6502.org Forum  Projects  Code  Documents  Tools  Forum
It is currently Wed May 15, 2024 1:08 am

All times are UTC




Post new topic Reply to topic  [ 271 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8 ... 19  Next
Author Message
 Post subject:
PostPosted: Wed Nov 30, 2011 6:29 am 
Offline

Joined: Sun Nov 08, 2009 1:56 am
Posts: 388
Location: Minnesota
Here's another crack at that plot routine. There are some changes that would need to be made to the set attribute routine to work with this version, but I hope those would be more-or-less self-evident.

A couple of the finer points:

- the check for 'clearing' is made here instead of the attribute routine and simply plots the first character pattern, which I assume is always a space char (all blank anyway)
- I got took out indirect addressing because a sixteen bit index register is already large enough to cover the address range of the character ROM (is it a ROM? - the character patterns in memory, anyway). Attribute setup just stores the low 16 bits of base address of the charset in use. Absolute addressing takes care of the high 16 bits
- after a char row pattern is shifted to the high eight bits I set a bit to its 'immediate right' that acts as a sentinel (value also set by the attribute routine). No index register or memory location is needed to track when all the bits of the pattern are done; instead the Z-flag does that

Code:
PLTCHR PHA             ; Plot Character Subroutine variable (1-7) H and V size
       TYA             ; save all reg's
       PHA
       TXA
       PHA       

       LDA #$2C        ; Prepare TFT to Plot
       STA DCOM

CACALC LDA PE      ; clearing all pixels ?
       BEQ pltspc   ; b:yes, plot space character (first in every charset)
       SEC          
       LDA CHR
       CMP #127+1   ; an ascii char ?
       BCS plotspc   ; b: no (could be $0080 - $FFFF, don't know)
       SBC #$20-1   ; first 32 ascii characters "undefined"
       BCS nnull    ; original char $20 - $7F now $00 - $5F (96 possible chars)
pltspc LDA #$00        ; make undefined char's a defined zero (space character)
nnull  ASL A           ; * 2
       ASL A           ; * 4
       ASL A           ; * 8
       ADC CHRBASE     ; add pointer to base either CA00 (C64)  or CD00(3x5) (carry clear)
       TAY

       LDA PATHGT   ; pattern rows either 8(c64) or 5(3x5)
       STA PATROW

loop7  LDA PLTHGT   ; plot row repeat count (1-7)
       STA PIXROW
loop4  LDA $FFFF0000,y   ; $FFFFCA00(c64) or $FFFFCD00(3x5)
       ASL A             ;
       ASL A           ;
       ASL A           ;
       ASL A           ;
       ASL A           ;
       ASL A           ;
       ASL A           ;
       ASL A           ; shift out upper 8 bits, don't care for 8-bit byte character font
       ORA SENTINEL   ; $0080 (C64) or $0800 (3x5)

       ASL A           ; get a pixel
loop5  PHA      ; save remaining pixel row data
       LDX PLTWID   ; plot column repeat count (1-7) (same as PLTHGT?)
       BCC xwnp        ; b: clear ('blank')

xwp    LDA PIXCOL1   
       STA DDAT        ; plot RED pixel TFT data
       LDA PIXCOL2   
       STA DDAT        ; plot GREEN pixel TFT data
       LDA PIXCOL3   
       STA DDAT        ; plot BLUE pixel TFT data
       DEX
       BNE xwp
       BEQ nxtpix   ; b: forced
       
xwnp   LDA SCRCOL1   
       STA DDAT        ; plot RED "blank" pixel TFT data
       LDA SCRCOL2   
       STA DDAT        ; plot GREEN "blank" pixel TFT data
       LDA SCRCOL3
       STA DDAT        ; plot BLUE "blank" pixel TFT data
       DEX
       BNE xwnp

nxtpix PLA      ; get pixel row data back
       ASL      ; another pixel to plot ?
       BNE loop5   ; b: yes (sentinel still hasn't shifted out)
       
       DEC PIXROW   ; repeat this row ?
       BNE loop4   ; b: yes

       INY
       DEC PATROW   ; another pattern row to plot ?
       BNE loop7   ; b: yes
     
