Author Topic: GTD: analysis of game verbs  (Read 6223 times)

Radiant

  • Mittens Knight
  • AGS Baker
  • Return once more to the Two Kingdoms!
    • I can help with publishing
    • I can help with story design
    • Radiant worked on one or more games that won an AGS Award!
    •  
    • Radiant worked on one or more games that was nominated for an AGS Award!
GTD: analysis of game verbs
« on: 21 Nov 2004, 01:51 »
Hi there!

I've been thinking about an aspect interface design, and would like some feedback and discussion on the matter. Just about every point-n-click adventure game I've played has a GUI that resembles Sierra classics (KQ5/6, SQ4/5, QfG3/4, etc), or one that resembles LucasArts classics (DoTT, TSOMI, Maniac), or a variation on the verb coins. For this discussion I'm not really interested in what shape it takes, but rather what verbs are on there.

The classic Sierra interface is rather limited, offering the five options of Walk, Look, Touch, Speak and Use Item. Barring some exotic cases such as Loom, I believe that using these five verbs goes without saying (with the possible exception that Touch can be split up).
Most verb coin interfaces use roughly the same commands.

The LucasArts interface has gone through some evolution. Their first game, Maniac Mansion, uses the following fifteen verbs:
Code: [Select]
Open Close Read
Push Pull Give
What_is Walk_to Pick_up
Turn_on Turn_off Fix
Unlock Use New_Kid

In Zak McKracken, Unlock and Fix were replaced with Put_on and Take_off; Indy3 and TSOMI refine the list a bit and add the missing Look and Talk verbs. The most often used variation contains
Open Close Look Push Pull Give Talk Use PickUp
and was used in many games from Monkey/2 to FOA, and even kind of adopted by Sierra for Gabriel Knight.

Now there are some verbs that are more interesting than others. However, on playing some interactive fiction (text adventure) games, I realized that they also tend not to use too many verbs. So it seems to me that there aren't that many verbs that actually offer puzzle potential (or immersion) in a game. And there are always verbs that could be simply replaced by 'use item on', if the item has an obvious function. Clearly 'use hammer on nail' requires no explanation.

Useless verbs
From Maniac Mansion, the verbs Fix and Unlock are rather useless. Since 1) you always require an item to perform these actions, and 2) the item in question is obviously meant for this verb (fix staircase with tools, unlock padlock with glowing key), these could be replaced by 'use tools on staircase'.
Zak's put-on and take-off are also pretty obvious, since they only apply to articles of clothing, which could be put-on by using them on the player character (or just using them, as in FoA, since in Lucas games you can't actually click on the player character). The same goes for eating edibles.
In Sierra's Conquest of the Longbow, the 'shoot' icon is rather superfluous since it means to 'use bow on <target>'. However it does add to the atmosphere of the game.
Along the same line, Larry adds the zipper icon. However, the smell and taste icons in space quest games are almost entirely useless and don't add much to the gameplay experience.
It does not seem to be the case that adding more verbs leads to a more difficult game. Judging by the size of the hints-n-tips forum, people can get stuck easily enough with just four or five verbs. Of course there's the hunt-the-verb problem of some old parser games, but that's just poor game design in almost all cases (e.g. the 'reach in hole' puzzle in Enchanter, or Larry 2's infamous 'put airsick bag in hair rejuvenator').

Spellcasting
Several IF games contain magic spells. Now for some games (enchanter trilogy) these could be easily handled by inventory items, since you 'carry' each spell on a scroll or spellbook. So 'frotz bread' could be done in Quest for Glory fashion. However in other classics, there are words that just happen to be spells ('plugh' and 'xyzzy' come to mind). This puzzle is very hard to pull off in a p'n'c game.

Verb redundancy
There is a difference between a graphical interface (Sierra) and one that uses words (LucasArts), because in the latter case, a sentence appears on a command line, and the sentence should make sense. While Sierra can get away with 'use door' to open it and 'use box' to pick it up, LucasArts would require the separate open/close/take commands. However, these do add to an occasional puzzle, where you have to open something that isn't obvious (the radio in MM) or pick up something that doesn't appear to be pick-uppable (the monkey and dog in MI2). I should point out that the latter tends to be frustratingly hard on most players, and I can't think of any circumstances where it was used in an entirely serious game.
In general I would say that Lucas games would not be off worse puzzle-wise if they had no open, close or pick up verbs.

