Jump to content

jum

Members
  • Content Count

    168
  • Joined

  • Last visited

Community Reputation

46 Excellent

1 Follower

About jum

  • Rank
    Chopper Commander

Contact / Social Media

Profile Information

  • Location
    Cape Town

Recent Profile Visitors

9,523 profile views
  1. See attached picture (blue wire). Looks like resistor R41 (on my 2-port 5200, I think 4-port is the same). Like the other guys say, may be a good idea to add a resistor to the audio out (although mine sounds fine and the level seems OK).
  2. Should work with the 5200. Let us know how it goes.
  3. Here's a quick hand-drawn schematic. Not 100% sure about the polarity of the 0.1uF cap, or the connections to the 2N2222 / 2N3904 transistor (you should attach the transistor as per the schematic in your post). You could also try the schematic in your post. Will be interesting to see which gives a better picture. Disclaimer: I take no responsibility for any dead 5200's resulting from the use of this mod.
  4. Wow, this is super impressive! Screenshot from Jum52 emulator:
  5. Yeah, pretty unusual fault. Lots of pondering over schematics and datasheets, and probing with an oscilloscope, was involved... Have now discovered the GTIA may be faulty - the player to player collisions don't seem to work! (Also unusual).
  6. Final install of the composite mod, with screenshots from a small LCD monitor:
  7. Finally some closure to this forum post Fixed my 5200 after 14 years - turns out it was a faulty inductor L3, which is on the SELECT line for the 74LS245s that multiple the address going into the RAM.
  8. Here are the pictures, although in person the images on the screen is not so bright and the ghosting is much less noticeable. Hopefully putting the mod onto a proto board and a bit of tweaking will give a slightly better image. Note: - RF modulator removed completely - Resistor R33 leg (audio) lifted off the board - Video signal taken from left side of L12 (green clip) - +5V taken from top of L13 (red clip) - Ground taken from top side of C40 - Transistor used in mod circuit is KSP2222N (2N2222?) - Resistor used in mod circuit is 100 Ohms These are just the first points I saw on the schematics that looked like they have the right signals, possibly there are better points to take the signal off.
  9. Just removed the RF modulator from my 2-port 5200 and wired in a simple composite connection using 1 transistor (2N2222?) and one resistor between video out and ground. Also cut the one leg of the resistor (R33?) that connects the audio to the video output. Have yet to connect the audio output. Very surprised at how good it looks (way better than 2600 and 7800 composite mods I have done). Will post pictures later. Tweaked the big pot on the lower right side of the board to get the colours correct. Video output may be a bit "hot", need to check that and maybe add voltage divider to the circuit.
  10. I was just testing an NTSC ANTIC in my PAL 800XL - It stretches the screen vertically compared to the PAL ANTIC.
  11. Fantastic project, great work Thomas. I am definately going to try build one of these (for PAL my 600XL and 800XL). Would love to contribute code / testing / 3D case designs etc, but will probably not get the time (too many other projects that I am obliged to finish first).
  12. Should this "fix" not be add to the standard cc65 repository: https://github.com/cc65/cc65/blob/master/libsrc/lynx/bootldr.s ??? (Just asking because this is the version of cc65 that I use).
  13. jum

    Millfork

    Yes, that is true for this specific example, but only because you have optimised it, not the compiler. It's not a good example, maybe a better example is something like: for (i = 0; i < 32; i++) { B[i] = A[i] * 2 + 1; // some processing is happening, not just a memcpy } As said before, the point of the comparison is to see how the the 2 compilers handle the same code construct.
  14. jum

    Millfork

    OK here's the cc65 code and ASM output for comparison. Note I used the cc65 "-Or" optimisation flag to enable register variables: cc65 code: void main() { uchar i; asm("nop"); i = 31; do { MIKEY.palette[i] = work_palette[i]; i -= 1; } while ( i >= 0 ); asm("nop"); for (i = 0; i < 32; i++) MIKEY.palette[i] = work_palette[i]; asm("nop"); } Assembler output (.s file): ; --------------------------------------------------------------- ; void __near__ main (void) ; --------------------------------------------------------------- .segment "CODE" .proc _main: near .segment "CODE" jsr decsp1 nop lda #$1F L0060: sta (sp) ldx #$FD lda #$A0 clc adc (sp) bcc L0044 inx L0044: jsr pushax ldy #$02 lda (sp),y tay lda _work_palette,y ldy #$00 jsr staspidx lda (sp) sec sbc #$01 bra L0060 L005F: lda (sp) cmp #$20 bcs L004E ldx #$FD lda #$A0 clc adc (sp) bcc L0058 inx L0058: jsr pushax ldy #$02 lda (sp),y tay lda _work_palette,y ldy #$00 jsr staspidx lda (sp) ina sta (sp) bra L005F L004E: nop jmp incsp1 .endproc Way longer/slower (for this example), and very hard to follow!
  15. jum

    Millfork

    Love that 8bitworkshop web IDE ( https://8bitworkshop.com/v3.4.2/? )
×
×
  • Create New...