 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 : PicoMite/VGA/WEB V5.07.08 release candidates
  Page 5 of 18   |
|||||
| Author | Message | ||||
thwill Guru  Joined: 16/09/2019 Location: United KingdomPosts: 4042 |
Hi Peter, The latest push to github has a typo in the README: ... all warnings in PicoMire code fixed ... ^ I can understand how you might feel that about the code at the moment  .Best wishes, Tom Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 9122 |
The mode file codec provides no information as to what is the end. It just plays in a loop (CMM2 is the same and I think the Maximite did the same with a different codec). If anyone can tell me how to detect the end I'll implement it. (github hxcmod) The hxcmod_fillbuffer function can return a tracker_buffer_state but I don't know enough about the mod file format to interpret it Edited 2023-10-01 19:54 by matherp |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 4042 |
Ack. @Martin H seems to know something about the format, so maybe he will be able to help. Tom Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
Martin H. Guru Joined: 04/06/2022 Location: GermanyPosts: 1113 |
Information about the File Format can be found here : https://www.ocf.berkeley.edu/~eek/index.html/tiny_examples/ptmod/ap12.html here : https://www.exotica.org.uk/wiki/ProPacker_2.1#File_format or here : https://www.eblong.com/zarf/blorb/mod-spec.txt If I remember correctly (it's been about 30 years) the trackers play the MODules until a user or the MOD to stop. Maybe hxcmod has a Flag, which tells what to do at the End of the Playlist. In a MOD, there may be jumps to other positions in the playlist, so some MODs play endlessly anyway. The end can just be detected by looking at the position in the "playlist" if the end of the last pattern is reached, and there is nou "jump to another playlist position"-command in the last pattern, this should be the end of the song. (The songlenth in Pattern should be found after the last Sampleinfo-block, at Offset 950, 1 Byte (Range is 1-128)). @Tom, you might look at the lose.mod or the win.mod from the PETSCIIROBOT InGame Music, I posted last Friday. This should not be repeated automatically, because the module jumps into an empty pattern at the end and stays there. I hope this helps.. Edited 2023-10-02 03:02 by Martin H. 'no comment |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 9122 |
I've hacked hxcmod and put a parameterised return if the table position loops back to 0. Attached is VGA test firmware USE: PLAY MODFILE fname$ to play on a loop as normal USE: PLAY MODFILE fname$, interrupt to trigger an interrupt and stop playback when the table address loops. Obviously only available in a program tested and seems to work although the last bars may not be quite correct - let me know PicoMiteVGA.zip Edited 2023-10-02 04:20 by matherp |
||||
crez Senior Member  Joined: 24/10/2012 Location: AustraliaPosts: 152 |
I have just tried 'play flac' on picomite 5.07.07 and it works fine for stereo, but a mono flac file is playing at what sounds like double speed. Using Audacity to generate the file. I have looked at the file header and it looks correct for a mono file. The mono file, as you would expect, is about half the size of the stereo one. |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 9122 |
Thanks - will fix |
||||
| Martin H. Guru Joined: 04/06/2022 Location: GermanyPosts: 1113 |
something has changed with the PIXEL(x,y) Function. @5.07.07 it returns the color of the Dot on the active Layer (Framebuffer write N/F/L), even if the Range was on an Invisible layer. (F or N even if something overlaps on L). @5.07.08RC4 PIXEL(x,y) returns the Color of the Visible dot, even if a other layer is switched to aktive (Framebuffer write N/F/L) 'no comment |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 9122 |
