Shop OBEX P1 Docs P2 Docs Learn Events
Retro II - An Apple II Compatible 6502 Computer - Page 2 — Parallax Forums

Retro II - An Apple II Compatible 6502 Computer

2

Comments

  • This post is coming from someone who happened to be in the market for the EXACT thing you just made:

    Put a socket to use an original BIOS ROM.

    I personally don't care about color AT ALL, and back in the day I never saw anyone using color because it looked like crp, for a couple different reasons.

    If it's a kit, everything must be through hole, so no P2. I imagine you're too busy to do anything other than a kit.

    Must be able to use original disc drive, because surprisingly after all these decades, all my old discs still work. (maybe have a utility so I can back them up to SD?)

    Sell for maybe $250? I'm trying to be reasonable because I am looking at the laser etched acrylic enclosure.

    If no keyboard provided, just specify which available part to use, so that I have the correct keyboard to nicely fit the frame

    (I assume expansion slots are the same as the original apple, so existing sound card and serial port card will work)

    Please have ready in three months, before Xmas

    I haven't seen anything like this, I've looked for one already, so you would have the niche totally cornered. And all/most of the work is already done?
  • There is a little fact about these Apple computers that I think most people don't know.

    I was going through boxes of my dad's old paperwork and saw original receipts for the Apple system he bought. He paid over $4K . I was flabbergasted. In 2020 dollars, that's basically the price of a new (cheap) car.

    That would explain why only schools had them, and none of my classmates did. All my classmates had the C64, and one had an Atari.

    How Apple made that business model work, must have been a schools-only strategy.
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2020-09-01 05:54
    Yes, that 4k back then was around 4 months average workers wages. I had an inbetween stint working in an Apple store fixing Apple II's and being able to play with the new girl "Lisa". An Apple II is what I cut my teeth on in learning Forth and I used it to interactively diagnose faults on the machine and align hard drives etc. When I left to design POS terminals I used Forth and 6502 based Rockwell 6511 chips in the design. For the next two decades I used 6502 based MCUs in a plethora of designs, onto the 16-bit 65816 and then shortly thereafter the M37702 which was a faster enhanced 65816 with all the peripherals and A/D etc, an MCU. Then the whole 7700 line was replaced with the C friendly M16C, what a total stuff-up that was. Bye bye Mitsubishi. Soon after they merged with Hitachi and formed Renesas which was then later taken over by NEC. Should have stuck to the very efficient 6502 based micros Mitsi, I say. They used those 7700s by the truck load in HDDs at the time.

  • Cluso99Cluso99 Posts: 18,066
    IIRC an IBM PC in the early days was $10K.
  • In the U.S. Apple ][ & later Apple 2e prices were around $1100 USD... Of course, you were adding displays, peripherals and memory cards to complete the system. A couple of floppy drives added about $900 USD. A 20MB hard drive was about $1200 USD. Makes today's computer prices seem reasonable!


    dgately
  • Yes, the price was pretty steep! Even as a kid I realized the incredible sacrifice my parents made to buy this. My mom was a teacher, so I'm not sure if maybe she got a special discount. My mom especially has always had this fascination with technology and I think she realized the impact the Apple II would have even way back then. Like Chip mentioned earlier, the quality of the machine was in a class of its own.
  • cgraceycgracey Posts: 14,131
    My dad started bringing home an Apple ][ from work on the weekends in 1979 when I was in the 6th grade. Ken would have been in the 4th grade. That was my introduction. I soon got a newspaper-delivery route so I could buy my own Apple ][. It took maybe six months from when we placed the order to when it arrived. Lots of anticipation. The base unit was about $2,000 then.
  • Don't forget the software bundle. Even basic utilities cost extra and you had to have VisiCalc which was the reason a lot of people bought these complicated newfangled computers.
  • kwinnkwinn Posts: 8,697
    Cluso99 wrote: »
    IIRC an IBM PC in the early days was $10K.

    Yes, it was. When the first clones came along I could assemble a system comparable to the top of the line PC's and sell it for about 6K and still make a reasonable profit.
  • Cluso99Cluso99 Posts: 18,066
    A friend and I used to drive from Sydney to Melbourne each weekend to pick up 10 Clones (and return one faulty clone) - a trip of 600km each way. This was a carload (station wagon full).
    We would burn in those clones for 48 hours which would yield 9 working units after we swapped all the faulty bits out. We then sold the 9 clones.
    The clone parts were never properly tested, hence the high failure rate. In those days, the parts came from Taiwan.
  • Cluso99Cluso99 Posts: 18,066
    The Apple ][, //e and /// got the attention of accountants who could do financials and modelling for a few thousand dollars using VisiCalc or supercalc.

