AMD developing reverse Hyper-Threading?


Recommended Posts

"Conscious that K8 architecture could not compete with the next high-speed motorboat of INTEL, all its hopes are for the moment based on a new 'revolutionary' technology (it is our opinion, not it his) on which AMD works in this moment for after-K8. This technology is in fact a kind of anti-HT: There or HyperThreading sought to emulate two virtual processors with a physical processor, it is a question for AMD of emulating a single virtual processor with two (or several) physical processors."

http://www.bit-tech.net/news/2006/04/17/am...hyperthreading/

:D

At first I didn't see the point, until I read:

If AMD can perfect the ability to make multiple cores appear as one, then it could take a massive performance lead in applications that are single-threaded - like the vast majority of games right now.

Imagine a Dual Core FX-60 2.6ghz running in 'single core mode' at 5.2ghz ! :cool:

:cool: that would be pretty amazing

I really hope AMD realease a driver so that Current X2 chips can do this. :D

Yeah, I hope so too, but as Hyper-Threading is a in-build function of processors, not driver related, I don't know if that will be possible

Who would of thaught that 6Ghz processors would actually be possible! It's just an idea at the moment, sounds plausable, but I would wait until some prototypes are actually made.

If it works, it would not increase latency. ie: 1x6Ghz computes twice as FAST SPEED as 1x3Ghz, while 2x3Ghz computes twice as MUCH QUANTITY as 1x3Ghz. 3Ghz is 3Ghz, and it can't get any FASTer than the clock. It would improve overall speed for large computations though (process as MUCH QUANTITY half the time). This can be done with Multithreaded Applications, so the biggest gains would be in applications not programmed as Multithreaded (and without the overhead of software Multithreading)

Why didnt anyone think of this before? :)

It has been and it's been dismissed, mainly by the virtualization people. This is a very difficult, if not impossible problem and I would be very impressed if AMD got this to work.

EDIT: I should mention that IBM does something sorta like this in its mainframes. They use two processors to execute the same instruction and then compares the results. If they're the same, everything's good. If they're different, then both processors are shutdown and marked for replacement since one of them is bad. Think of it as RAID 1 for CPUs. Different from what AMD is working on, but it still combines two processors to make one virtual on.

Edited by MrA

wow AMD are like the Soviets they don't have all the flash whitsles or bells intel has but god damm they just go 1000 miles ahead when the aposeing corps get that little foot ahead

read the post above yours....i wouldnt say 1000 miles.....not even one

why are you people acting like it will have the same effect as doubling the ghz? this wont give you a 5ghz machine from a dual core 2.5ghz, just like the regular dual cores right now are not "doubling" the speed...this isn't much different, it will speed things up but it wont double the ghz lol

why are you people acting like it will have the same effect as doubling the ghz? this wont give you a 5ghz machine from a dual core 2.5ghz, just like the regular dual cores right now are not "doubling" the speed...this isn't much different, it will speed things up but it wont double the ghz lol

It would essentially be the same as doubling the clockspeed. AMD's do 9 (i think) operations per cycle, so if you double it to 18 on a 2ghz cpu, that's the same as doing 9 instructions on a 4ghz cpu = 36Gips

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.