Candescence

Might as well show off something from what I've been working on. I've been working on the foundations of a Digimon side-scrolling game similar in nature to Digimon Battle Spirit, but as a Metroidvania rather than a fighting game. The movement mechanics and abilities for Agumon are fully implemented. The base mechanics are simple due to the Wonderswan/GBA's two-button implementation, though I gave the secondary attack and the dash their own buttons in this instance. My intended list of playable characters is: Agumon, Gabumon, Veemon, Wormmon, Patamon, Gatomon, Guilmon, Terriermon and Renamon. I have complete sprite sets for all of them except Guilmon and Wormmon (though there's the Extra Guilmon sheet I have that can be re-paletted to fit the normal Guilmon, though that seems to be more time-consuming than it seems, due to the eyes). I'm also kinda hoping there'd be someone who could make me a custom set of sprites for Masaru as well, but I'm not expecting that to happen.

The main problem with porting to phones in general is that they're designed for 3D rendering, not 2D, meaning you have to do a considerable amount of work to keep a game's framerate at a steady 30fps. The CPU and GPU budgets for mobile games are absurdly small, especially for 2D games, where it's extremely easy to run into problems, and thus you have to keep a rather conservative memory budget on the visual side of things. Now, the actual logic processing is another matter entirely - more often than not the game logic is much more efficient than the rendering, so generally it's not as much of a problem. A Sonic engine is a bit tricker to keep efficient, however - I've learned the hard way that Streak's engine is ridiculously un-optimized in C2. Creating a Sonic engine in C2 would likely have to be done completely from scratch, and by someone much more knowledgeable in that sort of thing than I am.

You can actually set up a manual debugger, just ask DW. I imagine they'll figure out how to create a built-in debugger eventually with HTML5, though Scirra has a pretty big to-do-list, though they've pretty much nearly got all the features from Classic implemented (if not all of them).

I've made a relatively simple Metroid-like map in Construct 2, which features map positioning of the player, indicating which rooms have been explored, and automatic creation of the map on runtime. Granted, it's kinda missing more advanced stuff like saving the map, special locations (like save rooms and the like) and other things, but it's a start.

I think it's probably best to not really argue about the Hedgehog Engine, but I do think it's long been been surpassed. However, the principle doesn't really change - the renderer picks out one voxel for every pixel on the screen. This is also true for animated models in this renderer. The challenge when it comes to moving objects is more due to the fact that very few people have actually bothered animating point clouds, so they had to do it from scratch, and completely independent of how polygons are animated. Yes, obviously animations will be a bit more intensive than static environments, but when you have a renderer that efficient (once again, it runs in software at a decent framerate), it doesn't make much difference anyway. Firstly, Bruce Dell has outright stated that they've been focusing on animation and physics ever since the previous demonstration was released, so, yeah, that'll be a given. He's even shown footage from earlier versions of the technology that feature rudimentary technology (as seen in the 40-minute interview vid you can find on Youtube), demonstrating that animation is quite possible. Secondly, they aren't asking for funding. They got it a year before the demonstration, a $2 million grant from the Australian federal government via the Commercialization Australia initiative, the largest funding ever granted by them. Hell, Bruce Dell has outright stated that they've been turning down offers from potential investors, since they've already got enough funding as it is, as well as for the sake of its internal secrecy. The company also has two former executives from Australia's largest IT company, Mincom, including its founder. Not to mention they're taking criticism in stride. Everything I've heard about Bruce Dell commenting on critics is mainly rebutting points from Notch, outright stating that he doesn't want to make enemies, and him mentioning that the other execs have remarked that having vocal critics helps exposure, especially when they can eventually prove them wrong. There may be the old adage "too good to be true", but I've seen enough evidence to figure that there's really no bullshit involved.

