generated a table to work with, based on the S1/S2 tables:
Quote
...package 'cgminer'
config 'asic-freq' 'default'
#option 'freq_value' '1486' #262.5M
#option 'chip_freq' '262.5'
#option 'timeout' '15'
#option 'freq_value' '1386' #250M
#option 'chip_freq' '250'
#option 'timeout' '16'
#option 'freq_value' '1306' #243.75M
#option 'chip_freq' '243.75'
#option 'timeout' '17'
#option 'freq_value' '1286' #237.5M
#option 'chip_freq' '237.5'
#option 'timeout' '17'
#option 'freq_value' '1206' #231.25M
#option 'chip_freq' '231.5'
#option 'timeout' '18'
#option 'freq_value' '1185' #225M
#option 'chip_freq' '225'
#option 'timeout' '18'
option 'freq_value' '1106' #218.75M
option 'chip_freq' '218.75'
option 'timeout' '18'
#option 'freq_value' '1086' #212.5M
#option 'chip_freq' '212.5'
#option 'timeout' '18'
#option 'freq_value' '0782' #200M
#option 'chip_freq' '200'
#option 'timeout' '20'
config 'asic-freq' 'default'
#option 'freq_value' '1486' #262.5M
#option 'chip_freq' '262.5'
#option 'timeout' '15'
#option 'freq_value' '1386' #250M
#option 'chip_freq' '250'
#option 'timeout' '16'
#option 'freq_value' '1306' #243.75M
#option 'chip_freq' '243.75'
#option 'timeout' '17'
#option 'freq_value' '1286' #237.5M
#option 'chip_freq' '237.5'
#option 'timeout' '17'
#option 'freq_value' '1206' #231.25M
#option 'chip_freq' '231.5'
#option 'timeout' '18'
#option 'freq_value' '1185' #225M
#option 'chip_freq' '225'
#option 'timeout' '18'
option 'freq_value' '1106' #218.75M
option 'chip_freq' '218.75'
option 'timeout' '18'
#option 'freq_value' '1086' #212.5M
#option 'chip_freq' '212.5'
#option 'timeout' '18'
#option 'freq_value' '0782' #200M
#option 'chip_freq' '200'
#option 'timeout' '20'
first off thanks to you.
second can you find a number between 200 and 212.50 ( 206.25 ) comes to mind
third can you find something for 215.625
my guess for 206.25 would be (use at your own risk):
option 'freq_value' '1006' #206.25M
option 'chip_freq' '206.25'
option 'timeout' '19'
With all my due respect for your time and effort, but this table is NOT very applicable to S3...
You can NOT apply the same S1 math/calculations... to S3, because these are using different chips with different PFD (PLL Frequency Divider...I guess), subject to different conditions...due to chips redesign...I guess...
BM frequency table - from BM1382 datasheet ...:
Work Input timing
Clock Freq.(M) Hash rate(G) Reg.value Ideal delay(ms) Recommenddelay(ms)
100 6.30 0783 42.6 38
125 7.88 0983 34.0 31
150 9.45 0b83 28.4 26
175 11.03 0d83 24.3 22
200 12.60 0782 21.3 19
225 14.18 0882 18.9 17
250 15.75 0982 17.0 15
275 17.33 0a82 15.5 14
300 18.90 0b82 14.2 13
325 20.48 0c82 13.1 12
350 22.05 0d82 12.2 11
375 23.63 0e82 11.3 10
400 25.20 08f2 10.6 10
These are the same in S3's "asic-freq" file...
Conditions for S3 valid frequency divider's values now are different...check them here...:
https://bitmaintech.com/files/download/BM1382_Datasheet_v3.0.pdf ...
Pay attention to ..." For normal operation the following constrains MUST be satisfied..."
I did NOT check Your value(s) on my S3's, but seems like they are NOT the same "animals"...
Cheers
ZiG
Agree Bitmain chart is much different for the same values...and they follow the equation..I wouldn't use the above chart values only the ones from the BITMAIN specsheet