• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

AMD's Dodeca-core processor

Joined
Nov 22, 2007
Messages
1,398 (0.22/day)
Location
Hyderabad,India
System Name MSI apache ge62 2qd
Processor intel i7 5700HQ
Memory 12 Gb
Video Card(s) GTX960m
Storage 1TB
Display(s) Dell 24'
AMD's planning a 12 core processor
AMD engineers reveal details about the company's upcoming 45nm processor roadmap, including plans for 12-core processors

"Shanghai! Shanghai!" the reporters cry during the AMD's financial analyst day today. Despite the fact that the company will lay off nearly 5% of its work force this week, followed by another 5% next month, most employees interviewed by DailyTech continue to convey an optimistic outlook.

The next major milestone for the CPU engineers comes late this year, with the debut of 45nm Shanghai. Shanghai, for all intents and purposes, is nearly identical to the B3 stepping of Socket 1207 Opteron (Barcelona) shipping today. However, where as Barcelona had its HyperTransport 3.0 clock generator fused off, Shanghai will once again attempt to get HT3.0 right.

Original roadmaps anticipated that HT3.0 would be used for socket-to-socket communication, but also for communication to the Southbridge controllers. Motherboard manufacturers have confirmed that this is no longer the case, and that HT3.0 will only be used for inter-CPU communication.

"Don't be disappointed, AMD is making up for it," hints one engineer. Further conversations revealed that inter-CPU communication is going to be a big deal with the 45nm refresh. The first breadcrumb comes with a new "native six-core" Shanghai derivative, currently codenamed Istanbul. This processor is clearly targeted at Intel's recently announced six-core, 45nm Dunnington processor.

But sextuple-core processors have been done, or at least we'll see the first ones this year. The real neat stuff comes a few months after, where AMD will finally ditch the "native-core" rhetoric. Two separate reports sent to DailyTech from AMD partners indicate that Shanghai and its derivatives will also get twin-die per package treatment.

AMD planned twin-die configurations as far back as the K8 architecture, though abandoned those efforts. The company never explained why those processors were nixed, but just weeks later "native quad-core" became a major marketing campaign for the AMD in anticipation of Barcelona.

A twin-die Istanbul processor could enable 12 cores in a single package. Each of these cores will communicate to each other via the now-enabled HT3.0 interconnect on the processor.

The rabbit hole gets deeper. Since each of these processors will contain a dual-channel memory controller, a single-core can emulate quad-channel memory functions by accessing the other dual-channel memory controller on the same socket. This move is likely a preemptive strike against Intel's Nehalem tri-channel memory controller.

Motherboard manufacturers claim Shanghai and its many-core derivatives will be backwards compatible with existing Socket 1207 motherboards. However, processor-to-processor communication will downgrade to lower HyperTransport frequencies on these older motherboards. The newest 1207+ motherboards will officially support the HyperTransport 3.0 frequencies.

Shanghai is currently taped out and running Windows at AMD.
source-Daily Tech
 
Toooooooooooooooooooooo many cores. I would rather see a 5GHz+ dual.
 
i want to see a crazy arch change...seriously..
 
yeah it will have 12 cores.....running at 2.0ghz and will only overclock to 2.2ghz rendering it lesser than the e8200
 
Okay, you have 12 cores on a single package, and server boards with 4 sockets already out. So....48 cores?

Those thinking, the wonder of Virtualisation is that at a hardware level, you can merge several cores to crunch one thread logically split and tagged. So no, a strong single/dual core chip isn't a match.
 
Okay, you have 12 cores on a single package, and server boards with 4 sockets already out. So....48 cores?

Those thinking, the wonder of Virtualisation is that at a hardware level, you can merge several cores to crunch one thread logically split and tagged. So no, a strong single/dual core chip isn't a match.

True.
An overclocked D@4.5 GHZ is still slower than a E2xxx series clocked @ 2.25~ GHZ.
 
If they can do what they did to the 3870x2(The card is seen as a single one but the card divides the work itself) then it would be good.
Also the thing to be noted here is they are finally doing what Intel is doing(using two die's) and they are now able to double the total no. of cores.
As for clock speeds they'll probably launch faster ones as time progresses.
 