The Show vs. Give dilemma
While using an item on a hotspot seldom implies more than one meaning, using an item on a character does. Click sword on soldier... am I trying to give him the sword? Show it to him? Ask him about it? Stab him with it?
Luckily the former three can be combined. Showing an item and asking about it tends to be the same (unless the item isn't around, in which case a notebook system like Laura Bow's would be useful). Giving can be included as well... clearly when you give an item to somebody, you show it to him first - and if you don't want your game to have walking dead situations, the player character must be unable to give the item away if he needs it later (or has to be able to get it back).
Stabbing him with the sword is a separate case, and could feasibly require an attack verb. However in most games there is an obvious distinction between 'good' people whom you cannot kill period, and 'evil' people whom you can't do much else with.

Physical actions
Certain actions may be required of the player character that simply cannot be expressed in other ways... for instance, jump/duck (Kq1), or run/sneak (QfG series). These actions tend to be rather seldomly used in puzzles, though (duck and run aren't necessary in either of those games, for instance, and jump is localized to a single locale where it is rather obvious). KQ3 has a puzzle that requires jumping, but logically Gwydion could also have climbed the crate since it isn't that high.
Some physical actions could still be done by adding an item that can be used for precisely that purpose, e.g. lockpicking or digging.

The Hand multifunctionality dilemma
The obvious problem with Sierra's hand cursor is that it does too much, and it isn't always clear whether you're going to open the door or knock on it, or to climb the gate or force it, or to pick up the cat or kick it.
LucasArts's push & pull verbs (which could be combined into one 'move' verb) make for a nice solution, even if they aren't used that often for puzzles (the knife thrower puzzle from FoA does come to mind).
An 'attack' verb would be useful in several cases (and is well-used in Grrrr, for instance).

Placement puzzles
The ability to drop inventory items does make for some interesting cases. In classic IF, there is a restriction on how much you can carry, forcing you to drop things in convenient places (but this could easily lead to dead ends if used carelessly, and people seem a bit spoiled these days with the limitless-pockets effect). But I've seen some good puzzles with item placement in Kyrandia, and the bait-the-dog puzzle in S301.
Of course droppable invitems is tricky to pull off in AGS, so that may stop some people from allowing each item to be dropped everywhere. If you want to save space, the 'get' icon could be changed to 'get/drop', depending on whether you click on the room or on your inventory.

Funky responses
Of course one disadvantage of adding more verbs is the fact that the game should have a decent response to many of them. LucasArts games seem to get a way with a lot of 'I cannot move it' and 'Nice tree', whereas Sierra games are almost expected to have a snappy answer to any look/touch/talk click. Clearly this is a tradeoff, but having too many generic responses does lower the game immersion.

So if I were to extend a Sierra verblist, or change a Lucas one, my first recommendation would be along the lines of
walk look
touch move use-inv
talk attack get/drop
(and Laura's notepad and Larry's zipper added in those games that could use it)
Interestingly enough an 'attack' verb can be extended to inventory items (to break them), and a move icon could shake inventory items.
A very good final addition would be the 'do something appropriate' verb from the Garfield game, which has such nice uses as 'kick odie' and 'save jon'. Also it could feasibly open doors, since I omitted the open/close verbs from my sample above.

Finally I would like to state the obvious case that a LucasArts conversation (with a tree of options) offers more puzzle potential than the Sierra way of clicking the talk icon on an NPC until it stops repeating itself, or the QfG way of giving a list of options where you have to ask about all of them.

A question I'd like to pose, is for examples of other verbs succesfully used in puzzles, in either IF games or parser graphics games (a la the older Sierra) or any verblist/verbcoin games that have a rare verb I have missed.

Snarky

  • Global Moderator
  • Global Moderator
  • Mittens Lord
  • Private Insultant
    • Best Innovation Award Winner 2018, for his numerous additions to the AGS open source ecosystem including the new Awards Ceremony client and modules
    • Snarky worked on one or more games that won an AGS Award!
    •  
    • Snarky worked on one or more games that was nominated for an AGS Award!
Re: GTD: analysis of game verbs
« Reply #1 on: 21 Nov 2004, 03:36 »
Radiant, is there any causal relationship between seguso's "manifesto" and this GTD thread?

You seem to be trying to define a "smallest expressive set of verbs". I'm not quite sure what the point is. Most people seem to think that the LucasArt set of verbs is rich enough, and I haven't heard anyone say that it has too many. What do you gain from reducing it further?

Since this thread is kind of about interfaces, I'm going to go ahead and post an article I wrote some time ago, for something else. Seems more appropriate than starting a new thread. I'd appreciate feedback, and be interested in earlier examples for the features I talk about, and particularly other types of controls and displays that I haven't covered.

