next »

[Thaumaturge]

Blog post (12th of November, 2018)
An Open Book It's Not

Summary: In which a collectible book is described; attendant elements are implemented; and some miscellaneous changes are made.

Greetings and salutations!

This week's screenshot shows a new collectible--a book that may be found in level two:



The week just past was a very slow one, I'm afraid. (In part due to my attempts--at last successful--to get an old computer up and running as a secondary machine.) Still, some work did get done:

The primary work of the week just past was the book shown above:

As the player explores the level, they may find, hidden away, a small cache of books. Now, these arguably belong to the adventurer who has claim on this section of the under-city--the person who our protagonist has come to meet. Having found the cache, our protagonist will report it to that adventurer--but she may also take a book as payment for the find.

The book in question is in a language that she doesn't understand, but it is clearly concerned with magic of some sort.

In addition to completing the book shown above, I made a few related changes: I added to the level a previously-made "book pile" model for the main of the cache; I produced a "closed" version of the collectible (based on the extant models for closed books, but modified to use the material of the collectible); and I implemented the logic for the appearance and removal of the "closed" model in the level.

Other than that, a few miscellaneous changes were made, too: The decoration of another (very small) house; some touch-ups to logical elements related to inventory items; a change in the mapping of a piece of level-geometry; and even a bit of (very nascent) work on a lore-entry.

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (19th of November, 2018)
A Tale of Letters

Summary: In which a new lore-tale is written; images are made for that lore; names are found for cities (and a coin); sunlight shadows are tinted; and a fade-to-black is added

Greetings and salutations!

This week's screenshot shows a new bit of lore: a light tale of literacy. (Although note that the text shown in the screenshot has been subsequently rewritten somewhat, if I recall correctly.)



The week just past was primarily a writing-and-art week--although a few other things were done, too:

The lore-entry shown above was perhaps the biggest addition to the game in the week just past.

This entry is associated with one of the "scenes" that the player might discover while exploring level two, which I believe that I described (in general) in a previous blog post. The story that it tells is... I suppose that it would be classified as "slice of life", perhaps: it's the story of a small family in a poor district of the city of old, and the place of literacy in their lives.

The writing of it took some time--and may see further revision in the future!

In addition to the text, an icon and page-image were made. The former represents the entry in the lore-collection (largely-hidden behind the page in the screenshot above); the latter, more-clearly visible above, is shown with the entry itself.

Choosing a page-image called for some thought. As mentioned above, the tale concerns literacy in a low-income family in the city being explored in level two. It was tempting, then, to depict the family members writing. But that would call for figuring out what they looked like, and what the clothing of that city in that time looked like--all of which will likely not be seen again in the game after this level.

Instead, I decided to take a more metaphorical approach, as can be seen above: letters, from first steps to their uses in adult life. And indeed, I'm quite happy with the result, I do think! ^_^

There was another challenge to be met in the writing of the story: I wanted to mention the name of the city in question--but I didn't yet have that name. In fact, I'd been somewhat putting off finding a name for both this city and another visited later in the game. For the past few years. More or less since I started working on the project in its current form, in fact. ^^;

So, at long last I sat down and looked for names for these cities, and at long last found them! This city, then, is named "Tenereth" (roughly: Ten-EH-reth).

Speaking of writing, I also added a name and description for a collectible coin that is found in level two. (The coin, by the way, is called a "Teketh".)

On the art-side, you may recall that I added a bluish tint to the darker areas under the player-light. In the week just past, I similarly added a blue tint to darker areas under the "sun"-light. While the player-light's blue tint is somewhat intense, however, this is a softer, paler blue. I think that it makes for a more pleasing aesthetic than the "sun"-lit areas previously had.

On the gameplay side, I added support for levels to temporarily fade to black in order to simulate lengthy tasks. This was put in place for use with the clearing of an obstruction that is encountered during level two: having it simply disappear in a moment seemed perhaps a little too effortless for so large a task. Furthermore, it may well prove useful for other such actions in later levels!

And finally, a few other changes were made that don't seem worth detailing here.

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (26th of November, 2018)
Letters for Lore

Summary: In which a lore-entry's scene is built; the logic for it is put in place; and additional sights are added to the level.

Greetings and salutations!

For this week's screenshots, a look at the in-level scene associated with the "Letters" lore-entry:





The week just past was primarily one of working on various sights of (hopeful) interest, small and large, in level two:

First of all, as shown above, I decorated the scene in which the "Letters" lore-entry (described in last week's blog post, I believe) may be found. It's a simple domestic scene, albeit one that's slightly more comfortable than many in this part of the city. Of particular interest, however, are the letters that might be spotted here, written in charcoal on the wall, and on a piece of paper on the table.

In addition to the visual elements, I worked on the logical elements by which the player gains said lore-entry: colliders for the examinable objects, and code to respond to them.

Other than that, much of the work of the week just past went into finding and adding to the level various little sights to be encountered: a bed stained with blood; a wooden figurine; scratches on a wall; and so on. Nothing the gives rise to anything significant, like a lore-entry, but little bits of additional detail and interest, I hope--much like the skeletal hand that was, if I recall correctly, shown in a previous blog-entry.

And there were a number of other changes, tweaks, and fixes done that don't seem worth elaborating on here.

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (3rd of December, 2018)
A Question of Secrecy

Summary: In which decorating is completed (sort of); some lore-logic is put in place; a secret is added--intended to lead to more; and the question is addressed of whether secret collectibles should be shown.

Greetings and salutations!

This week's screenshots show glass jars containing unknown substances, and another scene of interior decoration:




Quite a bit got done in the week just past, I feel! Much of it falls into the "lots of little tweaks and fixes" category, but there were a few more-salient changes made:

First of all, I finally completed the business of decorating the buildings in level two!

Well, sort of. I did subsequently discover that there were a number of issues to fix, including several objects that I'd apparently forgotten to properly attach to the scene. Furthermore, I might go back at some stage and enact some touch-ups, such as varying the colours of wall-hangings and carpets.

But at the least the objects in question are all laid out now, I do think!

(And indeed, I went through quite a through such buildings in the week just past.)

One of the buildings that I decorated has been shown once before, as I recall: I showed its upper floor, a small room with a cracked bookstand, a skeleton on the floor, and gashes in the stone wall. In the week just past I decorated its lower floor: a largely-empty set of shelves; a long single shelf holding wax-sealed jars with unknown liquids and powders, and a broken bone needle; nails in the walls, along with a scrap of paper; and a long, narrow carpet.

This was the work-place of a magic-user once upon a time, and is the location of a lore-entry. While that entry has yet to be written, I have now set in place the logic for acquiring it, I believe, with objects in place for the player to examine, and scripting to connect this investigation to gaining the lore-entry.

Another of these buildings is a little more difficult to access. Indeed, the means of getting into it are perhaps not terribly obvious at all. Which is appropriate, as it is the repository of the game's first secret.

The item that it holds is by itself nothing too special, perhaps, but it's intended to be the first of a set. Collecting all of that set unlocks in turn a final secret, a lore-entry otherwise not obtainable.

There was the question of whether slots for secret items should be displayed in the list of collectibles.

You see, the list of collectibles is visible to the player from the start--while it doesn't show uncollected items, it does show a full list of empty slots. This means that a player can easily see whether they've missed a collectible item somewhere. It also stands in contrast the lore-collection, which is intentionally a more-mysterious thing, with no on-screen indication of how many entries there are, or whether any have been missed.

But what then of secrets?

If I include a visible slot for a secret, then it gives away the presence of the secret.

Conversely, if I don't include a slot, then having a new entry appear in the list might leave it feeling less transparent, less trustworthy. Furthermore, it means that a player might think that they've found everything, supported by having a full list, when in fact they've missed items.

In the end I settled on showing these secrets. While I forget my thoughts at the time, I imagine that I weighed those last two arguments as heavier than giving away the presence of a secret.

Plus, the secret lore-entry to which they lead remains hidden, like other lore-entries.

In addition to the above, I made a lot of smaller changes that don't seem worth detailing here: work on a backpack model, as well as various changes, tweaks, and fixes!

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (10th of December, 2018)
The Troublesome Backpack

Summary: In which last week's decor issues are fixed; a grappling hook is seen lying on the street; and a backpack proves frustratingly difficult to get right.

Greetings and salutations!

This week's screenshot shows a place where, it seems safe to guess, something untoward once happened:



The week just past was perhaps not the most productive that I've had, but a few things did get done:

First of all, I believe that I mentioned last week that there were some issues remaining with the interior decorations performed during the preceding week's work. Indeed, there were objects incorrectly parented to the scene, issues with vertex colours of various stains (including the blood-stain above), a few superfluous tables that had ended up at the scene's origin, and a few other matters related to those changes. All of these have now, I think, been fixed!

(And in addition, there were a number of other such minor changes made that don't seem worth detailing here.)

A small amount of time went into modelling and placing a metal grappling hook, found lying near the entrance to the level. Its rope is damaged, so it's not of use to the player; it perhaps does, however, indicate how the adventurer being sought in this level usually gets around.

But perhaps the biggest change of the week just past was the modelling of a backpack. This might not seem like a major change, and indeed, I seem to recall that I went in thinking that it would be quick to make. Instead, it proved to be quite a pain. :/

At heart, I think that the problem was my UV-map, and in particular that I chose poorly where to split it. I don't know quite why I chose the specific mapping that I did, but I presume that it seemed like a good idea at the time...

So, I modelled the pack (which didn't take too long, as I recall), UV-mapped it, and in GIMP made colour- and normal-map- textures for it. It looked fine--save that there were ugly seams visible along some of the lines at which the UV-map was split.

I tried Blender's texture-painting tool, but the UV-map was ill-suited to it--especially as the map folded over itself in places (in an attempt to pack more into the texture).

I decided to try Blender's model-sculpting tool, intending to bake a normal-map from a high-poly sculpt. The actual sculpting went fairly well, I think. But I never quite got it to output a decent normal-map: the result was always either blank, or contained strange blank areas. (By "blank" I mean that it was of that blue colour that in normal-maps indicates "no change".) I was likely doing something incorrect on my end, but I don't know what. (I'm not at all experienced in Blender's sculpting and normal-map baking tools, or in the process in general.)

Along the way, I did try various alternate UV-mappings, as I recall, but those that I tried didn't seem to solve all of the problems that I had before me in the given instance.

Going back to the version of the normal-map that I painted in GIMP, I tried using Blender's texture-painting tool to smear normal-map colours across the edges. This seemed to sort-of work, but nevertheless, unsightly seams remained.

In the end, I primarily solved the problem by reworking my UV-map once again. The new map is no longer folded over itself in a manner that causes problems, and the sections whose split resulted in seams have been welded together. More painting, both in GIMP and a little in Blender's texture-painting tool, produced a normal-map that I think works fairly well. Furthermore, the geometry long which the seam ran was found to be a bit too sharp; bevelling it a little seems to have improved the pack's appearance further.

The result can be seen below. It's not perfect, but I think that I'm happy with it!



