Hi Glen,

On 09.03.2018 09:44, glen english wrote:
> Agreed.
> skip the F4 and go straight to the F7 , unless cost is an issue.
Porting the software is an issue as well. F7 needs some more care (it
has real caches for code and data), unless you stick to GPIO, a lot of
higher complexity peripherals have changed more or less slightly. It is
far from a simple recompile. I did the port of the UHSDR F4 to F7 (and
H7) so I gained some insights here. Using the STM32 HAL was helpful, but
even then it takes some effort. BTW, the CODEC2 part of the UHSDR
firmware was just a simple recompile :-)  . So YMMV... If the software
is based on the MMDVM_HS code, then you have the old StdPeripheral lib
code in, which is not available for the STM32F7xx
> CODEC2 was almost 1.7x the speed on the F7 compared to F4 for same
> clock .
Fits with my numbers
> The H7s are expensive, don't you think Danilo ?  I am going RT1050 for
> next design.. $5. k
Well, since there is more demand than supply, what do you expect? I see
prices from 14 Euros to more than 21 Euros for the same STM32H743ZIT6
part here in Europe (and most distributors have zero stock). Next
revision will be the first mass produced one and come much later this
year. Then prices will (hopefully) drop to be close to the F7 and then
there is no reason not to use the H7. Peripherals between H7 and F7 are
a lot closer then F4 and F7, at least from the HAL/CubeMX perspective.


Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
Freetel-codec2 mailing list

Reply via email to