Hi Richard
if you dont think you can stomach the changes of the RT1020 , suggest an
F7. Choose the one with lots of cache.
there are 4k, 8k and 16k cache parts. Use the 16k cache parts.
unless necessary, be sure that your compiler treats doubles as floats.
they are mostly pin compatible. the FPU is abotu 2x the speed of the F4.
My advice is dont use an F4, unless you are building a consumer product
that is down to the dollar...
F7s really only run at the same speed as F4s. (168 MHz) .The 216 MHz is
BS really.
-glen
On 5/02/2019 8:13 AM, Richard wrote:
Hi All,
As I said I don't have the software skills to jump to a significantly
different device.
Having had a better look at the F7 range it seems that to get more
FLASH and RAM than the STM32F405VGT6 whilst keeping to the "small" 100
pin package only actually leaves three options, differing only by
irrelevant peripherals. That makes the STM32F765VIT6 seem good, and
it's about £11 vs £9 for the STM32F405VGT6. So far so good.
Physical pin mapping is subtly incompatible, but that can be handled
as I haven't built any hardware yet. How much difficulty should I
expect getting the existing code to build for it?
Of course I could just proceed with the F4 like the SM1000, and just
accept that 700D may not end up being possible. Decisions. :-)
Thanks
Richard
<http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Virus-free. www.avg.com
<http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
On Mon, 4 Feb 2019 at 20:25, Richard <[email protected]
<mailto:[email protected]>> wrote:
Hi All,
Understood about the future modes, I was hoping to support 700D
(when released) so perhaps switching processor is desirable. My
skillset is 90%/10% hardware/software so both a complete rewrite
and porting to a totally different platform are out for me. I
might take a look at other ST options, several of you have
mentioned the F7 and a quick look seems promising. Does anyone
with more experience have a specific part number to recommend?
Low cost, whilst nice, isn't really a goal for me, but physical
size and power consumption are.
Thanks
Richard
<http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Virus-free. www.avg.com
<http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
On Mon, 4 Feb 2019 at 20:02, glen english <[email protected]
<mailto:[email protected]>> wrote:
Hi Don
I got codec2 working on F7 a few years ago, and I was getting
2x the
performance than the guys were getting on the F4. for the same
clock
speed...- 2 reasons- 1) the F4 code is completely unoptimized
and really
hurts the F4, and 2) the F7 is quite a bit faster if the cache
access is
optimal..
IMO the attempt to keep the code base portable between PC and
microcontroller is problematic and a limit.
suggest The ORANGE PI zero is a tiny inch square PCB and has
onboard
audio..... It's what I use.... quad core A7 etc and NEON.
H7 is th new kid ont he block , but STM32 is expensive !!!
Suggest Rt1050 series from NXP for low cost (couple of bucks)
and 400
meg F7 perfromance
-glen
On 5/02/2019 5:24 AM, Don wrote:
> I am currently working on getting the 700D mode to work with
the SM1000.
> It is not certain that this will be possible because of the
memory
> needed.
>
> Also Dave is looking at new compression modes with different
performance
> and bandwidths.
>
> I see the current firmware as best suited for experimenting
and developing
> on a desktop PC. To make a small, low cost embedded version
one would
> likely want to do a major re-write of the code. Lack
_______________________________________________
Freetel-codec2 mailing list
[email protected]
<mailto:[email protected]>
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
_______________________________________________
Freetel-codec2 mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
_______________________________________________
Freetel-codec2 mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
