[Aleks]

ck3D, on 13 October 2020 - 03:27 AM, said:

Aleks that's great, honestly I would almost frame something like this and put it on my wall if I were nerdy enough, I always appreciate funky 2D layouts. Now this one is only a few sprites short of being readable by my phone as a QR code. It's cool to see more people still focusing on that aspect nowadays, I feel like for a while 3D mode aesthetics were mostly the trending style in user maps (well most people skilled enough to pull that off in the first place usually had a clue about 2D level design too, but occasionally it'd take the back seat). Coincidentally I ran into the bunch of screenshots which you posted a few weeks back again just the other day while looking for another old post, and got even more excited for your project, promising stuff.

Thanks! Hahaha, when I showed this pic to my girlfriend this morning, she said it looks like an abstract painting :P It's mostly not really complex geometries there though (besides a couple places I guess), but the 2D mode challenge was to actually align each floor (bearing walls at same places, windows, ventilation shafts), filling each floor with rooms in their restricted area (every room of this building is accessible), seamless connection between the floors (the staircase creates quite a fake-TROR effect, as you can see both upper and lower floors from certain angles at the same time without HOM glitches) and making sure the "2-way windows" wouldn't show overlaying sectors from any angle (there's still a little bug here, but easy to mitigate and resulting from my too generous openings in one room).

Quote

Also progress on map 3 is ongoing with about 100 sectors, 1000 walls and 1000 sprites added yesterday. Trying to make the most out of some low times in work. I've been getting one sector glitching bug again which is minor (it's easily reproducible, but barely noticeable and only happens from one certain point and angle in the map which I can easily fix or block if all else fails), but intrigues me a wee bit as it's now part of a building I made months ago and never had problems with that's affected. Building is located at the very south of the map, and seems to partly glitch in and out of existence/display weird HOM when viewed from one particular point in a sector at the very north, both being separated by overly long roads for now which are sectors I have yet to split (I suspect the issue might disappear if I do). Then studying the glitch in-game with automap on (to check if anything weird was going on live in the layout at the same time) made me realize that the particular distance and angle needed to replicate the glitch was the exact one needed for fully zoomed-out automap to block that sector from appearing on screen (due to the size of the map). Not sure if there's a correlation or it's a coincidence but I thought that was funny. The building has full-bright sections and those seem to be the specific sectors acting up.

Also another funny thing is the north and south unreachable areas with the full-bright towers coincidentally seem to be just about the right distance from one another that it takes for Build to start struggling with rendering even tiles with shade value of -128 with the default visibility fog (as long as the player remains in the gameplay area it all looks fine, but from an unreachable area to the other, even the full-bright tiles fade to black).

Wow, that's going really fast! I was also trying to make as much as possible of my map during my 2 weeks annual leave recently, but got back to work this Monday and now the mapping time is a lot more tight as I got insanely much stuff on my head. The sheer size of your map already gives me goosebumps from what you're describing. And the automap is a great help, either for sector glitching or for checking if moving sectors (mostly ST23 or ST30) don't do weird stuff going out of mother sector boundaries etc.

[ck3D]

Yeah I reckoned aligning everything would be the challenge, at least you did it the best possible way I can think of by editing each floor in a different map file, and patterns in general are key, I used not to give a damn about numbers while mapping but for the past few years I've been using basic math for geometry a lot and it makes life a whole lot easier, especially once you've understood how much the editor liked counting in multiples of 8 for everything dimension (and of 3 for shade values), once one has grasped such logic then a lot more of the execution process becomes automatic so stuff ends up coming out a lot better and quicker.

And yeah for the past few days I've been mapping non stop because my work has been to wait on someone else to do their job first, so I've been more free than usual and since the whole map is laid out in my head (and the more ambitious structures out of the way already), progress is mostly a matter of raw materialization and execution, I no longer need to overthink much, only to lay terrain and decorative props. That being said, even some of that stuff is turning out pretty cool, I'll post screenshots soon, and also of that glitch I'm getting. I tried splitting the sectors I thought were problematic but it doesn't seem to fix anything, I guess I'll see about that with updated Mapster, otherwise it's not too problematic as again it only seems to pop up from one point at very specific angles.

Also map itself is large but not that large, it's mostly the surrounding unreachable areas that make it larger than fully zoomed-out automap, otherwise it'd be pretty much that size.

[Aleks]

ck3D, on 13 October 2020 - 09:48 AM, said:

Yeah I reckoned aligning everything would be the challenge, at least you did it the best possible way I can think of by editing each floor in a different map file, and patterns in general are key, I used not to give a damn about numbers while mapping but for the past few years I've been using basic math for geometry a lot and it makes life a whole lot easier, especially once you've understood how much the editor liked counting in multiples of 8 for everything dimension (and of 3 for shade values), once one has grasped such logic then a lot more of the execution process becomes automatic so stuff ends up coming out a lot better and quicker.

