Array
(
    [content] => 
    [params] => Array
        (
            [0] => /forum/threads/question-intel-first-full-custom-xeon-cpus.6180/
        )

    [addOns] => Array
        (
            [DL6/MLTP] => 13
            [Hampel/TimeZoneDebug] => 1000070
            [SV/ChangePostDate] => 2010200
            [SemiWiki/Newsletter] => 1000010
            [SemiWiki/WPMenu] => 1000010
            [SemiWiki/XPressExtend] => 1000010
            [ThemeHouse/XLink] => 1000970
            [ThemeHouse/XPress] => 1010570
            [XF] => 2021770
            [XFI] => 1050270
        )

    [wordpress] => /var/www/html
)

[question] Intel: First full-custom Xeon CPUs

opticon

New member
Last edited:
As long as you buy at least 10,000 of them...

I have no idea what their minimum order size will be, but if you are willing to put the money up for at least that many processors, custom work is on the Intel roadmap.
 
Advanced tech node tape-out costs millions. 10k units order should not justify a specific design optimization.
It could be done with FPGA integration. That actually increases the fun of life in my opinion.
 
Advanced tech node tape-out costs millions. 10k units order should not justify a specific design optimization.
It could be done with FPGA integration. That actually increases the fun of life in my opinion.

I have actually helped tape-out ICs when tape was still being used. I think the last one was 45 micron line widths. How long will the term tape-out live? I never would have guessed it would last this long.

I work with the latest FPGA's now and they are great tools, but there are a lot of things they cannot do that CPU's do easily. Anyone who has tried to implement a 64-bit CPU as a soft core that runs at 1GHz+ can tell you just how wide the gulf is.
 
Back
Top