Forum Rule: Always post complete source code & details to reproduce any issue!
Page 162 of 173 FirstFirst ... 62 112 152 160 161 162 163 164 172 ... LastLast
Results 4,026 to 4,050 of 4311

Thread: Teensy 4.0 First Beta Test

  1. #4026
    Senior Member+ defragster's Avatar
    Join Date
    Feb 2015
    Posts
    9,687
    Looking at qtmr_count.ino it is the self same Teensy - so GND is probably good

    There was an issue on timing self same Teensy. I'm running these loops against FreqCount with pin jumpers on one T4:
    This gives 23 MHz - which is what other thread Oscope showed:
    Code:
    		while (1)
    		{
    			digitalWriteFast(togPin, !digitalReadFast(togPin) );
    		}
    This bumps up to 30 MHz:
    Code:
    		bool tgl = HIGH;
    		while (1)
    		{
    			digitalWriteFast(togPin, tgl );
    			tgl = !tgl;
    		}
    And this drops down to 20 or under 2 when run with lesser delays:
    Code:
    		while (1)
    		{
    			digitalWriteFast(togPin, LOW);
    			delayCycles(220);
    			digitalWriteFast(togPin, HIGH);
    			delayCycles(220);
    		}

  2. #4027
    Senior Member
    Join Date
    Jan 2013
    Posts
    124

    FastLED 3.3 w/Teensy 4.0 support now up

    Quote Originally Posted by dgarcia42 View Post
    Sorry for the delays in getting this up and running - I've put up a branch here - https://github.com/FastLED/FastLED/tree/teensy4 - with Teensy 4.0 support for FastLED (after I pushed that branch I saw a comment someone made on the FastLED reddit pointing at a preliminary port someone here had done for it - apologies for having missed that!) - right now it supports clockless chips and uses hardware SPI when it can for the clocked chips - and I'm working on getting non-DMA'd parallel output up and running, and once I have that I'll merge it back into FastlED master and spin a new release for it.
    And just pushed/released FastLED 3.3 with Teensy 4.0 support (all 3 and 4-wire chipsets that FastLED supports are supported, uses hardware SPI for the 4-wire chips, non-DMA based parallel output available for the non-SPI based chipsets, will add in support for the smart matrix, WS2812 serial libraries in the future)

  3. #4028
    Senior Member+ defragster's Avatar
    Join Date
    Feb 2015
    Posts
    9,687
    Quote Originally Posted by manitou View Post
    Re: T4 frequency count

    I received two T4's today. I soldered male headers on one and plugged into breadboard. I ran my qtmrcnt sketch with pin 8 PWM jumpered to pin 9. With PWM at 40mhz count is 37.5 mhz as expected , and lower frequency counts are correct. But I don't get good counts with PWM @50mhz or 75mhz. Running same code on T4B2R, I still can properly count 50 and 75 mhz.
    The second new T4 I soldered long pin female headers. It too won't count correctly above 37.5 mhz ???

    all tests built with 1.8.8 1.47-beta5
    I github'd a mod to that test : https://github.com/Defragster/T4_dem...er/ManT_qTimer

    This is the process when PWM Freq is changed - valid?::
    Code:
    		if ( !(iiC % 4) ) {
    			iiF += fStart/10;
    			analogWrite(togPin, 0);
    			analogWriteFrequency(togPin, iiF);  // test jumper 8 to 9, max 75mhz
    			analogWrite(togPin, 128);
    			Serial.print("\nNEW Freq =");
    			Serial.println(iiF);
    		}
    Top lines set a start freq, and define the OUT pin to toggle==togPin.

    I have a Prod T4 with #18 to #9 and old PreWhite T4B2 jumpered the same. The results are different - though the T4Prod has lots of wires to twin SPI TFT's

