In defense of Pac-Man...

[+save2600]

Why does this Christo guy keep sticking " " in his posts?

 

Some weird HTML formatting... an artifact probably due to the way his browser handles replies or copying/pasting.

[zzip]

The first 3d games, like Doom, Heretic, Hexen, Quake, Descent, Duke3D, and Unreal are really high-quality and high class. What came after those are shit to be sure.

Some of those weren't truly 3D, especially Doom engine games. The enemies in those games were basically flat bitmaps that were always facing you. Also, the maps were flat in that you couldn't have rooms on top of other rooms. You could alter the height of the rooms and create stairs and towers to try to fake that it's really 3D, but if you ever tried to design your own level you quickly encountered this limitation.

[JBerel]

https://m.youtube.com/watch?v=1fQYphOUzvo

 

Was that the ad?

[Keatah]

I don’t know about that as an axiom. Hiroshi Yamauchi ran Nintendo pretty successfully and never even held a controller until 1996. I guess it’s more about trusting your dev teams to put out a polished product than having a gamer run the company.

 

 

Sent from my iPhone using Tapatalk

 

I'll give you that. At least someone who is interested in the product is required for great success. Or someone who has a love for the process of producing greatness and not just penny pinching.

[G-type]

https://m.youtube.com/watch?v=1fQYphOUzvo

 

Was that the ad?

 

That ad looks rad! Even has the Buckner & Garcia track!

[+frankodragon]

Had no idea until today that Will Lee a.k.a. Mr. Hooper from classic Sesame Street was in this Atari commercial for Pac-Man.

 

 

 

 

Now when I think about it, someone at Atari may have seen the arcade and looked at it sideways? Who knows.

[Keatah]

Some of those weren't truly 3D, especially Doom engine games. The enemies in those games were basically flat bitmaps that were always facing you. Also, the maps were flat in that you couldn't have rooms on top of other rooms. You could alter the height of the rooms and create stairs and towers to try to fake that it's really 3D, but if you ever tried to design your own level you quickly encountered this limitation.

 

I realize and understand that. They were simply amazing at the time and we considered them 3D however accurate or inaccurate.

 

Not sure when the "first era" of 3D games went south on the PC. Or at what point they all started to be the same..

[christo930]

Why does this Christo guy keep sticking " " in his posts?

I'm not putting it there, it's there on its own. I'll try deleting it.

 

I'm using Pale Moon as a browser with NoScript running and there is at least one "layer" of scripts not enabled. That might have something to do with it. It is irritating that you enable scripts that launch a bunch of additional scripts. You enable them and they launch yet more scripts. I can understand (I'm not referring to AA) one layer of scripts for screen formatting and some ads, but most websites are just way over-scripted and eating up resources for no real reason.

Edited October 9, 2017 by christo930

[christo930]

 

I realize and understand that. They were simply amazing at the time and we considered them 3D however accurate or inaccurate.

 

Not sure when the "first era" of 3D games went south on the PC. Or at what point they all started to be the same..

I would say it would be around the time Tomb Raider came out. That is one of the ugliest games I have ever seen. There was a whole slew of 'me too' games that came out right around the same time or a little later.

 

I agree that the very early games like Doom were good looking games. They likely benefited from the fact that 3d chips were not yet widely available. I also thought Duke 3d was pretty decent looking. Same with Decent, though I didn't particularly care for that game.

Edited October 9, 2017 by christo930

[Bryan]

https://m.youtube.com/watch?v=1fQYphOUzvo

 

Was that the ad?

Wow... listen to those sound effects.. which are totally not 2600 Pac-Man. Amazing what you could get away with.

[christo930]

Pretty much.

 

The VCS is an extremely limited economy. You have 128 bytes of RAM, 76 processor cycles on a line, twice as many if you update every two lines. You need memory for the dots, for two players, which takes almost all of your RAM, to say nothing of other bits of system state.

