[Moved to Milliways Section]

interesting about those 3 points. I would have thought the clue book written for the first version of the game. strange they would later retrofit the cluebook solution points into the final version of the game.

Codename Iceman could have been pretty aweome but it bogs itself down more so than even police quest with doing things in the 'right' order, and piloting the sub as the technical copyprotection really sucked.

BloodyCactus wrote:

Codename Iceman could have been pretty aweome but it bogs itself down more so than even police quest with doing things in the 'right' order, and piloting the sub as the technical copyprotection really sucked.

Time to decompile it and fix it. 😉

I had the Amiga version of Codename:ICEMAN with the hintbook and manual. I could never get past the torpedo part. It drove me mad.

The first part of Iceman is quite enjoable, but it gets worse as you play on. At the end the graphics look very void, I think they just wanted to finish that game and get over it 🤣

I would gladly decompile and check version 1.033 if I had access to it.

I also found that SHP contains save games for this version, so veryfing might be easier than you thought: http://www.sierrahelp.com/Misc/SaveGames.html

OK,
I managed to find version 1.033.
I decompiled all 3 versions (PC 1.023, PC 1.033, AMIGA 1.036) and found out that v1.033 is identical to v1.036.

I would have thought the clue book written for the first version of the game. strange they would later retrofit the cluebook solution points into the final version of the game.

I'm pretty sure that the cluebook solution points were always the intended points, and the first version is just bugged. The situation doesn't make any sense otherwise.

I also found that SHP contains save games for this version, so veryfing might be easier than you thought: http://www.sierrahelp.com/Misc/SaveGames.html

Thanks, that was a good idea. I just used those to check version 1.033.

This is weird. Installing the washer now gives you a point, but exiting the cavern still gives you only 9, and not 11 like the cluebook and Amiga walkthrough video show. So this version still has two missing points. Either that or there's a method to getting 11 points that I don't know about.

On the versions, an easy way to tell which version is the number of RESOURCE files included. The more common 1.023 version has ten of them, named RESOURCE.000, RESOURCE.001, and so on, up to RESOURCE.009. Version 1.033 has five, going up to RESOURCE.004, but each file is twice as big as one from version 1.023. You can also verify by opening the file VERSION in notepad.

Sierra's Codename: ICEMAN holds a special place in my heart. No other game so masterfully combines clichéd storytelling, shallow characters, political dullness, speed-sensitive action sequences, horribly idiotic "puzzles," and asinine vocabulary recognition to boot, while attempting to simulate the environment and inner workings of the most unexciting of all seafaring vessels - the submarine.

It's worth mentioning that the Atari ST received a 1.041 release. I haven't any idea what the differences or potential fixes of this version are, but the resources bundles are compatible with the DOS interpreter, in case you ever want to do a comparison.

thecrankyhermit wrote:

Installing the washer now gives you a point, but exiting the cavern still gives you only 9, and not 11 like the cluebook and Amiga walkthrough video show. So this version still has two missing points. Either that or there's a method to getting 11 points that I don't know about.

I'm 100% sure they have identical scripts. If you can tell me which room# is the cavern (you can use SciViewer from here: http://sci.sierrahelp.com/Tools/OtherTools.html), I'll try to see what is the logic behind the point distribution in this room.

thecrankyhermit wrote:

On the versions, an easy way to tell which version is the number of RESOURCE files included. The more common 1.023 version has ten of them, named RESOURCE.000, RESOURCE.001, and so on, up to RESOURCE.009. Version 1.033 has five, going up to RESOURCE.004, but each file is twice as big as one from version 1.023. You can also verify by opening the file VERSION in notepad.

The RESOURCE files have nothing to do with the version. Each game was distributed in many variations: low density/high density floppies, 5.25"/3.5". The individual resources (scripts, views, pictures, sounds, etc.) were packed inside the RESOURCE files, and their size was determined according to the capacity of the media. The different versions differed in the resources that were packed inside. I am 100% sure that the script resources in 1.036 and 1.033 are identical. I didn't check the rest of the resource types.

Cloudschatze wrote:

It's worth mentioning that the Atari ST received a 1.041 release. I haven't any idea what the differences or potential fixes of this version are, but the resources bundles are compatible with the DOS interpreter, in case you ever want to do a comparison.

I'll look up this version and have a look.

If you can tell me which room# is the cavern (you can use SciViewer from here: http://sci.sierrahelp.com/Tools/OtherTools.html), I'll try to see what is the logic behind the point distribution in this room.

pic.068 is the last room of the cavern.
pic.053 is the room outside of the cavern.

I tried looking at the corresponding scripts. No clue what's going on in them.

Here's a video of the 11 point gain:
https://youtu.be/LLyy1N57ttc?t=3h17m30s

Edit - holy crap, some of the later pictures are terrifying. Sharing the URLs to make the nightmares optional.

110
http://i.imgur.com/DvrnVxN.png

111
http://i.imgur.com/huNuB4b.png

The game starts really well and goes pretty nicely but all shit hits the fan when the main character gets inside that damn submarine. I don't mind so much when he gets inland in the final part, but i just hate the submarine part. Oh and driving the damn thing is impossible on a fast system.