    This is the OLDER T4B2::
    Code:
    T:\tCode\T4\ManT_qTimer\ManT_qTimer.ino Aug 12 2019 21:04:24
       18749016
    NEW Freq =22000000
       21428559   21428572   21428571   21428572
    NEW Freq =24000000
       24999984   25000000   25000000   25000000
    NEW Freq =26000000
       24999990   25000000   25000000   25000000
    NEW Freq =28000000
       29999980   30000000   30000000   30000000
    NEW Freq =30000000
       29999988   30000000   30000000   30000000
    NEW Freq =32000000
       29999987   30000000   30000000   30000000
    NEW Freq =34000000
       37148791   37140078   37135295   37162146
    NEW Freq =36000000
       37181885   37183125   37178189   37187305
    NEW Freq =38000000
       37211370   37180324   37188620   37192127
    NEW Freq =40000000
       37201151   37194338   37173911   37165855
    NEW Freq =42000000
       37161041   37174342   37167145   37182830
    NEW Freq =44000000
       49999954   50000000   50000000   50000000
    NEW Freq =46000000
       49999978   50000000   50000000   50000000
    NEW Freq =48000000
       49999979   50000000   50000000   50000000
    NEW Freq =50000000
       49999977   50000000   50000000   50000000
    NEW Freq =52000000
       49999978   50000000   50000000   50000000
    NEW Freq =54000000
       49999976   50000000   50000000   50000000
    NEW Freq =56000000
       49999978   50000000   50000000   50000000
    NEW Freq =58000000
       49999977   50000000   50000000   50000000
    NEW Freq =60000000
       49999977   50000000   50000000   50000000
    NEW Freq =62000000
       70   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =64000000
       No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =22500000
       21428526   21428571   21428572   21428571
    NEW Freq =24500000
       24999984   25000000   25000000   25000000
    NEW Freq =26500000
       24999990   25000000   25000000   25000000
    NEW Freq =28500000
       29999981   30000000   30000000   30000000
    NEW Freq =30500000
       29999988   30000000   30000000   30000000
    NEW Freq =32500000
       29999988   30000000   30000000   30000000
    NEW Freq =34500000
       37268497   37235479   37265578   37265407
    NEW Freq =36500000
       37267140   37274322   37280331   37267296
    NEW Freq =38500000
       37285092   37238056   37250963   37247350
    NEW Freq =40500000
       37296502   37267305   37265242   37256716
    NEW Freq =42500000
       37271816   37287583   37258870   37237759
    NEW Freq =44500000
       49999953   50000000   50000000   50000000
    NEW Freq =46500000
       49999977   50000000   50000000   50000000
    NEW Freq =48500000
       49999979   50000000   50000000   50000000
    NEW Freq =50500000
       49999978   50000000   50000000   50000000
    NEW Freq =52500000
       49999977   50000000   50000000   50000000
    NEW Freq =54500000
       49999978   50000000   50000000   50000000
    NEW Freq =56500000
       49999978   50000000   50000000   50000000
    NEW Freq =58500000
       49999977   50000000   50000000   50000000
    NEW Freq =60500000
       76   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =62500000
       No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =22500000
       21428526   21428572   21428571   21428572
    NEW Freq =24500000
       24999984   25000000   25000000   25000000
    NEW Freq =26500000
       24999990   25000000   25000000   25000000
    NEW Freq =28500000
       29999980   30000000   30000000   30000000
    NEW Freq =30500000
       29999988   30000000   30000000   30000000
    NEW Freq =32500000
       29999988   30000000   30000000   30000000
    NEW Freq =34500000
       37282862   37295037   37285695   37306321
    NEW Freq =36500000
       37296828   37286150   37276488   37287870
    NEW Freq =38500000
       37307910   37276325   37300998   37294131
    NEW Freq =40500000
       37310363   37292797   37296038   37273786
    NEW Freq =42500000
       37294024   37266358   37287666   37290043
    NEW Freq =44500000
       49999950   50000000   50000000   50000000
    NEW Freq =46500000
       49999978   50000000   50000000   50000000
    NEW Freq =48500000
       49999976   50000000   50000000   50000000
    NEW Freq =50500000
       49999979   50000000   50000000   50000000
    NEW Freq =52500000
       49999977   50000000   50000000   50000000
    NEW Freq =54500000
       49999977   50000000   50000000   50000000
    NEW Freq =56500000
       49999978   50000000   50000000   50000000
    NEW Freq =58500000
       49999976   50000000   50000000   50000000
    NEW Freq =60500000
       71   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =62500000
       No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =22500000
       21428527   21428571   21428572   21428571
    NEW Freq =24500000
    And here is the newer T4Prod on the same code - similar but less fidelity?:
    Code:
    T:\tCode\T4\ManT_qTimer\ManT_qTimer.ino Aug 12 2019 21:04:24
       18745668
    NEW Freq =22000000
       21428559   21428572   21428571   21428571
    NEW Freq =24000000
       24999984   25000000   25000000   25000000
    NEW Freq =26000000
       24999990   25000000   25000000   25000000
    NEW Freq =28000000
       29999982   30000000   30000000   30000000
    NEW Freq =30000000
       29999987   30000000   30000000   30000000
    NEW Freq =32000000
       29999988   30000000   30000000   30000000
    NEW Freq =34000000
       45   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =36000000
       No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =22500000
       21428529   21428571   21428571   21428572
    NEW Freq =24500000
       24999985   25000000   25000000   25000000
    NEW Freq =26500000
       24999989   25000000   25000000   25000000
    NEW Freq =28500000
       29999980   30000000   30000000   30000000
    NEW Freq =30500000
       29999988   30000000   30000000   30000000
    NEW Freq =32500000
       29999988   30000000   30000000   30000000
    NEW Freq =34500000
       47   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =36500000
       No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =22500000
       21428528   21428572   21428571   21428571
    NEW Freq =24500000
       24999984   25000000   25000000   25000000

