Jump to content
Have You Played Atari Today?
2600|5200|7800|Lynx|Jaguar|Forums|Store
Warnings about File Attachments - Please Read!

potatohead

Members
 View Profile  See their activity

  • Posts

    4,891

  • Joined

  • Last visited

  • Days Won

    16

 Content Type 

Blog Entries posted by potatohead

  1. potatohead
    Addons?
     
    Got my demo board today. It's a fun little beasty. Actually I've a coupla questions:
     
    Anyone care to comment on the video in capability of the 7800? Let's say I generate a NTSC signal and insert it. Assuming the necessary timing information can be provided by the 7800 (CPU maybe?), is this an overlay, meaning I can generate my own video on top of what the 7800 already does?
     
    The chip I'm tinkering with has onboard timers and video generation. (Composite video only takes a coupla resistors.) Thinking of an interesting blend where maybe I don't do much with the 7800 Maria chip and have the propeller chip do video generation or maybe just sound, like adding a POKEY currently does. Lots of fun possibilities that appeal to me because it would be both a hardware and software project.
     
    The end result would be programmable and not too expensive to replicate either as very few additional components would be required.
     
     
    My other question is along the lines of why bother? Is this going outside the classic computing aspect of things too much? I personally like the older hardware because of the kinds of games created. There are enough limitations to force the kinds of games I've always loved. This little project would very likely not exceed that boundary so much as to reset expectations beyond those in the norm here. Opinons?
     
    On the off chance that this idea sees some real hardware, how do you guys figure out what a target price would be?
     
    I know the core chip is the $14.00 or so. 5 resistors, maybe a buffer chip to handle the 3.3V to 5V conversion will cost some smaller amount, maybe a coupla bucks total. There is an on board timer that I don't know too much about yet, but would probably suffice to eliminate the need for an external oscillator. The 7800 could provide a baseline for programmers to assess what their particular timing variance is. The only real issue would be peak compute. Some kits would run faster than others, but all would run at a more than acceptable minimum.
     
    32Kb of RAM is onboard the propeller. Looks like I would need a cheap eeprom and socket for that too. 32Kb minimum.
     
    If I were to use the 40pin DIP package, a socket would be all that is needed to mate the chip to the board, thus making kits a viable option.
     
    How much are custom boards and who might make them?
     
    Programming happens via USB or serial. It would also be possible to feed data through the 7800 itself, but that has a lot of development issues. Programming components can be put on board, or an inexpensive adapter from the producers of the chip is available. The chip has onboard ROM that can communicate with a PC and can program the eeprom. One could just program it themselves too...
     
    Too many components to be viable at all?
     
    Thinking of three potential options:
     
    Fully assembled, plug and play board, with USB programming cable.
     
    Partially assembled board, with core components soldered, and a kit bag of goodies or let user aquire own or use stock.
     
    Completely unassemblied kit.
     
    I'm just musing right now, trying to consider what I might use this great little chip for. An early idea of price and hassle might tell me right away if this is not a direction I should go for anything other than personal reasons.
  2. potatohead
    Man oh man! Moving sucks! Well, getting a new and bigger home doesn't, but basically everything else does. One never realizes just how many little tech bits one has until they are all yanked and stuffed into a box! Basically nothing works anymore. My ReplayTV does not work with the new Directv sat receivers, DSL was hosed until today, computers are gonna take forever to reconnect, on and on and on...
     
    I think I'm going to be looking for a new 400. Hooked my old one up and it's flaky now. Tried reseating the chips and still it just goes nuts after about 10 minutes. The combination of old age and being moved has probably done it in. The 7800 still works great though! Most of the other computers doom won't be known to me for a while yet. Hoping the SGI isn't in any trouble or that's a trip to ebay for sure. Can't live without that one.
     
    While I'm here, I suppose I should post a coupla ideas I had for later:
     
    Was thinking about Ooze a bit and reread the comments collected so far. It's not quite a game, though it is fun. One thing that's been getting to me is the use of the joystick for controls. Mostly I did this because the Batari Basic environment limits keep the paddle from being part of the equation. I'm torn between taking the game into assembly land, adding the paddle and polish and just doing something new. Controls are at the heart of this because the lack of them prevent me (or so I think --feel free to chime in) from getting the core game elements where they need to be for a viable game. So, last night while unpacking my Atari stuff, I bumped into my driving controllers. I'm thinking those might be just the trick for Ooze.
     