    The alternatives were a software package on minicomputers which typically cost more than $20K (for only the software). Also when they ran this software it ground all other users to a grinding halt much to their dismay.

    This is why Apple really took off. These accountants spread the word quickly, and before long the mini software solutions were on the rails. Of course the EDP (electronic data processing) departments were not amused at losing control of their much protected and lucrative jobs.

    In those days, a typical mini-computer was worth $100K - $1M with probably the average around the $250K mark. A typical company with these would have a turnover of around $20M up.

    For reference, a VDU (video terminal) on a mini would cost around $3.5K and that was without any extras that may be required to connect it inside the mini. Think of a VDU as a specialised VT100 for the specific mini it connected to.

    And, maintenance on the minis cost typically 10%-15% of purchase price per annum.

    The best analogy of the computer industry is to the car industry. We would now get a car in the weeties pack and it would get 1,000,000 miles to the gallon! Of course, we would have to remove the engine and put it back for no reason at random intervals :(
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2020-09-01 23:41
    Cluso99 wrote: »
    The best analogy of the computer industry is to the car industry. We would now get a car in the weeties pack and it would get 1,000,000 miles to the gallon! Of course, we would have to remove the engine and put it back for no reason at random intervals :(

    That's if you used the "world's most popular" brand just like back when the world's most popular car color was black, because as Henry said "you can have any color as long as it is black" or words to that effect. Linux is free and fast and has a vast storehouse of software prepackaged plus the software store where the free software is also available with source code if you ever want to tweak it. After trying to keep a foot in both camps at the same time and not being successful in either, I completely changed over well over a decade ago and I haven't looked back.
  • This might be a great use of the P2 Blade
  • potatoheadpotatohead Posts: 10,253
    edited 2020-09-02 03:36
    BTW, the reason for two pixels being needed for white, or black, is the NTSC color cycle spanned two pixels.

    Wherever there is just one, its frequency exceeds the NTSC colorburst frequency resulting in that pixel being seen by the display as both intensity and color.

    Very cool project!

    I too learned a lot on Apple machines. First at school, then at home. Like Chip, I worked for one. And had an Atari machine while doing that work

    MECC is a huge part of why so many schools had Apple computers. Tons of reasonably licensed education programs and supporting material made it a great drop in solution for schools wanting to begin computer education.

    I have a //e machine and display on idle so my granddaughter can play Oregon Trail as intended one day here real soon. That one also has an accelerator and 65816 chip that will clock up to 16Mhz.

    The Apple community is still quite active. There would be interest in a machine that can take peripherals, and or one that has hardware ready to read disk images off some form of removable storage.






  • JonnyMac wrote: »
    This might be a great use of the P2 Blade

    I agree. Lay out the necessary connectors and such on a larger board and plug one in.
  • JonnyMac wrote: »
    This might be a great use of the P2 Blade

    That does look pretty neat. I really like the compact form factor.
  • potatohead wrote: »
    BTW, the reason for two pixels being needed for white, or black, is the NTSC color cycle spanned two pixels.

    Wherever there is just one, its frequency exceeds the NTSC colorburst frequency resulting in that pixel being seen by the display as both intensity and color.

    I didn't know that. That's pretty interesting. I had heard somewhere that Woz initially developed the Apple II to output monochrome, and only later figured out that he could manipulate the pixels to display color. I'm not sure if that's true or not, but it does seem to fit with how he liked to do a lot of the heavy lifting in software vs hardware.

    potatohead wrote: »
    MECC is a huge part of why so many schools had Apple computers. Tons of reasonably licensed education programs and supporting material made it a great drop in solution for schools wanting to begin computer education.
    Yes, I loved MECC growing up. The thing that is most fascinating to me for whatever reason is that MECC was a public agency, not private, that made educational software. I wish we had more companies like that today. A lot of my colleagues and I debate at work on how much we learned from these games, but I still contend that they had great educational content. The ones I remember the most were Oregon Trail of course, Odell Lake and Word Munchers.