Adventure Game Interfaces

A comprehensive discussion of all interfaces that have ever been used in adventure games is probably not possible, because the line between adventure games and other genres is blurred to an unusual extent. It's possible to make an adventure game in almost any game engine, so designers have created adventures in the tradition of other genres (first-person 3D, platform arcade…), using interfaces typical for those game types. This article, therefore, focuses on innovation in the mainstream adventure game tradition, not one-offs existing within other genres.

The evolution of adventure game interfaces has been an almost unbroken progression towards simplicity. While certain input techniques often go together with specific types of display, the innovations in each interface aspect have often happened independently. It is therefore clearest to discuss them separately.

Input
The majority of PC adventure games have used one of the following input techniques (listed roughly in chronological order): option-select, text parser, parser+arcade control, verb-select, modal cursor, simple cursor, or pop-up menu.

Option-Select (1972)
In a game with option-select controls, the player is presented with a list of available actions and selects one of them. The selection can be indicated by typing a keyboard shortcut, or by highlighting the list item with keyboard arrows or mouse cursor and clicking/pressing an action key (ENTER or SPACE, typically). Option-select was the input method used in the very first proto-adventure game, "Hunt the Wumpus" (Gregory Yobs, 1972). It is widely used for selecting dialogue responses in adventure games that allow the player to control conversations. It remains the input mechanism of choice for Japanese (often pornographic, hentai/ecchi) adventure games.


Option-select used for dialog in The Secret of Monkey Island 2

One problem with option-select controls is that it makes it difficult to design challenging puzzles, since there are no non-obvious actions. Unless some options are fatal, puzzle-solving is simply a matter of working through the alternatives. Consequently, option-select games are often trivial.

Text Parser (1975)
In text parser games, player input is entered as written commands. Examples of common syntax are "go north", "take stone". The classic "Adventure" (aka "Colossal Cave") (Will Crowther, 1975), which is generally regarded as the first adventure game, used text parser input. Text parsers were used in almost all text adventures and many graphical adventures throughout the 1980s. However, in the 1990s it was overwhelmingly abandoned in favor of mouse-based interaction techniques. One notable exception is "Leisure Suit Larry 7: Love for Sail!" (Sierra, 1996), which included optional text input as a nostalgia feature.

Playing a text parser game can be a frustrating experience, because of the limited capabilities of the parser to recognize natural language. It only recognizes certain words, so in order to perform an action the player has to find the right word to describe the action. The player cannot tell whether an action fails because it cannot be done in the game or because the game designer used a different word to describe it. Spelling errors (by the player – or by the designer!) can make a puzzle unsolvable. In text parser games, coming up with synonyms is often a bigger challenge than actually solving the puzzles.

Parser + Arcade Controls (1983)
In games that combine parsers with arcade controls, players move their character around (in order to get from one place to the next, and to avoid dangers) using arcade controls (usually the keyboard arrows, sometimes the mouse), and perform actions by entering commands into a text parser. This control model was introduced in "King's Quest" (Sierra On-Line, 1983), and used in many classic Sierra games until 1990.


King's Quest

Verb-Select (1984)
In verb-select games, players have access to a fixed list of actions that are available (talk to, pick up, push, repair, etc.). They select their action and indicate the target for the action, e.g. "Pick up rubber chicken with a pulley in the middle". Verb-select controls were first used in "Below the Root" (Windham Classics, 1984). Later, in "Maniac Mansion" (LucasFilm Games, 1987), Ron Gilbert took the verb-select concept and created the famous LucasArts SCUMM-driven interface. It allowed players to interact with objects in the game world by clicking a verb and then clicking the object. This interface, with some further tweaks, was used in LucasArts titles such as "The Secret of Monkey Island" (LucasFilm Games, 1990), "Indiana Jones and the Fate of Atlantis" (LucasArts, 1992), and "Day of the Tentacle" (LucasArts, 1993), and was widely imitated, for instance in "Simon the Sorcerer" (Adventuresoft, 1993).

The verb-select interaction model avoids the difficulty with text parsers of finding the right words to describe an action. If a player knows what to do, doing it isn't a problem. This does not render puzzles trivial, because the number of possible actions is usually large enough to make trying them all prohibitively time-consuming. However, a portion of the screen is usually permanently devoted to displaying the available actions and the player's inventory, leaving less space for game graphics.


