-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Aloha!
Hilarie Orman wrote: >> On 31 August 2016 at 20:48, Hilarie Orman >> <hila...@purplestreak.com> wrote: > >>>> From: Brian Sniffen <bsnif...@akamai.com> > >> The question is not "how much hardware?" but "price?" - with ARMs >> including h/w AES coming in at $2 for a single unit, its hard to >> explain why you\d want to use a less powerful CPU... > > > Power. > > Hilarie Did you look at the ARM Cortex M0+ Gecko Zero I pointed to? I'd recommend that you compare its power consumption to a PIC. The PIC is manufactured using larger geometries that consumes more power/gate/MHz. The Gecko Zero has more power modes allowing it to enable/disable different functions very fast, and is able to scale its own internal clock frequency very flexibly. The Gecko Zero (and other M0+ devices) can also do more/cycle so that total power up time is shortened, saving power. Specifically (since we talked about it before), the AES core in the Gecko Zero takes about 50 cycles to process one block (and the CPU core can be powered down at the same time). Googling for cycles to perform AES on PIC I found: Encryption PIC16F877 : 3834 cycles PIC16F84 : 7157 cycles https://edipermadi.wordpress.com/2008/02/09/an-aes-implementation-on-pic16f877/ So on the PIC you need to have the CPU core powered up and running about 80 times longer (in terms of cycles) than the Gecko needs to run its AES core. And even if you don't have an AES core, the ARM can do AES in fewer cycles. This one shows 2270 cycles for AES-128 http://www.cryptovia.com/ARM_Thumb_AES.html Selecting 8/16 bit MCUs like AVR, PIC, 8051 in 2016 for power reasons without looking at modern 32-bit MCUs based on ARM or MIPS is a mistake imho. - -- Med vänlig hälsning, Yours Joachim Strömbergson - Alltid i harmonisk svängning. ======================================================================== Joachim Strömbergson Secworks AB joac...@secworks.se ======================================================================== -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQIcBAEBCAAGBQJXzmPpAAoJEF3cfFQkIuyNW74P/3CuHjDaD+OnU4RHVemzDgKZ rRvDa/3ekLVDDKUQUP5UhPLolg8tn7L1pPWwT8MU+aY2T7TUyRmwG8dukQlTdk1G 7jZi7xE5efEY6vOHDH2s50nL166vRQAC2dBpSFqJQ4SXXM23qkViCW1JI64jvLmg KVryfU0LWSKZc0QB3Yta0g4nLwkBWIUywpxWOiGT0y54P6I6YjLKavnxvToOOnxw wdO6WDSNx8fBCOwf4Sb8cVmX77dYN+JywRgOUWfT9uweUWuQZQe3MujLRbNd8uNY +9qlwb7uG9I/OLypmu/7hHpwb5U/kbP04u6kbedG1h+TT/QxwU+vOwz9nRytvHDL kOj3VYZnmWQFgcr/fvmXMiUL3s9qQWubIH151JLylDQF1dC+QhvIPlHfGlKQXKWv 8+ZfDtLAAHIagDqdMNO7bX7I2NujqO5P7XmLgw0p6GMwgV2hrdX8Jw6t5sp/xEqc 9pv+hMJIYT3QzWm9XZbjPpkoSEmt4yHciZkQuzaZyShLf3M76mEC/HnpYCHdp6JQ 5YcKwfBTX6WpFQ/PPhEu9NcTVHu2z0WxIPsv4O3+FUdXWtBPzHto8D6k8m3hpwOw d/GwX0HMUIaldA2l1o+0ZWKmO4Ov81EDM3bCOeUzSjp4ZE33S3TkXgGG35MIX29T 7js19Jx4tVKQYimJbJRO =N7Q6 -----END PGP SIGNATURE----- _______________________________________________ TLS mailing list TLS@ietf.org https://www.ietf.org/mailman/listinfo/tls