    I want the game to be fast like Kaboom! That's where the trance comes into play and is the core, for me at least, of what makes really great old 8bit games. The joystick just isn't going to cut it because with speed comes the need for really fast player interaction with a far more dynamic and aggressive Ooze. Having not touched the thing for a while, it seems more like a great Bb tech demo than it does a game. (And the solid comments reflect that if taken as a whole.) For those of you out there, starting on the game journey I'm currently on, I highly recommend putting your work aside for a while. When you come back, your perspective is far more objective and valuable. You really get sucked in while working on the project. Core goals can get lost in the shuffle, thus blocking the creative process. At least I've discovered this to be true for me. Take that for what it's worth!
     
    Using driving controllers with Bb should be dead simple and they work a lot like paddles, but for the lack of absolute positioning. Maybe that's enough to get the experience where it should be. So I'm going to try that and see where it goes.
     
    My gut says it will be a coupla more weeks before things are in order enough to begin hacking on the game again. I'm gonna move everything over to the SGI. This will slow the project at first, but it will be worth it. If you have ever used an SGI, you know about the IRIX environment. It's insanely addictive and very pleasing to work with for long periods of time. On SGI, the emulator scene is good, but getting dated. I think it's time to compile the latest Stella and see if that's viable. If so, Atari game development on the platform will be solid again, which will make me very happy. There is something cool about using a 64Bit platform to develop for old 8bit systems. (That and the excellent screen fonts --easy on the eyes.)
     
    I'm not sure how I'm going to do controller interfaces however. Might be keyboard / mouse only for the moment. If push comes to shove and I need to do something else, I'll just use the Linux box for that and display the window on the SGI.
     
    Sorry for length, but that brings me to another "Why Unix and the X-Window system kick so much ass it's not even funny" post. I'm not sure the lions share of computer people fully understand just how big of a gift the X-Window system really was in it's time. Heck, that's still true today for those in the know.
     
    Basically it comes down to true multi-user computing. Not user switching, or a limited remote desktop environment, terminal server, etc... but true multi-user. That's what Unix has been from the get go. When networked computing became a reality for the masses, Unix and X were there in all their glory ready for what should have been a great time. Instead we got single user win32 boxes.
     
    Well, most of us did anyway. Those few who were running Unix in networked environments with the X window system took a different path...
     
    X allows any user on any machine to serve graphics to the user on the display they are located at. It's like VNC, but for any number of users and not just for the whole desktop, but for any application. It's possible, for example, to have the machine you are sitting at handle input and display graphics while another runs your window manager and a third machine to run the application. The implications for the desktop are powerful, if not so obvious. For me, this means I can use an SGI (once I get it hooked up again) for my primary desktop, and display Linux programs on it as well, thus having one nice environment. I will typically script the access to resources running on other machines so that my favorites are only a coupla clicks away. Done right with shared network storage and the average user will have no idea they are running stuff across a batch of machines instead of just one local machine.
     
    Sounds goofy I know, but it's very cool. Let's say my local machine is a bit slow (which my old SGI is), but a remote machine is much faster. Once I've asked for the program to be displayed on my box, it's pretty much like having a faster machine without the hassle of changing boxes, KVM, VNC, etc... From an IT standpoint this has serious advantages as well. Say you have a group of users who all need a very strong machine 10 percent of the time. Do you buy 10 strong machines, or buy one and let them share it as I've described above? Another example lies with shared applications. In this day and age, many software companies do not let their licenses float, depending on the single user win32 system to force folks to buy lots of copies of their software --even for casual users. With the X window system, this is not necessary at all. Just install one and let people share it to their local desktops as if they actually had the application running on their local box.
     
    Finally, there are lots of solid computing choices besides the sometimes problematic client-server approach. Client server is great, but does require some tinkering on both ends to work correctly. The more robust choice of application server can eliminate these problems. Have a difficult or expensive application? Put it on an application server and let all your users run it from there. One copy of the software, one configuration, etc... Or, maybe you want to manage some data. Keep it out of users hands. With the X window system, you can have sensitive data located on a server with the managed application that interacts with it. Users are limited to only those things the application allows when running over X, because they are not running code locally! There is no need for them to even be able to see the data, but through the application itself.
     
