Alpiner...

[Willsy]

Was just wondering, like... Anyone know??

[unhuman]

Whatchutalkinbout Willsy?

[OX.]

Dunno like but when are we likely to see a playable Manic Miner?

[Tursi]

It's a mix. You can tell using the Classic99 debugger - look at whether code is executed in the console ROMs or in the cartridge ROM, or take an educated guess using the heatmap (how much GROM is being accessed). Alpiner has 32k of GROM and 8k of ROM.

 

It actually sets up an interrupt routine pointing to ROM at >6026, looks like it may be handling the interrupt-driven speech routine. There's also some copy code in there (faster than the console MOVE for GROM->VDP).

 

The title page and player entry screens are all GPL, though at the beginning of each screen assembly code sets up the VDP.

 

The actual game code, and in fact after the screen listing all the mountains, assembly is frequently in use. But, the GPL interpreter is also still running. All movement seems to be heavily assembly (watching the heatmap shows a lot of CPU memory access when you move, it also shows the GPL interpreter getting a lot of use).

[sometimes99er]

I like and wonder if this is really true. Quote from www.videogamehouse.net:

Quote

THIS IS THE BEST GAME EVER!@$!@$

The guy below is incredibly lame.. who in their right mind would complain about a game that was created before all other games. EVER. It is one of 20 games created for this first every consumer PC computer.

IT's rocking! Seriously considering that no one had any type of measuring stick and no one had ever even seen anything other than a pinball machine at this point is is nothing short of amazing. A lot of fun to play.

I spent my entire 7th and 8th years of my life playing this everychance I got growing up.

We often compare TI games with the MSX and ColecoVision. The timelines including the TI-99/4 (without an A) are not all that coinciding. Loosely based on Wikipedia:

 

TI-99/4

06.1979 - 10.1983

 

ColecoVision

08.1982 - 1984

 

MSX

06.1983 - 1986

 

Edited March 22 by sometimes99er

[+adamantyr]

I like and wonder if this is really true. Quote from www.videogamehouse.net:

 

THIS IS THE BEST GAME EVER!@$!@$

The guy below is incredibly lame.. who in their right mind would complain about a game that was created before all other games. EVER. It is one of 20 games created for this first every consumer PC computer.

IT's rocking! Seriously considering that no one had any type of measuring stick and no one had ever even seen anything other than a pinball machine at this point is is nothing short of amazing. A lot of fun to play.

I spent my entire 7th and 8th years of my life playing this everychance I got growing up.

It seems the guy he is referring to has had his comment removed or something.

 

Bottomline. We often compare with a bit of envy to MSX(1) and Coleco(Vision), but the timelines including the TI-99/4 (without an A) are not all that concurrent. Loosely based in Wikipedia:

 

TI-99/4/A

06.1979 - 10.1983

 

ColecoVision

08.1982 - 1984

 

MSX

06.1983 - 1986 (MSX2)

 

The timeframe is a bit different, with arcades, ideas etc. developing rapidly. Speech and standard assembly access are other factors.

 

 

Also keep in mind the MSX systems had a LOT more programmers hacking away at it, finding ways around the technical limitations. Not to diss our 99'er community, but writing games was not something most of them were all that into...

 

Also, the fact we have a unique assembly language completely unlike the Z80 and 6502, and a scant amount of materials available about it, made getting anyone interested in programming it difficult.

 

Heck, I'm debugging the attack routine for my CRPG right now... thing is causing a VDP catastrophic failure, but I'm finding plenty of bugs on the way. Thank GOODNESS for Classic99's debugger. Remind me to send a bottle of Scotch Tursi's way sometime...

 

Adamantyr

[Tursi]

I like and wonder if this is really true. Quote from www.videogamehouse.net:

 

(quote removed)

 

Bottomline. We often compare with a bit of envy to MSX(1) and Coleco(Vision), but the timelines including the TI-99/4 (without an A) are not all that concurrent. Loosely based in Wikipedia:

 

TI-99/4/A

06.1979 - 10.1983

 

ColecoVision

08.1982 - 1984

 

MSX

06.1983 - 1986 (MSX2)

 

The timeframe is a bit different, with arcades, ideas etc. developing rapidly. Speech and standard assembly access are other factors.

 

 

Although the guy you quote is being a bit fanboyish, you make a good point about timelines. By 1980 (we'll say '80 to give the 99/4 time to get out there), video games were still relatively new. Spacewar (and the arcade clone Computer Space) had been out since 1971. But Pong was released /after/ that in '72, and was the first massive success. Odyssey was released in '72, it and the Odyssey 2 made some inroads.

