Thread: Multicore processor
View Single Post
Old 09-24-2017, 05:03 PM   #4
mdp
Wizard
mdp ought to be getting tired of karma fortunes by now.mdp ought to be getting tired of karma fortunes by now.mdp ought to be getting tired of karma fortunes by now.mdp ought to be getting tired of karma fortunes by now.mdp ought to be getting tired of karma fortunes by now.mdp ought to be getting tired of karma fortunes by now.mdp ought to be getting tired of karma fortunes by now.mdp ought to be getting tired of karma fortunes by now.mdp ought to be getting tired of karma fortunes by now.mdp ought to be getting tired of karma fortunes by now.mdp ought to be getting tired of karma fortunes by now.
 
Posts: 1,481
Karma: 9010563
Join Date: Jul 2013
Device: none
...benchmarks make you dream even further...

I tested the standard boards used by Onyx for most devices (i.MX 6SoloLite, or "ARM Freescale @ 996 MHz"), and consistently with other users I get the value of 285 in the Geekbench 3 scale (the Geekbench 3 scale uses as its parameter the Mid 2011 Mac mini with Intel Core i5-2520M @ 2.50 GHz, defined as value 2500). The "Onyx MC_C67ML_Carta" declares a "ARM RK30board @ 912 MHz" with values 275 single-core and 450 multi-core; the Kepler Pro should employ something better, but nobody run the tests.

Comparatively speaking, a high end mobile device could have values of 1000 single-core and 4000 multi-core; a high end desktop of 4000 single-core and 16000 multicore.

So, still dreaming: for example a Rockchip RK3288 would have values of around 800 to 850 single-core, around 2100 to 2500 multi-core. That's three times the current boards employed by Onyx, without even counting the multi-core.

I am counting the minutes. And I'd love, even before seeing it, to know what the future reserves.
Last edited by mdp; 09-24-2017 at 05:09 PM.
mdp is offline   Reply With Quote