    Ok, that's enough. Best get connecting or I won't be doing any of this stuff!
  3. potatohead
    I do and kind of went on a flashback tour....
     
    https://www.youtube....h?v=MJG5ewAgKhw
     
    https://www.youtube....h?v=okGfFYKCwYU
     

     

     

     

     
    https://www.youtube.com/watch?v=D2o99jGlZ2o
     
    https://www.youtube....h?v=ucQ6FMUszNY
  4. potatohead
    I'll go ahead and post up some images and such later. Didn't have a lot of time for that today, but I did have time to install the card and get it configured well enough to go and run some stuff.
     
    My Apple only has a single Disk ][ That's been a little limiting, simply because it's not enough storage for some of the better things you can run on an Apple. Don't get me wrong, an Apple ][ computer with ADT and a single disk is a lot of fun, and much is possible. I've been able to do some light programming and run a few favorite old games. Very good solution, though it is necessary to keep a few floppies around. What I ended up doing was to simply just write what I was interested in doing to one of the disks, then do it, recycling as needed. I have maybe 15 floppy disks that are in very good shape. Worked for me, though it was cumbersome to deal with.
     
    The CFFA ships with a nice, large PRODOS hard disk image already setup. Booting that takes about 3-5 seconds. It's just amazing, and more zippy than I remember hard disks being back when I got to use real ones. Love the storage, don't miss the fiddly hard disk at all.
     
    I chose to employ USB, simply because I didn't have a CF card handy, and an extension cable is easy to route out the back of the machine and onto the desktop where I can just plug it in. This rocks really hard. Very well done, and easily the most simple and robust 8 bit storage solution I've used, and I've used a few and on a few machines, Atari, CoCo, Apple, some tethered to a PC, some using SD Cards.
     
    Of course the Apple is so damn nice because of the slots and the ROM featuring the basic options needed to just use the devices. Now, when I boot the Apple, it drops right into PRODOS running off that large image. I don't have a lot in there yet, but I probably will as I do some projects. If I want to boot a real floppy, insert it and issue PR #6 to boot from that real disk ][. If I want to boot a virtual disk ][, that's in slot 5 on my apple, so it's PR #5. Simple. Pressing M after a boot gets me to the CFFA menu where I can deal with disk images and such, and for multi-disk programs, I can either load up a drive or two, or toggle among a set of images using the remote thing.
     
    That remote is just a board, so I've gotta enclose it somehow, just so it doesn't break easily. Other than that, it's just cake. Push the button to shift to the next sequential disk image setup in the menu, and it's just like moving a floppy.
     
    Highly Recommended. This device is very well executed, and I had a lot of fun today giving it a test run.
  5. potatohead

    Some stuff I did

    By potatohead,

    Was having a conversation about past projects only to find the project pages rotting out there. Since that conversation is happening here, I thought I would just make a quick archive.
     
    This is a STL file viewer I wrote. Two screening, one from a CAD system, and one from the viewer showing a reduced polygon file.
     
    The binary package contains compiled binaries for Windows, IRIX and Linux.
     
    The cup.jpg is from the CAD system. The other .jpg is the viewer displaying a low resolution polygon file. The cup.zip file contains an ASCII STL file to be rendered by the viewer.
     
    A source tarball completes the set. http://viewstl.cvs.s...ge.net/viewstl/
     
    cup.zipviewstl-binaries.tar.gzviewstl.tar.gz
  7. potatohead
    Scored a really great 80's portable TV. It can sit right on top of the Apple. Now, if I could only find one of those green screens... all in due time.
     
    Anyway, the TV is sweet. Required almost no real adjustment. I pulled the back and did the usual. Expanded the displayed frame, by pulling the yoke back just a bit. it's a newer one, so I didn't have to re-converge, which is a bitch let me tell you. Dealt with the video drives, linearity, etc...
     
    It's just about pixel perfect now. $2
     
    I used my Atari for the color tests, and the Propeller to determine where the overscan boundaries were, and to send it a PAL signal. That won't display color, but will display monochrome. The expanded frame makes that possible, and on this little set, no vertical hold adjustment needed! Sweet.
     