One of the last verb-select games, Day of the Tentacle

Modal Cursor (1990)
Games with modal cursor controls are primarily mouse-based. The player interacts with the game through the mouse cursor, which has different modes for performing different actions. The mode is indicated by the appearance of the mouse cursor, and players cycle through the modes using the right mouse button. The modal cursor interface was invented for "King's Quest 5: Absence Makes the Heart Go Yonder" (Sierra, 1990). The modes were "walk to", "talk to", "look at", "manipulate/pick up" and "use inventory item with", and these became standard (although some games varied them slightly, such as "Gabriel Knight: Sins of the Fathers" (Sierra, 1993), which had a separate "ask question" mode, and "Leisure Suit Larry 6: Shape Up or ..." (Sierra, 1993), which had an "unzip" mode). The modal cursor was even adopted by LucasArts, abandoning their own verb-select interface, for "Sam & Max Hit the Road" (LucasArts, 1993).

The modal cursor has a number of advantages over verb-selection. It is not necessary to devote a portion of the screen to a list or table of verbs (the inventory is usually accessed on a separate screen). That means game graphics can be full-screen. Cycling through the interaction modes is often quicker than moving the cursor to point to a verb, then back to the object on screen. However, this advantage in speed soon disappears as the number of interaction modes increases. For that reason, modal cursor games usually have fewer ways to interact with things than verb-select games. This leads to the main disadvantage of modal cursor games, that it is easier to "solve" a puzzle by trying all possible options.


Quest for Glory 4: Shadows of Darkness uses a modal cursor (here in "look at" mode)

Simple Cursor (1993)
Simple cursor interfaces work like modal cursor interfaces, except that there are no modes for the cursor to cycle through. "The Legend of Kyrandia" (Westwood, 1993) simplified the verb-select and modal cursor interfaces, creating this model. A click with the left mouse key is a command to perform an action. The game chooses an appropriate action depending on context. The right mouse key is used in some games, for example "Beneath a Steel Sky" (Revolution, 1994), to perform a secondary action (typically "look at"). Mouse strokes (dragging) can also be used in order to manipulate objects on the screen, or to move items between the main game screen and the inventory. Probably the most famous game to use the simple cursor is "Myst" (Cyan, 1993), although it is not always accepted as an adventure game. "Broken Sword: The Shadow of the Templars" (aka "Circle of Blood") (Revolution, 1996)  uses a two-button simple cursor that changes its appearance depending on what action will be performed, similar to a modal cursor. In some ways, it can be considered a modal cursor where the mode is determined by the game, not the player.

The simple cursor shares the advantages of modal cursor control. One additional advantage is that it is very easy to use (which may in part explain the success of "Myst"). However, the constraints on interaction make it very difficult to design challenging puzzles in the traditional adventure game style.


The Legend of Kyrandia 2: Hand of Fate uses a simple cursor interface

Pop-Up Menu (1992)
In a pop-up menu (aka “verb coin”) interface, the player clicks on an object to interact with it, which brings up a menu of the possible interactions, right by the cursor. The menu can be fixed, or change depending on context. This interface was first used in "Lure of the Temptress" (Revolution, 1992), and around the same time in the less well-known "Nippon Safes Inc." (Dynabyte, 1992). Since then it has been used in titles such as "Return to Zork" (Infocom, 1993), "The Curse of Monkey Island" (LucasArts, 1997), "Gabriel Knight 3: Blood of the Sacred, Blood of the Damned" (Sierra, 1999), and "The Longest Journey" (FunCom, 2000).

Pop-up menus save screen space and allow full-screen graphics, like modal cursors. They can support richer interactions than modal cursors can comfortably support, and certainly more than simple cursors can. However, they can be fiddly to use, and they obscure large parts of the screen while the action is selected.


Lure of the Temptress

Display
There are fewer types of adventure game display types than there are control types: text description, illustration, third-person view and first-person view.

Text Descriptions (1975)
This is what text adventures have. There are no graphics, only a written description of locations and objects. "Adventure" (aka "Colossal Cave") (Don Woods, 1975) introduced text descriptions. They linger on in many graphic adventures as the description players get when they ask their character to examine the background or scenery.

Illustrations (1979)
Illustrations are static pictures of things in the game, in addition to or as an alternative to text descriptions. An illustration cannot in general be interacted with, and it doesn't necessarily show all the important things in a given location. Therefore, they serve to add color and style to text adventures, but don't really serve any purpose in the interface. The first adventure game with illustrations was "Mystery House" (On-Line Systems, 1979), from the company that later became Sierra.