 

The Atari 2600 was released in 1977, seeing massive success in 1978 when Space Invaders became the first home licensed arcade game. So the 2600 had a 2 year head start, and many games were released before 1979. So saying that only pinballs had been seen before isn't even remotely true - we'd had video games for almost ten years, and well known ones for at least 7 years. Of course, what the TI could do was well beyond what the 2600 could put up on the screen, but other computers like the Apple 2 were well known (also since 1977), and could display the more advanced screens (admittedly with far fewer colors ).

 

So the game concepts were out there, and so were some of the games. But it's also fair to note that before TI's video chip, sprites were not well known. So it's fair to say game makers were still coming to terms with how to use the machine. I also have had conversations with some of the hacker/game makers from back in the day who said they always thought the TI was a neat machine, but that it was too expensive to get started programming with (compared to machines like the C64). So... we also fell behind for reasons like that, I think.

 

But I think you're right, and I never thought about that before. By the time the ColecoVision was out, sprites were well known. By the time the MSX came out, the VDP was almost 4 years old!

[Willsy]

Heck, I'm debugging the attack routine for my CRPG right now... thing is causing a VDP catastrophic failure, but I'm finding plenty of bugs on the way. Thank GOODNESS for Classic99's debugger. Remind me to send a bottle of Scotch Tursi's way sometime...

 

Adamantyr

 

Hell yeah... How the hell Wycove Forth was written on a real 4A is just beyond me. Just assembling the bloody thing would have taken eons, only to find that it crashes with no clue as to where. Thank God (or, thank Tursi for Tursi's debugger. I simply would not code 4A assembly without it.

 

Mark

[sometimes99er]

Thanks for good feedback - though maybe a bit off topic.

 

I think the Space Invaders arcade was 1978 and the 2600 version was 1980. ref.: wikipedia.org/Space_Invaders.

 

TI Invaders (1981) and Parsec (1982) sure made the TI standout at the time of release. Access to machine resources and quantity was the reason I quickly moved to the C64. But the TI was a bit ahead - on certain points.

 

I'm not sure, but I think the Sinclair (Timex) was the most successful computer around 1980 in Europe, we didn't see much of Apple - but I had a friends father bring one back from the US. He had a very tiny monochrome monitor - I don't think I saw much there except some text (Basic and adventures ?). Another friend has an ABC80 with matching 14" color monitor. The semi-compiling BASIC interpreter was quite fast.

 

I have to ask myself why I return to TI-99/4A. Firstly I gave stuff away to advance, and maybe make someone else happy. When the Amiga 2000 (with loads of extras) was returned to me, I turned it in at the recycle station (really stupid - or not). I guess back in 2004, Amiga emulation wasn't all that good or realtimish, and maybe after having done both Z80 and 6502, the 9900 was a programmers dream (and so was the 68000). Been some time since I last looked into the x86, but what a mess !?

 

Anyways, I guess I had the 2600 in 1979 and the TI-99/4A in 1981, and at that time the 99 was slick and had the possibility of good arcade emulation - TI Invaders and a few more proved that - with the later addition of Atarisoft titles - and I had my own go using MiniMem and 32K standalone (in 1982). I think that was what I loved back then - the C64 was released August 1982 and it was at least a year before price and games started dominate here.

 

[+Vorticon]

 

 

Heck, I'm debugging the attack routine for my CRPG right now... thing is causing a VDP catastrophic failure, but I'm finding plenty of bugs on the way. Thank GOODNESS for Classic99's debugger. Remind me to send a bottle of Scotch Tursi's way sometime...

 

Adamantyr

Amen!

[+OLD CS1]

So the game concepts were out there, and so were some of the games. But it's also fair to note that before TI's video chip, sprites were not well known. So it's fair to say game makers were still coming to terms with how to use the machine. I also have had conversations with some of the hacker/game makers from back in the day who said they always thought the TI was a neat machine, but that it was too expensive to get started programming with (compared to machines like the C64). So... we also fell behind for reasons like that, I think.

 

How do player/missile graphics compare to sprites? Back in the day I set out to learn about every home computer I could get my hands on. We had Apples in elementary, and the junior high had a computer lab filled with Atari 1200XLs. I learned the pithy graphics of the Apple, but my time with the Atari was cut short before I could fully absorb them.

 

I have an 800XL and 1050 floppy (not sure if Happy -- hell, I am not even sure I have any DOS disks) in my closet which I intend to pull out at some point, but my attention has been kidnapped by my TI setup.

