Madshrimps Forum Madness

Madshrimps Forum Madness (https://www.madshrimps.be/vbulletin/)
-   WebNews (https://www.madshrimps.be/vbulletin/f22/)
-   -   What you need to know about Intel's Nehalem CPU (https://www.madshrimps.be/vbulletin/f22/what-you-need-know-about-intel-s-nehalem-cpu-43470/)

jmke 9th April 2008 13:57

What you need to know about Intel's Nehalem CPU
 
In this article, I'll give a general overview of Nehalem, focusing on the major changes and big new features that the architecture will eventually bring to Intel's entire x86 processor line. A more in-depth examination of Nehalem from me will show up later in the spring; for now, read on for the highlights. Here's what you need to know about Nehalem.

http://arstechnica.com/articles/paed...ut-nehalem.ars

Kougar 14th April 2008 13:56

Great article. I didn't realize the changes were quite so numerous such as the loop stream detector.

Quote:

Nehalem widens the number of x86 instructions that can be macrofused in two ways. First, it expands four new compare + jump branch conditions to the list of macrofusable instruction pairs. Second—and this is major—it can now macrofuse 64-bit instructions
Macrofusion is partly what gives C2D it's performance advantage, it will be very interesting to see if this new 64-bit capable version means 64-bit OS's will offer better performance at long last?

I still am still betting QPI is going to show a significant performance improvement with the Nehalem generation thanks to all these changes. SMT should actually be useful this time around. :)

jmke 14th April 2008 14:24

doesn't 64-bit improvements come from applications written especially for 64-bit, which take advantage of the upgraded architecture ?

Rutar 14th April 2008 14:31

What we need to know? It made the CTO of AMD resign =P

Kougar 14th April 2008 21:19

Quote:

Originally Posted by jmke (Post 168066)
doesn't 64-bit improvements come from applications written especially for 64-bit, which take advantage of the upgraded architecture ?

Not sure the program must be natively 64bit to see an improvement. If I understand this Nehalem will simply act upon several specific 64bit instruction sets, and fuse them together for processing in half the time. Same principle as the current 32bit macrofusion. Except with Core 2 Duo 64bit instructions are first split into 32+32, then processed.

64bit applications on 64bit Operating Systems would no longer need to be split into 32bit+32bit. Instead, select 64+64 instruction sets would be processed in the same clock cycle previously taken to process a 32+32bit (A single split 64bit) instruction set. Or am I wrong? ;)


All times are GMT +1. The time now is 10:33.

Powered by vBulletin® - Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO