Home
JAQForum Ver 20.06
Log In or Join  
Active Topics
Local Time 17:24 29 Mar 2024 Privacy Policy
Jump to

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 : CMM2: V5.06.00b7 - 3D engine

     Page 6 of 6    
Author Message
matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 8516
Posted: 07:59pm 04 Dec 2020
Copy link to clipboard 
Print this post

Please try flashing this version - it forces a complete reset of options after flashing

CMM2V1.5.zip
Edited 2020-12-05 05:59 by matherp
 
MauroXavier
Guru

Joined: 06/03/2016
Location: Brazil
Posts: 303
Posted: 08:05pm 04 Dec 2020
Copy link to clipboard 
Print this post

When I use OPTION RESET appears in the first moment that is all right, but the cursor is corrupted yet... But when I use OPTION RAM, if I turn off and turn on the CMM2 it crashes completely, freezing at boot.
 
MauroXavier
Guru

Joined: 06/03/2016
Location: Brazil
Posts: 303
Posted: 08:18pm 04 Dec 2020
Copy link to clipboard 
Print this post

Unfortunately, it didn´t work. The things going the same, with the exception that now it keeps asking me the date and hour when I use OPTION RAM or OPTION CONSOLE SAVE.

But as it not saving correctly, it not crashes on the boot (but the cursor keeps corrupted).
 
TassyJim

Guru

Joined: 07/08/2011
Location: Australia
Posts: 5867
Posted: 08:35pm 04 Dec 2020
Copy link to clipboard 
Print this post

  MauroXavier said  Unfortunately, it didn't work. The things going the same, with the exception that now it keeps asking me the date and hour when I use OPTION RAM or OPTION CONSOLE SAVE.

But as it not saving correctly, it not crashes on the boot (but the cursor keeps corrupted).


This looks similar to the problem that toml_12953 has with any of the 5.06 betas
Everything seems to work OK for me on both my V and Y chip boards so can't offer any suggestions.

Jim
VK7JH
MMedit   MMBasic Help
 
matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 8516
Posted: 08:43pm 04 Dec 2020
Copy link to clipboard 
Print this post

The issue can only be that the RAM in the RTC isn't saving correctly. Check and replace the battery. Make sure NO links on the Waveshare
 
toml_12953
Guru

Joined: 13/02/2015
Location: United States
Posts: 323
Posted: 09:00pm 04 Dec 2020
Copy link to clipboard 
Print this post

  matherp said  The issue can only be that the RAM in the RTC isn't saving correctly. Check and replace the battery. Make sure NO links on the Waveshare


In my case, I did replace the battery and I don't have a Waveshare. I still get corruption whenever I change options CONSOLE, FLASH or RAM. There may be other OPTION problems but I don't change those so I didn't notice them.

I did notice that when the options look corrupted, if I do MODE 0 the screen clears and I get an error message about valid modes only 1-13 but then OPTION LIST shows no corruption until I power off and back on.
Edited 2020-12-05 07:02 by toml_12953
 
toml_12953
Guru

Joined: 13/02/2015
Location: United States
Posts: 323
Posted: 10:10pm 04 Dec 2020
Copy link to clipboard 
Print this post

  matherp said  The issue can only be that the RAM in the RTC isn't saving correctly. Check and replace the battery. Make sure NO links on the Waveshare


If that's the case, though, why do the 5.05 versions have no problem at all when changing options? Don't they use the same RAM that the betas do?
 
vegipete

Guru

Joined: 29/01/2013
Location: Canada
Posts: 1081
Posted: 01:34am 05 Dec 2020
Copy link to clipboard 
Print this post

When someone gets the swollen cursor, could he (or she) type "mode 1,8" and see what happens?

I saw something similar a few versions back when pressing F1 to exit the editor. Peter found and fixed a memory leak. The swollen cursor looked identical.

I also saw the swollen cursor once, just after flashing b12, I think. After the flash, options had changed slightly - to UK keyboard from my normal US. (I initially couldn't figure out why I kept getting ? "(100,100) ival when I wanted ? @(100,100) ival.)
Visit Vegipete's *Mite Library for cool programs.
 
TassyJim

Guru

Joined: 07/08/2011
Location: Australia
Posts: 5867
Posted: 05:10am 05 Dec 2020
Copy link to clipboard 
Print this post

Another play with mouse control.
This one lets you control a lot but there is still plenty of other variables to get my head around.

Using beta 13 as beta 14 gave me a few disappearing panels.

