Jump to content

Intel Core 2 Quad out! Better on a OSX86?


Synaesthesia
 Share

15 posts in this topic

Recommended Posts

Hey, as everyone probably knows, the new Intel kentsfield quad core processors are out, and they are supported on the 975x platform, so should be the best cpu for an OSX86 box...

 

This quote is from anandtech's review :

 

Using a Quad Core System

-----------------------------------------------------

Intel's first dual core processors, codenamed Smithfield, were actually the first 90nm Intel CPUs we actually recommended. The reason being that the real world performance improvement brought about by having two cores at your fingertips was simply too much to resist, especially if you were a heavy multitasker. A big reason for the improvement in real world performance came about because of inefficiencies in the way Windows XP's scheduler handled juggling multiple threads, especially on a single core CPU. The move to dual core got rid of many of those nasty Not Responding windows when Windows' scheduler would simply never properly allocate CPU cycles to a particular thread.

-----------------------------------------------------

 

So I take it, because of OSX better scheduling or whatever, that it would be the better os for this cpu?

 

Anyway, it's all academic, I don't have $999 now anyway!

Link to comment
Share on other sites

what the.. quad core.. wow.

 

i would like to see it in action compiling my whole gentoo system.. a rocket

 

anyways... yeah.. windows has got the worse task scheduling management i've ever seen. the cpu stuttering while min/maximizing windows is unacceptable

 

about mac os.. with the 8.8.1 kernel it should run great

Link to comment
Share on other sites

Quad core aint out, there gonna be released in 2007...

And they are not gonna be called Core 2 Qaudro... but i can't remember the name of them... Core 2 Duo is out though :construction:

The only way you could get qaudro is either put 4 singlecores in a 4 cpu mobo or 2 dual cores in a 2cpu mobo... or go for 8 cored with dual core in 4 cpu mobo :o

Link to comment
Share on other sites

Quad core aint out, there gonna be released in 2007...

And they are not gonna be called Core 2 Qaudro... but i can't remember the name of them... Core 2 Duo is out though :whistle:

The only way you could get qaudro is either put 4 singlecores in a 4 cpu mobo or 2 dual cores in a 2cpu mobo... or go for 8 cored with dual core in 4 cpu mobo ;)

actually they launched on the 2nd and will be available (to who i don't know) on the 14th http://theinquirer.net/default.aspx?article=35346

 

I'm waiting for the 32 core intel is planing on for next year..

Link to comment
Share on other sites

  • 2 weeks later...
  • 2 weeks later...

Hello again fellow testers. We (www.cdrinfo.com) actually have had a mighty quad core on our labs for testing and (yes) we were tempted and finally got it to test with OSX (10.4.8). Mifki core had some problems, while Semthex (10.4.8 Jas release) seems to work fine (both cases with no cpu identification on about box). I don't want to disappoint you but the engineering sample we had, produced very high temperatures under windows with no overclocking at all... (even though it had an unlocked cpu multiplier from 6-20).

I don't want to spoil the surprise, within the next few days I believe the owner of the site will post both my review on the 6600 dual core running OSX and his own experience on the quad-core from Intel. Stay tuned. :D

Link to comment
Share on other sites

Indeed... hehe. I have noticed a timing problem though with the 6600 today. After installing the SMBIOS kext to fix the about box problem, the realtime clock is ticking faster! (gains about 10' an hour) :P

Maybe it's not the smbios.kext that produces the error, but the cpu itself. I can't tell...

 

UPDATE: Mifki new core fixed it.

Thanks to this thread: http://forum.insanelymac.com/index.php?act...=33342&st=0

:huh:

Link to comment
Share on other sites

any recent semthex kernel should also fix your timing issues, just the one that came with JaS release didnt have any of the code to detect it.

 

Thanks for the info sg, I didn't know that :blink: . Actually I have found out another (and this is major) problem concerning the timing issues of OSX. This has to do with the assignment of the time_t variable as an unsigned integer within the mach kernel of OSX.

 

This has the effect on 2038.01.19 the computer to fail operating normally. I have reported this into the article (I promise it'll be out within the next few days) and sometime back I have mailed Apple about it too (never got a reply though). This is a problem I used to have on all my Macs (A G3 466 clamshell Special Edition, a G3 ibook dual USB White plus my G4 QuickSilver).

 

The weird issue here in my opinion, is that all these machines have a common thing: Their CPUs are 32bit. So there's a bit of 'expected' to have the time_t problem since they're not 'able' to count more than 2^31-1 = 2,147,483,647 seconds. But what about the 6600 intel core 2 duo; or my AMD64 X 2 CPUs? These are supposed to have 64bit extensions and thus the 32 bit word should have been extended to a 64 bit word...

 

...well, it isn't.

 

I just wonder if semthex or mifki kernel developers could override the issue by changing the way time_t variable is declared. That would be really fun to know and report to Apple...;)

By the way, excuse me about the :) post here. :compress:

UPDATE:Article is up you can read it here. Please excuse my English, it's not my native language. Also the Benchmarks are not done by me. :angel:

Link to comment
Share on other sites

 Share

×
×
  • Create New...