True.
An overclocked D@4.5 GHZ is still slower than a E2xxx series clocked @ 2.25~ GHZ.

ya but look at the arch diff netburst to core arch is non comparable
 
too late from amd! cuz intel become take markt's with quad core's cpu's
evryone use amd wait for this
 
"Don't be disappointed, AMD is making up for it," hints one engineer.

shame they didnt leave his name, address, telephone number or any other forms of contact so we can all call him & tell him how much he fails when AMDs efforts gets invovled in a 'hit & run' by the Intel war machine.

this is probably one of the dudes that said phenom was going to be amazing & probably one of the dudes that was promoting the thing at the event when phenom was released.....

IMHO if they were to release a 12core cpu right now ahead of intel only will people show some interest & start buying their stuff otherwise - what are AMD achieving that Intel hasnt already achieved & gone on to the next hurdle???
 
Yup, it's not about announcements, big talk or 'sneak peaks' but releasing something and then talking big about it. For example, when Intel used hafnium insulator layers, they advertised the technology after it was incorporated into a retail product and not announced it years in advance and talked big about.
 
AMD has to talk about it, they are loosing marketshare in just about everything except workstations where the Barcelona is faster than the Xeon, but thats about it
 
But if if you have twelve 1GHz cores wouldn't that be the same as a 12GHz chip? For exsample, those Gamelabs tests that determine if a PC can run a game or not said my CPU is the equivilent of roughly 6GHz... I know my CPU is NOT runnig at 6GHz, but @ 3GHz per core, if should effectively be running that fast... I'd say that once they figure out a way to give each core it's own memorey channel or atleast share one in between 2 or 3 cores then we will see multi-core CPUs start to reach their true potential.
 
OMG PIMP MOBILE SHANGHAI!! Japan?
 
But if if you have twelve 1GHz cores wouldn't that be the same as a 12GHz chip? For exsample, those Gamelabs tests that determine if a PC can run a game or not said my CPU is the equivilent of roughly 6GHz... I know my CPU is NOT runnig at 6GHz, but @ 3GHz per core, if should effectively be running that fast... I'd say that once they figure out a way to give each core it's own memorey channel or atleast share one in between 2 or 3 cores then we will see multi-core CPUs start to reach their true potential.

Uh no. Each core focuses on one process only. So 12 processes will each get 1 GHz. While a single core 12 GHz can focus all 12000 MHz on only one process.
 
Uh no. Each core focuses on one process only. So 12 processes will each get 1 GHz. While a single core 12 GHz can focus all 12000 MHz on only one process.

so a single core would get the same single task done 12 time faster?
 
so a single core would get the same single task done 12 time faster?

No. Other components interfere with the cpu. Like the ram, gfx, mb, etc....
 
so a 1GHz 12-core CPU would be faster than a 12GHz 1-core CPU in real life situations then...?
 
kind of old news
i have readed some where that amd had a 16 cored setup and got beated by intel 8 cored setup
now we will see it again
amd 12 cored cpu HT 3.0 vs nehalem 8 cored that can do 16 threads + quikpath interconect + way faster than anny amd got atm
 
The rabbit hole gets deeper. Since each of these processors will contain a dual-channel memory controller, a single-core can emulate quad-channel memory functions by accessing the other dual-channel memory controller on the same socket. This move is likely a preemptive strike against Intel's Nehalem tri-channel memory controller.

WHAT RUBBISH (at least, marketing rubbish).

For one CPU to try and use ANOTHER CPU's memory controller to get Quad channel! Well, let me see:
#1. The first CPU has to set up a communication protocol via HT to request data (latency), an receive it (more latency).
#2. The second CPU has to stand still on its own threads, to process the CPU1 memory access requests. Hence rendering CPU2 zero channel whilst CPU1 gets "quad channel"
#3. The data is out of sync due to #1, hence need to HALT CPU1 and the data from memory controller 1 until data from CPU2/MC2 is ready.

OMG!!! Quad channel indeed, but by significantly increasing latency and forcing a HALT on CPU2. WTF!!!

Intel skulltrail (Quad channel FB-DIMM) and Intel Nehalem (3 channel DDR3/4) FTW!
 
I don't think anything over 8 cores is going to get fully used in next two years. Nehalem and Shangai are going to be primarily server architectures. Game developers are going to struggle with giving 4 cores something to do, never mind 12 or 16! :shadedshu
 
yay, amd finnally got their shit together! or so it appears..

cant wait till the next cpu & gpu wars, amd is really heating up intel and nvidia

now if programs could be more multi threaded etc..
 
lol amd is not heating up intel XD
 
Back
Top