All Activity

Tried RYMcast, that didn't work. Tried VLC, and apparently it requires you to build a version of the program with the Game Music Emu, rather than just letting people throw the software into a plugins file. I think there was one other web-based VGM player that was no help at all. Honestly, I think some of these files are corrupted, or never worked in the first place. Some of them are maybe a couple of K. Others are... 18K? Something goofy happened, I think. A pity too, as I really liked those sounds! Can you test this with your equipment? Tell me if the files are valid on your machine. neat noises.zip

Well, I suppose that pennies count as money, too, so you're essentially correct. Like a pump & dump. You're special.

Problem is there are soooo many good games. I would have preferred a 1st, 2nd and 3rd choice option for the voting....... 👍 EDIT: Maybe we could have used a points system like 100 for 1st place vote, 50 for 2nd, 25 for 3rd.... Add up all the points at the end... But obviously the anonymity in voting would have been lost...

Back in the day om my Mega ST modified with Atari AJAX(8,16Mhz) chip high density floppy drives (One internal and One external) I needed to provide extra wire to it through a small 3mm jack.

NOTE: I did not test this yet I am assuming TOS4.4 has support for this. On Falcon there is no external floppy port but there is possibility of connecting second floppy internally and wiring the one missing gpio pin from Yamaha sound chip.

Updated rule 4 to account for the longer poll duration.

The panel of the contest composed of one person (me) has decided to extend the poll for one week more, in order to allow another weekend for people willing to test and vote. So the new closing date is May/14/2024 23:59 GMT

Highest Cosmic Chasm score ever? (as far as I know) Played for about 3 hours tonight and found a really great strategy for maximizing points, but also this last game I guess I was finally warmed up and played really well in addition to having a good strategy, and this is the result! 90,320 The really cool part is I got it on video as well, and will upload it to my youtube channel at some point

I guess first step is to let us know one of the players you are trying - give us the player and one of the files you want to try, and I'll take a look. That said, I wrote it for input to my compressor (and because VGM is a very easy format to parse and convert). So all the tools at https://harmlesslion.com/software/vgmcomp2/ can be used to import and manipulate these... but you'll need to be comfortable with command line tools. I tested the VGM files it generates with my toolset but also with AudioOverload and Winamp's VGM Input Plugin. VGMComp2 may or may not be overkill for your Coleco project if you're just playing short sound effects, but if you built them all up into a single sound bank you'd probably benefit some from the compression. If they are playing but just not playing correctly - there are some sounds that tool can generate that can't be played back on the SN at 60hz, which my export limits to. We'll have to see what you have.

I was going to do a SI clone until I saw source. Still haven't found a binary. Then I decided to do this, and committed. Then found MicroVaders Damnit! If anyone is planning a Breakout/Arkanoid clone, pls let me know, because I have ideas for that

Very cool. Thanks. Should I enter 340 seconds or do you want to run 10 iterations with "1899 primes" printed after each iteration.

haha... we'd get better results. I can run several of the large language models locally and they come off pretty similar to the online ones, so I don't think the chatbots are fooling us - at least not the older ones. And any human to create code in a language as old as TI BASIC as quickly as they do -- even someone from here -- feels unlikely.

**** UNICORN DANCE PARTY! **** The game's bonus round is now partially implemented and fully playable (and without using any extra RAM)! Welcome to the Unicorn Dance Party! You'll trigger a Unicorn Dance Party in Arcade Mode after successfully completing two levels, and then every three successful level completions thereafter. The object is to hop around the grid, trying to land on the lit square. Each time you do, you'll earn 1000 points, and you'll have 2 penalty points deducted. So if your penalty count gets high, here's a chance to bring it back down again! Joining Snuggy in the Unicorn Dance Party are his little sisters, Snowy and Rosey. One's a bit faster than the other (I'm too lazy right now to look at the code and see which one I made a tad quicker), but other than that and their color, there's no difference. Snuggy's a bit of a clumsy dancer, so you need to avoid bumping into your sisters. If you do, the music will speed up, and they'll turn red and run off. Be careful, as Snowy and Rosey aren't quite like the other enemies. Sometimes, they'll randomly reverse directions, making it a bit tougher to dodge them! The Dance Party ends when the music has cycled through 3 times, or you have bumped into your sisters 3 times, whichever comes first. You'll notice the top grid disappears during the bonus round. That's because something else will be going there once I'm done with it. But for now, the game is completely playable without it. I've made some other minor changes, and it looks like I've solved the problem with the Game Over tune not playing that I mentioned in an earlier post. I hope to have this game complete by early next week, then get it into the hands of my playtesters. If you'd like to be a playtester for this game, message me and let me know! I'll have to have a quick turnaround time though, and I need to get this thing all wrapped up by June so Al can get it ready in time for for PRGE! Frazzled_20240506.a Frazzled_20240506.bin

I saved 34 VGM files from that site, and none of them want to play on VGM players. Help!

