A new commercial game for MS-DOS

[JamesM]

Sorry, but that rationale while great for a hobby project does not a commercial venture make.

[Antti]

I should have avoided using the word "commercial" because it clearly invoked strong feelings. More suitable expression would have been something like "not-a-hobby project". However, the overkill plan of manufacturing boxes, manuals etc. would have required the real "commercial" aspect.

Why "not-a-hobby project", even it really would be a hobby? The rationale behind this can be very loosely be compared with the music and recordings. If I made an album with my band while doing all the production independently, the end product would be quite easily categorized as a "demo tape" even the production quality being very high. Meanwhile, the other band signs to a record label and releases a professionally produced album. Even if our "garage record" would be better in every field, it would be even hard to get a Wikipedia page for it (unless it being widely known). The other band gets their album to the public databases of published records, have a plausible ownership for their band's name etc. All the small things like these. Nevertheless, this is going to change in music industry. It has already changed.

The above comparison applies when the both bands are small and not widely known. No need for bringing out widely known indie bands that have gained all the things listed above.

As a conclusion, The Point I have been hunting down is that the "not-a-hobby" product will have more easily a reference in "history books". It is not The Point whether this reference is noticeable in this book or not.

[turdus]

Imho you should use some kind of scripting engine for your goals, like ScummVM. You can include a plugin for your data format if you like, or use one of Lucas Arts'. This vm runs old MS-DOS games (among others), so you'll have the same feeling without coding a single real mode code byte. Still have to design the game and look-and-feel though, which will be costy.

[Antti]

Imho you should use some kind of scripting engine for your goals, like ScummVM. You can include a plugin for your data format if you like, or use one of Lucas Arts'. This vm runs old MS-DOS games (among others), so you'll have the same feeling without coding a single real mode code byte. Still have to design the game and look-and-feel though, which will be costy.

My problem is that these real mode code bytes and programming techniques are interesting by themselves.

[sds2017]

Your idea is good for the most part. IMO here's what I think should change.

high-quality gamehigh-quality game

If your using DOS on a physical machine well that statement logically can't be possible. Although if you use it on a emulator such as DOS box you can make the setting high. That's what this guy did and he even used OpenGL! http://www.youtube.com/watch?v=vkUwT9U1GzA

The word "retro" is _not_ used anywhere

Well sorry but the odds of that project working are low unless you ported to a phone or tablet like posts have been saying. The reason is DOS is a legacy/retro operating system therefore most people who use it retro lovers which BTW there is a big market for retro games.

The open-source-hobby-project feel and appearance should be strictly avoided

Well unless you are going to port it to a phone/tablet people aren't going to wan't to spend the money. Also why not make it open source because there are plenty of commercial open source apps, games, and os's. ex. Red Hat.

[Antti]

If your using DOS on a physical machine well that statement logically can't be possible. Although if you use it on a emulator such as DOS box you can make the setting high.

There are so much more than a resolution or effects that define "the high quality game".

That's what this guy did and he even used OpenGL! http://www.youtube.com/watch?v=vkUwT9U1GzA.

I liked very much the pixelated look of the walking scene.

Also why not make it open source because there are plenty of commercial open source apps, games, and os's. ex. Red Hat.

As much as I like the open source in general, I did not include it in this concept. However, the open source game engine ports would be reasonable. The game data is the actual proprietary content and how the data is interpreted is secondary case. However, for me it is primary because I am technically oriented. A little bit confusing?

[turdus]

There are so much more than a resolution or effects that define "the high quality game".

Agreed. I like the old games better. They've provided a low quality visual, that's true, but that was compensated by marvelous game ideas and unbelievable programming techniques.

As much as I like the open source in general, I did not include it in this concept. However, the open source game engine ports would be reasonable. The game data is the actual proprietary content and how the data is interpreted is secondary case. However, for me it is primary because I am technically oriented. A little bit confusing?

No, not confusing at all, and it's not your idea. That's exactly how ScummVM work. And pretty much that's how every other Linux game work (just a few example: Doom, Quake, or Duke Nukem 3D. All of these howtos share one thing in common: "copy game data from purchased DOS version to your Linux box").

What I was trying to tell you with the ScummVM example is that your goal does not fit in a profit oriented project. There's no point in expecting money on a end-of-life OS. It's only the non-profit community behind that keeps 16 bit running. Even if they won't be pissed at you, and they are willing to buy your game, it's a not a big share of the market, isn't it?

[Antti]

No, not confusing at all, and it's not your idea. That's exactly how ScummVM work. And pretty much that's how every other Linux game work (just a few example: Doom, Quake, or Duke Nukem 3D. All of these howtos share one thing in common: "copy game data from purchased DOS version to your Linux box").