I checked ATARI v1.041, and there are some minor differences.
Many empty methods were removed. My guess is that they used a newer compiler that removed unused methods.
They seem to be using a slightly newer version of the SCI System scripts.
For example, the List.at method was changed from

1	(method (at n &tmp node)
2		
3		(for	((= node (FirstNode elements)))
4			(and n node)
5			((-- n) (= node (NextNode node)))
6		)
7		(return (NodeValue node))
8	)

to

1	(method (at n &tmp node)
2		
3		(for	((= node (FirstNode elements)))
4				(and n node)
5				((-- n) (= node (NextNode node)))
6		)
7		(return (if node (NodeValue node) else 0))
8	)

Also, when asking the girl on the beach for the date ("ask for date"), the year is calculated differently. Seems like the GetTime Kernel function in the PC version returned the number of years after 1920, and the ATARI ST used 1980 (like later games on the PC did as well - ScummVM has bug here - I'll need to report it).

I saw some different sounds being used in the volleyball game for some sound drivers.

I didn't see any change to the actual game logic.

thecrankyhermit wrote:

Edit - holy crap, some of the later pictures are terrifying. Sharing the URLs to make the nightmares optional. […]
Show full quote

Edit - holy crap, some of the later pictures are terrifying. Sharing the URLs to make the nightmares optional.

110
http://i.imgur.com/DvrnVxN.png

111
http://i.imgur.com/huNuB4b.png

without the animations, yeah they look empty

1Also, when asking the girl on the beach for the date ("ask for date"), the year is calculated differently.

Did not know you could do this. In DOSBox, playing version 1.033, she says it's 4/12/36, which is only off by exactly 80 years. Don't really understand the logic - what is counting the years since 1980 supposed to accomplish? Plus, the game's supposed to be set in 2004 anyway.

My scripts do not look anything like your example. They look more like assembly. For instance:

1code_0335:	pushi	$ef		; 239, has
2		push1
3		pushi	$8		; underBits
4		lag	global[$0]
5		send	$6
6		bnt	code_034e
7		push2
8		pushi	$35		; 53, '5', b-incr
9		pushi	$3		; y
10		calle	export255_0, $4
11		jmp	code_0357

Yes, what you have is disassembled and what Omer posted is decompiled code.

thecrankyhermit wrote:

Did not know you could do this. In DOSBox, playing version 1.033, she says it's 4/12/36, which is only off by exactly 80 years. […]
Show full quote

Copy code to clipboard

1Also, when asking the girl on the beach for the date ("ask for date"), the year is calculated differently.

Did not know you could do this. In DOSBox, playing version 1.033, she says it's 4/12/36, which is only off by exactly 80 years. Don't really understand the logic - what is counting the years since 1980 supposed to accomplish? Plus, the game's supposed to be set in 2004 anyway.

The logic in the PC scripts is returning [(YEAR - 1920) - 60]. ScummVM is computing it as [(YEAR - 1980) - 60] because it has wrong implementation for the GetTime kernel function.
The ATARI ST logic is [(YEAR - 1980) + 1990] which sets the game at a constant 10 years in the future.

thecrankyhermit wrote:

My scripts do not look anything like your example.

collector is right: You are looking at a disassembly, and I am using a decompiler to look at a higher level script. There is a decompiler bundled with SCI Companion: http://scicompanion.com.

I still hadn't had a chance to look into the cavern scripts. Maybe later today.

thecrankyhermit wrote:

This is weird. Installing the washer now gives you a point, but exiting the cavern still gives you only 9, and not 11 like the cluebook and Amiga walkthrough video show. So this version still has two missing points. Either that or there's a method to getting 11 points that I don't know about.

I found some clue in script# 304. Here's a snippet:

1(local
2	local98 =  15
3)
4(instance cavesRg of Rgn
5	(method (newRoom newRoomNumber)
6		(if (== newRoomNumber 53) (gGame changeScore: local98))
7		(if (< 0 local98) (-- local98))
8		(super newRoom: newRoomNumber)
9	)
10)

Room# 53 is the cavern exit room.
It seems that while in the caves region (cavesRg), whenever you are changing a room, the variable local98 is decremented by 1. It is initialized as 15.
When you finally exit the caves region into room# 53, you are awarded the remaining points in local98.
So to get the maximum amount of points, you have to minimize the number of time you change rooms.

I didn't test this though.

I just tested it, and I think you're right! Not only that, but it works in version 1.023 too.

Normally you use the flares to guide you through the maze. Using the flares gives you points too. But if you follow the path of the flares, you'll solve the maze in 6 steps instead of 4, and you'll miss two points. The optimal path is up, up, right, down, right.

So this makes the perfect scoring playthrough even stupider than I thought. You have to win an optional game of dice to win a jamming device. And then when you get to the magnetic scuba net, you don't use the jamming device at all, and instead take the long way through the cavern maze. And you're supposed to use all your flares to get 8 points, and then ignore them and take a different path instead.

And on top of that, the base version of the game is still missing one point.