Hah, I haven't really made myself clear earlier - I made floors 2, 3 and 4 in a different map, but they were all together. The trick to not mess up floor and ceiling alignments was moving the main sectors of their respective boundaries a multiple of largest build grid size, to a position where nothing else existed in the "main map". This way it was easier to coordinate tags correlated between the floors. And I could have done it in the main map too actually, but I made this one test "one way window" before, so Mapster kept pulling me warnings, which were kind of distracting. Now everything is in the main map, but not "sandwitched" yet. Another use of multiples of 4s and 8s ;) But yeah, that basic math is crucial to get a grasp on early during the journey with Build/Mapster, mostly for panning and alignment, but also for the tiny spriteworks to work good.

Quote

And yeah for the past few days I've been mapping non stop because my work has been to wait on someone else to do their job first, so I've been more free than usual and since the whole map is laid out in my head (and the more ambitious structures out of the way already), progress is mostly a matter of raw materialization and execution, I no longer need to overthink much, only to lay terrain and decorative props. That being said, even some of that stuff is turning out pretty cool, I'll post screenshots soon, and also of that glitch I'm getting. I tried splitting the sectors I thought were problematic but it doesn't seem to fix anything, I guess I'll see about that with updated Mapster, otherwise it's not too problematic as again it only seems to pop up from one point at very specific angles.

Also map itself is large but not that large, it's mostly the surrounding unreachable areas that make it larger than fully zoomed-out automap, otherwise it'd be pretty much that size.

Ah, probably same thing like there was with one of my sectors on the outside - it flickered for no particular reason at certain very specific angles, but it was only possible to spot from a Jetpack, which the player won't have, so in the end I just let it go after making sure it's not something serious and it's really hard to repeat. If it's similar in your case, then probably the wisest thing would be to just let it go and block the curious players from seeing this bug.

Hopefully I will reach the point where you are in with my map next week - the point of laying basic decorations to polish the outside part of the map. Currently fixing some minor bugs that appeared during playtesting the SOS part. In your words, that building will be kind of "level in a level" probably. Also found a very unlikely scenario where player's decisions would block him from progressing further with the map, but it will be easy and quite fun to mitigate and turn into something neat, but also very unlikely for anyone to notice...

[ck3D]

Completed the enclosure of the streets of map 3 today, I even started adding a few cars. I'm planning quite a lot of those that should turn out to improve the look, coherence and gameplay of the city significantly, so this might be some of the very last screens of the outside I'm sharing, as the next stages might resemble the final product too much to feel comfortable giving away. Two of them aren't shaded yet as they're still my main resource for copy-pasting. Each car only roughly takes up 50 walls which is sweet, I'm pretty much guaranteed I won't be flirting with the limit as closely as I did for map 2 where that was pure anguish, and will be able to implement most everything I wanted.



'tis one of the borders of the level with a big sprite bridge I made. Third screen is taken from the unreachable side.



Random scenery, red building on screen 1 serves as a focal point for the player to find their way back to the central area from the eastern region of the map. Screen 2 is of a random parking lot I've started (still car-free, that won't last). Screen 3 is a view out of some tunnel on what currently still is an empty-ish open area, and to be honest aside from a few trees and small detail I'm quite tempted to leave it mostly like that, just the shadow visible on screen 1 already does a lot of the job.



Also I documented the glitch I was bringing up earlier (first screen is normal, second and third can suck balls), upon further investigation it not only triggers from a specific point in a particular sector but also the player has to be looking at the problematic building at exactly a 45 degree angle. I kind of don't get it, if it has nothing to do with my construction (guess I'll triple check) then I wonder if it has anything to do with the game trying to draw a full-bright texture past a very distant, and complex mishmash of walls or something.



@Aleks - no matter the extent you did it, juggling between multiple map files is still a smart move, now how you did tackle it sounds like a happy medium between my idea of building the floors separately within the file of the main map only to move them atop of each other later (which really isn't practical as it's then easy to mess up non-relative floor alignment if you didn't respect the pattern you mentioned), and literally using copies of the same map file originally bearing nothing but the floor's outline (at the right coordinates from the get-go too, because why not), to then alter their heights and fill them in individually. That'd be one way to significantly speed up things; and now if you've added a 'resource bank' of assets (pre-made benches, doors, general sprite work...) to the template map, you're golden. Would probably risk resulting in redundancy if that mechanical approach was taken to an extreme though.

[MetHy]

Experimenting with Shadow Warrior's "Advanced" Room Over Room (official term)


https://i.imgur.com/wqmf6bD.png

https://i.imgur.com/IkLG3J1.png

https://i.imgur.com/qe1CfEk.png

I'll share some findings soon :)

[ck3D]

For those who keep up with the constant updates: map 3 got a huge boost today, I built like 40 cars at street level, bumping the spent resources to 2100 sectors/15000 walls/9000 sprites (I tried telling myself to stop adding walls in the 14000's but I couldn't), making it the largest map in the set so far by a microscopic margin. Minus some minor sprite detailing and shading here and there, the outdoor area is now complete, all remaining walls will be going to the handful of indoor locations I still need to add or polish. Framerate only really gets bad when looking at the entire level from out of bounds, so since I won't add much clutter it should play well. Earlier when I said I'd like to be done with this map before the end of the month I thought I was being overly optimistic but it's turning out to be a certainty now, all I need is a few more sessions and voilà. Already looking forward to working on the next levels.