There's no issue with utilizing better hardware for the sake of achieving a certain aesthetic, but good grief, the Hedgehog Engine isn't exactly the greatest of engines. It severely limits gameplay opportunities due to the way it works, and its capabilities were already outdated merely months after Unleashed was released. Not to mention it is horribly optimized, especially on the hardware it was designed for. But I digress. I think the next graphical revolution is gonna come from Euclideon's Unlimited Detail tech, rather than anything Square-Enix or Epic Games creates. Yes, I am aware that Unlimited Detail is a rather contentious subject, with many believing it to be a hoax or a scam. By conventional wisdom in 3D graphics, such real-time detail without a massive jump in hardware is unthinkable. However, I've actually done my research on my subject, and kept tabs on new developments, and I think Euclideon have indeed done what many would consider unthinkable - and in such a manner that it is far, far more efficient than polygon-based rendering could ever hope to be. The demo they showed back in August 2011 was entirely in software, on a modest laptop, for God's sake. I also think that Unlimited Detail could revolutionize the way developers create 3D assets and engines. The rendering method is so utterly efficient that rendering model geometry is utterly trivial, leaving a ton of space for other things for the GPU to do. As well, not only are artists no longer limited by polygon limits, but also can scan in real-world objects via laser scanning, which I imagine will have some very cool applications. More recently, Euclideon have been showing their geospacial application of this technology, Geoverse, to people in the geospacial industry behind closed doors. They've also been updating their website indicating their their proper website is coming soon. Bruce Dell has stated that when the next demonstration (which will likely feature animation, proper lighting, art assets by ex-THQ artists, and maybe physics, and be playable) is released, the Unlimited Detail SDK will be released along with it. Judging by his other comments, it's a fairly educated guess that it'll happen within a couple of months. Can't wait.

A bit of a something I've been working on, on and off a bit. It's not a Zero game, I'm using him as a placeholder because of his versatility. I want to replace him with something else later, when Spriter and its plugin for C2 is updated a bit. Currently, you can move, jump, use a single 'main' attack, use Castlevania-style subweapons, dash, duck, and wall-jump. Dashing and wall-jumping aren't perfect, the latter in particular, but, still. Oh, and there's an adventure-game style look mechanic, too.

It's been a couple of months, so... Hoo, boy. I haven't been working on this all that much, because of another disheartening setback (which I'll get to in a moment). Web playable link, and now an EXE download. Construct 2 .capx, C2 Release 106, requires Standard license (also requires the Function Plugin) Not much has really been done, aside from moving platforms (surprisingly simple) and more a proper level framework. Now, this project has been fraught with problems, but now that there's actually a level framework, it also came with a horrible realisation that even in WebGL, the framerate drops dramatically even with that bare bones of a level. Good grief, there should be no way that the engine should be THAT SLOW. It shouldn't even be a FACTOR in any kind of FPS drop. I'm at my wit's end, here, honestly. I don't know what to do, it makes me want to scrap the entire thing and start over, but with Streak pretty much inundated with work, and Dami hasn't responded to my PM, I'm pretty much lost. Considering the sheer instability of the project, a total restart is probably for the best, but I wouldn't know where to START, nor how to do things better than the framework I'm already using. Good grief.

