All Activity

I’m afraid there aren’t any left. PM me if you want to be on the waiting list in case of drop outs/potential next run list.

I see a few interesting things there. As already noted, the connector is designed to connect to the TI sideport--and has sawed pins 1, 2, 43, and 44 off of both connectors. Note that the card edge numbering is marked on the board, with pin 4 on one end and pin 42 on the other (closest to the RS-232 section). Note also that the board has two TMS9902s installed, likely one to drive each port, but one would have to carefully follow the traces to be sure. The memory chips are 64Kx1 DRAM chips, so the board has 64K installed, but it also seems to have sockets for another 64K of memory. That part of the board may be a partial sidecar implementation of the 128K memory card for the /4A. I have a bare PEB board of that one, along with the schematics and other documentation. @urbitealso helped me a bit with this one ages ago, and reconstituted the PALs for it. I have a "mostly" correct layout done for the PEB board, but it has a small number of corrections that it needs before one could actually test it out. That side may also be a sidecar implementation of the 64K ExtraRAM board as most of the chips match, although it seems to be missing the PAL. Mike Bunyard did the designs for the 64K ExtraRAM and the 128K SuperRAM card. Lastly, it looks like this board was mostly designed for this specific purpose, and fabricated as a one-off. Note there are a couple of spots that seem to indicate a board revision/date on them. TI 64K ExtraRAM Card PHA1268.pdf TI-99_4A 128K Memory Expansion Card PHA 1265 Schematics and Logic V2.pdf

Just for verifying your theory of course.

I usually do an AND #$3F (in your case, since the range is > 32 and <= 64) first (method #4). And then check if the value is in range size, else repeat (method #2). Since this can repeat very often, I usually add a counter for maximum repeats. If that doesn't help (or I skip the repeats) I simply execute an LSR. Add the end, the low number (14) is added. Yes, this means that the lower numbers occur more frequently, but depending on the range and use case, this is often not noticeable. I have found that in most situations, a certain degree of imperfection is perfectly acceptable. BTW: Theoretically a LFSR generates only one new bit per execution. But usually I use it, as if it generates a new number every execution.

Sorry you don't like the game(Ive never played it),the label is great! I used to love watching the muppet show as a child,Piiiiiiigs iiiin Spaaaaaaaaaaace 😀

LINES, the only lies are the spelling. I guess it could sit proud by about 10mm to fit the circuit board and components. Regards Arto.

HELP! I'm looking for the best way to get a random number between an upper and lower bound! EXAMPLE: Let's say I need a random number between 14 and 54. METHOD 1 - Power-of-Two Masking: Most examples I've run across use the method of adding up decreasing masked random powers of 2. For my example above you would be seeking a random number between 0-40 and then add 14 after to get 14 to 54. To get that you would then break down 40 into four AND'd 8 bit random rolls of 0-31 + 0-7 + 0-1 + 0-1, which together would result in a range of 0+0+0+0=0 to 31+7+1+1=40. I've read that problem is there an uneven distribution, with less results on the far ends and more in the middle, which is not great "randomness". METHOD 2 - Rejection Sampling: Another solution I've seen is to keep re-rolling the dice until it lands in the range you want. Using the same example above of 14 to 54, you would keep re-rolling a random number until it would fall within the range of 0 to 40 and then add 14 to that. I don't really find this a good solution as you could theoretically, but unlikely, be rolling forever. You'd also have to statistically roll more times for a smaller range. METHOD 3 - Iterative Subtraction: I think this is easiest and mostly fastest method to implement. Let's use the same example as above between 14 and 54. You would again subtract 14 from 54, giving you 40. You then roll for an 8 bit number and and keep subtracting 41 (40+1) until you're between 0 and 40 and then add the 14 to it. Say you roll a 133. You would then subtract 41 over and over until you're below 40. 133-41-41-41=10. Then add the 14 and you get 28, which is between 14 and 54. Working with the extremes of this example: You could roll an 82. 82-41-41=0, 0+14=14 and you've got the lower bound. You could roll a 204, 204-41-41-41-41=40, 40+14=54 which is the upper bound. METHOD 4 - Modulo Method? / Shift & Subtract?: I found this code generating website (rescued by archive.org), linked on this AA forum post, for solving my EXACT problem and it only calls the random subroutine once, just like in method 3. The code is also so incredibly compact and fast but it's also specific to only one case. Another problem is that I don't understand the code or the formula it generates. I've tried to look up the Modulo Method for 6502 processors giving you a remainder after a division, which is what I want, but I can't wrap my head around it. Is this the method used here? Now, given all four examples, I have some questions: 1) Is method 3 statistically sound that it's an even distribution across every range you might be looking for? A major downside I see is if you have a very small range and roll a very big number you'll be doing a lot of subtraction over and over. Say you need between 24 and 27, a range of 4 numbers, and you roll a 255, you'll be subtracting 63 times! 2) Is method 1 "close enough" to use in a game? I believe the times you'll be running the random number subroutine would be 8, which could eat up quite a lot of cycles! 3) In method 4 is there a generic way to implement the Modulo Method, if that's what this is, on the 6502 rather than using this website to generate specific code for a set range? Thank you so much! - James