Mystery House

Third-Person View (1984)
In third-person view, the player sees things from a point of view outside of the player's character, who is visible on screen. Third-person views can be broken down into 2D graphics and 3D graphics. In 2D, the point of view cannot move around freely (although the graphics may originally have been rendered in 3D). The first adventure game with third-person 2D graphics was "King's Quest" (Sierra On-Line, 1984). Possible views include top-down, side view, isometric and cinematic, and all of these have been used in many different adventure games. In 3D, the point of view can smoothly move around the environment, either freely or following the player's character. Possibly the first true third-person 3D adventure game was "Gabriel Knight 3: Blood of the Sacred, Blood of the Damned" (Sierra, 1999).

First-Person View (1985)
In first-person view, the player looks out of the eyes of the in-game character, seeing things as from their eyes. First-person views can also be broken down into 2D and 3D. 2D features static screens, and players move by jumps from screen to screen. The first adventure game to use this point of view was "Deja Vu: A Nightmare Come True" (ICOM Simulations, 1985). It was later used in games like "Dune" (Cryo, 1992) and "Myst" (Cyan, 1993). First-person 3D view, known from games like "Quake" (id Software, 1996), has been used in several games with strong adventure elements, with "System Shock" (Looking Glass Studio, 1994) an early and prominent example.


Dune gives a first-person view of the desert planet

DragonRose

  • Hey, look! Real skin!
    • I can help with play testing
    • I can help with proof reading
    • I can help with story design
    • I can help with voice acting
Re: GTD: analysis of game verbs
« Reply #2 on: 21 Nov 2004, 03:48 »
Yikes. Snarky posted while I was working on mine.  Some of my stuff might be redundant now... I haven't read his! But these were my first impressions.
Placement puzzles
The ability to drop inventory items does make for some interesting cases. In classic IF, there is a restriction on how much you can carry, forcing you to drop things in convenient places (but this could easily lead to dead ends if used carelessly, and people seem a bit spoiled these days with the limitless-pockets effect). But I've seen some good puzzles with item placement in Kyrandia, and the bait-the-dog puzzle in S301.
Of course droppable invitems is tricky to pull off in AGS, so that may stop some people from allowing each item to be dropped everywhere. If you want to save space, the 'get' icon could be changed to 'get/drop', depending on whether you click on the room or on your inventory.

In QFG1 VGA there was a similar idea.  In the inventory there were several options- look, select, use, drop and pick up. Whenever you dropped something in a room, the game recorded it.  If you returned to that room later and clicked "pick up," the item would be returned to your inventory.  However, there was no graphic to show that the item had been dropped. If you did something stupid like dropping your sword and forgetting where you left it... well, then you were screwed.

This worked very well, as the amount you could carry in the QFG games was determined by strength.  If you were carrying too much you could always drop something and go back for it later. In games without such stats, you might have to put an item limit or a size limit on items.  (Hmmm... that size limit idea is interesting- I might have to work on that a bit more).

Which brings me to another idea.  Different interactions with inventory items.  In SCUMM interface games you can use most of your interactions with your inventory. Some, such as "walk to" don't work, but it works as a generality.  In Sierra games, however, the inventory is in a completly different window, and has it's own interface.  Should a different set of interactions be applied to your inventory than to the rest of the game world? Having a full compliment of interactions, as was available in KQVI opens up some interesting ideas (talking to the Dangling Participle and the Rotten Tomato spring most readily to mind). However, it can also create the same problems as having too many interactions in the main world of the game. I don't recall there being any puzzle that actually needed to use the talk icon.

Quote
Funky responses
Of course one disadvantage of adding more verbs is the fact that the game should have a decent response to many of them. LucasArts games seem to get a way with a lot of 'I cannot move it' and 'Nice tree', whereas Sierra games are almost expected to have a snappy answer to any look/touch/talk click. Clearly this is a tradeoff, but having too many generic responses does lower the game immersion.

So if I were to extend a Sierra verblist, or change a Lucas one, my first recommendation would be along the lines of
walk look
touch move use-inv
talk attack get/drop
(and Laura's notepad and Larry's zipper added in those games that could use it)
Interestingly enough an 'attack' verb can be extended to inventory items (to break them), and a move icon could shake inventory items.
A very good final addition would be the 'do something appropriate' verb from the Garfield game, which has such nice uses as 'kick odie' and 'save jon'. Also it could feasibly open doors, since I omitted the open/close verbs from my sample above.

Putting on the devil's advocate hat for this one.