  4. #4029
    Senior Member
    Join Date
    Jul 2014
    Posts
    2,330
    Quote Originally Posted by PaulStoffregen View Post
    I've committed a fix for Serial1.write(0x00)

    https://github.com/PaulStoffregen/co...407b03eb2d91c1
    Any reasoning why you simply truncate the parameter to single byte?
    if API says that Serial.write only writes a single byte, the user should to the cast Serial1.write((uint8_t) 0x00); or proper declared variable.

  5. #4030
    Senior Member+ defragster's Avatar
    Join Date
    Feb 2015
    Posts
    9,687
    Quote Originally Posted by WMXZ View Post
    Any reasoning why you simply truncate the parameter to single byte?
    if API says that Serial.write only writes a single byte, the user should to the cast Serial1.write((uint8_t) 0x00); or proper declared variable.
    That looks like the Function Spec :: arduino.cc/ … /serial/write/

    Code:
    Syntax
    Serial.write(val)
    Serial.write(str)
    Serial.write(buf, len)
    Parameters
    Serial: serial port object. See the list of available serial ports for each board on the Serial main page.
    val: a value to send as a single byte.
    str: a string to send as a series of bytes.
    buf: an array to send as a series of bytes.
    len: the number of bytes to be sent from the array.
    UPDATE to p #4028 regarding FREQ - since there were two T4's involved - I added GND between them ( they are on the same Powered HUB ) - and crossed their #9 and #18 pin jumpers:
    Same code - results are similar with @manitou code base - I should switch to FreqCount in case that implements it differently.
    BTW: code on github shows a 'asm volatile("dsb");' in the count code - added/removed doesn't affect that.
    T4B2:
    Code:
    T:\tCode\T4\ManT_qTimer\ManT_qTimer.ino Aug 12 2019 21:04:24
       18548283
    NEW Freq =22000000
       21400249   21428588   21428589   21428589
    NEW Freq =24000000
       24962246   25000019   25000020   25000020
    NEW Freq =26000000
       25000009   25000020   25000019   25000019
    NEW Freq =28000000
       29947173   30000022   30000023   30000022
    NEW Freq =30000000
       30000009   30000022   30000021   30000022
    NEW Freq =32000000
       30000008   30000022   30000021   30000021
    NEW Freq =34000000
       33509806   33564621   33525288   33532800
    NEW Freq =36000000
       33503967   33535601   33546475   33543922
    NEW Freq =38000000
       33534967   33600300   33508717   33576723
    NEW Freq =40000000
       33529271   33549471   33540106   33538628
    NEW Freq =42000000
       33518965   33547986   33529751   33572524
    NEW Freq =44000000
       44670064   44737400   44740835   44794207
    NEW Freq =46000000
       44848616   44739622   44714609   44855103
    NEW Freq =48000000
       44693623   44804918   44776742   44737490
    NEW Freq =50000000
       44750509   44788554   44752883   44772425
    NEW Freq =52000000
       44870204   44720762   44715700   44798554
    NEW Freq =54000000
       44716659   44777870   44727823   44749753
    NEW Freq =56000000
       44833395   44753026   44766225   44732566
    NEW Freq =58000000
       44735253   44843989   44678385   44794801
    NEW Freq =60000000
       44766466   44770055   44754669   44732590
    NEW Freq =62000000
       504967   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =64000000
       No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =22500000
       21203165   21428584   21428585   21428584
    NEW Freq =24500000
       24962445   25000015   25000015   25000016
    NEW Freq =26500000
       25000005   25000016   25000016   25000016
    NEW Freq =28500000
    And Prod T4:
    Code:
    T:\tCode\T4\ManT_qTimer\ManT_qTimer.ino Aug 12 2019 21:04:24
       18778327
    NEW Freq =22000000
       21428554   21428555   21428554   21466297
    NEW Freq =24000000
       24999980   24999980   24999981   24999969
    NEW Freq =26000000
       24999981   24999980   24999981   25052793
    NEW Freq =28000000
       29999978   29999977   29999978   29999966
    NEW Freq =30000000
       29999978   29999978   29999979   29999966
    NEW Freq =32000000
       29999978   29999979   29999979   30049920
    NEW Freq =34000000
       34297245   34264009   34231580   34202908
    NEW Freq =36000000
       34298813   34255883   34260954   34237705
    NEW Freq =38000000
       34234538   34267935   34271865   34238147
    NEW Freq =40000000
       34280018   34216352   34315219   34253257
    NEW Freq =42000000
       34219135   34260212   34246907   34422531
    NEW Freq =44000000
       45993278   45924796   45921475   45966028
    NEW Freq =46000000
       45946379   45929427   45909103   46016454
    NEW Freq =48000000
       45871756   45996833   45934524   45919362
    NEW Freq =50000000
       45975953   45972560   45943774   45960958
    NEW Freq =52000000
       45898431   45953956   45938380   45858891
    NEW Freq =54000000
       45979354   45963744   45928652   45944390
    NEW Freq =56000000
       45971825   45952628   45972984   45941729
    NEW Freq =58000000
       45948425   45967750   45962427   45910153
    NEW Freq =60000000
       45894300   45971871   45949302   45481489
    NEW Freq =62000000
       No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =64000000
       No Ticks    0   No Ticks    0   No Ticks    0   225339
    NEW Freq =22500000
       21428558   21428559   21428558   21466098
    NEW Freq =24500000