    Time to play a little. Booted my ULTIMA V disks, created a character, and just wandered around a bit. I really wish I had the time to just play this one. Maybe do it in stages. Brought back a lot of fun memories --different from Atari memories. School was where I worked with the Apples for a long time. Managed to score one in the late 80's, and enjoyed it enough to want another one.
     
    Coupla screenies:
  9. potatohead
    Looks like I've gotta get hold of a board called "dracblade". It's a board using two props, some RAM, and other components.
     
    People have CP/M running on it rather nicely. Recent sound emulations of the SID and AY, led to a nice Speccy emulator that runs many games. It's not quite at speed, but fast enough to play the games.
     
    So, we've gone from a Manic Miner port a few years ago, to an emulated Speccy, running that title!
     
    It's getting time to do some 6502 computers. The Z80 guys are rocking hard! I'm thinking a VIC should fit in to a single chip...
  10. potatohead
    Let me explain that. For the last few years, I've really been struggling. Bad thing after bad thing has happened, and that's just the truth of it.
     
    After looking around, I realize that perhaps the string of ugly stuff is over. Mrs Potatohead and I have been caring for older parents and kids for a long time. (too long maybe)
     
    We will be down to just one kid, after the death of my Father in law, who I loved as a father. My own wasn't all that familiar to me, the two of us being VERY different people.
     
    Not sure where that's going to go, but I do know it's going to be different from where I've been, and that's something I think will be good news.
     
    My CoCo 3 Drivepak is coming finally!! It's a bootable cart, that contains a micro-sd card, filled with a bunch of disk images. CoCo 3 computers can also talk to PCs for storage through something called Drivewire. Now that I have the cart, I can set that all up, have the CoCo autoboot, and run lots of stuff on the real hardware.
     
    Now to get that and my Atari 800 XL setup with a SIO solution, and I'm retro computing set with two of my three favorite old computers.
     
    The Apple ][ is my third favorite, but is kind of big... I'll wait on that a bit to see how things shake out. A slightly longer term plan is to put Propellers in carts, or on cards with these machines and do stuff...
     
    Another plan is to start writing some code again. Looking forward to both.
     
    On the family front, my wife and I are transitioning into a more mobile lifestyle. Who knows what that will bring? It's exciting to think about though. Some free time...
     
    Deaths are hard things. This one was really tough, and I'm just drained, trying to find reasons to move forward and shake it off. So, there are a few of them. I'll find more, I am sure.
  11. potatohead
    There are basically two approaches to artifacting on the color computer 3. My blog entries from before use the 640 pixel, 4 color mode to generate artifacts. That's a good palette, with lots of texture, some bright colors, some dim colors, but lacking in red, and generally kind of coarse in places. When artifacting in 640 pixel mode, the 4 grey scale values are used.
     
    320 pixel mode artifacting is a bit different, because the computer has 4 bits per pixel avaliable, meaning colors can be mixed with other colors, and the intensity levels for a wider palette.
     
    The 4 screenies attached are some samples that appear to support the idea that somewhere around 1024 colors are avaliable on the machine, with 256 on screen at any one time, unless scan line interrupts are used to dynamically change color registers, like we would do on Atari machines.
     
    16 palette entries are available, each one can be set to one of 64 stock colors, meaning a scan line color change will impact multiple colors at once, not just one particular screen color.
     
    Jason Law and I have been working with the color computer and it's interesting video options, and I thought this expanded palette might be of interest to others here.
     
    One other very interesting software register in the machine is the color phase register. It's used to correct older school red / blue artifacts, where odd pixels might be red or blue, depending on the power up state of the machine at the time. The older Color Computers suffered from this, often featuring a color test & reset screen so users could get the right colors, or twiddle the tint knob on their display screen.
     
    When this register is toggled quickly, many artifact effects occur. This might prove interesting in the future, if used in a way that is synced with the screen.
     
    In these screen shots, they are in pairs, each one showing the phase bit set, and unset, which does not always result in the same color space. Some analysis of the resulting colors will need to be done to figure out whether or not the colors are actually the same, or if different, how many are different.
     
    There are some colors roughly the same between these screen captures, however, there are also a lot of color combinations not shown. Many combinations will be redundant, making for an interesting problem to sort out what really can be put on the screen. However that goes, it's 8 bits per pixel, for 256 choices per scan line, per palette register set. I don't know how many registers can be written between scan lines. I don't think it's the whole set, but I could be wrong on that.
     