       PLA       
       TAX
       PLA       
       TAY
       PLA         ;reload all reg's
       RTS


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Wed Nov 30, 2011 12:45 pm 
Offline

Joined: Mon Mar 02, 2009 7:27 pm
Posts: 3258
Location: NC, USA
Excellent I see and understand a few of your changes. I sorta knew I didn't need indirect indexed as I wasn't changing the high address pointer, but I couldn't think of how to do it at that moment. Thanks for pointing that and some other things out. I'll try it out this morning. Your method uses 3 more variables? PIXROW, PATROW, and SENTINEL. I can use my XWIDTH and YWIDTH, for your PATHGT and PLTHGT/PLTWID. The variable width and height are the same.
I'll try it out, but will at least take your pointers on the 1st half of the code. Thanks TT!


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Wed Nov 30, 2011 3:16 pm 
Offline

Joined: Mon Mar 02, 2009 7:27 pm
Posts: 3258
Location: NC, USA
I like how you used the value in SENTINEL as a counter. I'm going to use your entire modification, except including PE. I've quickly tested both fonts, sizes 1,2&4 and it works nicely! Thanks again!

The only problem I see is how you implement the PE into the PLTCHR to clear a character using a space. For example, if PXLCOL1,2,3=yellow, and SCRCOL1,2,3=blue, then when attempting to clear a character, instead of clearing it, there would be a yellow space on a blue background.

So I have removed that, and instead let the ATTBUTE routine manipulate TMPCOL1,2,3 to equal either PXLCOL1,2,3 or SCRCOL1,2,3. I'll post the 16bit version of the ATTBUTE routine when I get it done, hopefully soon.

BTW, I do see why you did this, I think. In order to not call ATTBUTE each and every time just to set a PE (plot or clear)...

What I need to do is incorporate some or most of ATTBUTE into the PLTCHR routine, using a 16bit register and test the individual 16bit settings everytime PLTCHR is called.

Code:
PLTCHR PHA             ; Plot Character Subroutine variable (1-7) H and V size
       TYA             ; save all reg's
       PHA
       TXA
       PHA       

       LDA #$2C        ; Prepare TFT to Plot
       STA DCOM

CACALC SEC           
       LDA CHR
       CMP #127+1      ; an ascii char ?
       BCS pltspc      ; b: no (could be $0080 - $FFFF, don't know)
       SBC #$20-1      ; first 32 ascii characters "undefined"
       BCS nnull       ; original char $20 - $7F now $00 - $5F (96 possible chars)
pltspc LDA #$00        ; make undefined char's a defined zero (space character)
nnull  ASL A           ; * 2
       ASL A           ; * 4
       ASL A           ; * 8
       ADC CHRBASE     ; add pointer to base either CA00 (C64)  or CD00(3x5) (carry clear)
       TAY

       LDA CHRYLEN     ; pattern rows either 8(c64) or 5(3x5)
       STA PATROW

loop7  LDA XWIDTH      ; plot row repeat count (1-7)
       STA PIXROW
loop4  LDA $FFFF0000,Y ; $FFFFCA00(c64) or $FFFFCD00(3x5)
       ASL A           ;
       ASL A           ;
       ASL A           ;
       ASL A           ;
       ASL A           ;
       ASL A           ;
       ASL A           ;
       ASL A           ; shift out upper 8 bits, don't care for 8-bit byte character font
       ORA SENTINEL    ; $0080 (C64) or $0800 (3x5)

       ASL A           ; get a pixel
loop5  PHA             ; save remaining pixel row data
       LDX YWIDTH      ; plot column repeat count (1-7) (same as PLTHGT?)
       BCC xwnp        ; b: clear ('blank')

xwp    LDA TMPCOL1   
       STA DDAT        ; plot RED pixel TFT data
       LDA TMPCOL2   
       STA DDAT        ; plot GREEN pixel TFT data
       LDA TMPCOL3   
       STA DDAT        ; plot BLUE pixel TFT data
       DEX
       BNE xwp
       BEQ nxtpix      ; b: forced
       
xwnp   LDA SCRCOL1   
       STA DDAT        ; plot RED "blank" pixel TFT data
       LDA SCRCOL2   
       STA DDAT        ; plot GREEN "blank" pixel TFT data
       LDA SCRCOL3
       STA DDAT        ; plot BLUE "blank" pixel TFT data
       DEX
       BNE xwnp

nxtpix PLA             ; get pixel row data back
       ASL A           ; another pixel to plot ?
       BNE loop5       ; b: yes (sentinel still hasn't shifted out)
       