  6. #4031
    Senior Member
    Join Date
    Jul 2014
    Posts
    2,330
    Quote Originally Posted by defragster View Post
    That looks like the Function Spec :: arduino.cc/ /serial/write/

    Code:
    Syntax
    Serial.write(val)
    Serial.write(str)
    Serial.write(buf, len)
    Parameters
    Serial: serial port object. See the list of available serial ports for each board on the Serial main page.
    val: a value to send as a single byte.
    str: a string to send as a series of bytes.
    buf: an array to send as a series of bytes.
    len: the number of bytes to be sent from the array.
    it may be Arduino definition, but IMHO at the same time the implementation stimulates bad programming style: Serial1.write(0x12345678) does not generate compile error but simply prints 0x12 (or 0x78 ?)\

  7. #4032
    Senior Member+ defragster's Avatar
    Join Date
    Feb 2015
    Posts
    9,687
    @mjs513 : Morphed the @manitou code to use FreqCount and showing similar results - on both T4's jumpered to themselves again.

    https://github.com/Defragster/T4_dem...er/ManT_fCount

    Results below with IDE 1.8.9 and TD 1.47 b6 - looks the same with TD 1.47 b5. The T4B2 keeps counting longer again when self triggered - P#4030 showed they both counted to the same point - when cross wired.

    New Prod T4::
    Code:
    T:\tCode\T4\ManT_fCount\ManT_fCount.ino Aug 13 2019 00:53:58
    