What I was trying to tell you with the ScummVM example is that your goal does not fit in a profit oriented project. There's no point in expecting money on a end-of-life OS. It's only the non-profit community behind that keeps 16 bit running. Even if they won't be pissed at you, and they are willing to buy your game, it's a not a big share of the market, isn't it?

It was not my idea but I would had ruthlessly used the same approach. My idea (hopefully) would had been to create "a classic MS-DOS game" in this decade. I did mention the "like abandonware" aspect so basically it would had been free to use (but still proprietary). Some commercial success might had happened in phone/tablet ports. The MS-DOS game just being the start of the game brand.

However, because the lack of my business skills, my company would had bankrupted and the game would had become _real_ abandonware very soon.

[JamesM]

The most profitable way to create "a classic MS-DOS game" is to write it in Objective-C, deploy it on iOS and rake in the millions.

There's no reason why the user's interaction with a system dictates the underlying implementation.

[MasterLee]

Cross-platform library for windows, mac, linux is fairly easy due to POSIX and OpenGL; crossing with DOS is non-trivial, DOS simply has nothing. I'm not saying it is impossible, it's actually doable and not difficult, but I'm certain it double the workload.

Anyway, good luck.

It is possible to use SDL with HX Dos Extender. (On this way you could even run DOSBox in DOS)

[Antti]

It is possible to use SDL with HX Dos Extender. (On this way you could even run DOSBox in DOS)

Running DOSBox in DOS with HX is not that senseless as it might seem at first glance.

That's interesting. If I wanted to do a system just for playing MS-DOS games with moderd hardware, this would be worth considering to. It would probably boot very quickly.

Nevertheless, maybe the Linux kernel with a minimal userspace environment and with Dosbox running without X (using DirectFB), would be possible to customize to be faster and more elegant.

[AndrewAPrice]

I still have a soft spot for many games from the late DOS era (Theme Hospital, Pizza Tycoon, Ignition, etc). The main limitation would be graphics acceleration, but when dealing with low VGA resolutions (VGA Mode X in all its 256 colour glory) you could do some pretty intense software rasterization on modern hardware, especially if you use a 32-bit extender.

I think you may be better off targeting a modern OS (Windows, OS X, Linux) because of four reasons:

• DOS programs have a fairly limited audience nowadays. 64-bit Windows cannot run real-mode DOS programs at all. Setting up DOS Box or rebooting into another OS is too much fiddling around, that the general user will consider it too much effort to give your game a try, no matter how awesome it is.
• Once the game is running, who cares what OS it is running on. The game is full screen and has the complete attention of the user, not the OS underneath it.
• Think of all the hardware support that is given to you for free by libraries such as OpenGL/OpenAL and the OS that we take for granted: Graphics accelerators, sound cards, USB mice, network adaptors, etc.
• Why does DOS appeal to you? Do you like to fiddle with your bits and draw pixels yourself? Use a library like PixelToaster that gives you that freedom. Do you like the low resolution 320x240 256 colour look? Nothing stops you from using a low screen resolution with OpenGL, SDL, PixelToaster, etc.

If the reason you want to target DOS is because you want to return to the olden days of game development with small teams and small budgets (because you can not afford to pay 20 3D animators, 5 level designers, and 15 programmers for 4 years) then I suggest you look towards the indie game movement where small-team small-budget 2D games are still strong and living. For example:

Braid:


VVVV:


RPG Maker:


Also popular is the 2D-in-3D pseudo-voxel look.

Fez:


Minecraft:

[freecrac]

Another challenge with write DOS game on modern machine is drivers.
You need ...mouse.....

I use the bios option "USB lagacy enable" for to become the mouse-informations of a USB-mouse like a PS2-mouse from the keyboard controller. (No USB-driver needed for an USB-mouse using DOS.)
Similar to the Cutemouse driver we can get the movements and button states from a PS2-mouse with a little help from our bios with a PS2 mouse handler that we can initialize to start.

Code: Select all

START:

CHECKPS2: int    11h                  ; get equipment list
          test    al, 3
          jz  short NOPS2             ; jump if PS/2-Mouse not indicated
          mov     bh, 3               ; data package size (1 - 8 bytes)
          mov     ax, 0C205h
          int 15h                     ; initialize mouse, bh=datasize
          jc  short NOPS2
          mov     bh, 3               ; 03h	eight counts per mm
          mov     ax, 0C203h
          int 15h                     ; set mouse resolution bh
          jc  short NOPS2
          mov     ax, cs
          mov     es, ax
          mov     bx, OFFSET PS2TEST
          mov     ax, 0C207h
          int 15h                     ; mouse, es:bx=ptr to handler
          jc  short NOPS2
          xor     bx, bx
          mov     es, bx              ; mouse, es:bx=ptr to handler
          mov     ax, 0C207h
          int 15h
          ret