 

Could Pac-Man have been improved? Sure, a blue maze, black background, but Atari did have a few guidelines against black backgrounds for space games. Put the maze holes on the right axis? sure, could have been done.

 

A flicker manager? Yes, this has been proven to be possible when you have more than 4K of ROM to be able to create more unrolled kernels.

 

I am sick and fucking tired of people, who quite frankly have _ZERO_ understanding on the difficulties of WRITING a VCS game, just shit all over "game x" .... When you have spent many weeks counting to 76 and carefully timing each instruction to make sure that you have enough time to reset each element of playfield, player, missile, and ball, to create something that can DRAW the game state you have...and you go through the difficulties of trying to carefully make sure that your game state calculations happen within the allotted time when things AREN'T being drawn.... then you'll have perspective...and will be much more likely to shut the fuck up about it.

 

-Thom

While it is true I have never programmed a 2600, I do have programming experience. A few weeks ago I was actually thinking about the Pac*Man conversion and I worked out how much ram you needed for all of the variables you mentioned and it really wasn't much, it was less than 1/2 and that would include 2 players.

You have the dots, basically 1 bit each

You have the x,y coordinates of each ghost and what direction they are moving (player dependent), the scores, the levels, the coordinates of Mr Man and direction of travel. A timer for the special, and probably 1 or 2 things I'm not thinking of. I don't recall the exact number, but it was well under 1/2 of the ram.

 

Don't get me wrong, I understand (more like vague idea) just how many limitations this machine has and the cleverness of the programmers and just how far this machine was pushed, even back then. There were probably fewer than 100 people in the world back then who programmed for this machine. Add the fact that they were also involved in creative work (which impressed me more as a kid than the technical aspects) and you have a pretty amazing person.

 

None of this makes criticizing their work somehow a personal insult or invalid.

 

I don't appreciate the attitude.

[+KaeruYojimbo]

Now when I think about it, someone at Atari may have seen the arcade and looked at it sideways? Who knows.

pac_man_tilted.jpg

 

I think I remember hearing Tod Frye say that's more or less what he was doing when he was adapting the vertical arcade maze to a horizontal TV screen.

[ZylonBane]

I think I remember hearing Tod Frye say that's more or less what he was doing when he was adapting the vertical arcade maze to a horizontal TV screen.

 

Except that's not what he did. He just threw in that sad repeating pattern that doesn't resemble the arcade at all.

 

Now this is what a sideways arcade maze looks like:

 

Edited October 10, 2017 by ZylonBane

[+KaeruYojimbo]

 

Except that's not what he did. He just threw in that sad repeating pattern that doesn't resemble the arcade at all.

 

Now this is what a sideways arcade maze looks like:

 

 

Well, we've established that he wasn't all that concerned with arcade accuracy. I think he meant it more as a description of his general approach to tackling the problem of putting a vertical maze on a horizontal screen, including putting the tunnels at the top and bottom, and not to say he literally turned the maze on its side. A sideways version wouldn't have worked using a mirrored playfield (though a stretched version of the arcade maze would have).

[NE146]

https://m.youtube.com/watch?v=1fQYphOUzvo

 

Was that the ad?

 

No.. the ad was pre-launch and was made to hype the upcoming release. It basically showed the arcade pacman.. I seem to remember a closeup of Pacman eating dots, and the basic jist of the voice was "this is Pacman.. the biggest arcade hit ever, and this summer (or spring, or whatever).. it's coming only to Atari". Or something to that effect.

 

And we all salivated guessing wildly how close it could be, but with the screens they showed, it HAD to be pretty close right? lol it wasn't.

[+Random Terrain]

Thanks Woolco! I pre-ordered Pac-Man and all I got was a lousy game and no T-Shirt.