     end setup
       No Ticks    0   18745733
    NEW Freq =22000000
       18749988   21428562   21428571   21428572   21428571
    NEW Freq =24000000
       21428571   24999989   25000000   25000000   25000000
    NEW Freq =26000000
       25000000   24999990   25000000   25000000   25000000
    NEW Freq =28000000
       25000000   29999986   30000000   30000000   30000000
    NEW Freq =30000000
       30000000   29999988   30000001   29999999   30000001
    NEW Freq =32000000
       29999999   29999988   30000000   30000000   30000000
    NEW Freq =34000000
       30000000   12   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =36000000
       No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =22500000
       No Ticks    0   21428537   21428572   21428570   21428572
    NEW Freq =24500000
       21428571   24999988   25000000   25000000   25000000
    NEW Freq =26500000
       25000000   24999990   25000000   25000000   25000000
    NEW Freq =28500000
       25000000   29999987   30000001   29999999   30000000
    NEW Freq =30500000
       30000000   29999988   30000000   30000000   30000000
    NEW Freq =32500000
       30000000   29999988   30000000   30000000   30000000
    NEW Freq =34500000
       30000000   10   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =36500000
       No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =22500000
       No Ticks    0   21428538   21428571   21428572   21428571
    NEW Freq =24500000
       21428571   24999989   25000000   25000000   25000001
    Older PreMod T4B2::
    Code:
    T:\tCode\T4\ManT_fCount\ManT_fCount.ino Aug 13 2019 00:53:58
    
     end setup
       No Ticks    0   18747551
    NEW Freq =22000000
       18749989   21428561   21428571   21428572   21428571
    NEW Freq =24000000
       21428572   24999988   25000000   25000001   24999999
    NEW Freq =26000000
       25000000   24999990   25000000   25000000   25000000
    NEW Freq =28000000
       25000000   29999987   30000000   30000000   30000000
    NEW Freq =30000000
       30000000   29999988   30000000   30000000   30000000
    NEW Freq =32000000
       30000002   29999986   30000000   30000000   30000000
    NEW Freq =34000000
       30000000   37499983   37500000   37500000   37500000
    NEW Freq =36000000
       37500000   37499984   37500000   37500000   37500000
    NEW Freq =38000000
       37500000   37499984   37500000   37500000   37500000
    NEW Freq =40000000
       37500000   37499986   37499999   37500000   37500000
    NEW Freq =42000000
       37500001   37499985   37499999   37500000   37500001
    NEW Freq =44000000
       37500000   49999973   50000000   50000000   50000000
    NEW Freq =46000000
       50000000   49999979   50000000   50000000   50000000
    NEW Freq =48000000
       50000000   49999979   50000000   50000000   50000000
    NEW Freq =50000000
       50000002   49999978   50000002   49999998   50000000
    NEW Freq =52000000
       50000000   49999979   50000000   50000000   50000000
    NEW Freq =54000000
       50000002   49999977   50000000   50000000   50000000
    NEW Freq =56000000
       50000000   49999979   50000000   50000002   49999998
    NEW Freq =58000000
       50000000   49999979   50000000   50000000   50000000
    NEW Freq =60000000
       50000000   49999980   50000000   50000000   50000000
    NEW Freq =62000000
       50000000   21   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =64000000
       No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0   No Ticks    0
    NEW Freq =22500000
       No Ticks    0   21428537   21428572   21428571   21428572
    NEW Freq =24500000
       21428571   24999989   25000000   25000000   25000000
    NEW Freq =26500000
       25000000   24999991   24999999   25000000   25000000
    Quick peek cross wired New T4:
    Code:
    T:\tCode\T4\ManT_fCount\ManT_fCount.ino Aug 13 2019 01:02:44
    
     end setup
       No Ticks    0   18785733
    NEW Freq =22000000
       18806157   21428561   21428561   21428561   21428562
    NEW Freq =24000000
       21503442   24999988   24999989   24999989   24999989
    NEW Freq =26000000
       24999979   24999990   24999989   24999990   24999990
    NEW Freq =28000000
       25104803   29999988   29999988   29999988   29999989
    NEW Freq =30000000
       29999976   29999988   29999989   29999990   29999989
    NEW Freq =32000000
       29999976   29999989   29999990   29999989   29999989
    NEW Freq =34000000
       30100705   34452460   34486685   34533091   34498351
    NEW Freq =36000000
       34452918   34475332   34511083   34504875   34532071
    NEW Freq =38000000
       34491837   34473190   34511156   34531833   34466558
    NEW Freq =40000000
       34528600   34481006   34518561   34449140   34489768
    NEW Freq =42000000
       34470106   34401271   34534648   34437410   34480096
    NEW Freq =44000000
       34709325   45730111   45590688   45597351   45714789
    NEW Freq =46000000
       45706985   45561751   45578076   45643308   45688938
    NEW Freq =48000000
       45544832   45567265   45603770   45626269   45669248
    NEW Freq =50000000
       45589620   45632617   45817167   45587613   45622446
    NEW Freq =52000000
       45478730   45653192   45567353   45739189   45547494
    NEW Freq =54000000
       45521952   45566411   45649358   45789426   45547383
    And similar T4B2 under TD1.47b6::
    Code:
    T:\tCode\T4\ManT_fCount\ManT_fCount.ino Aug 13 2019 01:02:44
    