LOL, I was trying to light a fire and your hands are full. I'm going over strategies. @ImaginaryRulebook has a good start with his templates for C64, Ill try and focus on A800 ans AST. For shits and giggles, I see theres a "puae_libretro.so" core on github. I dont know if its stable or not (probably not or you would have done it) but that might be first/next.

I did try, but it looks like I might have to swap out a few of the main chips to see what the problem really is. I have been too busy with the other projects to worry about it at the present. Thanks for asking, I will post when I see breath and a glimmer of life from the 99/22. Regards Arto.

What are the board's dimensions? Could it have been proto'd to fit in a sidecar case, with the connector for an external power supply.

Yep I entirely missed that EPROM, must've been too interested in the rs232 features:)

Already been said - it's an RS232 card with 64K of DRAM. I guess the whole left section is the refresh circuitry. Why do you need RAM on an RS232 card? Print spooler. Looks like it has a second DB-25 pad, but maybe the first one wires both ports (Y cable needed) like TI did. The E/A manual draft is pretty interesting, I wonder how many more comments it has. There are just a handful of people at TI who might have been reviewing it. I wonder if one is the builder of this RS232 card? One comment looks like "I'll send these to you on the Wang". Wang made word processing computers in the 1970s-80s.

@nobru Thanks for the copy bank 7 tip. I got Galagon working, but not RobotWar. Is your RobotWar version 3 or a different version?

I don't die in any of my runs. If I've died i know im not setting a high score. Its all about the battle with the clock... and the bats. Those dang bats!

Frankenstein's Monster: 3050 OMG! That was intense.

oh boy, it'd be interesting if they were shipped with v1.1 and the site has v1.0, and since nothing on the site talks about the version numbers then people thinking that they'd be upgrading their new unit will actually be downgrading them Thinking about the timing needed for completing them, packing them, and getting them shipped within Feb, and the post from Ben on Feb 3 about completing v1.1, deeply implies that the new run is also using v1.0 software. No biggie, but the claim, "it's disappointing that a system that is now on the 2nd manufacturing run is still running off of Alpha firmware" is still not correct. All the units were shipped with "official" software, and if someone wants to install the v1.1a or the v1.2a, then that is their choice. The v1.1a and v1.2a are Quality of Life improvements. The Sprint seems to work just fine on the original software, and no one is forcing fiudr, or anyone, to use Alpha software on their Sprint. Alpha means not ready for release, so if there are concerns then stay on the officially released software until there is another officially released version.

Yeah I know, just needed to get that off my chest. 🤣

Sounds like a great user made project 😁 😉

@EDFCentral Hi, did you get this resolved? I just tested my Haunted Adventure Trilogy cart from AtariAge in my Atari 7800+ with firmware version 2.0.1.6 (12/25/25) and it works. Were you able to test your cart with the newer firmware or an original 2600/7800? If not working, did you get a working replacement from the AtariAge store?

Yeah, joking, "that's the ticket." Joking

I believe I have everything supported in cfw 2.0 beta2 running now except ScummVm. Just havent gotten to it yet. When I say running, that goes for all consoles and handhelds, all decently/incredibly playable. Thanks Guys, You Rock!!! Computers all run well when you can control/start them, but are hit or miss with control allowing them to be playable. They look nice though! Atari ST's mouse even works (Star Glider), its a bit touchy but fixable. it will be a daunting task to set up all computer games individually so I would propose making, for each system, a template .opt and .sh for different genres like i.e. "basic 2 button games", "4 button games" "side scrollers", "run and gun" "Ultima style" "games that need controls extended to the keypad" "Star Raiders, Rescue on Fractalus" (those may all overlap somewhat but you get the picture) Some games would still need some minor tweaking after that, most likely. Then with a "templates made, just copy them and rename them to the appropriate control style of the romname. Other more complex games are gonna take some work

I'm still using FB4 & FB7, no problems here. For a while, I had a 9, because I was converting paddles for it.

Survival Run 475