[Aleks]

Man, you work and post updates too fast for me to keep up with commenting on all that stuff! :o Crazy amount of sectors, too, I doubt I'll be able to reach 1900 when I run out of walls myself. You're on quite a mapping spree, which is pretty great (at least from a mapper's perspective) if you managed to use your free from work time to get so much progress.

Now for the comments: I have a strong feeling from the first 3 shots (especially the first one) that either the wheels are too small or the police car is too wide, at least from that perspective. Also, my OCD begs to wrap the cars around in sprites (I'd use #1195) so their underside blends in with the street instead of leaving that brown-sand thing from the texture, but with 40 cars I imagine it'd be quite a hassle and I've no idea how many sprites that would use up and if that wouldn't play weird things with performance for you (plus again, that's just my OCD and something that should be barely noticable in actual gameplay).

It seems that screens 1 and 8 show the same part of the map (judging by how the buildings on the bounds of the map line-up), but from a different perspective, having these 2 screens together gives a nice sense of depth even before playing. Bridge looks pretty neat, always a fan of large-ass spriteworks that blend-in seamlessly with sectors, also nice use of textures (especially the tile on deck which is rarely used in this context I think).

As for the glitching building, well, you have my sympathy. Hopefully you can find some solution to keep it from player's view easily without spending too much time trying to figure a way to fix it.

As for my progress, I've been quite busy with work during the week, so the mapping sessions during last 3 days focused only on getting rid of the bugs and playtesting the level with SOS building overlaying for a test. The glitches with SOS itself were actually surprisingly easy to fix - there's not that much "2-way" windows, but there were some particularly vulnerable angles and places - now I think I've achieved a nice sense of "3D-ness" with that building, as parts of each floor can be visible from the outside and nothing glitches (gave it a really thorough tests, even for perspective normally unreachable for the player). One thing I found particularly cool was, in one place, I had to split a sector because of SOS-overlaying in exact location where a couple of moving sectors had their routes, so I joined the moving vertices to avoid collisions and it works perfectly, eliminating the glitching rendering and not interfering with the moving sectors.

The only bug I couldn't get rid of was one mirror starting to go HOM when looked at from the outside of the room it was located in, it only did it after I overlayed the other sectors on it. Nothing helped with it, so in the end I made it broken from the start, no idea what caused it - especially since there are also 2 more mirrors in this building, and I would rather expect them to glitch if anything, but they were perfectly fine. Another issue is with ambience sounds that were a bit too loud and hearable between the floors, but mitigated it as much as was possible I hope.

And now the biggest mapping satisfaction of the day for me, for the first time, I've played start-to-end through that large building to see if all the effects and sequencing works, and well... It does! Great feeling after meticulously tagging and testing each floor separately for past 2 months or so to only now see it in it's full glory like a clockwork mechanism (and testing different scenarios if nothing gets stuck if approached differently). In other words, I'm really happy with how it turned out now, with some interconnections and a nice balance between sequencing/effects/linearity and room for exploration/world-building. The whole place just gradually opens up, which is exactly what I aimed for. Fans of deep-exploration should feel rewarded here and there as well.

I've just noticed I have a total of 78 different version of the map made during various stages (along with "submap" used for making the SOS layers). Better safe than sorry!

[ck3D]

Yeah for the past few maps I've been working on, after a while I've always come to a certain point where the process would start feeling like it's all been dragging on for a bit too long, and that's when I usually realize that I'm actually super close to wrapping up the thing and that's it's worth putting in the effort as I eventually want to release this thing for people to enjoy, not just sit there and keep working on it for my own amusement so then it boosts me. I'm having tons of fun but I'm also fueled by the idea that the level won't build itself, and since like I was saying I have most of the project laid out in my head already I can just focus on execution, I've made it a personal rule to never ever launch Mapster without a specific idea of what exactly I'm going to add to the level this session to make sure that every single one is productive, and then I usually get sucked into building even more stuff.

About sector count IIRC map 1 has 2200+ which is a personal record, map 2 falls just under 2000 as I ran out of walls first. I remember Zaxtor posting about how it would be common for him to get into the 2500/3000 sector realm before running out of walls on single levels, I think that's crazy. It's kind of funny keeping track even just of the wall count because I commonly associate it with the corresponding number of space key presses, so let's say so far I've made three maps containing 16000 walls, that's at the very least 48000 space key presses, and that's all without counting everything else about the process, I like looking at things like that because it's pretty representative of the quantity of personal effort that's been put into a project by the author. Now numbers don't automatically equal quality or substance, but when you think about it the process of mapping is pretty crazy.