     end setup
       No Ticks    0   18456538
    NEW Freq =22000000
       18749998   21382606   21428582   21428581   21428582
    NEW Freq =24000000
       21428581   24938720   25000011   25000011   25000011
    NEW Freq =26000000
       25000011   25000000   25000011   25000010   25000010
    NEW Freq =28000000
       25000010   29914229   30000012   30000011   30000012
    NEW Freq =30000000
       30000011   29999999   30000011   30000011   30000011
    NEW Freq =32000000
       30000011   29999998   30000011   30000011   30000010
    NEW Freq =34000000
       30000011   33800286   33803690   33883209   33893663
    NEW Freq =36000000
       33843848   33787720   33880275   33927114   33824100
    NEW Freq =38000000
       33893277   33871991   33815755   33902384   33808807
    NEW Freq =40000000
       33893810   33903410   33854743   33818952   33889810
    NEW Freq =42000000
       33888990   33789926   33886778   33806839   33878802
    NEW Freq =44000000
       33850060   44908731   45015016   44912546   45114165
    NEW Freq =46000000
       44917259   44999080   44895286   45035951   45084984
    NEW Freq =48000000
       45107497   45111143   45004262   45030647   45027889
    NEW Freq =50000000
       45033932   44839045   45240846   44971732   44993880
    NEW Freq =52000000
       45062724   45025801   44964855   45153320   44943384
    NEW Freq =54000000
       44992165   44941310   44965215   45280268   44919103
    Last edited by defragster; 08-13-2019 at 08:25 AM.

  8. #4033
    Senior Member+ defragster's Avatar
    Join Date
    Feb 2015
    Posts
    9,687
    Quote Originally Posted by WMXZ View Post
    it may be Arduino definition, but IMHO at the same time the implementation stimulates bad programming style: Serial1.write(0x12345678) does not generate compile error but simply prints 0x12 (or 0x78 ?)\
    EEPROM.write(address, value) where value is also BYTE - so that is the Arduino rule followed for consistency across the device continuum … silly 8 bit world.

  9. #4034
    Senior Member+ manitou's Avatar
    Join Date
    Jan 2013
    Posts
    2,192
    Quote Originally Posted by mjs513 View Post
    @Defrag sterling and both had this problem when going from pin to pin 9
    But didn't see it with your standalone sketch. That was one of the reasons I went to the clock generator. Will retest it tomorrow as well. Oh BTW do you have gnd to gnd. Sorry had to ask.
    Actually, I was doing single T4 count test, so pin 8 to pin 9 on same T4. I thought it might have to do with breakout wiring or MCU batch number. Last number etched on T4B2R MCU was CTAA1848H, on latest T4 is CTZB1831D.

    Need short jumper.
    So, it works with T4 on breadboard with very short (less than inch) jumper between 8 and 9. I can get 50 and 75 mhz counts. Jumper i was using and worked on T4B2R was 7". I have worried about short ground-to-ground jumpers and actually was a little surprised I could get good 75mhz waveform with those longer jumpers. I don't have the EE training to know why it was working for T4B2R and why it takes tiny jumper for current T4. I think most of my testing with Adafruit generator was with the longer jumpers.

  10. #4035
    Senior Member+ mjs513's Avatar
    Join Date
    Jul 2014
    Location
    New York
    Posts
    4,101
    Quote Originally Posted by manitou View Post
    Actually, I was doing single T4 count test, so pin 8 to pin 9 on same T4. I thought it might have to do with breakout wiring or MCU batch number. Last number etched on T4B2R MCU was CTAA1848H, on latest T4 is CTZB1831D.