    Full screen color possibilities do exceed 256 colors at a time, but with some vertical limitations that have not been sorted out just yet.
     
     
    For now, here are some reference palettes, showing off what 320 pixel mode artifacting does. Effective resolution here is 160x200, just like it is for the 640 pixel mode, with the difference being two 16 color pixels being mixed together, instead of 4, 4 color ones.
     

  13. potatohead
    Edit: So I bagged on the mouse for now. Released version 1.0, text only to the object exchange. I'll attach a code archive here that auto-detects HYBRID, HYDRA and Demoboard, that includes a nice text demo. The release candidates serve no purpose now, so I pulled them. If you have been using this, please download the 1.0 version as that's stable going forward.
     
    I've attached a new screenie at bottom, in 40 columns, showing off what got done.
     
    PotatoText01.ZIP
     
    This whole VCS on Propeller thing is escalating. More on another thread later, but that's the reason for punting on this right now. Text only is excellent, and useful. Mouse pointer can come later. The hooks and core graphics capability is there now.
     
    I got stuck on this and it took way longer then I thought. Got it done though! 80 characters in color on composite video is a big stretch. Only some color combinations work well, but there is the S-video option. When I get my next capture device in, I'm going to run the S-video and check it out. Hoping it looks really good.
     
    Eric Ball did a very nice starter Propeller video signal template. I have struggled and continue to struggle with getting the signal characteristics I want, but there is more progress than not, so I'm feeling good about that.
     
    This driver has an 80x25 text display memory, uses an Atari font in RAM, and has a word per character for color definition. All in all, a nice memory foot print for this kind of thing, and lots of color choices. Was going to go with just a byte and use a palette, but doing that kind of thing is a lot more expensive than I realized. So, the lesson learned on Propeller is don't go for color indirection. It's too slow for useful resolutions, or takes too many COGs. One or the other will occur. (unless I'm missing something, who knows?)
     
    The cool thing about this driver is that most of the functions are live and can be changed after the driver is launched. Acts just like a graphics chip does and that's just fun. There are a few more things to add to the live functionality, but no worries there. Easy to do.
     
    Display modes are 16, 32, 40, 64 and 80 x 25 text display, with 256 character re-definable font in RAM. Signal characteristics are:
     
    "Scan Doubled" interlaced 200 line vertical display. This really helps with clarity on the color. It was surprising to me how much impact doing this had.
     
    Non-interlaced 200 line display.
     
    Color burst on, or off for getting the most out of a display device. If you don't need color, or maybe just need intensities, this is the way to go.
     
    Two color per screen mode for speed and smaller run time memory foot print.
     
    Interlaced video screens shown. 40, 64 and 80 columns.
     

     
    Here's the code archive. Viewable as text, or with the downloadable IDE.
     
    The last bit of functionality to go into this thing is a mouse pointer object. I did the video using 4 color mode on the Propeller video generator hardware. Text only uses two, leaving two for the mouse to use anywhere on the screen. So, that will end up working just like a sprite does. The font is converted from a 1 bit per pixel to a 2 bit per pixel bit stream on the fly, using a 256 entry table. %01001111 = 01000001010101 Color 00 = background, color 01 = foreground for text. The other two colors will be used by the mouse pointer and will be consistent across the entire screen.
     
    It's really fun to play with video displays like this. Almost like building one's own graphics chips.
     
    One other thing is the Atari font and readability. I tried some others, and they display way more artifacts than the Atari one does. That kind of thick, two pixels almost anywhere style really plays out well on higher resolution TV displays.
     
    I just got done uploading the other screen captures. One is two color mode, blue on white, showing off the Atari font at high character densities. The other is a color screen, with the color burst turned off. The final is a monochrome screen, with no color burst. Very high quality.
     

  14. potatohead

    Some progress...

    By potatohead,

    Here's a screenie of my WIP for what is titled Battle Pong. See blog entry from the past for some of the ideas surrounding this game.
     

     
    Notes for me:
     
    Playfield toggle as one power up to denote ball random velocity change as it passes through center net.
  15. potatohead
    Download Stella Here: http://stella.sourceforge.net/
     
    Download bB Base Build Here: http://www.bataribasic.com/download.html
     
    Download bB Latest Build Here: http://www.atariage.com/forums/index.php?s...p;#entry1752068
     
    It seems only the base build is on the Batari web site. If this entry is dated, ask where the latest can be found in the Basic Programming forum, and get that build, or ping Batari and see what the deal is.
     
