definition is synonymous with the Thermal Design Power (typical)
specification referred to in previous Intel datasheets. The Intel TDP specification is a
recommended design point and is not representative of the absolute maximum power the processor
may dissipate under worst case conditions.
Click to expand...

Depending on the chip, it's estimated TDP is between 75% and 90% of max power dissipation. (Not that you can't find max power dissipation, intel lists max core amperage for it's voltage regulator design schemes)

AMD, as you point out, defines TDP (at least on K8) as an absolute maximum rating (Idd_max * Vcc_nom) for an entire group of chips.
 
You must log in or register to reply here.
Back
Top