• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Do we REALLY need new CPU's?

Page 5 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
so i'd assume the 450m is a dual core, but supports hyper-threading. does that mean it behaves like a 1.2 ghz quad? i don't really get how multi-threading cores work
 
I need more cores. And more speed per core. More of everything, except heat, really. I'm a big Handbrake user at the moment, converting all my movies and TV shows onto my hard drive. So more cores are good. And a lot of the applications I use are really resource-intensive, but they're still not multithreaded (I'm looking at you, MS Office), so more speed per core is also good.

I think you're in the minority with your (completely anecdotal) assessment.

And yours isn't?
 
so i'd assume the 450m is a dual core, but supports hyper-threading. does that mean it behaves like a 1.2 ghz quad?

marketing of the product will want you to think that, but it is more lie than truth.

In real world use, hyperthreading is more likly to help overall system performance vs not having it by upto 20%. Though in some tasks, Hyperthreading can result in overall performance loss (about 5%). In the later case, it is more when running single threaded applications.
 
Not *everything* - i was talking about video editing.

Some things are done better and faster by serial processing - the CPU. However, the computing world is moving toward GPU-centric - a powerful GPU for parallel processing - the tasks that count (editing/graphics) and a relatively weak CPU to keep it fed.

in ten years Intel will be a much smaller company if they keep on in their current direction. They are going to fail in the growth industry - mobile - where the GPU and their competitors excel.
Except it isn't really. The most powerful supercomputers in the world today and those planned for years to come use POWER, x86 and, in the case of the current No. 1, rather surprisingly SPARC, CPUs for floating point operations - even though if you look at raw FLOPS GPUs could offer much higher numbers at a fraction of the cost - and that's because even the most demanding (and most parallelizable) tasks are often better served by CPUs than by GPUs. It remains to be seen if the future of parallel computing are actually heterogenous architectures as proposed by AMD and Nvidia or aggressive SIMD extensions to traditional CPU architectures (possibly along with Larrabee-like many-core processors) as laid forth by Intel. The latter approach is decisively favored by vastly simplified programming models, and in the end that makes a lot of difference (witness for example the failures of Itanium and Cell despite their impressive theoretical performance).
 
Back
Top