That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (17th of December, 2018)
Why I Broke a Table

Summary: In which audio effects are made from extant sounds; doing so proves tricky with a longish action; names and descriptions are written; a new lore-entry is begun; secret collectibles are identified; and side-tables are broken.

Greetings and salutations!

For this week's screenshot, an updated version of the blood-stain that I believe that I showed last week:



The week just past was a fairly busy one, I feel! There was audio work, writing, and the usual variety of visual touch-ups, tweaks, fixes, and suchlike:

First of all, I finally turned to some audio-work that I'd perhaps been putting off, making several new sounds for various events around the level.

What's more, all, I think, were constructed using elements from sounds that I had previously made.

This re-use proved quite difficult when it came to the sound used for shelving being pushed aside. As it turned out, the animation for that took longer than I'd thought, and I didn't seem to have quite enough samples on hand to make up the full duration.

In the end, with some editing and adjusting, I managed to use some of the samples more than once without--I hope--that fact being easily audible. Indeed, I think that I'm reasonably happy with the final sound!

That said, there are two sets of moveable shelves in the level, and the sound mentioned above was made for the more quickly-moved of the two. Moving the other took about half a second longer (I think it was), and I wasn't quite sure of what to do for it, as I recall. I tried simply having Panda3D reduce the playback rate of the sound, such that it became long enough--and it seems to have worked, I'm glad to say! ^_^

Concomitant with this came a little extra implementation of logic for the sounds in question--but nothing major, I believe.

As to writing, I set down or edited descriptions for a number of objects that can be viewed in the level, as well as two collectibles.

Furthermore, I've started work on the second lore-entry that may be found in the level. Where the previous lore-entry was somewhat of a slice-of-life, this one is a tale of magic in desperate times.

Speaking of collectibles, I've added a new "empty slot" icon for secret items in the collection list, as well as some simple logic to detect whether an item is "secret" or not. (In short, if its name-ID starts with the word "secret", then it's considered to be a secret item.) Concomitant with this, I remade the standard "empty slot" icon to be of a higher resolution.

Other than that, as mentioned above, I made a variety of changes, tweaks and fixes. Most of these don't seem worth detailing here, but there is one that I would like to mention:

In a few places around the level there are small side-tables. Like most objects, these are static, non-interactive things. (Well, aside from being solid objects that the player can potentially climb up onto, given enough space.)

The problem is that they are indeed fairly small--small enough that they look like something that the player-character might be able to pick up. In other words, they look like potential carriable items.

Now, I could make them so--but carriable items are dangerous to level design. Given enough of them and a sufficiently patient player, they can trivialise most traversal challenges. Thus I really don't want those tables to be carriable.

Left as they were, I fear that they could potentially damage immersion, and muddy the "language" of the game--the way that the game communicates to the player what actions are and are not available.

I could also just not include them--but there are places where I feel that it makes sense to have them.

My solution--one that I'd deliberated a few times before the week just past, as I recall--was simple: break them all. Instead of nice, sturdy-looking tables that look like potentially challenge-killing stairs, they become pieces of damaged scenery. They still more or less fulfil their purpose in the level, without endangering the gameplay of it.

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (23rd of December, 2018)
The City in Sunlight

Summary: In which a sunlit part of the level is built; cladding turns out to be shinier than expected; the level is named; the level's state of completion is indicated; a Christmas holiday is mentioned, along with its intended effects on blog-posting; and a holiday well-wish is given.

Greetings and salutations!

For this week's screenshots, the sunlit place above level two proper, in contemporary Tenereth:




The week just past saw a return to level-construction:

Almost all of level two takes place in the dark, in the under-city beneath Tenereth. Almost all, but not quite all: the part of the under-city visited here is accessed from a building within living, daylit Tenereth, and that building provides the entrance and exit to the level.

For quite some time this area has been prototypical: functional to a basic degree, but not pretty. In the week just past, then, I set about attending to that.

The building itself was simple enough: The floor-hole that allows access into the level proper called for some detailing, and the UVs of that hole were a little tricky, but for the most part things went smoothly, I think.

The view beyond the exit-doorway was a little more challenging, but not greatly so, as I recall. Through the open doorway we see the little buildings of this area across a quiet, broad-cobbled street; behind all that is a blue sky, with a few wispy clouds.

There were a few challenges: elements that didn't look right, or that weren't quite as I'd hoped. Some of these I managed to fix; some still aren't quite as I'd like, I fear.

Along the way, I discovered something odd: my "stone cladding" material, used quite widely in the under-city, had been using the normal-map intended for ceramics. Since A Door to the Mists stores "shininess" information in its normal-maps, and since ceramics are intended to be somewhat shiny, surfaces using this "cladding" ended up rather shinier than expected! I had presumably missed this somewhat in the under-city, under the "player-light", but in the light of the "sun" it was quite apparent.

On the other hand, I had been using that "ceramic" normal-map for some time, and liked the texture of it. Furthermore, the actual "cladding" normal-map turned out to be somewhat small. So, instead of using that "cladding" normal map, I exported a less-shiny version of the "ceramic" normal-map to replace it, and a using that in the level.

This area is not quite finished yet--at the least there are a few things that I still intend to do. But I do think that it's pretty close to done!

On a minor writing note, I now have a name for level two: "A Light in the Dark".

And as is often the case, a number of minor changes, fixes, and suchlike were done in the week just past that don't seem worth detailing here.

Overall, I think that the level is getting there; that it is, at last, drawing near to completion! ^_^

Christmas falls during the coming week, and I intend to take the week off (which is the reason for this entry being posted early). As a result, there will (presumably) be no blog post for that week; instead, I intend that the next entry be posted on either Monday the seventh or Tuesday the eighth of January.