This is a definite problem with not having a generic "use" verb.   If there is some action that will only be necessary once, such as untying a knot, for example, none of your verbs listed will do it, except the "Do something appropriate" (shortening that to DSA because it's a pain to type.)  The problem with DSA is that it wouldn't work in a serious game. Also, it becomes the same as either a generic use/interact or those annoying "smart cursors" from games like Kings Quest 7 and Torrin's Passage.  I honestly don't see how DSA is any different from use.  There is, of couse, the option of just adding an "untie" verb, but it will most likely just end up as a mostly useless extra verb.
Sssshhhh!!! No sex please, we're British!!- Pumaman

Radiant

  • Mittens Knight
  • AGS Baker
  • Return once more to the Two Kingdoms!
    • I can help with publishing
    • I can help with story design
    • Radiant worked on one or more games that won an AGS Award!
    •  
    • Radiant worked on one or more games that was nominated for an AGS Award!
Re: GTD: analysis of game verbs
« Reply #3 on: 21 Nov 2004, 09:22 »
Snarky -> well, yes. He did cause me to think about certain things. But I'm just trying to get some discussion on the verb matter in general. The point was not to define a smallest set of verbs (that would be one - see Myst). The idea was to see which verbs would be interesting to see on an interface. LucasArts would be a good place to start. Of course there's nothing wrong with the existing Sierra/Lucas styles but it would be interesting to see some variations (and of course there already are some in AGS)

Dragonrose -> good point, DSA is very similar to use. I don't recall talking to your inventory was useful in KQ6 (even if it was fun). In fact I didn't even notice it the first time I played :)


Pod

  • Zzzzzzzaap
    • I can help with proof reading
    • I can help with story design
Re: GTD: analysis of game verbs
« Reply #4 on: 21 Nov 2004, 17:00 »
Verb-Select
However, a portion of the screen is usually permanently devoted to displaying the available actions and the player's inventory, leaving less space for game graphics.

Modal Cursor
It is not necessary to devote a portion of the screen to a list or table of verbs (the inventory is usually accessed on a separate screen). That means game graphics can be full-screen. Cycling through the interaction modes is often quicker than moving the cursor to point to a verb, then back to the object on screen.

I have to say both these points are none issues. You can easily have a verb list that acts like the sierra roll down menu, and you can have a modal cursor mode that acts like LEC's constantly-on-screen gui.

Also you forgot to mention the right-click look/talk/open in LEC games, which is a very fast way of doing things. It's a lot faster than having to cycle to "look" if you're on the next mode along (meaning you have to cycle all the way around to get it). Maybe scroll wheel implementation could be benefited here?

Mats Berglinn

  • But now I know must go... Back... to the mansion!
    • I can help with backgrounds
    • I can help with characters
    • I can help with play testing
    • I can help with story design
    • I can help with translating
Re: GTD: analysis of game verbs
« Reply #5 on: 22 Nov 2004, 11:28 »
Quote
Pop-Up Menu (1992)
In a pop-up menu (aka “verb coin”) interface, the player clicks on an object to interact with it, which brings up a menu of the possible interactions, right by the cursor. The menu can be fixed, or change depending on context. This interface was first used in "Lure of the Temptress" (Revolution, 1992), and around the same time in the less well-known "Nippon Safes Inc." (Dynabyte, 1992). Since then it has been used in titles such as "Return to Zork" (Infocom, 1993), "The Curse of Monkey Island" (LucasArts, 1997), "Gabriel Knight 3: Blood of the Sacred, Blood of the Damned" (Sierra, 1999), and "The Longest Journey" (FunCom, 2000).

Pop-up menus save screen space and allow full-screen graphics, like modal cursors. They can support richer interactions than modal cursors can comfortably support, and certainly more than simple cursors can. However, they can be fiddly to use, and they obscure large parts of the screen while the action is selected.

Full Throttle does also have a pop-up menu in form of a tatoo but the unlike Curse of Monkey Island there were two kinds of Interactions, the foot and the hand. Often you use the foot kick things and persons which you often did in Full Throttle.

YOke

  • 26 orbits and beginning to get dizzy...
    • I can help with backgrounds
    • I can help with play testing
    • I can help with proof reading
    • I can help with scripting
    • I can help with story design
    • I can help with web design
    • YOke worked on one or more games that won an AGS Award!
    •  
    • YOke worked on one or more games that was nominated for an AGS Award!
Re: GTD: analysis of game verbs
« Reply #6 on: 22 Nov 2004, 12:50 »
I'll try to summarize my thoughts about GUI's here. I'll leave the parser discussions to those who know and love them. I don't.

In my opinion you need an option for all your senses.
- sight
- touch
- hearing
- smell
- taste
in addition you have to move around:
- walk

After all, that is how we humans interact with our surroundings.

