And now for something completely different to find out what CAP helps up to what SDRAM speed.
Evolving the test now with a 10 pF cap installed. Very surprising so many CAPS worked to such high speeds!
No Errors through 227 MHz with 10 pF!
Compared 2 .vs. 5 .vs. 25 ReReads and 5 seems to catch the trend.
When 240 MHz Errors start they are on passes 1-5 with FIXED and ALWAYS FIRST pass with Psuedo Random.
At 254 MHz: ReRead on the PRand Values. 25 does gives some proportion more (0.60% .vs. 0.54% ) than 5 ReReads { 9700 .vs. 1300
[*5) )
Next jump to 270 and 280 MHZ give extreme 19.6% and 20.7% so a clear out of bounds for usability - and with 10 pF both are in the same failure region.
When Errors appear at 254 MHz they are always on FIRST pass ReRead.
THE TEST: https://github.com/mjs513/SDRAM_t4/tree/main/examples/CapReadSDRAMTest
@Dogbone06 : This updated test (
new SKETCH NAME given old test name/code was usurped prematurely)
Given these possible values: uint32_t speedRange[] = {133, 166, 196, 206, 216, 227, 240, 254, 270, 288};
The values 196 to 227 work with NO ERRORS and 10 pF CAP in place.
To confirm BEST CAP is ideal need to be added and run across ALL MHz values.
Test uses EEPROM to store index into speedRange[] and it runs for ReRead number of times in each of the 57 tests, then RESTARTS to next value.
Test just altered to run from 227 to 288 MHz to allow test of "
A CAP of user choice" across those speeds. About 11 minutes.
Hold pin16 PULLED to 3.3V after UPLOAD to start test from beginning::
SDRAM testing PAUSED w/Restart Pin 16 HIGH : Will run when pin 16 goes low. It is PULLDOWN, so jumper to 3.3V and upload, then pull jumper and testing will begin.
Pin13 LED will 200 ms Blink when testing is complete, the EEPROM value is reset and it stops to blink, but pin #16 enforces a clean start at 227 MHZ with code above from github.
10 pF RESULTS:
> showing run speed and number of Errors in 57 tests at 227-288 MHz.
> 'a' or 'A' shows first read on that test showed Errors, ERRORS on subsequent passes 2-25 will be b-y or B-Y when used (z/Z over 25)
> below at 240 MHz shows Errors started on Fixed and PRand on 2nd pass with 'b' and 'B', so first pass worked.
Code:
SDRAM testing PAUSED w/Restart Pin 16 HIGH : Will run when pin 16 goes low.
Start 57 test with 5 reads 227.37 MHz ... wait ...:#############............................................
Test result: 0 read errors (0.0000%)
Extra info: ran for 125.52 seconds at 227 MHz
Start 57 test with 5 reads 240.00 MHz ... wait ...:#####a####b##AAAAAAAAAAAAAAABAAAAAAAAAAAAAAAAAAAAAAAAAAAA
Test result: 1383 read errors (0.0000%)
Extra info: ran for 123.26 seconds at 240 MHz
Start 57 test with 5 reads 254.12 MHz ... wait ...:a#a##a####a##AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
Test result: 49794255 read errors (0.5207%)
Extra info: ran for 121.23 seconds at 254 MHz
Start 57 test with 5 reads 270.00 MHz ... wait ...:aaaaaaaaaaaa#AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
Test result: 1873163227 read errors (19.5876%)
Extra info: ran for 119.36 seconds at 270 MHz
Start 57 test with 5 reads 288.00 MHz ... wait ...:aaaaaaaaaaaaaAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
Test result: 1978384746 read errors (20.6879%)
Extra info: ran for 117.24 seconds at 288 MHz
SDRAM Stop/Restart Pass Complete
@Dogbone06 : This more appropriate test should give needed details to test against various CAP pF values to see if any make a difference for better or worse to find the 'best' one for the current library code.
NOTE: The .begin() check_fixed_pattern() using the single FIXED value test is NOT triggering even at 288 MHz, but it should be removed to allow 'WARM Restart' data recovery for those that find it to work (still not tested here).
ReRun after array/index edit still with 10 pF and 5 ReReads - shows 240 MHz count of Errors at 992 .vs. 1383 above, and diff first fail 'd':
SDRAM testing PAUSED w/Restart Pin 16 HIGH : Will run when pin 16 goes low.
Start 57 test with 5 reads 227.37 MHz ... wait ...:#############............................................
Test result: 0 read errors (0.0000%)
Extra info: ran for 125.52 seconds at 227 MHz
Start 57 test with 5 reads 240.00 MHz ... wait ...:#####d#######AAAAAAAAAAAABAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
Test result: 992 read errors (0.0000%)
Extra info: ran for 123.88 seconds at 240 MHz
Start 57 test with 5 reads 254.12 MHz ... wait ...:a####a####a##AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
Test result: 61748626 read errors (0.6457%)
Extra info: ran for 121.23 seconds at 254 MHz
Start 57 test with 5 reads 270.00 MHz ... wait ...:aaaaaaaaaaaa#AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
Test result: 1869249094 read errors (19.5467%)
Extra info: ran for 119.36 seconds at 270 MHz
Start 57 test with 5 reads 288.00 MHz ... wait ...:aaaaaaaaaaaaaAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
Test result: 1978586473 read errors (20.6900%)
Extra info: ran for 117.22 seconds at 288 MHz
SDRAM Stop/Restart Pass Complete