       DEC PIXROW      ; repeat this row ?
       BNE loop4       ; b: yes

       INY
       DEC PATROW      ; another pattern row to plot ?
       BNE loop7       ; b: yes
     
       PLA       
       TAX
       PLA       
       TAY
       PLA             ;reload all reg's
       RTS


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Thu Dec 01, 2011 2:50 am 
Offline

Joined: Sun Nov 08, 2009 1:56 am
Posts: 388
Location: Minnesota
Glad you like it!

I confess I don't quite understand this:

Quote:
For example, if PXLCOL1,2,3=yellow, and SCRCOL1,2,3=blue, then when attempting to clear a character, instead of clearing it, there would be a yellow space on a blue background.



I think of a monochrome 'space' char as 'all pixels backround color' and the pattern for describing that as all zero bits in every row. If that's the case on your system, then I would expect that every shift of a bit from that pattern into the carry flag would clear it. If so, the code should branch to 'xwnp' every time through the loop, always skipping 'xwp'.

But you know your hardware better than I do, and it doesn't really make a noticeable difference in the run time of the PLTCHR routine either way.

Mmm, and this part I wrote:

Code:
CACALC SEC           
       LDA CHR
       CMP #127+1      ; an ascii char ?


Well that's just silly because there's no need to set the carry before doing a CMP, and I shouldn't have put that SEC there. It should be just:

Code:
CACALC LDA CHR
       CMP #127+1
 


You're right that I used a few more memory locations than you did. I did that to save space and time in the routine itself. However this does have the drawback of making it less amenable to being 're-entrant', and therefore less suitable to a multi-tasking environment.

Which was not really a problem with a 64K environment, where most programs could more-or-less count on having the whole machine to themselves. I do wonder a bit if that's a reasonable expectation in a 4GB environment.


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Tue Dec 06, 2011 12:26 am 
Offline

Joined: Mon Mar 02, 2009 7:27 pm
Posts: 3258
Location: NC, USA
teamtempest wrote:
...I think of a monochrome 'space' char as 'all pixels backround color' and the pattern for describing that as all zero bits in every row...

Ah, but this is not a monochrome display as I send three bytes (only 6 bits are used for each) to define 1 color of 65K. Many thanks for your help refining my 8bit plot routine for the 65Org16!

This completes my character out routine, with the possibility of some simple graphics like line, box, etc. for a "coolness" effect.

Tomorrow, I work on implementing the PS2 core once again, should be no problem. If it works, as in my wirewrapped 6502SoC project, then I will have my character in routine and I will be focusing on the software side of things, in fact I already have been trying to decide on 1 of 3 original 6502 Monitors originally made for the Commodore 64...


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Tue Dec 13, 2011 4:56 pm 
Offline

Joined: Mon Mar 02, 2009 7:27 pm
Posts: 3258
Location: NC, USA
Got the keyboard section working!

Also, I received the 800x480 TFT back yesterday. It was fully tested by the fine folks at NewHavenDisplay and was found to have no problems. Most likely my pullup resistor for the RESn was too high @4.7K, we shall see. So I will try to interface to it again since it has the touchscreen I would like to take advantage of, not to mention the resolution is better.


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Tue Dec 13, 2011 7:33 pm 
Offline

Joined: Mon Mar 02, 2009 7:27 pm
Posts: 3258
Location: NC, USA
ElEctric_EyE wrote:
... Most likely my pullup resistor for the RESn was too high @4.7K, we shall see. So I will try to interface to it again since it has the touchscreen I would like to take advantage of, not to mention the resolution is better.

This appears to have been the problem. My RESn pullup was not strong enough for the 800x480TFT, although it was good enough for the 640x480TFT... As suggested by a DS1818 datasheet footnote, "3. A 1 k external pull-up resistor may be required in some applications for proper operation of the
microprocessor reset control circuit.
", I've used a 1.2K pullup and the display is working consistently @40MHz E signal.