Sight, Look At or the Eye is perhaps the most important command in 2D adventures. There is a limited amount of visual information you can squeeze into a background so you need the extra help of a descriptive command. In 3D you can theoretically interact with an object using your Touch command, studying it from all angles. Then you can use the players own sight and make the experience even more involving by making the player make sense of what he or she sees for themselves.

Touch is interacting with your hands and is mostly split into Pick Up and Use, or simply the Hand in Sierra games. For all actions that obviously is performed with the hands, the hand is in my opinion self explanatory. Additionally you might have options such as Open/Close and Push/Pull. Only example I can think of using Touch as a sense is Indy: FoA when the lights are out in the digsite and in Atlantis.

Now we come to the rarely used Smell, Taste and Listen. The fact that theese are rarely used is more a sign of lack of imagination (or, more often the case, lack of will to add more descriptions). Let's say you make a Sherlock Holmes game, you can let the player smell and taste the evidence for clues. If you read the stories, they are filled with examples where vital clues are given by recognizing a smell or a taste.
Hearing can also be a good aid. What about putting your ear up to a door before busting through it, possibly interrupting a conversation that could give you some vital clues as to what is going on.

Walk really doesn't need any justification. On a side note one can mention the options of mouse or keyboard controlled walking.
Two notable examples of keyboard controlled games are Grim Fandango and Shadow of the Comet. Theese games are different in almost all aspects except for their interface. GF is 3D and has a higher resolution whereas SotC is low-res and operates on a small scale to boot. They have both solved the main problem of a cursorless game in the same manner. The problem being: "When I click USE, how can I be sure of what object the character will use?"

In SotC the game draws a line from the characters eyes to interesting objects, leaving no doubt as to what the character looks at. In GF they benefit from the higher resolution and non-static character sprites to make Manny's eyes follow objects of interest.
The main problem with this is that it's hard to place two objects close together without causing confusion or cumbersome twiddling of the controls, thus breaking the illusion for the player. This is the strength of the cursor.

This sort of takes us into a comparison of the Sierra and the LucasArts original interfaces and their strengths and weaknesses. Yes, there are more interfaces, and some are better, but I'll limit my discussion to theese since they are well known, and serve as good examples.

The strength of the Sierra interface is that they have boiled down the interactions to the essentials, offering all the same posibilities as LucasArts' but with less visual clutter.
The weakness of the pictographic cursors is, however, the lack of feedback. In some cases the cursor will change to indicate that you are over aninteractable object, in some cases it won't. If it does, you still might not know what the cursor is over unless you click on the item at get a verbal description. If it doesn't you either have to make the player click randomly all over the screen to find the objects, or you have to draw objects in such a manner that they stand out from their surroundings. In the latter case you might as well just put all the items in the players inventory as soon as he or she enters the room. This is where I feel the LucasArts statusbar does a great job. It gives a short description of the item your mouse is over and lets the player decide if interacting with that item will aid them in their current quest.

The downsides to the LucasArts GUI is the amount of space it takes on screen and the superflous commands. A good example of the confusion it can create is when Indy is operating levers in FoA. Push/Pull is the obvious choice for manipulating, but what isn't obvious is what direction the lever will move in when you use either command. It is rather pointless to supply the user with controls that do not offer complete control. They I would much rather prefer using just a hand, cycling the levers through their positions.

Yes... That's about it for now...

Enlightenment is not something you earn, it's something you pay for the rest of your life.

Dave Gilbert

  • Mittens Vassal
  • AGS Baker
  • Eye see you.
    • Lifetime Achievement Award Winner
    • Dave Gilbert worked on one or more games that won an AGS Award!
    •  
    • Dave Gilbert worked on one or more games that was nominated for an AGS Award!
Re: GTD: analysis of game verbs
« Reply #7 on: 22 Nov 2004, 14:44 »
I still think that the “best” type of verb interface was the parser used in the 80s Infocom games.  If you came across a rose in a modern point-n-click game, all you can do is click the USE icon on it.  You don't know what the result is going to be.  Your avatar might pick up the rose, touch the rose, take a petal off the rose, or even crush the rose to death.  The text games were more versatile.  If you wanted to pick up a rose, you'd type “GET ROSE.”  If you wanted to smell a rose, you'd type SMELL ROSE.  There were some downsides to this kind of interface, as well, if only because since the player can do anything, they WILL do anything. It's impossible to program every conveivable response to every action attempted within the game.  It's been awhile since I wrote text adventures, but I remember creating new verbs to be very simple.  Larry 7 merged the two interfaces pretty well, although it was kind of awkward. 