  • potatoheadpotatohead Posts: 10,253
    edited 2020-09-02 15:49
    In addition Woz, along with many others, abused the NTSC timings so that there would be a fast, stable display. Television, done the Apple way, offers about 160 "pixels" of color detail per line. The Apple showed 140 of those, due to that use of the high bit leaving 7 pixels per byte, 40 bytes per line.

    Other machines offered 160 pixels, but no 6 color high res display.

    Turns out, looking back, 6 colors is enough to do anything. In my view, that did hold the Apple back. Had it been less, we would have seen video cards more widely adopted. Had it been more? Likely too expensive to take off.

    So it all got done on that goofy screen addressing scheme.

    Abusing the signal spec some gave us 240p displays and color artifacts that all line up. Many machines that followed the Apple made similar choices, but also added color circuits to improve on display quality.

    A notable, very early trade off that I feel shows off how Woz would maximize functionally while minimizing circuits.
    Yes, I loved MECC growing up.

    Indeed!

    Fun times. I agree with you. Our British friends likely have similar thoughts about the BBC Micro and how Acorn and the BBC put together a nice education program.

    Then there were the magazines in the grocery store! I tell people about that, reading ROM listings, hacking on games and they almost always say the same thing, "wait! Grocery store?"

    What I got out of MECC was familiarity. And a craving to do more. Worked for most of us. Maybe it worked for all of us that it could work on. I don't know.



  • localrogerlocalroger Posts: 3,451
    edited 2020-09-03 01:09
    Yes, the price of home computers back then was very steep. An Apple ][ was $1100 with no peripherals, a TRS80 Model II was $800. And even the cassette deck was a peripheral -- and Commodore didn't even use a standard cassette player. That's why my first computer was an Interact Model One (or as Protecto Enterprises branded it, the model R). This was a 16K computer, thus a step up from the relatively useless TRS80 Model I, included the built-in cassette deck, and used your TV set as a monitor. It was a bit odd in not having a text generator so the text display was drawn in a 5x5 font on its 77 x 112 pixel graphics display, but they were being surplused for only $250 so we could afford it. We bought it from a classified ad in the back of Popular Electronics. Eventually we were able to get a second one so my father and I could computer at the same time.

    And it turned out to be great in that to do anything useful with it, I had to learn a lot about both general programming and about how that machine itself worked, which was in turn possible because the machine was pretty simple.

  • K2K2 Posts: 691
    Using a P2 to emulate an Apple ][ begs to be done. Wish I could remember the name of the 6502 assembler I first used. One of its identifying features was the use of ! rather than # to specify immediate addressing.
  • cgraceycgracey Posts: 14,131
    K2 wrote: »
    Using a P2 to emulate an Apple ][ begs to be done. Wish I could remember the name of the 6502 assembler I first used. One of its identifying features was the use of ! rather than # to specify immediate addressing.

    I remember using an assembler named L.I.S.A, but then had to switch to Merlin because that's what game developers were using and I was making an Apple ][ --> Commodore 64 downloader. The download was triggered by the USR mnemonic which was customizable.
  • I can't remember what I used either (they weren't very good) but by the 90's I was able to use a multi-target open-source macro-assembler from Alfred Arnold. It was so much better than anything else and I could program anything 6502/65816/M37700 etc as well as a plethora of other micros, all with a very flexible syntax and great macro capability.
    It is still maintained even to this day (incredible).

    btw. I wrote a 65xx simulator in TAQOZ Forth on the P2 that even runs my old Rockwell R6511 binaries!

    Here is the help listing of an older version of the assembler at the command prompt running under Wine in Linux.
    macro assembler 1.42 Beta [Bld 28]
    (i386-unknown-win32)
    (C) 1992,2003 Alfred Arnold
    Mitsubishi M16C-Generator also (C) 1999 RMS
    XILINX KCPSM(Picoblaze)-Generator (C) 2003 Andreas Wassatsch
    TMS320C2x-Generator (C) 1994/96 Thomas Sailer
    TMS320C5x-Generator (C) 1995/96 Thomas Sailer
    
