 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : [Picomite] Optimizing for speed
  Page 2 of 2 |
|||||
| Author | Message | ||||
| hhtg1968 Senior Member  Joined: 25/05/2023 Location: GermanyPosts: 123 |
speed optimization was/ is(?) my passion. in the 80´s i programmed (in assembler) several arcade games on z80 systems with 3.54 mhz. every tick is important for speed. very often in interpreted systems the length of the names of variables has a meaning: by using a compiler it is not importand for the speed in runtime. but often in an interpreter the length can low down the speed. i will test on picomite wether there is a difference in speed with the var "HowBigIsThisValueOnScreen" and "h". i do not know how mmbasic is programmed; perhaps geoffg or matherp can say something to the length of the names of vars or subs. sometimes i asked me, wether it would be cool to have 8bit or 16bit integer vars. by using "small" loops it can be better to repeat simply the order: if you code: for x=1 to 3:print "x":next most of the time the cpu will handle the loop itself, not the code inside the loop... if the code is not long (in the loop), it could be better to code a few orders without loop: print"x":print"x":print"x". a problem in the 80´s: If you code a hires scroll in a memory with bit shifting commands you will get speed if you code 256 shift commands (with carry flag) instead of coding a loop. in early times there are so called macro assemblers to help. i will have time the next weeks to test the mmbasic speed optimization... |
||||
fmafma Newbie  Joined: 30/06/2023 Location: FrancePosts: 31 |
Thank you all for these advices! I will make a few changes in my code to see if I can get a few ms. Then, I will overclock the CPU. Frédéric |
||||
| lizby Guru  Joined: 17/05/2016 Location: United StatesPosts: 3150 |
Of course this: print"x":print"x":print"x" does not give the same result as this: for x=1 to 3:print "x":next But your point is well take--in some critical situations, unrolling a loop will help. With MMBasic, it's been largely onward and upward with respect to speed (but not for the PicoMite, which has many other advantages, including availability). Comparison between versions from this thread BM1 BM2 BM3 BM4 BM5 BM6 BM7 BM8 80's computers ABC 800 single 0.9 1.8 6.0 5.9 6.3 11.6 19.6 2.9 ABC 800 double 1.2 2.2 10.0 10.6 11.0 17.8 26.4 14.4 IBM PC 1.5 5.2 12.1 12.6 13.6 23.5 37.4 3.5 Apple III 1.7 7.2 13.5 14.5 16.0 27.0 42.5 7.5 VIC-20 1.4 8.3 15.5 17.1 18.3 27.2 42.7 9.9 ZX81 "fastmode" 4.5 6.9 16.4 15.8 18.6 49.7 68.5 22.9 MMBasic Maximite 0.016 0.144 0.196 0.205 0.354 0.512 0.721 0.310 Maximite w. #s 0.016 0.131 0.193 0.194 0.245 0.393 0.582 0.241 Maximite w/o #s 0.016 0.111 0.173 0.173 0.192 0.336 0.525 0.220 MicroMite 40MHz 0.028 0.18 0.285 0.289 0.644 0.892 1.346 0.376 MicroMite 48MHz 0.023 0.15 0.237 0.24 0.536 0.744 1.121 0.313 MM+ MX470 120MHz 0.013 0.082 0.135 0.135 0.265 0.380 0.597 0.174 MMX 198MHz 0.006 0.045 0.07 0.08 0.141 0.201 0.287 0.126 Picromite ZeroW 0.014 0.058 0.093 0.102 0.184 0.298 0.354 0.127 Armmite H7 0.003 0.023 0.038 0.042 0.067 0.098 0.146 0.065 ArmMite F4 0.011 0.079 0.14 0.149 0.249 0.354 0.528 0.257 MMBASIC DOS 0.0002 0.0017 0.0028 0.0028 0.0073 0.0092 0.0117 0.0058 CMM2 0.00176 0.0118 0.01903 0.01728 0.03611 0.05062 0.07582 0.02851 PicoMite 0.0166 0.1096 0.1800 0.1840 0.3325 0.4715 0.6688 0.3193 Linux 0.0000 0.0020 0.0040 0.0040 0.0080 0.0130 0.0190 0.0070 Other Recent Basics ByPic MX170 40M 0.001 0.010 0.023 0.020 0.027 0.041 0.080 0.043 BBC_V Pi Zero 0.0001 0.0007 0.0016 0.0014 0.0015 0.0030 0.0030 0.0036 Annex RDS ESP32 0.032 0.223 0.384 0.338 0.463 0.714 1.224 0.631 C-Language Arduino UNO 0.010 0.010 0.058 0.043 0.043 0.043 0.045 0.284 Arduino DUE 0.003 0.003 0.006 0.007 0.007 0.007 0.106 0.014 ESP32S 0.00007 0.00007 0.00030 0.00011 0.00016 0.00016 0.00016 0.02413 I think Peter published the numbers for MMB4W, but not in that thread. ~ Edited 2023-07-05 05:11 by lizby PicoMite, Armmite F4, SensorKits, MMBasic Hardware, Games, etc. on fruitoftheshed |
||||
| fmafma Newbie Joined: 30/06/2023 Location: FrancePosts: 31 |
Hi, I have another question about optimization: for audio reproduction, what does require less CPU: using a MPC4822 (SPI), or direct PWM (with low pass filter)? Frédéric |
||||
| cicciocb Regular Member  Joined: 29/04/2014 Location: FrancePosts: 70 |
Hi, just for the precision, these are the results for Annex RDS ESP32: Benchmark 1 0.009 Benchmark 2 0.084 Benchmark 3 0.161 Benchmark 4 0.159 Benchmark 5 0.198 Benchmark 6 0.278 Benchmark 7 0.56 Benchmark 8 0.277 ESP32-S3: Benchmark 1 0.008 Benchmark 2 0.065 Benchmark 3 0.124 Benchmark 4 0.119 Benchmark 5 0.149 Benchmark 6 0.218 Benchmark 7 0.413 Benchmark 8 0.211 |
||||
Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 6798 |
PWM is a bit less of a load on the CPU than driving the DAC, but the sound quality is inferior, especially at the higher frequencies. PWM also uses more components and more PCB area. :) Edited 2023-07-25 07:44 by Mixtel90 Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Page 2 of 2 |
|||||
 Print this page |