To all of you who read this blog, if you observe or celebrate anything during this time of year (or have just recently), I hope that it's wonderful (or was so, in the parenthetical case). If not, then I hope that the period is nevertheless a happy one for you. Furthermore, I hope that the year ahead is one of great happiness for you!

That's all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (8th of January, 2019)
Trouble With Walls and Levels

Summary: In which wall-edges prove troublesome; level-importation is slow; and street-edging is begun.

Greetings and salutations!

I don't have a screenshot for this entry, I'm afraid: the week just past was not a hugely productive one, I fear. Between a disruption to my usual routine (a pleasant one, at least), a day off, and difficulties with the work itself, not an awful lot got done. A little bit, but not a great deal:

To start with, I began work on UV-mapping the edges of walls.

I forget whether I've mentioned this issue previously, so let me explain:

UV-mapping the broad sides of walls is fairly straightforward. There may be some minor discontinuities at corners, but I think that they're nothing too serious.

The problem arises when a wall's narrow edges are visible. For example, imagine a ruined building, its missing roof exposing the tops of its walls. Now, the UVs on one side of a wall closely match those of the other. This means that the UVs of the edges between one side and another will end up pretty much flat, crossing few-to-no pixels, and thus result in the texture being "smeared" across the edges.

So, how does one align the UVs in this case?

One idea that I had was to cut the edges down their middle, and then offset those middle UVs. This would allow the edges to use more of the texture, and thus reduce smearing, at the cost of being "mirrored" along the mid-lines.

In the week just past, I tried it, and... it didn't work. Aside from being slow and tedious, it turned out that the normals didn't function as expected when mirrored across the mid-line, producing a visible artefact.

I have one or two ideas that I might yet try--we'll see how things go...

Another issue that I've been dealing with is that, as levels become larger, the process of exporting them from Blender and importing them into Panda3D becomes very slow. On my computer, level two can take hours to import. This interferes with development at times.

In the week just past I put some work into addressing this issue. I don't want to simply pass the time off to players--even a small increase in loading times might add up across many players.

But, having asked for help on the Panda3D forum, I may have at least one potential solution to investigate: it seems that it may be possible to export levels piecemeal, then merge them into a single file once they're ready.

Finally, I started in on adding curbs to the streets of level two. This seems like it might not be as slow and unpleasant as I'd feared, and I think that it improves the appearance and verisimilitude of the level. However, it's also a little too nascent for me to be comfortable in showing it just yet.

And that is all for this week! Stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (14th of January, 2019)
Curbing the Streets

Summary: In which streets gain curbs; and a script helps with this.

Greetings and salutations!

This week's screenshot shows the new street-curbs of level two:



The week just past was a bit of a slow one, perhaps, and a bit of a short one, too. As a result, not a lot got done--although somewhat did:

The main thing done in the week just past is shown above: the streets of the under-city in level two now have curbs.

I had been a little hesitant about undertaking this: it seemed likely to be a slow, tedious affair, and I wasn't confident that the result would be worth it. And the task wasn't exactly fun--but it wasn't nearly as bad as I feared, either.

To start with, I quite liberally duplicated my curb-models from place to place, speeding up the process of laying them out. Since many of the buildings are of comparable sizes, only minor adjustments were frequently called for.

But even so, the streets of level two have, in general, a rather distinct downward tilt, and can otherwise be somewhat uneven. This means that, in many cases, curbs will call for editing in order to match the shape of the street that they follow. And manually editing each curb to that end could very well become tiresome and time-consuming.

So I wrote a script to do the job for me. In short (as far as I recall offhand), it does the following for each curb: First it examines the geometry of the curb and classifies it into a set of "rings". It then, for each "ring", uses Blender's built-in ray-casting feature to find the height of the street below, and adjusts the position of the "ring" to match. Finally, it determines which street-mesh was most commonly hit, and makes the curb a child of that street-mesh.

It seems to work quite well, I do think!

As to the overall effect of the curbs... I think that they help with the look of the place. It doesn't look right to me to have the streets just suddenly become vertical walls, I think; it feels unfinished.

That said, I also think that there is some adjustment and tweaking that I want to do.

The curbs aside, I made a few other changes and fixes within level two that don't seem worth detailing here.

That, then, is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (21st of January, 2019)
Walls Edging In

Summary: In which curbs are given touch-ups; "charcoal smears" are added; a solution to texturing wall-edges is found; a bit of writing is done; some decals are added; and some geometry is tweaked.

Greetings and salutations!

This week's screenshots show some "charcoal smears", and--at last!--UV-mapped wall-edges!




The week just past was a more productive one than its predecessor, I do feel! Perhaps not my most productive ever, but I do feel like I'm finding my feet again somewhat. ^_^

To start with, I touched up the curbs that I believe that I described last week. In particular, I found a number of places in which I felt that the curbs ran too obstructively past a doorway, and (hopefully) improved those.

I also adjusted the positioning of a piece of traversal-related geometry so that it no longer intersects a curb. This makes the related jump a little bit easier, perhaps, but not so greatly that I'm terribly concerned about it.

On the decor side, and as shown above, I went about and added some "charcoal smears" to a handful of places in the level. The goal is to give the impression that there were children at play here, scrawling on the walls, once upon a time. Or at the least provide a little bit of dirt here and there.

I do worry that these marks are a bit too sparse--perhaps I should add more still.

Perhaps more exciting to me, and also shown above, I finally have an answer to wall-edging! :D

It's a much simpler solution than my previous one: In general, I just re-map the edges such that they fit into a single row of blocks on the wall-textures, and then more-or-less line up the grooves.