    Download Visual bB Here: http://www.atariage.com/forums/index.php?showtopic=123849
     
    1. Install Stella
     
    I just took the defaults so that program lives at c:\program files\stella\stella.exe
     
    2. Create directory c:\Atari2600\bB
     
    3. Extract the contents of the bB base build into the new directory.
     
    If using the Microsoft deal, point the extract wizard at c:\atari2600\bB
     
    4. Set the following environment variables: (Control Panel, System, Advanced, System Environment)
     
    bB=c:\Atari2600\bB
     
    This sets the install directory for Batari Basic.
     
    path=%path%;c:\Atari2600\bB (Append ;c:\Atari2600\bB to your existing path statement, paths are separated by the semicolon. Just tack it on to the end of the mess that's probably already there!)
     
    This insures the bB compiler can work from any command prompt.
     
    5. Do a test build of the sample progggy!
     
    start, run, cmd [enter]
     
    cd c:\atari2600\bB\samples
     
    2600bas sample.bas
     
    It should say build successful!
     
    dir *.bin should show sample.bas.bin
     
    6. Run this binary in stella
     
    7. Make two new directories under c:\atari2600
     
    Projects and VisualbB
     
    8. Copy the contents of the Visual bB archive to the new VisualbB directory created above.
     
    9. Run VisualbB.exe with a simple double click.
     
    In the configuration wizard, specify where things live. The batari basic default is fine. You will have to specify stella.
     
    Stella lives in c:\program files\stella\stella.exe by default, as noted above.
     
    That's the setup!
     
    When in the nice batari IDE, start a project and create a directory under the projects directory created above, and start to building stuff.
     
    I've a PS2 to PC USB device that I'll use for Stella in the shorter term. Since I'm gonna be working with paddles this time around, I think I'll also get setup to run stuff on an actual VCS. For now however, this is a solid setup.
     
    Thanks to the creators of the IDE, DASM and bB. This whole deal is SO much better than it was long ago. Nicely done! I think I'll have a bit of fun!
  16. potatohead
    It was bad ass.
     
    Multiple video generators rendering to the same video output pins gives hardware layers and translucency. Nicely done. Fractal mountains too! 32K bootable program size. 48K on chip RAM.
     
    I'll link up a video and EEPROM file here later when the results come out.
     
    Nice job Linus!
     
    The attached .eeprom file is for Propeller Demo Board. No source, until some reverse engineering happens.
     
    turbulence.zip
     
    Video is up now!
     
    http://www.pouet.net/prod.php?which=53003
  17. potatohead
    [dials phone]
     
    "Hi! Welcome to the automated help system! Press 1, if you need help."
     
    [presses 1]
     
    "If you feel better about it, press 0, otherwise press 1"
     
    [presses 1]
     
    "Hi! Welcome to the automated help system! Press 1, if you need help."
     
    [presses 1]
     
    "If you feel better about it, press 0, otherwise press 1"
     
    [presses 0]
     
    "I'm glad we could be of service! Acceptance is a powerful thing. Don't hesitate to call again."
     
    *click*
     
    I have this programmed into our phone system. Have been debating for quite some time whether or not I should take it live for April fools day. Each year, I pass...
  18. potatohead

    Serial Numbers

    By potatohead,

    The VCS I has as a kid was a Sunnyvale model. That one was lost and had to be replaced. One $5 craigslist dea, and a coupla hours doing repair, I've got a near perfect looking Taiwan heavy 6.
     
    I always liked the heavy models, just because that alum shielding is so damn cool! My Atari 400 is built the same way too. While fixing the heavy 6, I opened that up, just for kicks. Everything looks just fresh and new. Love that.
     
    So, I got to thinking how close was I to getting an actual Made in the U.S.A. video game? Turns out the answer is harder than I thought. I'm confused with all the prefixes, this number and that.
     
    All I do know is the replacement one looks exactly like the original one, right down to the switcher board.
     
    The number is 5xxxxx something. Don't have it in front of me at the moment, but it's in the low 500K range. Not interested in value, so much as the whole thing just got me to thinking how many were made here, as opposed to somewhere else?
  19. potatohead

    My 6 switcher works!