[Willsy]

Player missile graphics suck compared to the sprites on the TI. The "missiles" are literally a single (low resolution) dot, and the "players" (there are 4 of them) are stripes that run the entire length of the screen. Your "sprite" occupies a small portion of this stripe. It means a sprite can be the length of the screen if you want (but why would you want to do that). The down side is that it is horrendously expensive to move your sprite up and down the screen, because you literally have to copy the sprite data up and down memory.

 

However, the graphics modes on the A8 are very nice, and you can write "display lists" - kind of assembly language - but written in the assembly language of the graphics processor and you can make it do something different on every scan line if you want (IIRC). It is very very easy to mix different graphics modes on the screen on the A8 - it's designed to do it, right there into the chip. Nice.

 

Mark

[Tursi]

Player missile graphics suck compared to the sprites on the TI. The "missiles" are literally a single (low resolution) dot, and the "players" (there are 4 of them) are stripes that run the entire length of the screen. Your "sprite" occupies a small portion of this stripe. It means a sprite can be the length of the screen if you want (but why would you want to do that). The down side is that it is horrendously expensive to move your sprite up and down the screen, because you literally have to copy the sprite data up and down memory.

 

I disagree some with your assessment. The players running the entire vertical length of the screen, while a minor pain for vertical movement, is great for making larger objects. I had occasionally wished the TI could do it -- to do a full screen object we'd need to use 12 sprites (6 if magnified).

 

A full screen object is only 192 bytes, and most are much smaller. It doesn't take that long to move the average 9 bytes of data up and down. (Also remember that on the A8 it's in CPU memory, so there are no delays, memory ports, or wait states involved like with the TI. It probably takes close to the same amount of real time to move an 8x8 sprite on the A8 as on the TI ).

 

On the 2600, they actually don't even do that much, though. They just turn the players on and off on a per-scanline basis, so no scrolling needed. Do any Atari800 games do that?

[mäsäxi]

I'm not sure, but I think the Sinclair (Timex) was the most successful computer around 1980 in Europe,

 

Hello!

 

I am not sure if you are talking about Sinclair ZX80 (1980) or Sinclair ZX Spectrum (1982)?

 

If you mean the first one, and if you are comparing it to TI-99 or TI-99/4A, there is not much point to compare them, except for very different price range.

 

ZX80 had black&white screen, with alphabets and numbers and some ready-defined characters in ROM. 1Kb of RAM. No sound. It overheated easily. If computer had to do something, screen went blank, until task was done, so no arcade games, or at least not arcade games in the form we understand them. And you all probably guessed already, that it had no sprites.

 

ZX81 (1981) was a BIT better version of same computer, and screen didn´t went blank anymore, so arcade games were possible (made from ready-made graphic craracters, of course). This was somewhat popular, as it was cheap and (almost) everyone could buy it.

 

Sinclair ZX Spectrum (1982, but we had to wait to late 1983 here in Finland) had 8 colours (each with two different brightnesses, so it was possible to have 15 colours), 256x192 resolution, maximum of two colours on same 8x8 character, no sprites and it had simple ten octave single channel beeper, which halted the computer when it played, just like on Apple. This computer became really popular in Europe. (at least in some countries, including U.K. of course, Finland, Spain, and some other countries too)

Edited June 13, 2011 by mäsäxi

[+RetroElectroDad]

I'd give the advantage to the TI sprite system against Atari's PMG in straightforward use for the types of game being released when they competed against each other. The PMG are lower resolution,

narrower, more expensive memory-wise in normal use and there are fewer of them. However, I do think that PMG has advantages in some areas - the larger areas of colour when scaled up can act as an underlay to the playfield graphics and soft-sprites in front of them, it's easy to create large vertical overlays such as cockpit struts for a flight sim, screen area borders that you don't need to redraw when scrolling etc.

 

The real power comes when you combine PMG with the Display List and in particular the Display List Interrupt. This lets you change anything about the graphics - mode, colours, character set, scrolling offsets, with scan-line accuracy and without having to count the scan lines in a kernel as you would need to in the 2600. If you wish, and I think some early 2600 conversions did, you could write to the player register directly in a DLI or kernel without memory access and change the player hardware bytes, colours or enable/disable them on each scanline but the computer hardware was engineered to make this unnecessary.

 

For PMG players using DLIs you can shift a single player around, changing position and/or colour of the player stripe on different mode lines down the screen so that it bahaves like a number of sprites. If you consider something like Miner 2049er then it's likely that in some if not most levels all of the enemies are made from a single player stripe chopped around and colour-changed. For demo purposes, you could split the player on each mode line and combine the missiles into a 5th player to give you 5 players x 24 player splits = 120 apparent sprites if you only want horizontal movement. Not very useful or practical I know put possible.