I'm gonna post my own brief analysis of certain design aspects of Metroidvanias and what should be used and avoided, which should hopefully be a springboard for some good discussion on the topic. For a while, now, I've been contemplating how to design a good Metroidvania-style game. I have attempted before with Aria of Destiny, but that was much more rudimentary and, well, crude, in my honest opinion. I could do so much better. In my examinations, I've come across this particular article examining the ins and outs of Super Metroid's design, which I think does a marvelous job of demonstrating why the game is a pinnacle of its genre. Super Metroid, followed by the Metroid Prime games, are, in my opinion, the pinnacle of "Metroidvania" design. The Castlevania games in this style, in my opinion, while they can be fun, they do have some things that fundamentally hold them back in some respects. I also think Sonic could easily be adapted to a 'metroidvania' style. Hell, there's already a game of this style in the series, Tails' Adventure, which provides its own spin on the 'genre', utilizing a structure of self-contained areas that act as levels on their own, but can be explored with new items to uncover hidden goodies later on, but I'll get to that later. Firstly, let's explore the distinction between Metroid and its Castlevania 'cousins'. Metroid and Castlevania Castlevania's first foray into more open-ended level design started with Castlevania 2: Simon's Quest, was an intriguing experiment that, much like Link's Adventure, was rife with issues, from the more perplexing (even in Japanese, the NPCs were incomprehensible, and the bosses could be SKIPPED, and even if you chose to fight them they were piss easy) to the simply annoying, such as grinding for hearts to get cool stuff, and the lack of a map to see where the hell you were going. Castlevania: Symphony of the Night tried again, this time with more level heads and another incredibly good game to work off of - Super Metroid. It pretty much avoided nearly all of CV2's problems, and a couple of its own issues would be fixed in sequels (such as being forced to equip consumable items to one of Alucard's arm slots to actually use it, which was silly). It was one of the most beautiful and atmospheric games on the PSX, and with good reason. Alucard's animations were fantastic, the game was packed with detail, the music was amazing (which is the standard for Castlevania, really), and there was a ton of enemy variety. You had several subweapons to utilize, dozens of swords and other weapons, spells, familiars, all that jazz. Now, where it fell short? A couple of things - firstly, it didn't nail that sense of loneliness that Super Metroid excelled at, and its atmosphere, while haunting and gothic, wasn't quite up to that level either - the (amusingly) bad dialogue was probably a factor in that, as well. And, as well, it inherited a problem from Simon's Quest - grinding. No, I'm not talking about level grinding, the "Castleroids" are actually generally decent in that respect (Dawn of Sorrow in particular has a perfect leveling curve), but grinding for anything else? Hoo, boy. The most broken sword in the game, in particular, is dropped (with a very small chance, I might add) by a rather small and weak enemy only encountered in the inverted library. And that got me thinking - did the Castleroids NEED RPG stats and items? Honestly, I think not. Those games generally allow you to play as a normally unplayable character after you've beaten it, and these alternate characters only had the stuff you start with, and some of them don't even have levels. And these alternate modes are very fun in spite of the lack of leveling and whatnot that the main game has, if not moreso in some respects. I think the main issue is, Metroid hides its upgrades and hidden goodies across the world, and relies on that for 100% completion. Problem is, Castlevania doesn't go to such lengths - you don't have anything like the morph ball, bombs, or missiles to uncover hidden items. The 100% completion is mainly reliant on filling out every part of the map. And then Portrait of Ruin and Order of Ecclesia made things worse by adding in fetch quests. While some of the games after Symphony had some inventive mechanics, many of them have relied on grinding. The card system in Circle of the Moon is especially frustrating. It's a rather unfortunate result of trying to stand out from the thing it's inspired by. I believe that the Castleroids would be able to stand on their own without the need for such RPG elements. Tails' Adventure Tails' Adventure isn't strictly a 'Metroivania' - key word here being 'strictly'. The game is structured into several self-contained areas linked by an overworld, that are structured like linear levels but feature hidden items that the game encourages you to find by returning with other new items to access them. Castlevania: Order of Ecclesia uses a similar overworld style, but sadly doesn't feature nearly as much hidden goodies. Tails can utilize a whole assortment of various equipment and items to progress on his quest, and even acquires Chaos Emeralds, which act as health/energy upgrades, a staple of the genre. In one respect, it succeeded where Simon's Quest failed - providing structure to levels that provide challenge and can be carefully designed without having to compensate for the player having to go both left and right, one of Egoraptor's main complaints about the game. That's not to say it's designed like a 'traditional' Castlevania game, but you get my point, though TA can be rather difficult at times. And it even did it while allowing multiple routes - Polly Mt. 1 has two routes, one of them being a difficult route that only exists to find the Knuckles item. The way Tails goes about solving his problems is very "Metroidvaniac". There are often conspicuous details that the player will notice and will want to use items on later. Small spaces can be navigated with the remote-controlled robot, much like the Morph Ball. It's like playing with Samus as a gageteer genius rather than a futuristic bounty hunter (granted, such a game would suit Agatha Heterodyne of the Girl Genius fame much better, who happens to be have been using a rather excessively large powered armor with a death ray, lately). Tails' Adventure wasn't perfect (it had some issues, mainly with the interface and inventory) but it was one of the better Game Gear Sonic games. How to potentially apply that to Sonic? The Adventure games had upgrades that characters could acquire. "Classic Sonic" already places an emphasis on branching paths and exploration, as well. Use rings as a measure of health like in Tails' Adventure, and we're in business. It's not rocket surgery. After that, you need atmosphere and careful planning in your design. The Super Metroid article provides a potent example of how it goes about doing things, try and see how you can apply certain methods to your design and how well they work in the context of a Sonic game. There are some fantastic design choices that Super Metroid utilizes. All things considered, I think there's so much in the way of opportunity. The key is figuring out what works.