 

youtube.com/watch?v=IRLDPSYI19s

https://www.youtube.com/watch?v=IRLDPSYI19s

 

 

The video below has changed my mind. Atari 2600 Pac-Man is clearly better and more fun than the arcade version:

 

youtube.com/watch?v=rR71hEbWNmk

https://www.youtube.com/watch?v=rR71hEbWNmk

 

 

Who want's an arcade-like maze when you can have 4 boxes on each side of the screen with a couple of holes punched in them? Boxes are fun. Ask any cat.

 

youtube.com/watch?v=9rDCdqeTD0o

https://www.youtube.com/watch?v=9rDCdqeTD0o

 

 

Atari 2600 Mazes

 

Speaking of mazes, here is an Atari 2600 random-ish maze test program I was working on in 2011/2012 for the Atari 2600. Run it with your favorite emulator and press Reset any time you want to see a new maze:

 

random_maze_test_16x16x16_with_fast_data_32x23_2012y_11m_22d_0322t.bin

 

They're not the greatest mazes in the world, but at least they are more Pac-Man-like than what we got with Atari 2600 Pac-Man. To be clear, I'm not saying that Atari 2600 Pac-Man should have had random mazes, I'm saying how about creating a maze that has a Pac-Man style to it instead of that boxy thing we ended up with.

[Nukey Shay]

While it is true I have never programmed a 2600, I do have programming experience. A few weeks ago I was actually thinking about the Pac*Man conversion and I worked out how much ram you needed for all of the variables you mentioned and it really wasn't much, it was less than 1/2 and that would include 2 players.

You have the dots, basically 1 bit each

You have the x,y coordinates of each ghost and what direction they are moving (player dependent), the scores, the levels, the coordinates of Mr Man and direction of travel. A timer for the special, and probably 1 or 2 things I'm not thinking of. I don't recall the exact number, but it was well under 1/2 of the ram.

 

You left out a whole lotta stuff (which not only apply to this game, but many others). Each player uses 23 bytes (that's all the dots on their own boards, current difficulty, remaining lives, and score)...over 1/3 of the available ram just for that! (i.e. the program stack eats up 10 bytes due to the deep nesting). There's stored colors to make the attract mode work, bits here and there to keep track of game selection, console buttons, etc.

 

Is it that difficult to take a peek at Dennis Debro's reverse-engineered source file that describes what ram is used for?

 

Of course, it's relatively simple to combine variables that don't use a full byte...but that costs romspace to separate the bits when you update or test them. More optimization time needed (but didn't have due to the tight schedule). Modern homebrewers can take months/years to create a game from scratch...and that is with the benefit of unlimited computer time, tools light years ahead of anything that existed BITD, and an entire library of advanced programming methods to draw solutions from.

 

Aside from that, this was coming at a time when the "make the port match the original or bust"-mentality just wasn't there. I think that Breakout might be the closest one for that (and even a couple of liberties were taken in that cartridge).

Edited October 10, 2017 by Nukey Shay

[Nukey Shay]

Except that's not what he did. He just threw in that sad repeating pattern that doesn't resemble the arcade at all.

Reading though the program, I know what he was trying to do (cut down the space needed for a fully mapped layout). The repeated pattern *nearly worked in that regard, but there were too many exceptions needed in the kernel for the energizers, bonus item, and ghost pen. The top/bottom tunnel would have added ANOTHER exception to the pattern, it was just easier to put the gap before and after the maze is drawn.

 

* = "nearly" because if he had taken the time to look more closely after everything else had been added in, he might have noticed that the repeated boxes did not save any space whatsoever. Same goes for the hardware collision vs. hitbox choice. Possibly the reflect registers being used, as well.

[zzip]

I realize and understand that. They were simply amazing at the time and we considered them 3D however accurate or inaccurate.

 

Not sure when the "first era" of 3D games went south on the PC. Or at what point they all started to be the same..

Agreed, even though they were taking shortcuts, they were mind-blowing at the time.