    Need short jumper.
    So, it works with T4 on breadboard with very short (less than inch) jumper between 8 and 9. I can get 50 and 75 mhz counts. Jumper i was using and worked on T4B2R was 7". I have worried about short ground-to-ground jumpers and actually was a little surprised I could get good 75mhz waveform with those longer jumpers. I don't have the EE training to know why it was working for T4B2R and why it takes tiny jumper for current T4. I think most of my testing with Adafruit generator was with the longer jumpers.
    Think we talked about having to have short jumpers at some point. One of the things I was just thinking about was that maybe the "drive strength" for the pad configuration on those pins have something to do with the problems we are having. The clock generator may have a "stronger" signal than what is coming from pin 8 for the PWM signal. Have to clean my desk to make some room but between what @defragster just showed that at around 30-40Mhz with pin to pin testing you get strange results. Wonder what the testing would show if we used something like a T3x to the T4. Can't remember if we did that configuration or not.

  11. #4036
    Senior Member+ KurtE's Avatar
    Join Date
    Jan 2014
    Posts
    5,418
    Quote Originally Posted by WMXZ View Post
    Any reasoning why you simply truncate the parameter to single byte?
    if API says that Serial.write only writes a single byte, the user should to the cast Serial1.write((uint8_t) 0x00); or proper declared variable.
    I think the simple answer is, that is what the did before... That is what a T3.x would do...

  12. #4037
    Senior Member PaulStoffregen's Avatar
    Join Date
    Nov 2012
    Posts
    20,556
    I've uploaded 1.47-beta7. As always, links on msg #2.

    This includes Kurt's fixes in ILI9341_t3 and Touchscreen, Daniel's new FastLED 3.3 release, and the minor fix for Serial1.write(0).

  13. #4038
    Senior Member+ KurtE's Avatar
    Join Date
    Jan 2014
    Posts
    5,418
    Hi @Paul,

    Beta 7 installed fine on my Windows machine. The 32 bit image is not installing on my RPI4 board. It complains about exec format error. (Running Raspian...)

    Edit: Also does not install on RPI3

    EDIT2: I think I had a corrupt download (i.e. downloaded the wrong one ) Downloaded the linuxarm version and appears to run

    And I was able to compile the ili9341_t3 example graphictest for T4
    Last edited by KurtE; 08-13-2019 at 04:26 PM.

  14. #4039
    Senior Member
    Join Date
    Dec 2015
    Posts
    125
    Question
    When updating from 1.47-Beta n to 1.47-Beta n+1 should I apply over the exiting Arduino or a un-altered fresh copy?

  15. #4040
    Senior Member+ mjs513's Avatar
    Join Date
    Jul 2014
    Location
    New York
    Posts
    4,101
    ArduCAM for T4

    With an assist from @KurtE (Thanks Kurt) I managed to get ArduCAM working on the T3.x's (tested 3.2 and T3.6) and the T4 as well. I tested with the ArduCAM Mini 2MP and the ArduCAM Shield V2 (w/LCD on it). Both are streaming to the desktop and the shield now displays to the LCD as well. Posted here since this is still a WIP and code needs clean up. But if you want to give it a try the library is posted here: https://github.com/mjs513/WIP/tree/master/ArduCAM_t4

  16. #4041
    Senior Member PaulStoffregen's Avatar
    Join Date
    Nov 2012
    Posts
    20,556
    Anyone see bugs that really should get fixed before calling this 1.47?

    Of course there's still features missing and libs not ported, but those will wait until 1.48, which I hope to release in a few weeks.

  17. #4042
    Senior Member+ mjs513's Avatar
    Join Date
    Jul 2014
    Location
    New York
    Posts
    4,101
    Quote Originally Posted by PaulStoffregen View Post
    Anyone see bugs that really should get fixed before calling this 1.47?

    Of course there's still features missing and libs not ported, but those will wait until 1.48, which I hope to release in a few weeks.
    Going over the list in post #6 and playing with different things on the T4 I don't think there are BUG issues that would need to be fixed to prevent it from being called 1.47.

    Like your last line said there are still things to be worked on but nothing to prevent 1.47 baseline from being created.

  18. #4043
    Senior Member+ KurtE's Avatar
    Join Date
    Jan 2014
    Posts
    5,418
    I agree with @mjs513, I think we are probably pretty good. Obviously there are always things that would be great to have and probably something we did not fully test. But as you suggested, I am guessing we will have another release in the near future, with probably additional USB support.

    So personally think it would be good to get an official release out, for all of those people who are now receiving the nice new Shiny things

  19. #4044
    Senior Member+ MichaelMeissner's Avatar
    Join Date
    Nov 2012
    Location
    Ayer Massachussetts
    Posts
    3,252
    And in the 1.47 announcement, it might help to have a list of things that are known not to work on T4.