NOPS2:    stc
          ret
;----------------------------------------------------------------------------
PS2ON:    call PS2OFF
          mov    ax, cs
          mov    es, ax
          mov    bx, OFFSET PS2IRQ
          mov    ax, 0C207h           ; es:bx=ptr to handler
          int 15h
          jc  short NOPS2
          mov     bh, 1               ; set mouse on
          mov     ax, 0C200h
          int 15h
          ret
;-------------------------------------
PS2OFF:   xor     bx, bx              ; set mouse off
          mov     ax, 0C200h
          int 15h
          xor     bx, bx
          mov     es, bx
          mov     ax, 0C207h          ; es:bx=ptr to handler
          int 15h
          ret
;----------------------------------------------------------------------------
 org START + ((($-START)/16)*16)+16 ; we want a code alignment
;----------------------------------------------------------------------------
PS2IRQ:   push    ds
          pusha
          mov     ax, DATEN
          mov     ds, ax
          mov     bp, sp
          mov     bx, [bp+22+6]   ; status byte(buttons,Sign,Overflow)
          mov     cx, [bp+22+4]   ; X movement
          mov     dx, [bp+22+2]   ; Y movement
          mov     ax, [XACK]
          test    bx, 10h         ; Sign X  Mouse goes right
          jz short MOGOR
          neg     cl
          sub     ax, cx
          cmp     ax, [BHMIN]     ; (Hmin) Mouse-pointer to much on the left side?
          jb  short IY0
;          test   bx, 40h          ; Overflow X
;          jz  short IY0
IX1:      jmp short IX2           ;  write X
;---------
MOGOR:    add     ax, cx
          cmp     ax, [BHMAX]     ; (Hmax) Mouse-pointter to much on the right side ?
          ja  short IY0
;          test   bx, 40h          ; Overflow X
;          jz  short IY0
IX2:      mov     [XACK], ax
;---------------------------------
IY0:      mov     ax, [YACK]
          test    bx, 20h         ; Sign Y Mouse goes down
          jnz short MOGOU
          sub     ax, dx
          cmp     ax, [BVMIN]     ; (Vmin) Mouse-pointer to high ?
          jb  short IIZ
;          test    bx, 80h         ; Overflow Y
;          jz  short IIZ
IY1:      jmp short IY2           ; write Y
;---------
MOGOU:    neg     dl
          add     ax, dx
          cmp     ax, [BVMAX]     ; (Vmax)  Mouse-pointer to deeply ?
          ja  short IIZ
;          test     bx, 80h        ; Overflow Y
;          jz  short IIZ
IY2:      mov     [YACK], ax
;---------------------------------
IIZ:      and     bx, 3           ; only buttons, remove Sign + Overflow
          mov     [TACK], bx
;---------
          popa
          pop     ds
PS2TEST:  retf

;--------------DS-SEG----------
XACK    DW MXpos  ; Mouse X
YACK    DW MYpos  ; Mouse Y
TACK    DW 0      ; Button,Sign,Overflow

BHMAX   DW Hmax   ; Mouse limits
BHMIN   DW Hmin
BVMAX   DW Vmax
BVMIN   DW Vmin

Dirk

[cassioraposa]

For those asking why would a person want to play something that requires an old machine or some kind of emulator, here's the proof of concept when applied to video-games: have you ever hear of Beggar Prince or Pier Solar? Both games designer for Sega Genesis/Mega Drive and released like one or two years ago.
All Pier Solar units sold out once, twice, a third time... The game is a full boxed genesis game, includes manual, etc. Now...

Why can't the same be done for computers?

The answer is simple: Of course it can be done and commercially speaking. It's just not conventional but it can be accomplished.

I was thinking about creating a Computer game for some outdated OS like MS-DOS or Amiga just because I want to prove that today society is “junking” hardware too fast and there is no need for that, since computers have suppressed our need for new technologies, at last for consumers, long ago. Such a game would be an example of a masterpiece that doesn’t need 8 cores and 16GB of ram to run.
People should concentrate in technologies that last longer or our environment won't last – is that so hard to understand?

Antti, I'm not an OS hobbyist but I'm skilled in pixel-art and audiovisual game resources in general. Should you want to team up in your project, just PM me, k?

*I don't get why people get so upset to see someone trying something different. What does it change in their lives anyway?

[Antti]

Antti, I'm not an OS hobbyist but I'm skilled in pixel-art and audiovisual game resources in general. Should you want to team up in your project, just PM me, k?

Thank you for your offer. However, I am not very interested in this idea anymore. Enthusiasm is very important if doing something like this and I do not have it enough. It is also because I have no necessary skills for this (programming or business). If I had, I am sure we could get things rolling. With high quality teasers like pixel-art "screenshots", it would definitely be possible to attract attention and get more developers for the team.

If you are really interested in this idea, I recommend that you find old-school programmers (who have made MS-DOS games) and bring this idea up. As fun this could be, I have no intention to be involved in. It goes without saying that you can freely use every "idea" I have mentioned in this thread.