 OPTION EXPLICIT
 OPTION DEFAULT FLOAT
 MODE 1,8
 DIM INTEGER viewplane = 500
 CONST CAMERA = 1
 CONST mouse_port = 2
 DIM q(4)
 DIM yaw=RAD(1),pitch=RAD(2),roll=RAD(0.5)
 DIM INTEGER nv=8, nf=6 ' cube has 8 vertices and 6 faces
 'array to hold vertices
 DIM v(2,nv-1)=(-1,1,-1,  1,1,-1, 1,-1,-1, -1,-1,-1, -1,1,1, 1,1,1, 1,-1,1,  -1,-1,1)
 MATH SCALE v(),100,v()
 ' array to hold number of vertices for each face
 DIM INTEGER fc(nf-1) =(4,4,4,4,4,4)
 DIM INTEGER cindex(9)=(RGB(RED),RGB(BLUE),RGB(GREEN),RGB(MAGENTA),RGB(YELLOW),RGB(CYAN),RGB(WHITE),RGB(BROWN),RGB(GRAY),0)
 DIM INTEGER fcol(nf-1)=(9,9,9,9,9,9)
 DIM INTEGER bcol(nf-1)=(0,1,2,3,4,5)
 'array to hold vertices for each face
 DIM INTEGER fv(MATH(sum fc())-1)=(1,5,6,2, 1,0,4,5,  0,3,7,4,  5,4,7,6, 2,6,7,3, 0,1,2,3)
 DRAW3D CREATE 1, nv, nf, CAMERA, v(), fc(), fv(),cindex(),fcol(),bcol()
 DIM INTEGER c
 DIM INTEGER camx, camy
 DIM FLOAT inc_yaw, inc_pitch, inc_roll
 PAGE WRITE 1
 CLS
 CONTROLLER MOUSE OPEN mouse_port
 'draw3d diagnose 1,0,0,1000
 GUI CURSOR ON 1,MOUSE(x), MOUSE(y)
 BOX 0,0,MM.HRES-1,MM.VRES-1
 camx = MM.HRES/2 - MOUSE(x)
 camy = MOUSE(y) - MM.VRES/2
 inc_yaw = 0
 inc_pitch = 0
 inc_roll = 0
 DO
   IF MOUSE(w) THEN                    ' wheel button > reset box position
     camx = MM.HRES/2 - MOUSE(x)
     camy = MOUSE(y) - MM.VRES/2
     inc_yaw = 0
     inc_pitch = 0
     inc_roll = 0
     viewplane = 500
     yaw=RAD(1):pitch=RAD(2):roll=RAD(0.5)
   ELSEIF MOUSE(r) THEN                ' right mouse button > camera movement
     camx = MM.HRES/2 - MOUSE(x)
     camy = MOUSE(y) - MM.VRES/2
   ELSEIF MOUSE(L) THEN                ' left mouse button > roll and zoom
     inc_roll = (MM.HRES/2 - MOUSE(x))/4000
     viewplane = MAX(viewplane + (MOUSE(y)-MM.VRES/2)/50,0)
   ELSE                                ' no button pressed > yaw and pitch
     inc_yaw = (MOUSE(x)-MM.HRES/2)/4000
     inc_pitch = (MM.VRES/2-MOUSE(y))/4000
   ENDIF
   
   GUI CURSOR MOUSE(x), MOUSE(y)
   DRAW3D CAMERA 1,viewplane,camx,camy
   MATH Q_CREATE roll,1,1,1,q()
   DRAW3D SHOW 1,0,0,1000
   MATH Q_EULER yaw,pitch,roll,q()
   DRAW3D ROTATE q(),1
   
   TEXT 10,10,"Yaw  "+STR$(yaw,3,3)+" "+CHR$(241)+STR$(inc_yaw,3,3)
   TEXT 10,30,"Pitch"+STR$(pitch,3,3)+" "+CHR$(241)+STR$(inc_pitch,3,3)
   TEXT 10,50,"Roll "+STR$(roll,3,3)+" "+CHR$(241)+STR$(inc_roll,3,3)
   TEXT 10,70,"VP   "+STR$(viewplane,7,0)
   
   INC yaw,inc_yaw
   INC pitch,inc_pitch
   INC roll,inc_roll
   PAGE COPY 1 TO 0
   PAUSE 15
 LOOP UNTIL INKEY$ <>""
 ' tidy up
 DRAW3D CLOSE ALL
 CONTROLLER MOUSE CLOSE mouse_port
 CLS
 PAGE WRITE 0
 CLS


Jim
VK7JH
MMedit   MMBasic Help
 
matherp
Guru

Joined: 11/12/2012
Location: United Kingdom
Posts: 8516
Posted: 08:21am 05 Dec 2020
Copy link to clipboard 
Print this post

Mauro and toml_12953

Let me know when you are at your computers today and we can continue working on this. toml_12953, if you PM me with your email address we can do this off the TBS.