    By potatohead,

    In a move, I lost a fair amount of stuff. One was my original VCS from when I was a kid. I ended up scoring one for $5 bucks that needed a little attention. Some soldering later, and I'm playing KABOOM! Kind of missed that game.
     
    It's nice to have the machine back! Same exact model too. Silver Atari logo, funky wood grain, black.
     
    Now it's time to think about a mod. The RF really is horrible.
     
    After that, it's time to go 800XL or 130XE hunting, along with one of those cool SIO interface devices.
  20. potatohead
    Spent some of this snow building a software 6502 on the Propeller. Just musing about how fast it might be in assembly language.
     
    It's over 1Mhz for the demo case, which consists of mostly the longer instructions.
     
    http://forums.parallax.com/forums/default....33&m=310828
     
    The work now is to flesh out the CPU. Get the instructions coded, and work out the multi-COG instruction processing needed to have it run at speed, and do all the opcodes. Also need to do some assembly optimization too, but that can wait a bit.
     
    Demo case is a simple number counting program. Propeller runs it over 1Mhz, so that's Apple ][, maybe VCS speed. Not quite there for A8 bit speed however. Not to shabby for a little micro, considering it can do that, display a text screen to debug on, and still have 6 CPU's remaining.
  21. potatohead

    SNOW!

    By potatohead,

    Yep, we got lots of it. I now live east of PDX. Snow is better here for sure.
     
    Hey, did you know this "#" is called the octothorpe? I didn't. Now I do. Fun huh?
     
    Also, from the same source (some chat on a Propeller forum), tilde is also known as til. ---> "~"
     
    Up late, kids home, just surfing...
  22. potatohead
    Got a test, partial 6502 emulated, with a quick test code bed. This weekend, it's some assembler work to do a speed check, and figure out the details. Just by way of a comment, I don't think you really understand a CPU, until you work on stuff like this. I'm enjoying the experience. Pretty sure the Prop can get to VCS 6502 speed. Gonna find out real quick.
     
    Now, the coolness is this little project!
     
    http://www.parallax.com/tabid/708/Default.aspx
     
    The Propeller sits between the 6502 and 64K of static RAM. It also contains 32K of it's own internal memory as well, some of which is used to run the 6502, obviously. The Prop feeds the 6502, meaning it controls what happens on a read or write to the RAM, can do address translation and all sorts of other tricks!
     
    Several of the Propeller COG's remain, unused, meaning video chip emulation and other things can be done that would essentially make this a software configurable computer! Apple ][ emulation would be a no brainer. Just load the ROM's and setup bitmaps for the graphics pages. Probably could get this thing to run lots of other 6502 computer code, with only software to provide the I/O and chipset emulation on the Prop.
     
    I'm gonna build one of these, or buy a board, if he sells them!
     
    Could do cool tricks too. Ever wonder what a given computer would be like, if it had this feature or that, or perhaps a different CPU? You can find out on this bugger!
     
    ...thinking about the potential for this, if put in a cart. Lots of good stuff might be possible. Anyway, just fun stuff. Had to share!
  23. potatohead
    Despite the horrible times, I'm slowly gaining ground. We've whittled life down to the point where it's doable no matter what. As things wind to a close, short sale on the old house, medical, etc... my free time has slowly been coming up.
     
    On the Propeller, I've made progress on KABOOM! Think I've got the driver issues worked out. Dammit, it was a PITA to build the display on the fly. Should have went for the frame buffer... Oh well. Learned a lot. I think that will all be acceptable this week, meaning I can start on game logic and code to a game pad for now. Eventually, it's gotta be paddle, but that's another minor league project that I need breadboard space I don't have to work out.
     
    There is some talk of getting a 6502 core up and running. One COG just might be fast enough to hit the VCS speeds. So, I'm collecting docs and looking around at what's out there. Found a couple of them written in C, instead of x86 assembly. For some reason, I just have a hell of a time with x86. It's painful. The C ones look good, though. Makes for a nice start. One guy ran one through a filter, and left a rough SPIN translation on the net. Don't know if it works, but it's a start.
     
    The plan is to load up an Atari style text display driver. Standard 40 char, 8x8 display, then run some simple 6502 code on both the Prop and my old 400. Found my copy of MAC / 65. I've no storage medium at the moment. (lost in move, with my original 7800 and VCS...) Ok for short benchmark, testing code though.
     