I can see what you think looks off about the wheels of the car and you're right, I think it's just how I like to build cars though, thinking about it now the ones in every map so far have that 'cute' look - I don't especially aim for realism when I'm building objects (although I will use visual references quite often), if it looks like what it's supposed to be and most importantly plays well then stylization is fine by me. I don't think I'll fix the dimensions but a lazy way of doing it would also be to just paste wall-aligned sprites of that single tire tile from Atomic over the small tires, I've done it before and it can actually look alright. I think I remember I used to have that OCD about the brown on those textures too, a while back, which I had completely forgotten about, but I've definitely done the black sprite thing to cover it up before. Won't be doing it here though, I've come to accept and even like the look of that tile, it is what it is as a pure Duke 3D texture; I'm not really into altering the style of the original tiles anymore with visual tricks (on the other hand I've been super into trying to figure out fresh ways to pan them). Same as using transparent sprites everywhere to alter the look of wall textures, I used to do that a lot but these days I'm really into honoring the base textures to make levels that directly reference the original game, and feel like a coherent extension of it a bit better (at least for this specific project).

Looking forward to your project, great to hear about the progress, making me curious now. 78 back-ups sounds cute to me though, just for map 3 I think I must have had 2000, I just Save As an awful lot, especially when copy pasting. Whenever a filename gets too high up in the something hundreds (usually around 200 or whatever it takes to feel like the level has reached a considerable new stage) I just start from scratch again with a slightly different variation on the filename. Then way later when the time comes to organize all those files I just delete most of them but the latest revisions of each individual 'version'.

[Mark]

ck3D said " the level won't build itself"

Darn it. I was just about ready to check on some projects I started years ago and I left them alone hoping they would finish themselves. GAH!!! :angry:

[ck3D]

I mean in general, you'd be surprised to realize the amount of time many people think they're spending creating vs. the amount of time they're actually spending creating. I can spend hours in Mapster but if I fire it up it's with a goal in mind and while I enjoy the process, I also make sure it's spent more constructively than two hours floating around 3D mode like it's 1996 and I just discovered the jetpack, thinking of how good my level is eventually going to be when one day I wake up and it's magically done. We've all been there, it's the worst creative trap and a big reason why a lot of things often get discussed but never done.

[Mark]

I'm guilty of the above behavior these days. ;) It wasn't a problem when I was working with a team.

[Sanek]

Something that's on my desk right now and my future plans (if they ever come to fruition):

I didn't work on my map at all over the last 1,5 months (well, I had a pretty good reason for it). I'm sorry to say it, but it looks like I don't have time to work on it anymore. The map I was working on all summer is basically frozen now. It's like 50% complete right now, but still there's a lot of work ahead. Sure, I can finish it someday but I'm afraid there'll be another case "Sanek-itis" making my maps worse. So right now I have the idea to let someone else complete the map and make it overall better. I have one likely candidate right now, but we'll see. But other than that, the project's closed righ now.


Future:

I really, really, really want to start making a classic 3DR-style maps that Fernando's making. Maps that's medium in size (like a majority of the original ones) and doesn't require too much details yet it can be detailed on par with the originals. It also looks like I have to make the episode instead of a separate user maps. Perhaps I should release it as a map pack (like what's Mr. Sinister's doing right now. If I'll release a single map with basic 3DR style, it'll pull my overall Hot Maps' average down.
_________________________________

I wanted to make a CBP with Doom-like design but it didn't happened. Like some members suggested, I should've propose to make the episode of such maps instead of one giant map. I'll try to make one map to set it up as the example. Perhaps that's what I can make easily and then create Thread №2, hoping to find mappers who's interested.
_________________________________

And for cryin' out loud, I really should start making that Ion Fury map that Oasiz promised to help me with, someday!

[ck3D]

How about you merge the map with the other abandoned one you were trying to submit into CBP2020? Wouldn't work?

The original Doom CBP thread could have worked, in fact I'm sure people would still be down, honestly the only reason why that thread died is you stopped replying in it. It's cool to hear you're still thinking about it though. And congrats on your game.

[Aleks]

Sanek, congratulations on your game being published! That's definitely a priority over a nerdy Duke 3D project one does for a hobby ;) And yeah, remember that - making Duke maps is a form of recreation, you're doing it for yourself, so you should never feel guilty for not mapping or put some pressure on yourself to push it as planned. There are no deadlines here, plus it's a creative form of activity, so you can't just stress yourself out to sit and make it, you need inspiration to build. Just shelve the project and come back to it whenever you feel like it or whenever you have some free time for it. I imagine if you're becoming a "pro game dev", making user levels might not exactly be the form of activity you want to spend your free time on, too.

[Aleks]

Worked some on the terrain on the left side of the picture during the weekend and pretty much got the boundaries of the map ready. That area still need proper shading and some polishing, but the shapes are quite OK. Experienced some sectors flickering there as well, but only at unreachable areas and from perspectives not accessible for players without cheats, so there's nothing I can do about it I guess. Examined the problem and found some rotten egg sector which got removed already, but with the complex geometries (multiple lines joining in a single vertex, massive sloped sectors) it's kind of predictable. Also from the perspective visible on the screen, some sprites are flickering, but again, it's unreachable from the player. When I'm finished with all the outside area, I'll have to see if there are any flickering problems from player's perspective that are serious enough to require intervention, then I might lower the visibility a bit - but I'm pretty happy with how it looks, so hopefully that won't be necessary.
https://cdn.discordapp.com/attachments/309341856729268228/767442485411577876/duke0036.png

[ck3D]

Great to hear about the progress Aleks (and sorry to hear about the glitches, but if the player can't spot them from the gameplay area I'd say it's really no problem), your screenshot isn't showing for me though.