VGA, PicoMite, What did you do first etc.? |
||||
| Martin H. Guru Joined: 04/06/2022 Location: GermanyPosts: 1113 |
PicoMiteVGA MMBasic Version 5.07.08RC4 'test MODE 2:Font 1 FRAMEBUFFER layer FRAMEBUFFER write n CLS Text 0,0,"A" FRAMEBUFFER write l Box 0,0,16,16,,RGB(green),RGB(Green) FRAMEBUFFER write n Font 7 Text 0,32,"" For y=0 To 11 For x=0 To 7 If Pixel(x,y) Then :Print "1";:Else : Print "0";:EndIf Next x Next y Option list 'Save image "5.07.08RC4.bmp" if you REM out the BOX on Framebuffer L, it works as it should.  On PicoMiteVGA MMBasic Version 5.07.07 it was correct Except that the green square on the LAYER isnt saved by "save image"  Edited 2023-10-02 23:22 by Martin H. 'no comment |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 9122 |
Wasn't this a change I made for you? It now reports the OR of N and L when the layer is active to allow SAVE IMAGE to capture the actual display |
||||
| Martin H. Guru Joined: 04/06/2022 Location: GermanyPosts: 1113 |
No Volhout asked for a posibility, saving everything from the Screen.. But that's not the point The green Square was also @5.0707 on the Screen (Layer) and the PIXEL Function still returns the correct values from the aktiv Layer (N). What it does not on 5.07.08RC4 So now we can save everything from the screen, but the pixel function only works to a limited extent. Edited 2023-10-02 23:36 by Martin H. 'no comment |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 9122 |
OK understand - will fix |
||||
TheMonkeys Regular Member  Joined: 15/12/2022 Location: AustraliaPosts: 59 |
Hi All, a little thought for the day. How (not) to use BitBang: My Library code (excerpt) Option explicit Option default integer Dim dog=10 ' in case unattended freezboot Const con=1 ' CONsole Const lgf=2 ' logFiles Const trc=4 ' TRaCe/TRaCk Const lds=8 ' LEDs Dim debug=lgf+con '~~~ledstrip' setup Const cort=0 '~~~ cortex, r=Hrs, g=Mins, b=Secs Const serv=1 '~~~ servient, r=404, g=connection#, b=ftype Const parse=2 '~~ parseReq, r=POST. g=parseing, b=paramcount Const logshed=3 '~~~ logfiles, 15 min update, r=state, g=day, b=week + schedule reminder: gn to rd as reminder reached, bl on completion Const therm=4 '~~ thermostat, r=on,g=off,b(off=thermo,on=forced) Const ledpin=21 SetPin ledpin, DOut Const striplen=5 ' led strip length - 5 LED bar Dim ledbar(striplen) Sub RGBled(ledind, rv, gv, bv) ' Solid - no LeDS means dimmed display ledbar(ledind)=Choice(debug And lds,RGB(rv, gv, bv),RGB(rv/60,gv/60,bv/60)) Bitbang ws2812 O, ledpin, striplen, ledbar(0) End Sub My main loop WEB tcp interrupt servient main: ch=Val(Field$(Time$,1,":")): cm=Val(Field$(Time$,2,":")): cs=Val(Field$(Time$,3,":")) ' pre-calculate rd=ch*5: gn=cm*2: bl=cs*2 ' Time of Day If (Timer Mod 1000)<200 Then ' Cortex Stable RGBled(cort,rd,gn,bl) ' H,M,S If cs=8 Then done=0 ' once If cs=10 Then shedAct() ' a minute Else RGBled(cort,rd/8,gn/8,bl/8), ' h,m,s EndIf If remend > 0 Then ' do we have a... ct=ch*60+cm ' mins since midnight If remend>ct Then ' reminder. Only handles oaat(One At A Time) gn=(remend-ct)/(remdur+1)*120 rd=120-gn bl=0 RGBled(logshed,rd,gn,bl) ' reminder state Else remend=0 remdur=0 logIt("reminder complete",con,3) RGBled(logshed,0,0,16) ' reminder complete EndIf EndIf usage=round((1-(MM.Info(free space)/MM.Info(disk size)))*100,1) Pause 6 updateheater RGBled(cort,0,0,0) ' Cortex twitch enhance maxmin Pause 6 If cm Mod 15=14 And cs > 50 Then ' *:14:5*,*:29:5*,*:44:5*,*:59:5* busy=1 ' seems to help If cs=51 Then fact$=getShed() If cs=52 Then logFiles(0,maxtemp,mintemp) If Time$="23:59:54" Then maxmin(1) logFiles((Day$(Now)="Sunday")+1,currenttemp,currenttemp) logIt(Date$+" Day roll-over",con+lgf,3) EndIf busy=0 ' /seems to help EndIf If dog Then WatchDog (dog*1000) Else WatchDog off GoTo main 35 minutes and 43 seconds of bit banging later, is it any wonder that it freezes? I commented out the bitbang line in the subroutine and all is good. 50708b11 was the last time that I was getting away with this. Cheers, Chris. Edited 2023-10-03 15:19 by TheMonkeys |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 9122 |
