Upload Smir 3, 1?

[Harry Potter]

Hi!  I have a text adventure called Smir 3, 1.  It is a Sci-Fi/Fantasy text adventure for the C64 where you must free your friend Talnon and other prisoners from a person from another dimension who wants the secret to fusion.  Currently, its two main problems are that it has a few typos and is nowhere near completion--only the first chapter is written.  If I finish it, I plan to port it to other platforms, including the Plus4, C128, Apple3enh and 8-bit Ataris.  Should I upload it as is?  Or should I wait?

[carlsson]

Make it playable, iron out bugs and typos and then perhaps release the first part. See if you get any attention, and then decide what to do with the rest. To me, C64/128 is just about the same computer except the C128 has more memory. I don't know how big the Plus/4 scene is, in particular for text adventures.

[Harry Potter]

Thank you.  So far, it appears that there are some typos.  They occurred because I manually compressed the text using tokenization, and occasionally I used the wrong token.  I currently see no other bugs in the code, but I only tested it quickly.

[Harry Potter]

The typos were worse than I thought.    I need to fix them before publishing the program..

[carlsson]

Sorry for the laughing reaction, but going from "ready to publish" to "oops, multiple typos and only tested briefly" in the matter of 15 minutes made me click on that one.

[Harry Potter]

I just fixed the typos in the first two rooms. 

[Virtualsky]

On 1/1/2023 at 4:42 PM, carlsson said:

Make it playable, iron out bugs and typos and then perhaps release the first part. See if you get any attention, and then decide what to do with the rest. To me, C64/128 is just about the same computer except the C128 has more memory. I don't know how big the Plus/4 scene is, in particular for text adventures.

I find that the C128's 80 column screen is much more accommodating to text adventure games.  Much more comfortable to read that way.

[Harry Potter]

Today, I fixed some more typos, but, when I type "get map," I get garbage, and the command doesn't work.  The garbage is in the form of actual text in the code but the wrong text.  I think the routine that prints messages is getting the wrong address.  Should I post the code that might be responsible for the printing?

[Harry Potter]

I fixed a lot of typos today, but I forgot about the "get map" issue. 

[Harry Potter]

I updated one of my text adventure codes today.  I have mainly some bug fixes and a text compression technique.  The option to disable string compression is currently not working properly.    PET support was improved: the character set is now changed properly.  You can find it at c65 additions - Manage /game at SourceForge.net.  The file is called AdvSkelVic65-011.zip.  The hi-end version online is currently not working properly.    I got most configurations to work but am getting the monitor on the 128k Plus4 version when I enter "get key."

[Harry Potter]

I just emailed the program to zimmers.net.  Let's see if he'll upload it.

[Harry Potter]

Good news: I just optimized it by converting the code that returns whether there is a match between a noun given and the nouns in the database to assembler and gained .04k optimality. 

[Harry Potter]

Smr 3, 1 is doing better: I've been fixing the typos, but now, I am having a problem: the code that displays a standard message is giving garbage.    It is currently in assembler, but the C version works properly.  Also, the code is being put in zeropage, and when I move it to the default code segment, it gives me a blue screen with no text.    Following are the C version of the function:

------------------------

static void __fastcall__ printmsg(unsigned char m)
{printtok ((void*)Message[m]); printcr();}
---------------------

and the assembler version:

-------------------------

.segment    "ZP2CODE"

_printmsg:
    asl
    tax
    lda    _Message,x
    pha
    lda    _Message+1,x
    tax
    pla
    jsr    _printtok
    jmp    _printcr
 

[carlsson]

Why do you end up with code in the zero page to begin with? Is Message located higher in memory, so it renders a 16-bit address? I'm thinking if you end up with LDA $NN,X and X wraps around zero page.

[Harry Potter]

Actually, because I put it there.  It's part of my Cubbyhole optimization technique.  The game uses some ZP for variables, and the unused section of ZP is used to store extra code such that it is not included in the main file.  Message is an array of pointers to strings that's located in main memory.

[Harry Potter]

Yesterday, I fixed the bug in and heavily optimized the text adventure yesterday.  The bug was that the version of my CBMSimpleIO's printu() function had a bug in it that was causing corruption of two zeropage locations.  I replaced it with another version, and now it works.    Now, I need to continue to fix all the typos.

[Harry Potter]

BTW, I recently updated my AdvSkelVic65 program.  WHile you're waiting for another version of AdvSkel65, you can try AdvSkelVic65.  It supports the PET, the Vic20, the C16 and cartridges for the Vic20, the C16 and the C64.  It is at c65 additions - Manage /game at SourceForge.net.  Try it out!  If you like it, I ask for ways in which I can improve it. 

[Harry Potter]

I just updated my AdvSkel65 program.  Its main updates are heavy source code clean-up, optimizations and debugging.  I also documented a feature allowing room text to be printed depending on conditions such as a task done or a player's possession.  It is in the same location as AdvSkelVic65.  Try it out!

[Harry Potter]

I optimized Smir 3, 1 today, even though I need to fix a lot of typos first.

[Harry Potter]

I recently optimized my text adventure codes and knocked the 64k C128 version of AdvSkel65 down to 3.06k.    I also fixed a bug with AdvSkel65 where the C128 versions didn't display the score board properly.

[Harry Potter]

I have an update to AdvSkelVic65 online.  It now supports C64 and Atari 800 floppies and one-part programs for supported CBM targets other than the PET.  It was also heavily optimized.  It has support for text compression, but such support is limited and requires manual compression.    You can disable it if you want--the instructions are in the manual.  It is in the same location as the earlier versions.  I actually came back here because there's a bug in the function of my text adventure: it's not handling spaces properly: when I press Space, the emulated cursor disappears, and no new text is displayed until I press Return.  The main program loop doesn't see the text after the space either but sees the verb.  I attached the function here:

------------------------

unsigned char __fastcall__ GetInput (char* In, unsigned char InLen)
{
    //unsigned char i;
    CurPos=0;
    printc ('>');
    while (1)
    {
        printc ('_');
        i = getkey ();
        printc (20);
        if (i==13){ /*return*/
            In[CurPos]=0;
            printcr ();
            return CurPos;
        } if (i == 20 && CurPos) { /* delete */
            --CurPos;
            printc (20);
        } if ((i&0x60) && CurPos<InLen) {
            //In[CurPos]=i;
            printc (In[CurPos]=i);
            ++CurPos;
        }
    }
}
---------------

In is the array that receives the output, InLen is the size of the buffer, excluding the NUL,i is a temporary ZP unsigned char variable, CurPos is also in ZP and reports the current pos. in the buffer, and the use functions are from my CBMSimpleIO library.

[Harry Potter]

I found the problem: silly me!  My emulator was in Warp mode.    It works now.