Seems the advantage for F3 is it has LQFP48 package where the smallest for F4 is LQFP64. I thought about F3 but since it can only go up 72MHz, which made me hesitated.
LQFP48 of F3 is very attractive but the RAM size is limited to 48 kbytes. I need more RAM region to calculate the shortest path to the goal precisely. This is the reason why I focus on F4 device.
On the other hand, clock of 72MHz is enough, I think, because F3 also has floating point unit. It will accelerate the calculation of my control program.
Yeah, ram is a big issue,unless you write everything very conservatively. Glad Kato can survive with 20K Ram on his tetra :) I am exciting about the high clock speed of F4 even with only internal OSC, it literary double the compute speed of my path generator and flood fill since I on purposely made everything integer(in this case FPU doesn't help at all), still F103 is good so far. I am looking forward to see your new half size mouse with AS5040 or similar encoder working!
Seems the advantage for F3 is it has LQFP48 package where the smallest for F4 is LQFP64. I thought about F3 but since it can only go up 72MHz, which made me hesitated.
返信削除LQFP48 of F3 is very attractive but the RAM size is limited to 48 kbytes. I need more RAM region to calculate the shortest path to the goal precisely. This is the reason why I focus on F4 device.
返信削除On the other hand, clock of 72MHz is enough, I think, because F3 also has floating point unit. It will accelerate the calculation of my control program.
Yeah, ram is a big issue,unless you write everything very conservatively. Glad Kato can survive with 20K Ram on his tetra :) I am exciting about the high clock speed of F4 even with only internal OSC, it literary double the compute speed of my path generator and flood fill since I on purposely made everything integer(in this case FPU doesn't help at all), still F103 is good so far. I am looking forward to see your new half size mouse with AS5040 or similar encoder working!
返信削除