Would help if we could actually post topics. I sure as hell can't, it seems, haha.

Edit: Yes, please, widescreen would be ace. It would help immensely in a game like this. Just unlocked Carol... And holy hell, I love playing as her. It's like I'm playing as a Mega Man X character with Sonic physics, it's brilliant. Though wall-jumping is a bit more difficult than it really should be. And I love how she says "DIDJA SEE THAT THAT WAS AWESOME!" at the end of the level, and considering I finished the boss by hitting him out of his own attack, I can't really disagree. The voice work in this game is great, btw. There's also very little subtlety involved in newer games these days, something that some older games were absolute masters of, such as Super Metroid's absolutely masterful methods. Of course, there's the occasional instance in games where the developer either gives no hints whatsoever or their attempts at subtlety don't work - the infamous barrel, for instance. You know the one.

I beat it in 6:20. No unlock.

I'd rather keep the difficulty as it is. It's the first level, but it's refreshing to have a Sonic-style game that isn't depressingly easy (well, that is, one that doesn't rely on bottomless pits for its difficulty). The health bar alone does wonders for the game's difficulty, much better than the rings system. Also, how the heck do you unlock Carol? I've already gotten all the cards...

I know, took their time, but they had better things to do. The last three releases were mostly devoted to adding image editor features, they said they're not trying to replace Photoshop or anything, but it's better than MS Paint. Granted, I don't really see much use in it aside from prototyping visuals, and I already have a bunch of open source placeholder images for that.

Alright, since we're in the run-up to SAGE, I figured I might as well try and continue working on this thing and polish it up as much as I can first. So, yeah, new release. Links are two posts up, as before. Release 99 is required. Additions: - All monitors work properly now, work-around made so they can be solid - Bubble shield works now - Insta-shield implemented - Several sound effects added, as well as music - Water fully implemented (bubbles, drowning, etc.) It's not using the "water level" implementation like in Worlds, but it's a bit more flexible in some ways, as you're not limited to a body of water for your level design. I'll probably include the original water level implementation at a later date, though it's hardly necessary right now. - Speed shoes implemented - Invincibility implemented - Super Form implemented To-Do List: - Moving platforms - Enemies - Tails - Knuckles - Other gimmicks - Probably changing Sonic's sprites to Sonic 3 sprites rather than Advance sprites, for consistency And in other Construct 2 news, Release 100 is coming in August, which will feature: - Full native iOS and Android export options (probably) - WebGL shaders and CSS filter effects - Windows destop export (using Awesomium, even faster than running games in Chrome), and possibly Mac and Lunix desktop application export as well. Which will come with an increase in the license prices, mainly due to C2's increasing capabilities and, as of R100, release to a rather diverse range of platforms. All things considered, I think it's rather reasonable. Oh, and the most recent releases now have actual image editor features. Sweet.