 

I think it all went to hell when 3D hardware started to be released on the PC (after consoles like PS1 started adding 3D hardware). The wave of Dos games like Doom, Heretic, Descent, etc did all that without 3D hardware. The early 3D hardware sure helped with rendering 3D objects, but wasn't powerful enough to make them look nice

[christo930]

 

You left out a whole lotta stuff (which not only apply to this game, but many others). Each player uses 23 bytes (that's all the dots on their own boards, current difficulty, remaining lives, and score)...over 1/3 of the available ram just for that! (i.e. the program stack eats up 10 bytes due to the deep nesting). There's stored colors to make the attract mode work, bits here and there to keep track of game selection, console buttons, etc.

 

Is it that difficult to take a peek at Dennis Debro's reverse-engineered source file that describes what ram is used for?

 

Of course, it's relatively simple to combine variables that don't use a full byte...but that costs romspace to separate the bits when you update or test them. More optimization time needed (but didn't have due to the tight schedule). Modern homebrewers can take months/years to create a game from scratch...and that is with the benefit of unlimited computer time, tools light years ahead of anything that existed BITD, and an entire library of advanced programming methods to draw solutions from.

 

Aside from that, this was coming at a time when the "make the port match the original or bust"-mentality just wasn't there. I think that Breakout might be the closest one for that (and even a couple of liberties were taken in that cartridge).

I did account for the things you mentioned, but there were other things I wasn't sure of for a number of reasons. I figured there were only a few things the non-current player needs stored like board layout (which dots have been eaten), men left, score and board # (how many mazes have been completed). I wasn't sure how many bytes were allocated to score keeping.

I wasn't sure if the TIA needs anything in ram or can simply access stuff in the rom (which is presumably just on the address bus as if it were ram as in a program that gets loaded into ram on a different computer where everything is in ram). I actually had it all down in a notepad file and was going to post a few questions I had, but I decided not to, I lost the file and was just typing off the top of my head while writing that reply yesterday. I was actually a bit surprised about how little ram this could be done with.

 

It really is impressive what was done with this machine. Though it has been a LONG time since I read it, I think I recall the idea that the 2600 was going to have maybe 10 - 20 cartridges released for it based on what was happening at the time. Defender II and Millipede come to mind as examples of excellence even though they both have the super-chip with the extra 128 bytes of ram. Pitfall is another one and even Space Shuttle.

 

My 12YO self knew that games were never going to look as good as on the arcade version. A lot of the games that didn't come out in the arcade had the advantage of having no comparison. Even games that I would praise, like Ms Pac*Man weren't arcade clones. It's really not about 'it doesn't look and play exactly like the arcade game' so much as it is 'god, this game is ugly, has poor and inappropriate sounds' and now-a-days gives me a headache.

 

But really, I just don't believe these guys are beyond criticism, which is what I was arguing against. You have every right to say a game sucks just as much as to say how good game X is.

Edited October 10, 2017 by christo930

[Tod frye]

yeah. well. so.

does anyone know where I can find a disassembly of the good old 2600 PacMan?

lots of details i don't remember

Edited October 11, 2017 by Tod frye

[G-type]

Wow! It's the man himself! There's not much I can add to what I said in the opening post other than thanks for the many years of entertainment!

[JBerel]

Based on the way PacMan has aged, I don't think he needs any defense anyway.

 

[zzip]

My 12YO self knew that games were never going to look as good as on the arcade version. A lot of the games that didn't come out in the arcade had the advantage of having no comparison. Even games that I would praise, like Ms Pac*Man weren't arcade clones. It's really not about 'it doesn't look and play exactly like the arcade game' so much as it is 'god, this game is ugly, has poor and inappropriate sounds' and now-a-days gives me a headache.

 

Right, even though Ms Pacman wasn't arcade perfect it was clear they tried, and that was good enough. With Pacman it was clear they didn't even care about accuracy!

[Bryan]

Hi Tod! It's awsome when Atari celebs show up here (and don't get scared off by a bunch of fanatical man-children). Do you read AtariAge or did someone contact you about this thread?

 

I'm sure there's got to be some disassemblies of Pac-Man around here.