This isn't a perfect solution: the grooves don't always match well, and the process sometimes involves adding vertices in order to convince things to line up. Furthermore, there are discontinuities where the edges meet the sides, and sometimes other edges.

But for the most part it seems to work well enough: the discontinuities don't seem to be too terrible, and the grooves seem to look reasonably connected, I think.

Furthermore, while the process is a little slow and fiddly at times, it's nowhere near as tedious as my former approach!

I've started in on edging the various exposed wall-edges in level two, but it's still very much a work-in-progress.

Otherwise, I did some writing, working on a lore-entry that may be found in level two; I added some scratch-decals to a few places--but I feel that some of these don't work well; and I tweaked various bits of geometry.

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (28th of January, 2019)
Considering a Menu

Summary: In which a tutorial-in-the-making results in a short week; wall-edging continues; geometry is cleaned up; lore-writing is done; and changes to menus are considered.

Greetings and salutations!

The week just past was a bit of an odd one: Partway through the week I decided to take on the project of making a "beginner's tutorial" for Panda3D. (Panda3D being the game-engine that I use, for those not aware of that.) As a result, for A Door to the Mists the week just past was only a half-week:

I did get some work done: I'm making progress on dealing with the wall-edges--it's still a slow process, but I think that I'm getting better, and perhaps faster, at it. (And during the process, I cleaned up some iffy geometry that I encountered.) I also did some writing-work on a lore-entry that may be found in level two.

On the design side, it would be nice, I think, to be able to access uncovered lore-entries from the main menu--especially in the case of the intended "secret epilogue" entry, which may only be discovered right at the end. For that reason, I've been thinking about adding to the main menu a button that provides access to all lore-entries discovered during play.

The problem is that of where to put it, without disrupting the look of the menu. I had previously considered a few potential solutions, but felt unhappy with all of them, I believe.

During the week just past a new idea came to me: The "save" button is only really used during gameplay--before a game has been begun or loaded, it's superfluous. So I might perhaps remove that button from the main menu--thus freeing up space for a "lore" button--and instead place it in a new, in-game "pause" menu.

I'm still considering this possibility: while it solves the problem of the "lore" button, and arguably makes for a more-elegant use of the "save" button, it does incur some additional work, possibly including new menu-logic.

Finally, I currently estimate that the tutorial that I'm working on will take another half a week, so next week's blog-post may once again be short. We'll see how things go!

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (4th of February, 2019)
The Week of the Tutorial

Summary: In which tutorial-making expands to (more than) fill a week.

Greetings and salutations!

You may recall that last week I reported that I was working on a tutorial for Panda3D; this week's screenshots show some of the work done for that:





I believe that in last week's blog-post I estimated that it would take about half a week to complete my Panda3D tutorial.

It did not work out that way:

Indeed, the tutorial is still a work-in-progress! As a result, for the moment A Door to the Mists is on hiatus. I hope that this hiatus will be brief--but once again, we'll see how things go!

As to the progress of the tutorial, it's coming along, I think. The tutorial-game itself is functionally complete, I believe, but still wants for some assets. (Indeed, I'm currently working on a little environment model for use in it.)

Once the game is done (and the code perhaps cleaned up and commented), there remains the task of constructing a tutorial from it. My intention at the moment is to use it as a basis for a set of "steps", each producing a working program (if not a full game), and building up towards the full, final code of the game. These would be paired with a written tutorial, explaining what is being done in each.

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (11th of February, 2019)
The Expanding Tutorial

Summary: In which a WIP main-menu for a side-game is shown; the tutorial continues to take time; and key-mapping complexities for A Door to the Mists are contemplated.

Greetings and salutations!

The week just past was once again taken up by my "Panda3D Beginner's Tutorial", so once again A Door to the Mists remains on hiatus. Having no screenshot for it, let me instead share an image from a side-game that I've been developing during off-hours. It's a short "ghost corridor" of sorts; a set of little spooky set-pieces for the player to encounter. It's titled "Night River", and this is the work-in-progress main menu for it:



As I said, the week just past was taken up by work on my tutorial. I'm making progress, I believe--I'm partway through the writing process at the moment--but it's proving to be a much bigger task than I had anticipated!

Indeed, I've already written nearly as many lessons as I had originally planned--but I'm only about halfway through the set of lesson-programs that I had set out.

Still, I'm hoping that the pace of things will speed up a bit: much of the expansion of the tutorial came from discovering that some of the most fundamental elements of Panda3D seemed to call for more explanation than I'd realised. In addition, some of the upcoming lessons may be less Panda-related than the early ones, and can perhaps be somewhat skimmed over. We'll see!

With all that said, A Door to the Mists hasn't been forgotten. Indeed, with Panda3D having recently added support for gamepads (and other input-devices, I think), I've been giving thought to how I might handle support for those.

There's nothing terribly complicated about any one input system, I think. The complexity lies in unifying them, and upgrading my keymapping-module to handle them in a manner that's elegant and convenient for both me as a developer and for players interacting with its UI.

For example, right now A Door to the Mists uses button input for movement: the player presses the button assigned to "step left", an event is called, a flag is set, and when the player-character is updated, its movement code checks that flag and responds accordingly.

But a gamepad thumbstick doesn't generate events in that way. Instead, one is expected to poll an axis.

I don't want to complicate my player-code by including multiple approaches to handling input. Instead, I want my keymapper-module to present a unified interface, so that the player-code can just query it and respond.

And then there's the player-end of it: how should this be presented to the player? Should there be different key-mappings presented depending on which input-method is selected? Should I allow players to mix-and-match input-methods as they like? And so on.

I have things to think about here, I feel.

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (18th of February, 2019)
Tutorial Complete!