    calling convention : asw [options] [file] [options] ...
    --------------------
    
    options :
    ---------
    
    -p : share file formatted for Pascal  -c : share file formatted for C
    -a : share file formatted for AS
    -o <name> : change name of code file
    -olist <nname> : change name of list file
    -shareout <nname> : change name of share file
    -q,  -quiet : silent compilation
    -cpu <name> : set target processor
    -alias <new>=<old> : define processor alias
    -l : listing to console               -L : listing to file
    -i <path>[;path]... : list of paths for include files
    -D <symbol>[,symbol]... : predefine symbols
    -gnuerrors: error messages in GNU format
    -E <name> : target file for error list,
                !0..!4 for standard handles
                default is <srcname>.LOG
    -r : generate messages if repassing necessary
    -Y : branch error suppression (see manual)
    -w : suppress warnings                +G : suppress code generation
    -s : generate section list            -t : enable/disable parts of listing
    -u : generate usage list              -C : generate cross reference list
    -I : generate include nesting list
    -g [map|atmel|noice] : write debug info
    -A : compact symbol table
    -U : case-sensitive operation
    -x : extended error messages          -n : add error #s to error messages
    -P : write macro processor output     -M : extract macro definitions
    -h : use lower case in hexadecimal output
    
    source file specification may contain wildcards
    
    implemented processors :
    
    68008     68000     68010     68012     MCF5200   68332     68340     
    68360     68020     68030     68040     
    56000     56002     56300     
    PPC403    PPC403GC  MPC505    MPC601    RS6000    
    MCORE     
    6800      6301      6811      68HC11K4  
    6805      68HC05    68HC08    
    6809      6309      
    68HC12    
    68HC16    
    H8/300L   HD6413308 H8/300    HD6413309 H8/300H   
    HD6475328 HD6475348 HD6475368 HD6475388 
    SH7000    SH7600    SH7700    
    6502      65SC02    65C02     MELPS740  6502UNDOC 
    65816     MELPS7700 MELPS7750 MELPS7751 
    MELPS4500 
    M16       
    M16C      M30600M8  M30610    M30620    
    4004      4040      
    8008      
    8021      8022      8039      8048      80C39     80C48     8041      
    8042      
    87C750    8051      8052      80C320    80C501    80C502    80C504    
    80515     80517     80C390    80C251    
    8096      80196     80196N    80296     
    8080      8085      8085UNDOC 
    8086      80186     V30       V35       
    80960     
    8X300     8X305     
    XAG1      XAG2      XAG3      
    AT90S1200 AT90S2313 AT90S4414 AT90S8515 ATMEGA8   ATMEGA16  
    AM29245   AM29243   AM29240   AM29000   
    80C166    80C167    
    Z80       Z80UNDOC  Z180      RABBIT2000Z380      
    Z8601     Z8604     Z8608     Z8630     Z8631     
    KCPSM     
    96C141    93C141    
    90C141    
    87C00     87C20     87C40     87C70     
    47C00     470C00    470AC00   
    97C241    
    16C54     16C55     16C56     16C57     
    16C64     16C84     
    17C42     
    ST6210    ST6215    ST6220    ST6225    
    ST7       
    ST9020    ST9030    ST9040    ST9050    
    6804      
    32010     32015     
    320C25    320C26    320C28    
    320C30    320C31    
    320C203   320C50    320C51    320C53    
    320C541   
    32060     
    TMS9900   
    TMS70C00  TMS70C20  TMS70C40  TMS70CT20 TMS70CT40 TMS70C02  TMS70C42  
    TMS70C82  TMS70C08  TMS70C48  
    370C010   370C020   370C030   370C040   370C050   
    MSP430    
    7810      78C10     
    75402     75004     75006     75008     75268     75304     75306     
    75308     75312     75316     75328     75104     75106     75108     
    75112     75116     75206     75208     75212     75216     75512     
    75516     
    78070     
    7720      7725      
    77230     
    SC/MP     
    8070      
    COP87L84  
    SC14400   SC14401   SC14402   SC14404   SC14405   SC14420   SC14421   
    SC14422   SC14424   
    ACE1101   ACE1202   
    SYM53C810 SYM53C860 SYM53C815 SYM53C825 SYM53C875 SYM53C895 
    MB89190   
    MB90500   
    1802      1805      
    peter@peter-OptiPlex-9010:~/PBJ300D/!D/ASM/ASX-new$ 
    
    
  • cgracey wrote: »
    K2 wrote: »
    Using a P2 to emulate an Apple ][ begs to be done. Wish I could remember the name of the 6502 assembler I first used. One of its identifying features was the use of ! rather than # to specify immediate addressing.