V5.07.08RC5 https://geoffg.net/Downloads/picomite/PicoMite_Beta.zip Fixes bug introduced in reading the VGA screen when the framebuffer L is enabled Fixes bug in the editor when a file is edited without an extension being specified (defaults to .bas) Fixes bug in playing mono FLAC files (untested) Changes LIBRARY DISK SAVE so only used memory is written to disk Changes LIBRARY DISK LOAD so files smaller than the full library can be loaded Together these changes allow libraries to be stored on, for example, a PicoMite, to be used on a PicoMiteVGA PLAY MODFILE fname$ [, intsub] If the optional interrupt parameter is set it will be called when the modfile has played through a single loop and playback will be terminated. If not set the modfile will play on a loop as before. Because of the way mod files work the end of playback may not be perfect with a few additional notes from the next loop played. OPTION PICO OFF/ON (default ON, permanent option, N/A WebMite) If set to OFF GP23, GP24, and GP29 are not initialised allowing them to be used freely on non-Pico boards. GP25 remains under the control of OPTION HEARTBEAT ON/OFF BLIT MEMORY address%, x, y[,col] If the word at address% has the top bit set the sprite will be treated as compressed as though BLIT COMPRESSED was used WebMite WEB TLS functionality is removed. The maximum program size is increased to 96Kb (from 80Kb) WARNING: installing this version on the WebMite will delete the A: drive and all options |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 4042 |
Thank you Peter, it looks like it also includes the fix to deal with the SD card access display artifacts during a background framebuffer copy  .Could I trouble you for a pointer to what it is (in the source) that triggers this behaviour (not specific to this case but in the general PicoMite firmware update procedure) ? Best wishes, Tom Edited 2023-10-06 22:03 by thwill Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
| Martin H. Guru Joined: 04/06/2022 Location: GermanyPosts: 1113 |
@Tom, Peter I just updated the VGA Version, no option was deleted, Pixel Function is working correct again, thank you 'no comment |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 4042 |
Yes, Peter was clear it was only the WebMite which would experience this. I'm just hoping for a hint as to what governs when this does and does not happen so I don't have to unpick this piece of the source-code without direction. Best wishes, Tom Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 9122 |
1. I want it to - change of magic number in Configuration.h 2. Change of the heap/program/variable table size in Configuration.h 3. Change to the option structure - shouldn't be an issue going forward NB: also, if new commands are added then programs stored in the library or flash slots may become corrupted - incorrect tokens. In this case I try and remember to do 1 In addition the A: drive will be automatically re-formatted if corrupt (e.g. after OPTION MODBUFF is set to a different size) and the options will be reset if corrupt Edited 2023-10-07 02:02 by matherp |
||||
| carlschneider Senior Member Joined: 04/08/2023 Location: South AfricaPosts: 158 |
Hi Peter Thanks for releasing RC5. Is it just me or have you restricted the OPTION DISPLAY column width to 100? I have some longish lines of code and have been using 157 up to RC4. Cheers Carl Retirement is tough on Hobbies without a day job |
||||
| Page 5 of 18 |
|||||
 Print this page |
