Why is Core i7 920 better than Phenom 2 955

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
hyperthreading is not a gimmick dork! Read up on hyperthreading before you call it a gimmick. Ok, let's make it fair, Hypertransport is a gimmick.
 


The problem with exclusive caches is that the hypertransport bandwidth has to be used for snooping in the other cores' caches. See HERE :

Nehalem’s 8MB and 16 way associative L3 cache is inclusive of all lower levels of the cache hierarchy and shared between all four cores. Although Intel has not discussed the physical design of Nehalem at all, it appears that the L3 cache sits on a separate power plane than the cores and operates at an independent frequency. This makes sense from both a power saving and a reliability perspective, since large caches are more susceptible to soft errors at low voltage. As a result, the load to use latency for Nehalem varies depending on the relative frequency and phase alignment of the cores and the L3 itself and the latency of arbitration for access to the L3. In the best case, i.e. phase aligned operation and frequencies that differ by an integer multiple, Nehalem’s L3 load to use latency is somewhere in the range of 30-40 cycles according to Intel architects. The advantage of an inclusive cache is that it can handle almost all coherency traffic without disturbing the private caches for each individual-core. If a cache access misses in the L3, it cannot be present in any of the L2 or L1 caches of the cores. On the other hand, Nehalem’s L3 also acts like a snoop filter for cache hits. Each cache line in the L3 contains four “core valid” bits denoting which cores may have a copy of that line in their private caches. If a “core valid” bit is set to 0, then that core cannot possibly have a copy of the cache line – while a “core valid” bit set to 1 indicates it is possible (but not guaranteed) that the core in question could have a private copy of the line. Since Nehalem uses the MESIF cache coherency protocol, as discussed previously, if two cores have valid bits, then the cache line is guaranteed to be clean (i.e. not modified). The combination of these two techniques lets the L3 cache insulate each of the cores from as much coherency traffic as possible, leaving more bandwidth available for actual data in the caches.
 


On the other hand, due to inclusive technique, Core i7 also has less cache miss than Phenom II (theoretically speaking). High rate of cache miss means in the case core cannot find the file in its L1 data, it will then seek data in L2, L2s in other cores, and L3s (also segmented among cores), before finally hitting the RAM. This causes outrageously high latencies.
 
Not really.

K10's latency:
cm3d-phenom-ii-x4-940.gif


Core i7's latency:
cm3d-i7-920.gif


Not only that, the chance of Core i7 having a cache miss (which warrants a read from the RAM) is much lower than Phenom II due to inclusive cache.
 
^ +1. Also note that for Istanbul, AMD is including a snoop filter that will look in the other 5 cores' caches for data before heading out to main memory, in an attempt to catch up to Intel. Of course, this will consume some of the hypertransport bandwidth.
 
hey tomshardware didn't notify me about the replies to this thread....

yes I only want to play a lot and lot of games and I will be using Vista then will change to windows 7. Am planning to use a 22 inch monitor.

Thanks for the answers. Really appreciate them.
 


22-inch very likely equals 1680 x 1050, right? :??:

2 x HD 4770s have looked really good at that rez - and I was surprised how well they have mostly performed at 19x12, too.

It's a tough call (but) if I were gonna spend $200 today I'd snag an HD 4890 1GB for $200 AR ---->>
http://www.newegg.com/Product/Product.aspx?Item=N82E16814102830

And later on this year when you want to reward yourself (and the price continues to drop) I'd snag another one (so you can play all the upcoming games really fast on that new 50-inch 1080p plasma you get for the holidays - LOL)

 

Hes got speed stickers on the side panel helps overclock with the magic green AMD lighting meaning the overclocking gnomes are working hard.

Word, Playa.