My thoughts after sleeping on it:

FACTS
Both your computers work properly with previous versions
b14 works perfectly for most people
There are no changes in b14 that could explain the issues
It is nothing to do with Waveshare vs all-in-one or 400MHz vs 480Mhz versions of the processor

IRRELEVANT
The cursor bug in b12 was specifically limited to leaving the editor with F1 when the firmware didn't tidy up properly and has been fixed

THEREFORE
The only conclusion is that it is the flashing process that is going wrong.
Why could that be?
the b14 image is considerably bigger than say RC22
b14  programs from &H8000000 to &H80B2D9C
RC22 programs from &H8000000 to &H80AC9f4
The difference is important as b14 goes into another FLASH page starting at
&H080A0000

So the approach will be:
Cut a bunch of commands out of b14 to get it below &HA0000 in length.
If that works try and find what it is in your environments that is causing STM32CubeProgrammer to not work. If it doesn't - think again  
 
JohnS
Guru

Joined: 18/11/2011
Location: United Kingdom
Posts: 3641
Posted: 10:35am 05 Dec 2020
Copy link to clipboard 
Print this post

Or pad it (0xff) to be significantly bigger - on the assumption (!) part is not getting flashed.

John
 
toml_12953
Guru

Joined: 13/02/2015
Location: United States
Posts: 323
Posted: 01:18pm 05 Dec 2020
Copy link to clipboard 
Print this post

  matherp said  Mauro and toml_12953

Let me know when you are at your computers today and we can continue working on this. toml_12953, if you PM me with your email address we can do this off the TBS.



I would like to thank you publicly for your work on this. You took the time to solve the problem even though it only appeared on two systems (so far)

You also saved me some money. I was just about to order another CMM2 since I was becoming convinced my hardware was bad.

 
 
TassyJim

Guru

Joined: 07/08/2011
Location: Australia
Posts: 5867
Posted: 08:14pm 05 Dec 2020
Copy link to clipboard 
Print this post

@toml
I am glad that Peter has squashed the bug for you.
It is very difficult to find when you can't reproduce it yourself so I was pleased when Mauro reported something similar.

Why only a few systems had the problem is a mystery but it was probably lurking in the background ready to strike us all.

I have fixed the problem I was having with disappearing faces in my cube program a few posts above.
Starting with beta14, I had to change the order of the vertices for two sides so the solid was on the outside.

Line 19 is now
  Quote    'array to hold vertices for each face
 DIM INTEGER fv(MATH(sum fc())-1)=(1,5,6,2, 5,4,0,1,  0,3,7,4,  5,4,7,6, 2,6,7,3, 3,2,1,0)


It was definitely a case of thinking outside the box.

Jim
Edited 2020-12-06 06:14 by TassyJim
VK7JH
MMedit   MMBasic Help
 
toml_12953
Guru

Joined: 13/02/2015
Location: United States
Posts: 323
Posted: 08:40pm 05 Dec 2020
Copy link to clipboard 
Print this post

  TassyJim said  @toml
I am glad that Peter has squashed the bug for you.
It is very difficult to find when you can't reproduce it yourself so I was pleased when Mauro reported something similar.

Why only a few systems had the problem is a mystery but it was probably lurking in the background ready to strike us all.



I'm very grateful to Peter for taking all that time to fix a problem that only happens to a very few systems.

Now I know why the computer would start correctly a bunch of times then hang for a bunch more. RAM was powering up in a different state each time and just happened to come up with bad values some of the time.
 
TassyJim

Guru

Joined: 07/08/2011
Location: Australia
Posts: 5867
Posted: 09:07pm 05 Dec 2020
Copy link to clipboard 
Print this post

  [ said  QUOTE=TassyJim
Starting with beta14, I had to change the order of the vertices for two sides so the solid was on the outside.

Line 19 is now
  Quote    'array to hold vertices for each face
 DIM INTEGER fv(MATH(sum fc())-1)=(1,5,6,2, 5,4,0,1,  0,3,7,4,  5,4,7,6, 2,6,7,3, 3,2,1,0)


Jim

RC2 has it back the way things were.

Jim
VK7JH
MMedit   MMBasic Help
 
     Page 6 of 6    
Print this page


To reply to this topic, you need to log in.

© JAQ Software 2024