[Aleks]

ck3D, on 18 October 2020 - 10:37 AM, said:

Great to hear about the progress Aleks (and sorry to hear about the glitches, but if the player can't spot them from the gameplay area I'd say it's really no problem), your screenshot isn't showing for me though.

Seems there's some problem with imgurl having it's certificate overdue, so that might be the problem. Edited the post now with the image link from Discord.

[ck3D]

Screenshot is cool, I like how cohesive the whole place looks with all its little individual elements, it actually looks like a miniature model of a real-world place which barely surprises me now knowing what you do. I like how you made four watchtowers yet ours on Duke4 still might not like the map, lol. I'm looking forward to seeing the place shaded, looks like those cliffs could use a lot of contrast to look the best (no doubt you have that in mind already but it's my first general impression). Bridge looks like a good opportunity for some sweet shading in the water too (unless you want it destructible). It's hard for me to tell on the central building which windows are one-way and which aren't but that you managed to make it work is really cool. You actually overlaid some lines in 2D mode for the SOS or were you talking in general? I'm always paranoid of doing that, in case walls or vertexes would connect, all my stuff is always one or two pixels off just to make sure.

Also I've been keeping working on map 3 but my past two nights have been spent doing nothing but shading all the cars individually. I still have a few to go, some of them actually bear quite a few sprites (text, etc.) that the editor duplicated during all the copy-pasting (so that there are two instances of each sprite atop one another) and I have to manually remove in the process, so now I'm tearing every individual vehicle apart, fixing/shading every element and putting them back in place as I don't really trust copy-pasting those sectors again. Also, just making the shadow sectors for the cars ate up 600 walls, so I should be able to make it but just barely. Also, just doing away with all the 'invisible' duplicates of the sprites slightly boosted performance up - that's cool.

[Aleks]

ck3D, on 19 October 2020 - 04:22 AM, said:

Screenshot is cool, I like how cohesive the whole place looks with all its little individual elements, it actually looks like a miniature model of a real-world place which barely surprises me now knowing what you do.

Thanks, that's a great compliment! To be honest I'm quite surprised with how cohesive the composition of that whole area looks despite not keeping that strict to the brown palette and throwing some variation within the rock formations.

Quote

I like how you made four watchtowers yet ours on Duke4 still might not like the map, lol.

Hah, there are even 5 of them, one behind a building - but yeah, I don't think it's THE Watchtower style map ;) Actually I named these just a few days ago to make the navigation easier, unfortunately the naming pattern didn't include a "Nansci Watchtower" - maybe I should change it...

Quote

I'm looking forward to seeing the place shaded, looks like those cliffs could use a lot of contrast to look the best (no doubt you have that in mind already but it's my first general impression). Bridge looks like a good opportunity for some sweet shading in the water too (unless you want it destructible).

Yeah, the "other side" of the bridge has just all same set shade value so far, gonna be a lot of work with it - same for the bridge. But that's the last part to consider, I wanted to make sure the geometry is 100% ready. The cliffs there will be mostly dark to match the shading pattern and directions of the whole area. Probably will have to keep it a bit simpler than on the island itself (some shading highlight can be seen here on the grey rock in the bottom right corner of that pic) due to wall restrictions - now I'm at like 900 left only and still have plans for a couple of stuff to do. However, I lately found some emergency wall supply as there are parts (mostly just details) in the very first building created for this map that can be easily scrapped, so I could probably scavenge like 300 more through there...

Quote

It's hard for me to tell on the central building which windows are one-way and which aren't but that you managed to make it work is really cool. You actually overlaid some lines in 2D mode for the SOS or were you talking in general? I'm always paranoid of doing that, in case walls or vertexes would connect, all my stuff is always one or two pixels off just to make sure.

The building only has the ground floor "installed" so far, other ones are waiting, but that will be one of the last things to do, so no windows from the upper floors are really functioning yet. The green pal on some of them is my working notation for the windows which I plan on making 1 way, but not sure if all will be possible. And the only lines overlaying in 2D mode when all SOS is merged are staircase entrances and ventilation shaft connection going straight from 1 to 4th floor.

[ck3D]

Wee update on map 3, all the cars are finally shaded (that took forever) and tonight I added 75% of the detail I want at street level, which really isn't all that much but still makes for a pretty strong difference when it comes to just visually structuring and cleaning up the otherwise empty look of some of the larger areas (even just with the road marking, lamp posts, signs, hydrants etc.). Sprite count took quite the bump again, and will keep taking a few more till I'm done, then I'll have a few last rooms to build out of the few hundreds of walls I have left and that'll be it for this one, hopefully by sometime next week.

Ah, also I kept experiencing one more map insta-nuke while insisting a little to split some (unremarkable in size and design) sectors for the road markings, too. Splitting one bit of a sector in particular would turn all the cars in a completely different sector/region of the map into invalid player space/white walls. What may or may not be interesting is that sector where the cars would get messed up is exactly in the middle of the distance between that glitching building I was mentioning earlier, and the spot where the player needs to stand to get it to glitch, so there's probably something random that's subtly off about it, even though the problem is barely noticeable, I'll see about giving the anticorruption tools a try out of curiosity.