Summary: In which the "Panda3D Beginner's Tutorial" is at last complete; a link to it is given; and more thought is given to handling non-mouse GUI-inputs in A Door to the Mists.

Greetings and salutations!

For this week's screenshots, a look at the game that's built up in my "Panda3D Beginner's Tutorial":



Once again, the week just past was eaten up by work on the "Panda3D Beginner's Tutorial"--but this time, I actually completed it!

This tutorial was a project that turned out to be much larger and more draining than expected. I'll confess that I'm happy that it's done! Indeed, although I finished on Friday, and thus could perhaps have worked on Saturday, I decided to take that day (more or less) off as I was feeling rather fatigued.

Still, I think that I'm glad that I embarked upon writing my "Beginner's Tutorial"--especially as I've already had some rather positive feedback on it! ^_^

As to the tutorial itself, I've added an entry for it on the "Projects" page of this website. For the sake of convenience, should you want to read it, let me also link to it here:
https://arsthaumaturgis.github.io/Panda3DTutorial.io/

Regarding A Door to the Mists, I've been giving more thought to how I might handle alternative input-methods (e.g. gamepads) in the UI. It's looking tricky, but I have some thoughts as to how I might approach it, at least.

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (25th of February, 2019)
Level 2's Modelling... Complete?

Summary: In which I return to A Door to the Mists; wall-edging is (I think) completed; geometry is touched-up; the streets of level two gain some detritus; more detailing is considered; level-modelling may be completed; what remains to the level is mentioned; and conversion to Python 3 is undertaken.

Greetings and salutations!

I don't have a screenshot this week: while work was done on level two that might be shown, the level has yet to be exported from Blender, then imported into the game, and Blender doesn't show it as I'd like. (Exporting and importing is a lengthy process, I fear. I've been given a potential solution for that--but it will likely only be attempted after this level is done.)

As you may gather from the above, the week just past saw me return, at last, to work on A Door to the Mists! Most of that work was done in the modelling of the level itself--in applying polish, specifically--but there was also a technical matter to attend to:

To start with, I returned to pretty much where I left off when I began the tutorial: fixing the narrow-sides of various damaged walls. For the most part this went fairly smoothly, I think, and I believe that I now have all of them done. (Save one set that the player isn't expected to reach or clearly see, and which I've thus left.)

I do wonder whether there isn't a better way of approaching this issue in later levels--it was a rather slow process, and a tedious one, too. Perhaps more-extensive use of prefabs for such broken walls? I don't know, but it's perhaps worth giving consideration to, going forward.

I also touched up various bits of geometry, fixing iffy sections and bevelling sharp edges.

Next, I moved on to a bit of detailing. Specifically, I scattered various little props about the streets: little chunks of stone; bowls; stone sherds; and the occasional bucket or broken chair. Nothing that affects gameplay, but which hopefully makes the streets feel a little less pristine.

It's tempting to go further: I would love to have the walls of the city splashed with stains and marks; to place an edging of sand and dust on the streets; or to add various scratches, nicks, and scrapes to the place. But conversely, this level has taken a long time to bring to this point, and adding those things might take significant time further. Are they worth it? Will they improve the level enough to justify their inclusion, given that I'm not working with a team across whom to spread the workload?

It's tempting... but right now I'm leaning towards leaving them out.

And if I do indeed leave them out... then I think that the modelling of the level itself... may well be done! I have yet to export the level and check it--it's entirely possible that I've missed something, such as leaving a prop incorrectly attached to the scene. But if the level is done, then this is a bit of a milestone! :D

Does this mean that the level is complete? Not quite: there are still a few things that I have in mind to be done. Off the top of my head, I have some audio-work; some things to be done in the level-editor; some writing; and perhaps most saliently, the cutscene that introduces the level.

Still, I feel like I'm getting there! ^_^

And finally, I mentioned a technical matter at the start of this post:

If I'm not much mistaken, support for Python 2.7--in which I've been working--is ending in the not-too-distant future. Furthermore, Panda3D seems to be leaning more on Python 3 these days, with occasional issues cropping up in its support for 2.7.

As a result, I've switched over to developing in Python 3.

For my little side-project, Night River, this wasn't a major issue. There's a tool called "2to3" that handles some of the basics of converting one's code, and which seemed to do most, perhaps all, of what was called for.

But A Door to the Mists is a larger and more complex project. The "2to3" tool made a solid start (once I discovered how to get it to convert all Python files in my directory structure in a single command). But there were still a number of things to convert by hand. The changes to strings in particular seemed to be a pain, now calling for additional encoding, decoding, and prefixing. The change to integer division wasn't in itself a problem--the trick there was finding all divisions that were intended to be integer. (And I hope that I found all such divisions!)

In the end, however, I think that the conversion is done. At the least, the game seems to load levels and run as expected!

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (4th of March, 2019)
Text and Fixes and Night River

Summary: In which level two's model sees more work; various bits of progress on the level are made; more writing is done, both lore and otherwise; some convenience debug features are added; a number of bugs-fixes and minor changes are made; and a side-game is released!

Greetings and salutations!

Once again, I have no Door-to-the-Mists-related screenshot this week: the work of the week just past was again focussed on a great many little changes, fixes, and suchlike, none of which really make for much of a screenshot, I feel.

Instead, have a release trailer for my side-game, Night River:




The week just past was a busy one, I feel! Not only did I do quite a bit on A Door to the Mists, but I also released the above-mentioned little side-game, Night River! :D

Do you remember that I mentioned last week, I believe, that I thought that the modelling for level two might have been done?

I was wrong! :D

As it turned out, there were a great many little fixes, changes, and touch-ups to do. There were broken walls that had not yet been edged. There were curbs that had visible gaps when seen from some directions. And a variety of other little things to be done.

