-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Nehalem won't need a Northbridge
- Thread starter Teaf
- Start date
I think that's only the value processor, the performance parts will have double the cache in the place of the integrated GPU.
Jokester
Intel have been doing that since the P4 launch, I was expecting something different.
I'm pretty sure that Nehalem will be 45nm to start with and then 32nm not long after.
There was more info on Nehalem arc. in this thread and it's links in case you're interested.
So in lamens terms (for me. lol) does this basically mean, yes, there is still a northbridge but just that some of the features which are currently handled by it are going to be ran direct from the CPU instead? Basically.... reducing overheads?
Cheers.
Cheers.
So in lamens terms (for me. lol) does this basically mean, yes, there is still a northbridge but just that some of the features which are currently handled by it are going to be ran direct from the CPU instead? Basically.... reducing overheads?
Cheers.
Might actually cause more problems, if Intel stick to the FSB solution. More traffic concentrated on the CPU, instead of between SB and NB.
The latter. But unlike AMD, Intel are also putting the graphics controllers (IGP or PCI-E) on the CPU for the value range.
I was referring to *current* AMD solutions. But point taken.
So let me recap my shoddy knowledge on the northbridge . . . . it acts as a kinda hub between all the high speed components (Memory, Graphics what else?).
So by placing the northbridge in the CPU itself the high speed data all goes direct through the processor and latencies are reduced hmm. . .
I can also see some downside in as much as when you swap out one of these CPUs for another one the memory controller could act differently, if the chip gets to hot (from overclocking) you may run into memory problems or in the worst case scenario you could fry the memory controller?
Whats it all about then in laymens terms? reduced costs and lower latencies?
So by placing the northbridge in the CPU itself the high speed data all goes direct through the processor and latencies are reduced hmm. . .
I can also see some downside in as much as when you swap out one of these CPUs for another one the memory controller could act differently, if the chip gets to hot (from overclocking) you may run into memory problems or in the worst case scenario you could fry the memory controller?
Whats it all about then in laymens terms? reduced costs and lower latencies?