[Merlijn]

Great shot of the area, Aleks! It looks really cohesive.

Quote

then I'll have a few last rooms to build out of the few hundreds of walls I have left and that'll be it for this one, hopefully by sometime next week.

That's almost a shame. ;) I liked those regular updates, you sort of guided us through your mapping process which is really cool.

[ck3D]

Thanks, yeah, it was quite the captain's log. Really though, it was pretty fun sharing the progress on that one, I guess I felt comfortable doing it because it's going to be one of the biggest maps, taking place late in the episode ('à la Derelict' like Sanek was describing it in another thread), so it was easy just showing the evolution of the size/scale/scope without spoiling any detail. I think I want two more city maps in that style (just smaller as I want the huge maps to feel exceptional) and then the rest of the episode should be more conceptual stuff, as time goes more and more of the storyline is naturally coming together in my head, were I gifted of ubiquity I'd probably be working on eight maps simultaneously right now with the same amount of stoke. Now I don't know how much the conceptual aspect will allow me to spoil/share then. But I know people are into studying mapping styles and the process of crafting something in general is fascinating so I'll probably be keep reporting every time I shade two new rooms and find new ways to one-hit-delete half my map by touching a funny vertex.

[ck3D]

Worked on some indoor areas for map 3 tonight, trying to detail the inside of a little train station amongst other things. I'm already in the 16100 walls which kind of sucks as there are a few more interiors here and there I need to make, but I don't feel too pressured as there's a couple of relatively wall-hungry rooms I never detailed that I can still cut if needs be to earn the resources back (I kept those areas relatively empty till the last minute just so that I could always easily turn them into secret places if I didn't hit the wall limit, or scrap them if I did). Building almost everything out of sprites might be making me beat my personal record, I can't remember whether or not I ever used this close to 10000 (and then I'll have to add most of the SE's, C9's etc. and that's without the monsters and items, I hope I don't start catching glitches in the end). The aforementioned train station itself will be a rather small area, so I'm trying to spend at least some effort on it, for instance out of a few things I just made a couple of satisfying-looking gates to the platform.

Making my other former post lie by posting... more of the same I guess; sections I've shown before just with the recent extra little detail, new indoor stuff is still too WIP to be presentable. I'm estimating the map should be done within the next two, three sessions or so now.

Attached thumbnail(s)

• 
• 
• 

[Aleks]

@Merlijn, thanks! I'm hoping to post another one during the weekend when I finish the shading. I'm a bit surprised myself at how well this composition clicked there altogether.

@ck3D, love the last bunch of shots (nothing new), addition of details and cars made the city look really alive! Especially the 2nd shot is my favourite, that bright area at the end of the street, behind the hotel sign, gives such an intense feeling and depth there that it reminds me of GTA games.

And about the corruption thing, I doubt the default corruptcheck in Mapster will be of any help there, as it doesn't seem to recognise flickering sectors problems (even with one abruptly joined and overlapping sectors which I manually found and fixed a couple days ago, which looked seriously fucked up in 2D mode). It's good for mostly the minor stuff from what I've noticed, but still does a good job at it. I think the problems with flickering sectors (as we both seem to be experiencing some of that with our current maps) result from the fact that we're really stretching out the possibilities of Build with large sectors with multiple vortices, multiconnections at single vortices (no idea how many walls can join at a single vortex, but I'm trying to be careful with that thing, which is sometimes difficult to keep up when drawing terrains) and high, complex slopes (this has been largely fixed with Eduke I believe, I remember Duke straight out crashing when entering higher slopes, e.g. "Damned Dam" map). Then again, with the complexity of 2D layouts that we draw and the weird nature of Build at times, there is always the possibility we could have joined some sectors in the wrong order, split/inner looped something out of place or made other small mistakes, which would backfire at us 900 sectors later.

Just after reading your post yesterday I playtested my map I got shit scared when I saw the rocks going highwire with HOM/sector warping right from the temp-starting position, then I noticed the motorboat just warping through the rocks and realised that despite carving the locator route very carefully, I didn't account for the fact it would just push its way to locator 0 before being entangled in the actual path. Player normally wouldn't have a chance of spotting that and I don't think it would have any consequences at the boat's or its surroundings behaviour later on (as I probably spent a couple of hours by now just travelling on that boat and making sure everything's ok both in automap and FPP view), but just changed numeration of locators to make sure anyway. It also seems that a couple of outer sectors are flickering black on automap from very certain angles, but it can't be seen in FPP view and seems quite unlikely to have any consequences.

I also hit the "milestone" of reaching 1 MB filesize with the map, still got about 900 walls to use. Sprites are far in 10 500s here and I still just barely started on the vegetation (which will have to be restricted a bit to make sure there's no flickering on the outside part - good thing it's mostly a desert after all!). This week I hardly had time to map (except yesterday really) and mostly worked on some minor stuff like cameras (for the first time I dug deeper into mechanics behind them, namely the order in which they show on viewscreens - it's the spritenum in descending order). Shading is quite a tedious job, I'm again using my meticulous technique of counting X and Y offset of each vortex of the cliffs basing on it's elevation, but the results are spectacular and look far more natural than if I tried just randomly drawing the shadow.

[ck3D]

That's great, I'm looking forward to seeing the place shaded with your technique, I keep saying it but shading and playing with contrasts is so crucial in the process of mapping to me I really appreciate it when people focus on it and I love it when someone goes as far as developing their own tricks and style. Subway sector coasting over to the first locator shouldn't be a problem at all as long as the player isn't around, if anything, remember you can roughly determine its navigation even when it's doing that, by changing the value for the starting locator, maybe there's a more convenient one than the one you're currently using that would make your ship cross less stuff. Automap getting glitchy for no reason sounds intriguing.

About big slopes yes I've noticed that buggy temperament before, in my map The End Of The World there's one train the walls of which are actually extremely steep slopes that I remember having to make, well, a little bit less extreme or I'd get weird texture behavior and HOM from certain angles and distances (and IIRC the problem still persists to a barely noticeable extent in the final release, which I only caught later). And now in this map I noticed the game had problems with texture alignment when working with very large slopes, at least in my version of Mapster/eDuke in 8-bit mode I've had perfectly aligned neighboring sloped floors/ceilings from up close that would start looking more and more jagged with distance, or floor tiles that look like they 'overflowed' into the next sector and the like. Still love them though.

About that sprite count, don't forget that in addition to all your decoration and effects the game will have to render all the projectiles, bullet holes, gibs and whatnot too, making it prone to crashing if you're really flirting with the limit. Not sure if there's a direct correlation but it's one thing I constantly keep in mind personally. Just like struggling with wall limit on map 2 has taught me one is better off never actually getting close to reaching it; I feel like as soon as one is in the 16300's there, more weird shit than usual is likely to happen.

[quakis]

It's nice to see so much progress on all these projects posted here as of late.

I've been slowly cooking up something. I know how the overall structure should look like and the players' goal from start to finish, but when it comes to designing actual gameplay geared toward these goals and inbetween those spaces I'm often left blank on how to approach it for such a long time, so progress bogs down to a snails pace until something comes up that both feels right and is satisfactory to play. I'm too orientated around visual/aesthetic mapping and struggle the most with the gameplay side of things. At the same time I want the areas I create to be designed with gameplay in mind, not an afterthought, but I still need a set piece or somthing in the first place before that process can begin despite know what my general outline is.

The ability to think up a gameplay scenario and building around that is something I'd like to improve on in general.

[Mark]

I think you have been doing great so far. Your method is working. ;)

[Aleks]

ck3D, on 23 October 2020 - 03:52 AM, said:

That's great, I'm looking forward to seeing the place shaded with your technique, I keep saying it but shading and playing with contrasts is so crucial in the process of mapping to me I really appreciate it when people focus on it and I love it when someone goes as far as developing their own tricks and style.

Yeah, shading is crucial for making the area look natural and detailed, also can put a nice bit of diagonal lines to more blocky rooms. I sometimes basically focus on some eye-candy shadow instead of going overboard with details to avoid making some room look "boring".

Quote

Subway sector coasting over to the first locator shouldn't be a problem at all as long as the player isn't around, if anything, remember you can roughly determine its navigation even when it's doing that, by changing the value for the starting locator, maybe there's a more convenient one than the one you're currently using that would make your ship cross less stuff. Automap getting glitchy for no reason sounds intriguing.

About big slopes yes I've noticed that buggy temperament before, in my map The End Of The World there's one train the walls of which are actually extremely steep slopes that I remember having to make, well, a little bit less extreme or I'd get weird texture behavior and HOM from certain angles and distances (and IIRC the problem still persists to a barely noticeable extent in the final release, which I only caught later). And now in this map I noticed the game had problems with texture alignment when working with very large slopes, at least in my version of Mapster/eDuke in 8-bit mode I've had perfectly aligned neighboring sloped floors/ceilings from up close that would start looking more and more jagged with distance, or floor tiles that look like they 'overflowed' into the next sector and the like. Still love them though.

Yeah, I just re-numbered the locators, so locator 0 is directly in front of the boat.

I got one weird HOM/disappearing child sector glitch last night when working on the shadows, but I caught pretty much the exact moment when it happened and can't recall doing something wrong that would mess it like this - so I'm pretty sure this has to do with complex geometries/multiconnection vertices and Build engine having problems with handling the geometry, perhaps even in the intermediary process of joining sectors - guess that's one more thing to be really careful with.

As for textures on slopes, it's funny, because just few days ago I've read this interview with Ken Silverman: https://www.arcadeat.../ken-silverman/ (found in Supreme Topic of Miscellaneous Knowledge), where he admitted that one of the things he wished he's done better in Build - along with ROR and multiplayer code - is texture mapping on slopes. Wonder if that's what he meant or something more noticable.

Quote

About that sprite count, don't forget that in addition to all your decoration and effects the game will have to render all the projectiles, bullet holes, gibs and whatnot too, making it prone to crashing if you're really flirting with the limit. Not sure if there's a direct correlation but it's one thing I constantly keep in mind personally. Just like struggling with wall limit on map 2 has taught me one is better off never actually getting close to reaching it; I feel like as soon as one is in the 16300's there, more weird shit than usual is likely to happen.

I know, I suppose I'll end up before 13 000, which should be ok. The maximum Qsize value (number of active bulletholes, blood etc.) is 1024, I guess it's related (I still keep it much lower). Didn't know about walls though, so it's nice you gave me heads up about that!

Quote

I've been slowly cooking up something. I know how the overall structure should look like and the players' goal from start to finish, but when it comes to designing actual gameplay geared toward these goals and inbetween those spaces I'm often left blank on how to approach it for such a long time, so progress bogs down to a snails pace until something comes up that both feels right and is satisfactory to play. I'm too orientated around visual/aesthetic mapping and struggle the most with the gameplay side of things. At the same time I want the areas I create to be designed with gameplay in mind, not an afterthought, but I still need a set piece or somthing in the first place before that process can begin despite know what my general outline is.

Good to hear you're working on something, as for the gameplay - judging from Flaming Shipwreck you're doing pretty good then! I actually think some people tend to overfocus on that, usually gameplay seems to come quite naturally with design, and unless it's really messed up with stuff like super-cramped spaces, frustrating enemy placement, completely unintuitive puzzles or difficult to navigate design (now this could also go for general design flaw), most gameplay ideas work well if the overall design is good.

Anyway, just finished shading these damn cliffs, so it's a good time to post some more revealing screens:

[ck3D]

Really cool, my favorite detail is the foam behind the motorboat, love the idea. Also is that shadow from the bridge sector- or sprite-based? Sprite-based would be the easy go-to but part of me wonders if it's actually sector-based and you figured out a way so that the boat can cross it (à la Gambini's trucks in Blown Fuses).

And yeah the general warning about getting too far up in the 16300's for walls (and I guess sprites) is as soon as you 'accidentally' break that limit, it seems to cause an overflow in the values and lots of stuff in the map stops making sense and gets corrupted for good (I've described some of the weird behaviors I'd get with map 2 a few months ago in this thread). I just usually recommend staying low because once in the 16300's it's easy to miscalculate and accidentally go over the limit (for instance with an Alt-S), also when you get too close to it it seems like the editor (or my version of it at least) struggles even to just buffer the values correctly, for instance on map 2 I'm 10 walls short of the hard limit and just the calculation of joining two sectors (to try and earn walls back) will fuck things up, that's my naive impression at least based on what I get, past mapping I only know so much about them fucking magnets.

Speaking of which I just deleted all my superficial sectors/potential secret areas and earned 300 walls back when I was expecting a lot less, I'm stoked, now I'm back in the 15800's and will be semi-comfortable finishing everything. Map also just made it past 10K sprites and 1Mb file size too (which makes sense as for my past two mapping sessions I've been building using nothing but sprites). Been touching up the train station and its exteriors. Also caught a different variation of the disappearing sector glitch once somewhere else in the map which freaked me out for half a second, but this one seems so elusive I couldn't even find the right frame and angle to replicate it, so nothing important if it's bound to be a one-off.

[Aleks]

ck3D, on 24 October 2020 - 01:46 PM, said:

Really cool, my favorite detail is the foam behind the motorboat, love the idea. Also is that shadow from the bridge sector- or sprite-based? Sprite-based would be the easy go-to but part of me wonders if it's actually sector-based and you figured out a way so that the boat can cross it (à la Gambini's trucks in Blown Fuses).

Thanks, that foam took a while to figure how to make it look decent and not eat up too much of the walls! And yeah, the shadow of the bridge is made with sprites, there might be a slight difference in tones between sector-based shadow and sprite based, but it's far less noticable than the glitch that would occur with something crossing the lines. I remember the trucks in Blown Fuses as I played it quite recently and there was some "warping vertices" effect with one of the truck, but I've never really looked it up in Mapster. Speaking of Gambini, the plants right under the bridge (visible on 3rd screen) are made with upside-down Octabrain frames and it was inspired by this MSDN screenshot from "It Lives" (I'm yet to play the map, but loved the idea here!).

Quote

And yeah the general warning about getting too far up in the 16300's for walls (and I guess sprites) is as soon as you 'accidentally' break that limit, it seems to cause an overflow in the values and lots of stuff in the map stops making sense and gets corrupted for good (I've described some of the weird behaviors I'd get with map 2 a few months ago in this thread). I just usually recommend staying low because once in the 16300's it's easy to miscalculate and accidentally go over the limit (for instance with an Alt-S), also when you get too close to it it seems like the editor (or my version of it at least) struggles even to just buffer the values correctly, for instance on map 2 I'm 10 walls short of the hard limit and just the calculation of joining two sectors (to try and earn walls back) will fuck things up, that's my naive impression at least based on what I get, past mapping I only know so much about them fucking magnets.

Ah yeah, Build playing shennanigans when trying to get the resources back might indeed be an issue to be wary of that I didn't take into consideration yet... I'm at like 15 600 so might not be that bad anyway.