I should probably make a new topic for this now, because it's becoming its own beast (in a way) and advancing beyond its Classic counterpart, but I've decided to leave a few issues behind, such as "snagging" on some curves, the inability to roll after landing on some curves, and stuff. Links in the above post, but now the capx. requires R95, which is the latest and stable release. Additions: - Z+Down allows you to fall through platforms. - Springs - Sonic can now be hurt (With S3 animations rather than Advance, yes, I'm a bit lazy with that, but I'd rather be coding than doing animations) - Spikes - Rings and ring loss - A few sound effects - A "Inner_SubStepLimit" fix where the Player would stop moving at speeds higher than that value (seems to cause a couple of minor issues of its own, though, oddly enough) - Monitors, some of them working fully (the icon doesn't float up and disappear yet, and monitors aren't solid, see issues list) - Shields with special abilities (Bubble shield ability doesn't work yet for reasons listed below) - Bumpers, both from Spring Yard and Carnival Night Zones, working more or less perfectly - Balloons from Carnival Night Zone - Shoved all the Sonic-specific stuff into its own event sheet, for the sake of organization New issues: - I need to figure out how to make solid monitors and the Bubble Shield bounce work without making Sonic automatically go into 'idle' mode upon landing. - Sonic's sprite lags behind the player object a tad now in some circumstances, for some reason.

There's also Text Blitter and String Parser... ... And the extension pack installer won't let me install either. Terrific.

Hey, could someone upload the requisite plugins to get Sonic Worlds to work? I need to get the assets (and look at the events) so I can properly port stuff over to Construct 2, but the plugins aren't included in the new mirrors.

I made some rather simplistic 'interior' tiles for a building of some kind. Heavily inspired by Iji.

Wow, that looks sweet! I sympathize with your woes, though, animating that manually looks like it's gonna be a massive pain in the ass.

This isn't nearly as nice as the ones above, but... Threw together some tiles from an open-source tilesheet. Tried to give off a feeling of erosion and decay when modifying the tiles to make the slopes. Oh, and you can see them in action in-game, too, in your browser.

Arguments aside... Progress! Playable Thingy! (Controlled with arrow keys and shift) Construct 2 .capx, release 90, requires Standard license (also requires the - Finally added DW's debugger thingy. - I've also fixed up a couple of slope detection events that I realized were done a bit wrong, and is making the angle detection work much better. You can now go up large curves and loops properly! With enough effort. Regular slopes work perfectly, right now. - Platforms now work. Known problems: - For some odd reason, you can't roll upon landing. - It's easily apparent that the player doesn't rotate fast enough. This results in some awkward results when trying to move up slopes or loops. - In certain slope situations, Sonic spontaneously stops for absolutely no reason. It may or may not be the result of certain sensors mucking things up when they shouldn't be. - I just discovered a limitation of the Classic engine that carries over to this one - if you don't have a gap of wall between a curve slope and a floor, you will immediately move onto the floor once you've gone up the slope rather than flying off. Granted, this can be easily worked around with level design tweaks, but I still call shenanigans.

Well, I'm re-doing the tower defense prototype in C2, with a bit more organization for placement. The enemy and turret visuals are placeholders, and the semi-visible boxes off to the right are 'sensors' for placing turrets and traps to check for valid locations.

Auditions, huh? Interesting. Might be able to throw something together for Spade's voice. Not sure if my acting will work for him, but, eh, might as well give it a shot. Figuring out a good tone of voice for the third line is a bit tricky, IMO, but I think I've got a good idea of it - Spade seems like the type of hot-shot poker player who gets openly aggravated when his "cunning bluffs" fail repeatedly, whether it be because of his own skill, or his opponent's.

Can't say I object you your choice. R-no71 is a fantastic artist, his animations are wonderful, and, well, a Blaze fan is an added plus.