Now I have to stay up until I get 31 😂😂

Put me in for Chivalry too.

So I was busy looking for information on old Win 3.1 games, and came across a reference to a cancelled "Text Adventure Maker" for Antic from 1982, by a 15 year old Brian Goble. A little more research found this interview ( https://blood-wiki.org/index.php/User:Tchernobog/archive/boodint/goble ) in which Brian says: "I eventually got an Atari 800 computer and played a lot of games on it and also programmed a lot of games (and I could actually save them-but don't ask about the battles between me and my cassette recorder). At one point, I had written a "Text Adventure Maker" program that let you make text adventure games. It was fairly cool so I submitted it to an Atari magazine called Antic. Then, I received the letter that would change my life forever. Antic replied saying that they had accepted my program for publication in their upcoming book of games and that I would get a small percentage of royalties from the book sales. This was very exciting for a 15-year-old and I was bouncing off the walls. Now that I knew I could make money from programming, I had found my destiny and from that day forward, I only focused on writing programs that I felt had a chance to make money. (the book of Atari games eventually got cancelled and all I really got was a free T-shirt...but, the course of history had already been changed)" Do we know what this proposed Antic book of games was? Is there any chance any of the content is preserved somewhere? It could be, and I'm just unaware of it, but this was interesting...

I wasn't able to find anything related to using a modded GBS8200 with a composite modded 7800, so I figured I'd provide a description of my experience. Originally I was using one of those AV2HDMI converters but I was really dissatisfied with the processing of 280p as 480i. Purchasing something like a RetroTink is just too expensive. After Canadian customs, shipping and exchange rates the 2X Pro is more that $200 CAD. So after a bit of shopping around I found an 'OSSC Add-on board' on Amazon that transcodes composite to component with the option of line-doubling to 480p for about $35CAD with next-day shipping. This solution alone worked great for my Television which as component in. However, I want to use the 7800 with my desktop monitor, so I picked up a GBS8200 for about $30 CAD and modded it following the GBS-Control guide. I did not bother with the external clock. This gives a nice solution to output resolutions that work with my monitor: 720x480 or 1920x1080. My monitor is 1440p, so I'm using 720x480 with the scaler on the GBS-C tuned to give an integer scaled 6x image. I gotta, say for composite the picture looks very nice and I enjoying the composite blending with games like Rikki&Vikki. The whole package costs me less than $70CAD, which is significantly less than the cheapest RetroTink and is giving me a better picture than the 2x Pro. I've only run into 2 games that gives me trouble: Asteroids for the 2600 and the cut scenes for PacMan 8K. Asteroids will very briefly lose sync semi-periodically which is annoying and it's only the title and cut-scene from 8K that are lost, so I think I can live with the results. That's my experience so far, and maybe someone else will find this useful.

It is kind common in many cases of using blwp routines to have a data statement right after which is used to pass the parameters to the routine from the main code. In this case the vdp screen address for the input line, the cpu version of what you want displayed and edited, then finally the length with if you want a starting offset useful for path entries. For example: ....... example BLWP @KB DATA 5x40,path,>0520 ...... path TEXT 'DSK3. ' So above would display path string on line 5 of the screen if 40 column mode and have the cursor blinking after the period . With a maximum input length of 32 characters. For allowing you to input a filename, etc. and if you want backspace to change the dsk3 part. And his routine handles all the various fctn keys for insert, delete, etc.

I, for one, would be happy to have a RAM card for my 74 and 95. It seems impossible to find an original one these days.

