While I'm thinking of ways to improve the Creation Kit, let me add a few I forgot in my last post:
.
Firstly, the TESCS/GECK/CK tools desperately need an accessible dialogue system where:
- Non-voiced dialogue is automatically subtitled at a default display speed of 100 words per minute - which can be adjusted by the modder
- Existing dialogue statements automatically offer the option of selecting from the range of available and corresponding voice-overs.
- There are modular voice-over sets:
A.) based on a neutral reading of Ogden's 800 or, if you're daring, Ogden's 2000 word vocabulary.
B.) With a voiced dialogue assembly window displaying the modulation curve of "volume" (amplitude) & "pitch" (frequency).
C.) And with the option to set modification limits on a portion of modulation curve which can be modified, within those limits, by dragging up or down with the mouse.
Secondly, it would be better if non-modular objects were tied to the heightmap during heightmap editing. In other words, things like rocks, actors, trees, stand-alone buildings, etc. should have their altitude relative to the terrain mesh (i.e. immediately above or below their coordinate point) preserved during heighmap editing so that most existing objects do not wind up being buried or left hanging in the air by a little topographical tweaking. I can imagine, quite vividly, how not having to individually reposition every tree, rock, shrub, and loose pebble -every time the heightmap is edited- could be quite the time saver.
.
Thirdly, I haven't fiddled with
Papyrus yet, but I think it a good idea to include some tools for creating books which are editable in-game. This would be particularly handy for in-game challenges involving basic codes and ciphers, because cracking a cipher is very much like filling out a crossword. You start with the obvious and, using this as a basis for elimination, you deduce the imponerables. This usually involves adding your own notes and findings to the document in question. Another aspect of old literature is that of marginal notes. These can be very important, and would ultimately enrich the game. However, implementation would require separate but linked documents to be rendered seemlessly - and always at a somewhat odd angle to oneanother.
.
Finally, something I've always wanted to see in these games is some form of automated diary. The advantage of an autodiary, if it is recorded to a text file, is that snippets can be copied to modders and developers to demonstrate issues - in much the same way as screenshots are used to demonstrate graphic issues. For example, there were times in
Morrowind and
Oblivion when the wandering beasties just wouldn't let up - and the loot collected on the journey returned considerably more gold than the loot yielded by the quest. A snippet from an autodiary record would demonstrate the issue nicely. Also, if well developed, the appearance of autodiary entries on the web would point back to the Elder Scrolls and expand exposure, and possibly sales, considerably.
.
In a modular environment, the autodiary is very much easier to produce than it sounds. On any given day, conditions (e.g. weather) and surrounds (e.g. mean topographic variation i.e. in player Z ordinate position, modal near object class, etc.) can be characterised on a watch by watch basis with events and major cell changes recorded sequentially while non-aggressive actions are grouped on an hour by hour basis towards the end of a given watch paragraph. Describing the surroundings automatically may sound challenging, but it's fairly easy as a concept. Initially, no diary is generated until a statistically significant number of non-aggressive action deviations are recorded. By non-aggressive action deviations, I refer to where an object, such as a harvestable plant or container, was positioned in the player's field of view relative to the cross hair just before the player changed direction to harvest the plant or access the container. With enough of these coordinates, a modal map bounded by statistical or modal inflection can be used to estimate where a player is actually looking relative to the cross hair at any given time - based on the players dominant visual attenuation. Once this dominant visual attentuation is established for a given player, the classes of closest object passing through this zone can be counted over a watch period (e.g. forenoon watch) with the dominant two or three classes used to describe the surroundings encountered during that period. This can be repeated, for a given cell, whenever the player transitions into a cell not already described that watch - in much the same sequential order as the events of the watch. Events can be described (e.g. hostiles encountered, estimated force deployment in terms of squad, section, platoon, company or battalion strength; dialogue recording who said what) followed by their intensity (e.g. extended or brief, offensive or defensive) along with their outcomes (e.g. victory or escape; severe, light or no injury; diseases contracted; perhaps even loot taken, etc.). All it would take is for a modder to be able to come up with something like this is if
Papyrus has the capacity for large scripts or for scripts to directly invoke oneanother - along with persistent variable arrays and functions which allow the temporary recording of player position and crosshair orientation through time) as well as functions which can list objects within given radius of a selected point - not to mention the ability to record a log file in a sibling directory to the save games directory.
.
The advantage of an autodiary in the game for someone using the game's modding environment is that, if set to debugging mode, it could also be used to record AI anomalies which may not be noticed during game testing (e.g. actor path deviations, Actor AI pathing failures, exact scripting fail or termination points by script name and line number, etc.) and it could even act as a receptacle where all lines of script which get run are dumped sequentially and grouped by script. An in-game debugging log based on a system like this would allow testers to identify exactly where less visible problems are occurring (both in the sequence of play and in the mechanics of the game), which could streamline the testing phase of development considerably - especially where more complex situations are being represented within the game.