Boston Linux & UNIX was originally founded in 1994 as part of The Boston Computer Society. We meet on the third Wednesday of each month at the Massachusetts Institute of Technology, in Building E51.

BLU Discuss list archive


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

cpu core counts (was: iPad)



On Mon, Apr 12, 2010 at 11:44 PM, Dan Ritter <dsr-mzpnVDyJpH4k7aNtvndDlA at public.gmane.org> wrote:
> On Mon, Apr 12, 2010 at 07:48:35PM -0400, Mark J Dulcey wrote:
>>
>> But in these cases, the disabled parts of the chip may have been turned
>> off for a reason. They don't work at all, they work but fail under some
>> operating conditions such as extreme temperature, or the chip overall
>> fails to meet its power consumption and thermal specs if all the cores
>> are enabled. Then again, they may be perfect chips that the company is
>> selling off as the less capable versions because they don't have enough
>> of the not-fully-functional parts to meet demand at the reduced price.
>> So... is your 3-core CPU an honest sale if the fourth core is defective
>> but a ripoff if the fourth core works?
>
> My cousin the CPU ceramic packaging engineer says that yields are always
> being pushed on the lines to the point where finding 1/4 dead chips is
> perfectly normal, and it's quite rare that they dump fully armed and
> operational chips as inferiors.
>
> He's surprised that AMD and Intel haven't announced 5-core
> spoilage variants of the new 6-core chips... yet.

I wouldn't be surprised if they just took them and sold them as
quad-cores. One production line for both the 6-core and 4-core chips,
less marketing hooey. :)

-- 
Jarod Wilson
jarod-ajLrJawYSntWk0Htik3J/w at public.gmane.org






BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!



Boston Linux & Unix / webmaster@blu.org