Something that I notice with Jim's code that I never realized before, is that pulling multiple parameters from another workspace is the same amount of code you would use if you made a stack. MOV *R14+,R0 RETURN = R0 contains length MOV *R14+,R1 of input MOVB *R14+,R3 MOVB *R14+,R2 I have a POP macro in my assembler that emits this same code with: R0 POP, R1 POP, R2 POP, R3 POP,

Thanks again for your help and advice. I am certainly continuing to look for and scrutinize old trade show photos and videos in the hope of finding something to prove your theory. For sure, I really want to solve the mystery. Or perhaps at some point someone will see this thread who either owns or has seen another one of these "plastic posters," since the video game store owner whom I got it from has confirmed to me that he had a few others of them in the late 1990s. If I can find any more info, I will certainly update this thread. I'm curious though about your opinion of its value. If I can definitively determine its provenance, especially if it really was used by Atari at trade shows, you don't think this item would be valuable to a serious collector? I bet it would look super-cool displayed on a wall in an actual lightbox! If it's not really valuable that's OK too; I'll be happy enough just to finally be able to prove its origins.

A while back I went chasing the goal of making my VDP driver faster. I made it faster at the expense of about 100 extra bytes from what I can see. I have decided 100 bytes could be used for better things. I simplified my scroll code by a few bytes. It is about 50 bytes to scroll the screen in chunks versus about 100 bytes in Assembler, but quite a bit slower. The speed can be tweeked by using multi-line buffers. 4 lines is about the optimum compromise. The smallest scroll is fast and only 34 bytes, but uses a full screen sized buffer which eats into code space when memory near full. In the process of trying a lot of different options to reduce the size of the code I found one little word that is 6 bytes in Forth but only 4 bytes in Assembler and I used it twice. The word lets you increment a variable and return the new value. This is used as you bump the column and row values when you write a character or go to a new line. \ : ++@ ( addr -- n) DUP 1+! @ ; CODE ++@ ( addr -- n) *TOS INC, *TOS TOS MOV, NEXT, ENDCODE I went down the rabbit hole to trying to make an EMIT word that operates the way we do in things Assembler, using auto incrementing indirect addressing and VDPs auto incrementing. In theory this should be faster but in Forth the extra setup required, the slower loops and Forth tests for scrolling, bought me very little improvement on the SEVENS benchmark. Here is how TYPE looked using that idea. *Notice how many Forth words are in the name DUPC@EMIT and it is one instruction in the 9900 instruction set. Wow. CODE VDPWA! ( Vaddr --) WMODE @@ BL, TOS POP, NEXT, ENDCODE CODE DUPC@EMIT ( addr -- addr') *TOS+ 8C00 @@ MOVB, NEXT, ENDCODE : TYPE ( addr cnt --) PAUSE VPOS VDPWA! \ set vdp address to cursor position 0 DO DUPC@EMIT \ read byte and write to screen VCOL ++@ \ update video column and return new value C/L@ = \ compare to columns per line IF CR THEN \ CR if needed LOOP DROP ; But the idea is still pretty cool for writing fast to VDP from Forth. In theory I could use it to make VWRITE (VMBW) in Forth and save more bytes at the expense of speed. : VMBW ( addr Vaddr len -- ) SWAP VDPWA! 0 DO DUPC@EMIT LOOP DROP ; \ 14 bytes \ versus 30 bytes CODE VWRITE ( RAM-addr VDP-addr cnt -- ) TOS R0 MOV, \ cnt in R0 TOS POP, \ vaddr to TOS R2 POP, \ cpu addr to R2 WMODE @@ BL, R3 VDPWD LI, \ vdp addr. in a reg. makes this 12.9% faster BEGIN, R0 DEC, OC WHILE, *R2+ *R3 MOVB, \ write byte to vdp write port REPEAT, 2 LIMI, TOS POP, \ refill top of stack cache register NEXT, \ return to Forth ENDCODE So in the end the spoiler has the console driver I will commit to in the next release. It's fast enough for most purposes. and if you really need fast typing there is a VTYPE library file that just blasts text to the screen with VWRITE and only updates the column variable. It doesn't scroll.

Hi! Not at the moment - for simplicity I kept it as before, adding the clearing of $12 would conflict with the code for the 5200 (that only uses the integer implementation). I documented in the manual that you should not use %TIME and TIMER together. Have Fun!

Yes, the price point is more than fair, no issues with that. My only gripe is with the randomness or blind luck element involved with trying to purchase one. And that’s not impatience talking on my part either, as I don’t care if I missed the first batch, or end up missing the second, or even the third, etc. Just worried that this becomes an item that is always sold out for months on end and is never easily available to buy as I absolutely MUST have one of them by this November for a yearly video game show that I participate in.