Is it finished now? I don't know--the next export-import of the level from Blender and into the game may tell...

Beyond that, I made a bit more progress with the level, adding or fixing things for which I had been waiting to have the updated level. Few of these are worth detailing here, I feel, but it is nice to have them finally done.

One that might be worth mentioning is that the level now has an actual "exit" object. This means both that the player can now move to the next level once the objective in this one is done, and that the player can no longer step out of the bounds of the level and fall into the void. ^_^

Perhaps more interestingly, I did a bit of writing in the week just past.

Some of this was small: adding a better description to the "distant light" in level two, or adding a line that I had missed in level one.

I also added text for a guard-log that may be found in level two. It's nothing terribly exciting, but I feel that it adds a bit of "life"--or a sense of life that once was--to the level.

But perhaps most saliently, I think that I've finished the second and final lore-entry for level two: a tale of desperate magic against a plague. What was the result of that magic? Well, if you discover the entry in the level you may find out...

On the technical side, I added two debug features to the game proper (that is, as opposed to the level-editor; the version that has the main menu, and plays cutscenes, etc.) for the sake of convenience:

With three levels in the game now (including the prologue), it can become tiresome to play through everything before just to test something in a given level. To that end, the game now allows me to press one of the number-keys to skip to the desired level!

When debugging, it can be a problem that the game-proper confines the mouse to the boundaries of its window. (Especially as this confinement seems to persist even when the window is no longer in focus.) Thus, I've added a key-press that toggles mouse-confinement, thus freeing the mouse for debugging tasks.

And through all of this, I encountered and attended to myriad bug-fixes and little changes! Most of this is once again not worth detailing here, I feel. One particular issue--a rather puzzling inconsistency in text-nodes--is still being investigated.

Finally, as mentioned above, I released something this week! That something is my little side-game, "Night River". It's a short, simple, spooky experience, in which the player boats down a black-glass river and discovers the various ghosts that may be seen along the way.

If you're interested, you should find it here: https://thaumaturge-art.itch.io/night-river

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (11th of March, 2019)
Illustrated Lore

Summary: In which level two sees more work; document-display is (hopefully) improved; a lore-entry gains icon and page-image; title-screen art is reconsidered; and Night River receives an update.

Greetings and salutations!

For this week's screenshot, a lore-entry:



The week just past was... a bit of an odd one for me, and occasionally a little rough. As a result, not all of the week's usual work-time was given to work.

Still, some things did get done:

To start with, level two's model received further touch-ups, fixes, and changes. Once again, I think that I may have it finished at last!

I also worked on an issue related to how I display document-text. Specifically, some paragraphs in my lore-entries were ending up spaced further from their predecessors than others.

After investigating, this turned out to be a mistake on my part: I was adjusting the word-wrapping of my lore-entries after placing their paragraphs, resulting in some of those paragraphs becoming one line shorter than expected.

My solution was to simply do away with that post-hoc adjustment, and to use a single word-wrapping setting for all documents. With that, and some concomitant adjustments to my document-margin settings, the problem seems to be solved. ^_^

Perhaps more interestingly, I made a page-image and icon for the second lore-entry that may be found in level two. The former may be seen in the screenshot above (along with the newly-aligned text). (The latter is hidden behind the document itself.)

I've also been giving thought to my title-screen art. While I like it, I feel that it's perhaps a little too downbeat for the tone of the game itself. Thus, I've been trying to come up with something more fitting. I've experimented with recolouring the current art, and I've considered other possible designs, but either way haven't yet found anything satisfactory; more thought may be called for here...

And finally, Night River received an update in the week just past. This update increased the dead-zone used for thumb-sticks, hopefully making them more usable, and made the behaviour of the game's window more reliable.

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (18th of March, 2019)
What's on the Menu?

Summary: In which level two's model sees yet more work; a sound for "the city by day" is made; the "distant-light" is logically renamed; level two's intro cutscene is advanced by a step or so; lore may be accessed from the main menu; an in-game menu is added; and the main menu is somewhat redesigned.

Greetings and salutations!

For this week's screenshot, some changes to the main menu:



The week just past was a varied one: a week of level-touch-ups, art, menu-work, and more:

To start with, the final touches on the level-model for level two have continued, as I found several more things that I felt called for fixing. Once again, I'm hoping that I've made the final such edit!

In addition, I turned to a final piece of sound-work for the level: the sounds of the living city outside of level two, standing above the under-city which composes the main of the level. Instead of the complex, dynamically-arranged sounds of the previous levels' surrounds, this is simply a single, five-and-somewhat-second looping sound-file: I don't expect players to spend much time in the area from which it is heard.

On the logic-side, the name and description for level two's "distant light" is now changed when the player reaches a certain point. (Specifically, a point at which it's reasonable for the player-character to easily see what that light is, and so name and describe it more precisely.)

And on the art-side, I've taken some preliminary steps towards the level's introductory cutscene.

Moving away from the level, the game's menus saw some work in the week just past:

I've been thinking for a while now about giving the player access from the main menu to their collected lore.

For one, it would be convenient, allowing them to read or re-read lore-entries without loading a saved game. For another, I have it in mind to give one last lore-entry--if certain conditions are met--at the very end of the game. If I do so, however, it might not be feasible for the player to access this entry from the in-game collection. Thus, it would be handy for the player to be able to access their lore from the main menu.

Plus, before a game is begun there is no player lore-collection to access--so lore is now additionally stored in an overall collection, which the main-menu button gives access to. This has the bonus effect of allowing players to potentially accumulate lore over multiple runs, should any be missed in a given playthrough.