[sometimes99er]

I am not sure if you are talking about Sinclair ZX80 (1980) or Sinclair ZX Spectrum (1982)?

I did think of the ZX80. And yes, it does not compare with the TI. And so I shouldn't have brought that one on.

 

Alpiner was apparently released as late as October 1982, so it's really not an example of something being ahead of its time (I somewhere along the line assumed it to be from 1980).

 

RadioShack TRS-80, Apple II, Atari 400/800 and VIC-20 was probably comparable with the TI in 1981. Games up to that point in time does not seem to put the TI ahead. The slick TI Invaders, released November 1981, may be an exception !?

 

[mäsäxi]

If you consider something like Miner 2049er then it's likely that in some if not most levels all of the enemies are made from a single player stripe chopped around and colour-changed.

 

 

If you are eager to know more, the whole thing is explained very clearly on following site by the author of the original game.

 

 

http://www.bigfivesoftware.com/atari/Miner/Technical/technical.html

 

 

And yes, you can find much interesting Big Five Software information from the same site!

[+thegoldenband]

I have a different question about Alpiner: has anyone ever completed all three loops of the game? Heck, has anyone even beaten Everest on the 3rd loop (after accessing it via the Shift-838 cheat code)? I gave it a shot, and I don't see how it's physically possible to anticipate or dodge the falling objects that come at you nonstop (except when a skunk is onscreen).

 

Or: has anyone ever disassembled the game? I'd be curious to see what's supposed to happen after you beat Everest for the third time.

 

I really like Alpiner, and was pretty pleased to summit Everest for the first time ever yesterday -- but my enthusiasm for trying to beat the game completely was dampened by seeing how unfair it gets on the third loop.

[S1500]

I only ONCE got to Everest. I was playing in a slightly tired state, which somehow put me "in the zone". I normally wouldn't get past McKinley. Was surprised to see the snowman skiing by.

 

I usually cheated by just splitting my character in half by climbing on the screen outer edges.

[RXB]

We had gaming sessions with some members of the TI 99er club PUG (Portland Users Group) and we had two players that could beat the game.

 

I was the best at Frogger setting records.

 

We are talking 1984 or 1985 for a time frame of how long ago.

[+thegoldenband]

I only ONCE got to Everest. I was playing in a slightly tired state, which somehow put me "in the zone". I normally wouldn't get past McKinley. Was surprised to see the snowman skiing by.

 

I usually cheated by just splitting my character in half by climbing on the screen outer edges.

 

That's not cheating, it's totally fair game! The manual even talks about it a little bit. I think that, with practice, I could summit Everest about 50% of the time -- but only on the first loop.

 

We had gaming sessions with some members of the TI 99er club PUG (Portland Users Group) and we had two players that could beat the game.

 

Wait, they could really loop it three times?! Wow if so, as I can't figure out how anyone could possibly withstand the insanity of what I saw on the 3rd loop of Everest. What did the ending screen look like?

[Tursi]

I didn't realize the Alpiner HAD an end screen, so this sounded like an interesting challenge. After I did it, I did a video that demonstrated some ways to find cheats in Classic99's debugger - I show how to find and stop the timer, how to change the number of lives, and how to get infinite lives (I talk briefly about how you might find the code patches for invincibility, but I think that's a big job).

 

The video will be up in about 20 minutes here:

 

However, I'm going to spoil the ending, there's no end screen, it just loops forever. Did it say somewhere there were only three rounds?

[+thegoldenband]

^Awesome, Tursi, that's fantastic! I can't wait to watch it after work today.

 

Regarding three rounds, the manual says that "Alpiner has eighteen different levels of increasingly difficult game play to challenge you. The eighteen levels of play are divided into three rounds of six levels each. Each level corresponds to one of the six mountains you must climb." And then it has a chart of the six mountains, divided into three rounds.

 

Between that, and the fact that the 838 cheat code lets you select levels (or "difficulties") 1 through 18, it certainly seems like three rounds was the idea. The manual doesn't say anything about an ending, though, and this was right on the edge of the era when "fixed length = some kind of special ending sequence" became a normative part of gameplay.

[Tursi]

Ah, okay, it probably just stops getting harder. it may be one of those cases where someone said "the last level is impossible anyway".

[RXB]

Tursi we did kinda cheat, we had a CPU IDLE SWITCH soldered onto the 9900 CPU.

 

That way we could take a breaks more often.