Pod

  • Zzzzzzzaap
    • I can help with proof reading
    • I can help with story design
Re: GTD: analysis of game verbs
« Reply #8 on: 22 Nov 2004, 17:50 »
I still think that the “best” type of verb interface was the parser used in the 80s Infocom games. If you came across a rose in a modern point-n-click game, all you can do is click the USE icon on it. You don't know what the result is going to be. Your avatar might pick up the rose, touch the rose, take a petal off the rose, or even crush the rose to death. The text games were more versatile. If you wanted to pick up a rose, you'd type “GET ROSE.” If you wanted to smell a rose, you'd type SMELL ROSE. There were some downsides to this kind of interface, as well, if only because since the player can do anything, they WILL do anything. It's impossible to program every conveivable response to every action attempted within the game. It's been awhile since I wrote text adventures, but I remember creating new verbs to be very simple. Larry 7 merged the two interfaces pretty well, although it was kind of awkward.

I've always felt a verbicon/Lure of the tempress style interface to be a perfect combination of both. Right click on a rose and you could (possibly, depending upon the programmer) get options of smelling, picking, crushing etc.

Scummbuddy

  • Mittens Knight
  • Give Stylish Confetti to BoZo - Wheee!
    • I can help with AGS tutoring
    • I can help with animation
    • I can help with backgrounds
    • I can help with characters
    • I can help with proof reading
    • I can help with scripting
    • I can help with story design
    • I can help with voice acting
    • Scummbuddy worked on one or more games that won an AGS Award!
Re: GTD: analysis of game verbs
« Reply #9 on: 22 Nov 2004, 20:43 »
New Verb: Punch

("It was just an example!1! I didn't mean for every game to have 'punch' as a new verb.")(Mittens Quote)
- Oh great, I'm stuck in colonial times, tentacles are taking over the world, and now the toilets backing up.
- No, I mean it's really STUCK. Like adventure-game stuck.
-Hoagie from DOTT

Ghormak

  • Hum hum hum
    • Ghormak worked on one or more games that won an AGS Award!
    •  
    • Ghormak worked on one or more games that was nominated for an AGS Award!
Re: GTD: analysis of game verbs
« Reply #10 on: 22 Nov 2004, 20:49 »
I've always felt a verbicon/Lure of the tempress style interface to be a perfect combination of both. Right click on a rose and you could (possibly, depending upon the programmer) get options of smelling, picking, crushing etc.

The problem with that is that puzzles can become quite obvioius. Let's use the rose as an example.

We right-click on it and are presented with the options "look", "smell" and "slowly remove the petals and thorns and then eat the stalk, keeping the thorns and petals".
You can be damned sure that at some point the solution to a puzzle is to slowly remove the petals and thorns from a rose, eating the stalk and keeping the thorns and petals for later use.

An exaggerated example of course, but I hope you see my point.
Achtung Franz! The comic

Babar

  • Creator, Mutator and Defecator
    • I can help with proof reading
    • I can help with scripting
    • I can help with story design
    • I can help with translating
    • Babar worked on one or more games that won an AGS Award!
    •  
    • Babar worked on one or more games that was nominated for an AGS Award!
Re: GTD: analysis of game verbs
« Reply #11 on: 24 Nov 2004, 15:41 »
Radiant, you are judging most of the verbs from the point of usage. For example the smell and tongue icons from SQ4 had no USE, but they were really funny, and added enjoyment to the game. I did like the idea of the Lure of the Temptress GUI, provided it can be recreated in a less cumbersome way than it was in the game. There should not really be the problem you mentioned Ghormak. Even important actions you would need to do would be broken down and placed into the list to appear just like any other. Of course, it would be great if there was an option to "slowly remove the petals and thorns and then eat the stalk, keeping the thorns and petals", just because it would be so funny.
I don't really think it is necessary to limit verbs. There is no need to limit them, even if they ARE used only once. In fact it would be a great point if there was a verb to be used only once, perhaps at the climax of the game.
Of course, if you want to limit verbs, a further limitation could be to have a "sense" icon. Clicking on it would tell you what you see, hear, feel, taste and smell all at once.
The ultimate Professional Amateur

Now, with his very own game: Alien Time Zone

Re: GTD: analysis of game verbs
« Reply #12 on: 24 Nov 2004, 21:26 »
Think of Zork Grand Inquisitor. There was one spell which was only used once, but didn't detract from the game. It was just like: 'Ah! So that's what its for!' I think the number of verbs is only really an issue if the GUI is permanently onscreen. Most of the time I do prefer fewer verbs, because some games just stump me, and it gives less possibilities to try. But sometimes its nice to have more verbs to play with - particularly if there are interesting and amusing responses to the actions, or just amusing actions themselves.