The problem was that of where to put the button for new lore-collection: the main menu was already somewhat full. My solution was to make space by removing the "save" button, and to instead implement a new in-game menu that allows manual saving. After all, there's no point in saving when a game isn't underway.

For a while now, I think, I've been a little dissatisfied with the background art that I use for the main menu--specifically, I've felt that it's perhaps a bit melancholy in tone.

Thus, I've spent some time attempting to design a new backdrop. And a major problem that I kept bumping into was the layout of the menu: with the buttons centred, and an unknown amount of space to either side, I found it hard to design something satisfying.

Moving the buttons to the left would solve this problem: the backdrop would no longer be bisected down the centre, thus allowing more space given to a single image.

I hesitated to do so: I quite like having the menu centred, and there were some questions as to how best to place a left-aligned menu. However, as you can see above, in the end I relented. And I'm glad that I did: even the now-old art is, I feel, rather better shown without the menu blocking it.

As to the art itself, I actually rather like the design of the old art. However, reworking it to be less melancholy would be tricky, and fiddly, in part due to my--alas--keeping rather a lot of it on a single layer. Thus I've begun work on a new version, designed with the new menu-placement in mind--and using rather more layers this time!

And finally, Night River saw one more update in the week just past: it now has both full-screen support and a (very rough, and amateur) Korean translation!

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (25th of March, 2019)
Title Art Reworked

Summary: In which new main-menu art is made; a bug is fixed; and IndieDB is joined.

Greetings and salutations!

This week's screenshot once again shows the main menu--but now with a new backdrop!



The week just past was, for the most part, an art week:

As shown above, the main menu has some new background art. And that background was the work to which the majority of the week was given.

It was a tricky piece to put together, as I recall, in composition and colouring. (Which may be one reason that it took nearly all week to reach the point shown above.)

A particular issue was that the orange section originally stood out rather too much, I felt. Next to the various greys and pale browns of the other sections, it was eye-catchingly vibrant.

My solution was to dim much of it.

The scene was originally lit as though most sections were open to the sky, and thus lit by the sun--meaning that there was light on much of the exposed surface area. To dim that bright orange, I changed the lighting such that "interior" sections--which includes the orange one--were instead lit as by the player's lantern-light, thus shadowing much more of them. Furthermore, I introduced the blue shadows used in-game, which not only cooled the orange, I feel, but perhaps also helped to connect the various sections a little more.

I have a few changes that I still want to make to the image as shown above. One of these I simply forgot about before exporting the piece. Others came about thanks to critique given on Twitter, for which I'm grateful. ^_^

One non-art thing done in the week just past was the fixing of a bug in my resolution-setting code. In short, I found that certain sections of code were not being run if at startup the selected resolution was the first in the list. This was likely harmless for the most part--but did result in my main-menu buttons not being placed as intended. The bug should now be fixed, I believe!

And finally, in the week just past I signed up on IndieDB. I now have a page there, as well as entries for both A Door to the Mists and Night River.

If you'd like to visit my page, you should find it here:
  https://www.indiedb.com/members/thaumaturge-art

The games themselves should be at these links:
  https://www.indiedb.com/games/a-door-to-the-mists
  https://www.indiedb.com/games/night-river
(Or by the "games" button on the main member-page, to which I first linked.)

That then is all for this week--stay well, and thank you for reading! ^_^

[Thaumaturge]

Blog post (1st of April, 2019)
A Thread to Save Time

Summary: In which the title-screen art is updated; level one sees some tweaks; game-saving is threaded; and a cutscene is begun.

Greetings and salutations!

For this week's screenshot, one more look at the title menu, with further revisions to its art:



The week just past was fairly productive, I feel! It saw some work on visual art, level-tweaks, a technical improvement, and more!

To start with, as shown above, I made further tweaks to the art shown on the title menu.

Some of these changes were things that I'd forgotten in the previous version, like distance-haze and the inclusion of a signature. (The latter being something that I'm still not accustomed to. ^^; ) Others were changes that I felt improved the piece, such as fixing the perspective on the orange-stoned section, or reworking the scale of certain structures. And still others came from critique given to me on Twitter, such adding as elements on the right-hand side to prevent the eye from being drawn past that border.

And overall, I'm much happier with it now, I do believe! ^_^

I also made some tweaks to level one. Two of these were minor: I moved some pages that had been, I think, made less visible by changes in the game's lighting, and I made the distant mountains look a little better, I feel.

Perhaps most significantly, however, I removed the "portals" that allow visibility between the "mummification" room and the upper barrow-corridor. The door that connects them isn't openable, so I see no good reason to incur the expense of portals there--especially as the "mummification" room contains another portal looking into a third room.

And indeed, removing these portals does seem to have improved performance in that section. I may even have removed a nasty hitch in frame-rate that could be experienced when exiting a certain room into that upper corridor!

On the technical side, I finally attended to a task that I fear that I had been somewhat putting off: handling the process of saving the game in a separate thread.

In short, what this does is allow the computer to work on saving the relevant files without stopping gameplay to do so. The result is that saving no longer has a significant effect on the feel of the game (on my machine, at least). ^_^

There were a few tricky bits to this, such as how to handle the "quicksave" button being pressed multiple times, or what to do about multiple types of save being triggered in quick succession. Nevertheless, for the most part it actually went surprisingly smoothly, I feel! (... Presuming that no terrible bugs are lurking unawares, of course...)

And finally, returning to the art side of things, I've started work on the introductory cutscene for level two. Thus far, I believe that I have its text written, the first scene completed, and a start made on painting the backdrop for the final scene.

That then is all for this week--stay well, and thank you for reading! ^_^