What Is Penryn Bringing To The Table?
If you were thinking it was anything but a serious upgrade, then perhaps a quick look at some of the features and how they'll help Intel fight AMD's upcoming Barcelona might change your mind.
As always, until Intel officially releases the Penryn family of CPUs treat this as nothing more than interesting rumors/speculations. That said, here are a few of the most interesting tidbits:
New to the
Advanced Digital Media Boost technology is the inclusion of a new
Intel SSE4 instruction set. SSE4 introduces 47 new instructions to
improve performance of video accelerators, graphics building blocks
and streaming load. Intel claims a 2x performance gain in video
acceleration tasks. There are 14 new instructions for video
accelerator performance enhancement. Intel improves compiler
auto-vectorization performance with 32 new instructions.“
The large L2 cache(s) could certainly offer some performance gains, especially for anyone doing a lot of heavy multi-tasking. For users who are running only one program, it will still likely offer some performance gains, but just how much will vary from program to program.
Despite the enhanced size of the L2 caches, they are also rumored to have lower latencies. This means that programs that occupy a large portion of the cache could receive an even larger performance boost. From an engineering standpoint, it's quite hard to increase the size and reduce the latency. Generally if cache sizes go up, latency stays the same or takes a small performance hit. If Intel can accomplish both at once while keeping the price reasonable, we'll be impressed.
While these two feature are the easiest to talk about, there are a few others that could end up being very important. Chief amongst these are the enhanced division capabilities (4 bytes/cycle vs. 2.), improved Intel VT (with lower latencies) and improved power management which should help keep those electric bills down.
All things considered, the Penryn sounds like it will a rather serious upgrade. We can't wait to get our hands on a few of them and see how the handle our battery of benchmarks.
