• 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.

First Haswell engineering sample?

Is Haswell even taped out yet? Zero chance an engineering sample (and importantly, an engineering sample motherboard capable of running benchmarks) would even exist today.
 
grambo said:
Is Haswell even taped out yet? Zero chance an engineering sample (and importantly, an engineering sample motherboard capable of running benchmarks) would even exist today.
Click to expand...


Haswell certainly has ES's out. I know someone writing graphics drivers for the IGP.
I don't know anything more then he's got a gimped chip that allows him to do his work, but no *real* testing.
 
Indeed, I believe Intel announced that Haswell taped out by IDF 2011. In April 2012 bigpao007 (a poster on chiphell who has repeatedly proved to have access to unreleased products) posted a Haswell screenshot, indicating that Haswell has an Extended Family number of 3C, and a consequent CPUID of 306Cx, where x is the stepping number.

This purported screenshot indicates an Extended Family number of 4A. Such a Family would suggest a 14nm processor, not 22. 1x Families were 45nm, 2x families were 32nm, and 3x families are 22nm processors.

The other thing wrong in this image is the cache size and associativity. 20-way associativity indicates 2.5MB cache slices per core. On an eight core processor that would give 20MB of L3. Assuming similar cache design, you would need 32-way associativity to get 32 MB of L3 cache on an 8 core processor.

All signs point to this being a decent photoshop by someone with limited knowledge of Intel CPUID parameters and cache characteristics.
 
it can easily be an ES why is it impossible? every cpu has an ES, thats how thet get tested and reviewed
 
Sorry, I was blatantly wrong. I should have said I doubt anyone would have these chips "in the wild" and be playing with them at home with CPUID. I would thing at this stage of the game it would be in-house development on the chipset/motherboard side where they would have controls over leaks.
 
I call BS. Early ESes don't know how to report the model numbers.
 
You must log in or register to reply here.
Back
Top