Set variables

[Duggle]

I'm using NG V1.025 which adds support for set-able variables.

The idea is 3 encoders which set the midibox's  Channel, instrument Part, and Port.

The default.ngr contains the following:

if ^section == 16

set ^chn ENC:67
set ^ins ENC:68
set ^dev ENC:69
    
exit
endif

When uploaded to the core I get the following:

[30825.918] [MBNG_FILE_R:95] ERROR: invalid id 'set ^chn'!
[30825.919] [MBNG_FILE_R:97] ERROR: invalid id 'set ^ins'!
[30825.920] [MBNG_FILE_R:99] ERROR: invalid id 'set ^dev'!

I'm not sure what's wrong.

Thanks for any help.

[TK.]

This works at my side, could you please check with the latest version?

http://www.ucapps.de/mios32/midibox_ng_v1_036_pre2.zip

Best Regards, Thorsten.

[Duggle]

Hi TK,

I should have mentioned the midibox is a LPC17 core.  (I do use F1 and am playing with F4)

The binary for LPC17 core not in the zip?

Many thanks.

[Duggle]

Just flashed the latest from uCApps V1.035.

It fixed the error with the variables (which is awsome) but it has brought back this issue for me (non-awesome):http://midibox.org/forums/topic/20232-bootloader-changelog

I gather flashing the the newer firmware has replaced the bootloader with a newer one. (The change causes display CLCD #3 to #24 not responding). If I reflash the earlier one which worked V1.025 the display problem is still there. I really need to make my midibox HW compatible with the latest booloader.  

TK, you kindly did the file diff for me, but there is lot of code to parse to understand what needs to be changed in hardware: Do you happen to remember or have hints for what changed with the way multiple CLCDs need to be wired? 

[TK.]

No changes are required for the hardware, but there are software optimizations which might require a better hardware setup, such as buffer ICs for the LCD control signals which are shared between multiple displays.

Major optimizations in universal/app_lcd.c:

• APP_LCD_ExtPort_SerDataShift: add more "stretch" operations to slow down the serial clock
• APP_LCD_ExtPort_UpdateSRs: strobes J28.WS, you could duplicate APP_LCD_ExtPort_PinSet(2, 0) to stretch the pulse
• APP_LCD_Data: uses APP_LCD_PollUnbusy(2500), was APP_LCD_PollUnbusy(10000) before - maybe a timeout happens with this low cycle number?

If software modifications are not successful, please give more details about your hardware setup. This could bring up ideas for improvements.

Best Regards, Thorsten.

[Duggle]

I've made each of the 3 changes you suggested to app_lcd.c separately and together with no difference.

V1.035 (from uCapps) also does not display the RGB matrix with correct colours in half of the array.

V1.036 (which I compiled) also nearly all RGB LEDs are off or incorrectly illuminated.

V1.025 (which I have used a lot lately) displays all 64 RGB LEDs correctly.

Would you like a complete wiring (specification or diagram)? It will take some effort here, but would be good to have anyway. I didn't do a diagram at the time I built it 3.5years ago (it all seemed so "obvious" at the the time :-).  

 

[Duggle]

So I just reflashed Bootloader V1.012 after the flash procedure MIOS Studio displayed LCD #3 to 24 as not responding. I repowered the midibox and the bootloader displayed its test text on all LCDs correctly.

I then flashed  NG v1.025 and the midibox is working all correctly again.

So it seems later changes are impacting the RGB LED matrix as well.

Let me know what specific information I can provide to help work through this.

Many thanks.

[TK.]

We continue here: