Raap

Yet again people would not spend $500 if they can use a $0 unit for capturing. I disagree, the Engineer has plenty of other perks in APB. If HW in particular was a concern with this, then it's simple enough to add some mechanics that only Engineers can deal with to make capturing easier. And making capturing done via another weapon means too many niche utility keybindings to press. But eliminating a team's inability to capture objectives should be the focus here.

Based on tester feedback, I solved the problem where the Siege Cannons had a hard time hitting their target. It should now be easier to hit your target using the new shrapnel blast alternate fire mode. Edit: And true to my previous words, the above image is already outdated again as I've created a new projectile and projectile trail effect (explosion remains the same as it is good), as well as an extra smoke effect from the cannon itself.

Ideally Danpaul can enhance the script logic to make room for an optional second warhead, so that technician tools can also register.

Add a 2-4 extra credit tick to the naval building while you're at it, the level gameplay at times can be too slow, much slower than intended, especially when a team lost their barracks, their silo, and their ability to capture abandoned silo's. The defenses were initially there to slow down the match and make infantry pushes in the early game harder. Having less defenses will make that tactic too strong. Generalcamo once suggested interior defenses when I was working on the map. I'm not a fan of the idea as it breaks a lot of consistency rules, but perhaps it is the only option.

That gives me an idea that would spice up map tactics... Instead of having base defenses be in stationary positions every time the map comes up, why not have defenses (Gap Generator excluded) spawn in 2 or at most 3 different possible positions? The same logic as used with power-ups. A Flametower being a few meters apart from another position might not seem like a lot but it will change a number of things and always makes players wary when approaching the enemy base (as they should be), rather than always knowing the exact spot where a defense can be safely out-ranged. Food for thought, @Pushwall.

That frame rate though. I'd record some gameplay myself, but I don't currently have a copy of solid editing software like Sony Vegas for example, and raw footage clips directly tossed to YouTube is really terrible.

I now have a tentative date for delivery to Pushwall mid-August. Pushwall reserves the right to make any changes he sees fit once I hand over the files, and it's up to him to decide when the level is released past this point. The level so far has taken a lot more time than anticipated due to a number of reasons, but what it comes down to is me trying to ensure the level is fun to play, while also attempting to raise my personal bar in regards to aesthetic design, which is the reason why Siege uses so many custom textures. For now, I've decided to stop giving further previews since each time I make a preview image, it becomes outdated 2 hours later (the above image is already missing a ton of changes I made since I posted it). If enough interest exists, I'll go into writing a "blog style post" explaining how some things were made, but due to scheduling concerns I cannot make promises on that.

The light rays, shouldn't they travel from the cone of the lamp rather than the light source itself? Other than that, the renders look interesting, but I am unfamiliar with the tools used. Am I correct in assuming most assets are not yours?

That's because before Delta, all EVA messages were played simultaneously. Delta staggers most EVA messages so you can actually hear them separately and so the chat doesn't become so overloaded that the chat box vanishes entirely. EVA messages spilling over to the next match is an unforeseen bug that never got spotted in testing and likely can't be fixed without Chronojam but it doesn't hurt the game so who cares On the topic of EVA audio, if it cannot be scripted, then add a silence period at the beginning of the "battle control terminated" audio file. Currently that line is spoken over whatever ended the match, which in nearly all cases is a single building destruction, so the silence period should be long enough to cover that one message (two at most). In the event a time-out ends the match, there is no harm in the "battle control terminated" line to be spoken later into the score screen.

Don't shy away from 1024x1024 resolutions for most terrain materials and 2048x2048 for materials that see heavy tiled usage (however, keep it realistic, a barrel prop doesn't need a high resolution texture for example). Something you'll want to apply to all your terrain (from grass to buildings) is a "noise filter" to remedy some of the tiled appearances. You add it to materials in 3DS. If you need to know the details just ask (I'm sure you know how it works though).

I've updated the image from the above post, the below image more accurately displays the art style. As you can see, I've seemingly taken a liking towards custom texture solutions.

It works now. As for feedback, I'm not sure what the model is based on, but if you aimed to make the model fit with Renegade assets then I think you did a good job of making it aesthetically consistent with the original assets of Renegade. Your level terrain on the other hand, could really use a texture pass. The UVW scaling alone is quite poor, and likely the texture files are low in resolution, even for Renegade. Especially UVW scaling should never be a problem since it has no impact on performance.

The link doesn't work.

I completed another major terrain mesh overhaul, this time the castle and walls. This is very time consuming work, so I don't have a lot to show besides a glimpse of what to expect when it is finally done; -snip- This is currently what the entire inward facing wall looks like. Still a bunch of stuff missing, such as foliage and some more things (like a scaffolding bridge to cross the wall gap). But this is the visual direction you can expect.

Don't get me started on that one. An automatic culling system would be amazing. But like so many hard-tech things for W3D, the complete show stopping fact is that we lack programming contributors.

You might not be do dependent on VIS then either. However VIS primarily serves to hide vehicles and infantry you cannot render, and especially vehicles are some of the most taxing things on W3D. And in this situation VIS will always work based on the actual VIS set-up and not the terrain. Besides, in a perfect world, a culling system would decide to render any polygons regardless of the object they belong to.

Another important thing is that there can only be so much going on in a single mesh before LevelEdit just refuses to load a terrain model. I don't know the exact limit and whether it's a limit of polys, tris, vertices or what, but I've always found I can't go over ~4500 polys per mesh without that happening. So keep that in mind if you're merging a lot of props or have a particularly defined backdrop (which Raptor's appears to be) - you'll have to split the extras off into separate meshes. That's always a big problem when trying to optimize a level. I run into it especially often with grass and bushes despite the polygon count being fairly low. Another killer in this regard is the shipping container model, it's really poorly optimized (frankly, a lot of the props are, including small crates and oil cans) and merging just two is already enough to encounter the problem of LE, the game, or W3D viewer crashing upon trying to load it. So it's certainly an issue within the W3D file architecture. Any chance Saberhawk can give it a look? If we didn't have this limitation then we could cut down the number of draw calls dramatically on each map and essentially merge all meshes that share the same material.

Currently they function more like an artillery with their handling and damage. I expect to be tweaking it some more before the level is released.

Finalized it to 8 texture variations, although due to UVW scaling concerns I've opted to go for a different high resolution wall foundation texture; I don't think I've quite spent several hours on a single wall before, but hey, first time for everything? Edit: And don't worry, no more boring texture updates from now on.

Here is one example. The first image is the regular wall texture, and the second image is one of several variations of it. It's not quite as simple as just placing some images on top of another. Color balancing and shadow effects are added as well, and the brick pattern is adjusted. At this point I'm not sure how many variations will exist, it depends on how many I can realistically create (I'm going to run out of plant and brick pattern variations). Either way this route is much more performance friendly than having actual meshes for the plants and what not, but the in-game wall will offer both, especially near the lower part of the wall, where infantry can walk. But yeah, work continues.

Yeah that can be real tricky. Yeap, fortunately the wall is a man-made object and not, let's say, a cliff or anything else that is part of the core terrain and intended to look smooth in terms of material transition. In this particular case I've created several (still working on the exact final number) "copies" of the wall texture, but with alterations. Using the same UVW coordinates I can seamlessly mix these textures across the wall to create variation, as well as additional geometry tweaks to make sure it all looks different. That's just the outward facing wall for now. The inward facing side of the wall will be more traditional in terms of material detail, since making several textures for each section of the level would take too long. The good news is, I suppose Siege is nearly an "urban/industrial" map in some ways, due to the amount of man-made terrain as cover. And that was a very popular request for APB maps.

Ya know, every time I can think of a dev claiming this, it's been followed by, some time not long after, a "never underestimate players" comment when it's figured out within days. We will see about that in this case. Even if you data-mined information, who can say you can puzzle the pieces in the correct places? Worth noting is that the idea of any bonus area, is to have it be fun to try reach it, not impossible. If something is impossible to reach, then I failed as a content designer. Rest assured however that the bulk of my time goes into developing the main area of the map itself. I just had some space to fill and old assets to make use out of. I've already had to make several new textures for the map as well. The problem with having a large object like a castle wall covering such a large percentage of screen view is making sure it doesn't look too tiled, so I've had to use a specific method on making it look better, without going overboard on more system taxing visuals.

I can't really make them smaller because then the infantry model would look strange in the seat. Cannons IRL are not small either. The turning platform and ability to raise the cannon angle is obviously not based on real word designs, but the rest is.

Slow progress continues, as I've had a chance to put a few hours in over the past three days. Still no ETA, but it's slowly getting there. If I did my job correctly, you'll have a hard time recognizing the map when you get to play it. As for the alternate victory condition, I've had to change the original plan for "king of multiple hills" due to not having anyone to program such functionality. But with all the dramatic changes to the map, a true alternate victory condition might no longer be required. Rest assured, there will be a solid amount of map-specific content including the cannons, as well as a number of secrets (including one very extensive secret that will take players months to figure out).

In a decade of W3D experience, I've never seen that happen before. I always thought the previous match information is wiped clean once a new level loads. That leaves me thinking, it would be interesting if we could change certain things based on the results of a previous match... *crazy thoughts*