    First pass will be my assembly "hello world" deal. That's counting numbers from 000000 to 999999 with a timer to see where the speed is at. I've done this on every platform that had a display capable of it, as my first step into assembly. Have all the pieces for that done on the prop, but for the CPU. Being on a little multi-processor is kind of cool. Basically, all that needs to happen is setup the display, fire off the CPU, and include the sample executable as a file to be compiled / assembled into the finished program image to be loaded. Easy stuff, from that perspective.
     
    What I can't see yet is how to make one chip talk to another in this environment. I think it's gonna be about having each one output it's state, all looking at a master clock, and each reading what it needs to read, while the others are writing what they need to write. If so, that's a small block of shared memory for a BUS of sorts. If not, well... I'll need some help. Probably need some anyway.
     
    Been looking at the 6502 from a different perspective because of this. It's a cool chip! (love the 6809 more, but it's not seen so wide of use...) I think I'm starting to understand an aspect of this era of computing that I didn't before. Looked at the TIA too. Same kind of coolness. I've a lot of admiration for those people who went the distance, grokked the schematics, and got their emulators up and running. I've got a lot more for those people able to build from that kind of environment too. Today, there is a lot of patch work, lego style development. The designers of these older chips just flat knew their shit, and got it done the hard way, with a drafting board, state diagrams and such. Kick ass.. I've got some schematics printed E size to look at. I know more about them now, and that makes them worth hanging up in the office.
     
    Maybe someday we can get a 4K 2600 game up on the thing...
     
    Pulled the 400 out and my newer plasma TV will display it. Sweet! I dimmed the lights, cranked the contrast on the thing, and played for an hour or so. Freaking love that game. Always will.
     
    That's about it. I've got a reason to pull the 400 out now. It's getting a bit senile, as some games will fail after 30 minutes of play. Probably bad caps or something. (Star Raiders always seems to work though.) That is the last original bit of hardware from when I was a kid. Well, that and the supercharger. It's gonna be craigs list, or here for some purchases next year. Hate not having a VCS. Emulators are ok, but there is just something cool about firing up the actual unit for some simple game action.
     
    (and It's killing me not to play Medival Mayhem! That's probably the best purchase I've made for the VCS. Homebrew rules. That is the work I show off right now to the unbelievers. Most of them are sold after an hour or so of game play.)
     
    Projects then are continuing to work on KABOOM!, and starting on the 6052 core. There are a few people interested in seeing that go, so it's not a solo thing.
  24. potatohead

    Settling in!

    By potatohead,

    Well, got the major stuff done! There is at least another weekend or two of hard work, but the core move is over!
     
    Feeling good about it already.
     
    That's it! Short and sweet. Hope to be retro-computing and gaming soon.
  25. potatohead

    Closing the Loop

    By potatohead,

    I'm currently writing up some how-to stuff for the Propeller. That's got KABOOM! and a coupla other things on hold, but that's not really what this blog entry is all about.
     
    Over time, I've been facinated with the idea of closing the loop on a conversation. What I mean by closing the loop is say you and I talk about something. I post this up, you read it, then maybe comment later on. That's a closed loop. Another example is somebody at the radio station says something and somebody else phones it in to talk about it. That's another closed loop.
     
    This thread:
     
    http://www.atariage.com/forums/index.php?s...=100537&hl=
     
    is a most excellent and valuable closed loop! Long ago Donald wrote Gaunltlet and many of us played it. We all had some stuff we would have liked to say to the author, but the loop just remained open for a lot of reasons. Recently, Don shows up and that loop gets closed up, friends made and experiences shared after all this time.
     
    It's one of the very coolest things about the Internet. If you've not read through that thread, you really should. It's been an interesting ride.
     
    We now have people exchanging info with Don, and each other, in the context of some shared experiences many years ago. What a wonderful way to learn, share and enjoy the hobby.
     
    Thanks go to Don for his game, willingness to share and most importantly, his getting something out of all his experience through sharing it with us, in the gaming context. I'm totally learning and enjoying the experience. Thanks Don, and thanks to the others giving that thread the energy it needs to carry on.
AtariAge
Forums
Store
General
Follow AtariAge

Copyright ©1998-2023 AtariAge

×
×
  • Create New...