    I remember using an assembler named L.I.S.A, but then had to switch to Merlin because that's what game developers were using and I was making an Apple ][ --> Commodore 64 downloader. The download was triggered by the USR mnemonic which was customizable.

    Gadzooks! I remember using both of them. By the time I switched to the PC, an AT&T PC 6300, I was extremely good at both writing in R6502 assembler and making use of it, to combine with BASIC on the host.

    Now I'm trying to track down an out of work Apple 2C for a PI to Apple 2C project I've been putting on backup mode for the past two to three years now.
  • btw. I wrote a 65xx simulator in TAQOZ Forth on the P2 that even runs my old Rockwell R6511 binaries!

    Awesome! That is very impressive. I remember one of the recent "early adopter" videos mentioned TAQOZ Forth in it and I remember being very intrigued. I thought I heard that it was actually built into the P2, is that right? I'll definitely check it out now, thanks! I've got a few hopefully small items to finish up with this current P1 version of the board but after that I'm really looking forward to starting in on the P2 version.
  • The P1 surprises me again. I was not expecting to get full color working for this project, but here it is! Thanks again everyone for your help. As a kid all I knew was the green monochrome screen, but this is pretty neat! It's like discovering all these games again in a new way.

    I may post a new video soon detailing some of the changes. There's still some work to do on adding back in the monochrome screens as display options, but hopefully that shouldn't take long. The other unexpected outcome of this is that now that I'm not dealing with the color bits and the 7 bit wide columns translations in my hires routine, it doubled my frame rate! So it's clocking in at 106/107 fps vs 50.
    4096 x 3072 - 250K
    4096 x 3072 - 299K
    4096 x 3072 - 1020K
    4096 x 3072 - 457K
  • Look how bad the text looks in each of the four photos. This would totally clean up in monochrome mode. And all of this just for six garish colors?

    This is the same from what I remember as a kid. But I assumed at the time it was only because I was using a TV and RF modulator for color, instead of a proper color monitor. Now I know better.
  • Saw the Youtube comment that you might offer a PCB and BOM. Is there any chance you will offer the clear enclosure? Very few people can do that on their own I assume
  • The_Master wrote: »
    Look how bad the text looks in each of the four photos. This would totally clean up in monochrome mode. And all of this just for six garish colors?

    This is the same from what I remember as a kid. But I assumed at the time it was only because I was using a TV and RF modulator for color, instead of a proper color monitor. Now I know better.

    You're right, by today's standard's it's pretty bad. I think the way Woz added color to the Apple II was essentially a hack, which explains the clashing and fringe effects. It also is essentially a reduction in resolution going from monochrome because it's going from a 1bpp to a 2bpp video.

    But still, I think it's pretty cool. As a kid I think I would have thought it was neat to see some of these games in color. There's details that I hadn't noticed before in monochrome that are showing up in color which is cool.
    The_Master wrote: »
    Saw the Youtube comment that you might offer a PCB and BOM. Is there any chance you will offer the clear enclosure? Very few people can do that on their own I assume

    To get the enclosure cut out all I did was send the svg file to ponoko and they ran it through their laser cutter. It took them a few weeks to get the parts sent back to me. I think at the time they were ramping up production of face shields for the hospital workers back in April/May. They may be back to a shorter turn around now.
  • potatoheadpotatohead Posts: 10,253
    edited 2020-09-16 05:51
    Hey, that is working pretty good! (and ignore my next few posts, if you are happy that's all that counts. I'm not wanting you to do more work on this. Maybe you want to though. )

    The text is going to look poor no matter what, but it can look better (and does on a real machine).

    On the real hardware a single color pixel gets the color and remains a single pixel. As it stands right now in your implementation, a single pixel ends up being two, and there are patterns on byte boundaries that won't work out, unless both bytes are considered too.

    That is what gets hard about apple graphics. It isn't really two bits per pixel in all cases.

    For many cases, that kind of conversion works.

    The reality is the color really is just the position in the scanline + whether it's an even or odd byte + the high bit shift to determine which of the two artifact color sets the pixel color can be.

    This is also why single line wide fonts generally look bad on an Apple 2.

    I will drop a couple pictures here in a minute.
  • Here are a couple of single line text. One showing each high bit combination:

    3024 x 4032 - 4M
    3024 x 4032 - 4M
Sign In or Register to comment.