  20. #4045
    Senior Member+ mjs513's Avatar
    Join Date
    Jul 2014
    Location
    New York
    Posts
    4,101
    @defragster - @el_supremo - @Frank B

    I just tried to use imxrt-size but on compile and load it did not print anything. I added this to the platforms.txt file:
    Code:
    recipe.hooks.postbuild.4.pattern.windows=cmd /c "{runtime.hardware.path}\..\tools\arm\bin\arm-none-eabi-gcc-nm -n {build.path}\{build.project_name}.elf | {runtime.hardware.path}\..\tools\imxrt-size"
    in the "## Post Build - inform Teensy Loader of new file" section.

    I also changed the teensy id to 0x24 in imxrt-size.c and recompiled using gcc. However, it still does not print anything? Any thoughts on what I am doing wrong.

  21. #4046
    Senior Member+ defragster's Avatar
    Join Date
    Feb 2015
    Posts
    9,687
    Just got to install the TD 1.47 B7 on yesterday's new unzipped IDE 1.8.9 that got B6.

    Install looks good - a couple of sketches upload as expected - including ILI9341_t3.

    Also note - the NEW install 1.47 did indeed remember the active board was the T4 - like yesterday's b6 seemed to - but 2nd time is proof.

    Even Example BLINK works! Also BlinkWithoutDelay - has about the same memory footprint as this from Blink.ino::
    Sketch uses 12128 bytes (0%) of program storage space. Maximum is 2031616 bytes.
    Global variables use 18528 bytes (1%) of dynamic memory, leaving 1030048 bytes for local variables. Maximum is 1048576 bytes.
    Indeed - if Release not already DONE - seems like a safe starter for all those New Teensy 4.0's looking to get warmed up.

    @mjs513 - imxrt-size was a Frank B special - I saw it work before but have not looked recently. I'll need to get my Hookup back to SubLime w/TyComm working - after suffering the IDE past weeks.

  22. #4047
    Senior Member
    Join Date
    Nov 2012
    Posts
    1,154
    @mjs513: Post your imxrt-size.c and I'll give it a try.

    Pete
    P.S. I've just realized that it doesn't appear to have done anything since I updated to b7. I'll look at it.
    P.P.S. I had forgotten to copy imxrt-size.exe into the new distribution at arduino-1.8.9_1.47_b7\hardware\tools

  23. #4048
    Senior Member+ mjs513's Avatar
    Join Date
    Jul 2014
    Location
    New York
    Posts
    4,101
    Quote Originally Posted by el_supremo View Post
    @mjs513: Post your imxrt-size.c and I'll give it a try.

    Pete
    P.S. I've just realized that it doesn't appear to have done anything since I updated to b7. I'll look at it.
    P.P.S. I had forgotten to copy imxrt-size.exe into the new distribution at arduino-1.8.9_1.47_b7\hardware\tools


    Anyway here is the zip of the windows exe and the .c file (hope its the right one I found).
    imxrt-size.zip

    Thanks for checking.

    Mike

  24. #4049
    Senior Member+ defragster's Avatar
    Join Date
    Feb 2015
    Posts
    9,687
    Quote Originally Posted by mjs513 View Post


    Anyway here is the zip of the windows exe and the .c file (hope its the right one I found).
    imxrt-size.zip

    Thanks for checking.

    Mike
    I pulled the ZIP/EXE down - I see it is REM'd out of my SublimeText TSET.cmd system - will go through that again with the new edits to TD files as needed and see what I see.
    With no Speed .menu item - that edit will be the same … will have to see if the board name matches the build and … sometime in next hours.

  25. #4050
    Senior Member+ mjs513's Avatar
    Join Date
    Jul 2014
    Location
    New York
    Posts
    4,101
    Quote Originally Posted by defragster View Post
    I pulled the ZIP/EXE down - I see it is REM'd out of my SublimeText TSET.cmd system - will go through that again with the new edits to TD files as needed and see what I see.
    With no Speed .menu item - that edit will be the same … will have to see if the board name matches the build and … sometime in next hours.
    @defragster - thanks - wanted to use it to check out uncanny eyes memory sizes on each of the test cases I have to see if I can track down the issue

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •