ATTENTION: You are viewing a page formatted for mobile devices; to view the full web page, click HERE.

Main Area and Open Discussion > Living Room

should intel concentrate on HD access instead of pumping up CPU Ghz?``

(1/1)

urlwolf:
Interesting read at joel's

Xenonym:
Well, mass storage access rates ain't going to sell more processors, since in a lot of people's minds, GHz is directly proportional to speed.

Lashiec:
As various readers commented in that discussion, Intel makes the fastest disks around. And, coincidentally, they also make the fastest processors. Intel is a really big company, so they can easily focus their efforts in various fields, and still be at the top of all them, which mind you, it's what they're doing. Case closed :P

It should be said that even if Intel focused all its muscle on disks, it's practically impossible they could improve the access and data retrieval from them on a serious number overnight, radical improvements in technology take time. Not to mention it's impossible to have 1 TB of disk space with the access time of DDR3 RAM, you can't have short access time and big capacity at the same time with an acceptable cost for the customer, barring gimmicks like Gigabyte i-RAM or A-Card RAM Disks (which are not even as spectacular as they could be, blame bus speeds). Besides, there are several other SSD makers which are continuosly improving their disks performance, promising even more impressive results than Intel offerings, so don't throw all the weight on Intel shoulders.

Also, the GHz race has long died down, CPU speed has remained flat for a couple of years now (including AMD CPUs), yet they continue to improve performance via other methods, whether the task is single-threaded or multi-threaded. CPU speed is not the sole variable by which CPU performance is measured, and actually it rarely plays an important role nowadays.

Deozaan:
Interesting discussion. I don't know enough about the topic to contribute, but I am enjoying the conversation. :)

Navigation

[0] Message Index

Go to full version