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Thu Dec 15, 2011 5:45 pm 
Offline
User avatar

Joined: Thu Dec 11, 2008 1:28 pm
Posts: 10800
Location: England
Hi EEye, to summarize your dev board progress: you have got 65Org16 code running, but only from block RAM, not yet from external RAM. Is that right?

Cheers
Ed


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Thu Dec 15, 2011 7:18 pm 
Offline

Joined: Mon Mar 02, 2009 7:27 pm
Posts: 3258
Location: NC, USA
Correct... I wish I could make quicker progress. But the progress I have made I can say it is solid.


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Thu Dec 15, 2011 10:37 pm 
Offline
User avatar

Joined: Thu Dec 11, 2008 1:28 pm
Posts: 10800
Location: England
No, no criticism intended - my own progress isn't fast. I was just wondering what we've managed to do so far with the core. It seems we've both had it working on silicon - spartan 3 and 6 - with various peripherals, but not yet with external memories. (I've previously had the extra difficulty of only having an 8-bit wide memory available, but now I've got 16-bit wide I might get somewhere with that. I do need to figure out the DDR interface.)

Cheers
Ed


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Fri Dec 16, 2011 2:03 am 
Offline

Joined: Mon Mar 02, 2009 7:27 pm
Posts: 3258
Location: NC, USA
BigEd wrote:
No, no criticism intended - my own progress isn't fast. I was just wondering what we've managed to do so far with the core. It seems we've both had it working on silicon - spartan 3 and 6 - with various peripherals, but not yet with external memories. (I've previously had the extra difficulty of only having an 8-bit wide memory available, but now I've got 16-bit wide I might get somewhere with that. I do need to figure out the DDR interface.)

Cheers
Ed

Good luck with the DDR core. That's some unventured territory!

In a way, the TFT I am working with can be considered external memory although it is being written to in a successive manner, i.e. one sets a beginning and ending address then sends the 3 consecutive 8bit data to define the color of each pixel. O2, A0, CSn, RWn, and D0-D7 are all used just as if a typical SRAM would be used. O2 is going through a DDR FF. All other signals through a regular FF.

This devboard is setup for an SDRAM though, not DDR. I was hoping to take advantage of Arlet's SDRAM core...

I would like to have some sort of "OS" in the form of an assembler/dissembler present first though, at least 8bit to start with. That way I don't have to keep reprogramming the FPGA PROM.


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Fri Dec 16, 2011 10:17 am 
Offline
User avatar

Joined: Thu Dec 11, 2008 1:28 pm
Posts: 10800
Location: England
Yes, getting a monitor program working is very worthwhile. Whatever progress we make on hardware, we still have lots to do on software as well.

(Edit: I think my DDR problem is easier than it would be for you, because my spartan6 is a BGA type which allows use of a memory controller block. I'm hoping I can borrow chunks of tactics from the supplied examples.)

Please could you put your dev board parts list/BOM in a post here? (Preferably text, not screen grab, so it's searchable. Ideally the head post of this thread or a new thread, so we don't need to look through 17 pages!)

Cheers
Ed

Edit: good point about the LCD peripheral. It means we've had success with on-FPGA and off-FPGA memory-mapped peripherals.


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Fri Dec 16, 2011 9:37 pm 
Offline

Joined: Mon Mar 02, 2009 7:27 pm
Posts: 3258
Location: NC, USA
BigEd wrote:
...Please could you put your dev board parts list/BOM in a post here? (Preferably text, not screen grab, so it's searchable. Ideally the head post of this thread...

I need to update it, and I'll post it at the header next few days...


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Fri Dec 16, 2011 9:44 pm 
Offline
User avatar

Joined: Thu Dec 11, 2008 1:28 pm
Posts: 10800
Location: England
That would be great - thanks!


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Sun Dec 18, 2011 12:44 am 
Offline

Joined: Mon Mar 02, 2009 7:27 pm
Posts: 3258
Location: NC, USA
No problem. I added it, but it is incomplete... Most of the data was from V1.0. I did try to update all the IC's.
Will update the formatting and listing as time permits.


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 271 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8 ... 19  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 6 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to: