zendorf's Forum Posts

  • So if I understand everything correctly all we need is:

    1. Remember the original source of imported animation frames and have a "reload from source" option

    2. Make an easier way to replace existing audio files

    Does that basically cover everything?

    That would be great if it was added! Some extra notes:

    • Reload and Replace should be separate options depending on what you want to do. A reload is fine if your updated assets retain the same filename, but if I want to replace current assets with another with a different filename or source location you would use Replace. This standard practice in something like After Effects or Premiere.
    • Would also be good to have a reload/replace options for the audio files, so that it would have the same functionality as animation frames.
    • An option to hide the ogg/m4a audio in the project window as it just clutters up the list.
    • Please don't rename the animation frames upon import. At least then there is also the option of replacing files directly in the C2 native folder structure.
    • multi select (either marque or shift select) of frames in the animation editor. A more advanced feature would be to multiselect some frames and then a right click menu to choose "replace frames". This way it would be more flexible in that you could replace a single frame, multiple frames or all frames. This would also be useful for deleting, duplicating and shifting frames around.
    • multi select in the project bar would also be very handy!
  • Ashley , The problem with replacing animation sequences directly into the C2 created folders is that the images have all had their names replaced to a generic 000.png - xxx.png. If I have my run animation originally created as run000.png - run012.png , am I expected to export all of my new animations and images with the exact same file names sequence of 000.png - xxx.png ?

    If you don't want to change the IDE to allow external asset reference then one easy solution would be to allow a multi select of all of the animation frames, which would then allow you to delete them , and then reimport the new sequence. Currently you can only select one frame at a time to delete. Hence why I create a new animation, import the updated frames into this, delete the old one and then rename it to the original. Yes, it is as tedious at it sounds. At least implementing a multi select via a marquee select of frames would be a halfway measure.

    The problem with updating sounds in the folder is that we have to replace not only the original wav file but also the ogg and m4a files. This means to you have to use C2 to import the original wave file, but for any updated audio files, the user must manually convert to ogg and m4a to replace into the folder structure. At least the wave file has not been changed by C2 upon import.

    One possible way around this would be have the option for C2 to only do the conversion to ogg/m4a on preview/export, so only the wave file needs to be replaced. Yes this would slow down the preview process, but would be good to have as an option. Even better is have C2 only reference the audio files from a folder of my choosing, then I can replace the originals at will.

    Another issue with replacing sounds is that if I delete a sound , any actions using this sound are automatically deleted without any warning to the user! This caught me out quite a few times initially. So the workflow becomes import the updated sound (which of course gets renamed!), have it converted, changed the actions to reference the newly named sound, then delete the old ones. To compound this, since I can't multi select sounds in the project view, every single sound needs to be put into a folder so that all 3(wav,ogg, m4a) are in one folder to make it easier to delete. Yes , it is a currently a nightmare. My current C2 game is a musical one, where every sound has each 7 notes of a C major scale, so each time I need to update a sound it is quite a nightmare!

    On a related note, is the general way the C2 needs all assets within its own directory structure. If I want to iterate my project to the next version (which I tend to do several times a day), this means that all assets have come along for the ride. If the project has 250mb worth of assets, each time I save my project it is saving all the assets, where ideally all I want changed is the project file that will point at the same assets. What is I want to go back to a version of the project that I did the previous day , but with my updated assets?

    Hopefully, you can begin to see that separating project files from assets is the only way to go for a decent sized project. Sure, there is a certain elegance in having everything contained in one zipped capx. This will work fine on a crappy bird clone, but for a project with hundreds of sounds and sprites it becomes a logistic nightmare.

    While I don't have the game industry experience of stuatk , I have over a decade of experience in the broadcast animation business and have worked on many high profile jobs. All compositing and editing packages only ever reference media files for good reason. It makes it easy to update/reload files at their source and the small project file is easy tranportable.

    Just the fact that someone of stuatk's calibre is even trying out C2 is noteworthy of itself. C2 has been turning heads, mainly because of its event system which is the jewel in its crown. Since it is so good, users expect the rest of the software to be of a similar quality. Sadly the asset management, exporting problems (reliance upon 3rd party solutions) and non modularity of code are the main issues which will continue to plague the product unless they are addressed.

    Hopefully the Scirra management will take note these issues in its long term business plan, as there is now plenty of competition among engines now. Most offer a free (Unity and GMS) or very low priced(UE4) version and they don't tend to have the same issues. Although there is nothing out there that is as easy to use as C2, there are plenty of contenders such as Stencyl, GameSalad and Game Develop (which was recently open sourced and is very close to C2 in terms of workflow), who would happily eat Scirra's lunch. I would prefer to see a more pro orientated version (C3 Professional) with haxe or similar underpinnings (ala Stencyl) be developed in tandem to the current C2 which would stay as a html5 engine.

  • Actually I have brought up this issue several times on the forum and not really had any replies. To me asset management is by far the weakest aspect of C2. The amount of times I have wasted replacing animations and sounds has really driven me to the point where I have mostly moved onto other engines. If you are building even a medium sized game, art assets will need to be updated numerous times and you shouldn't have to jump through hoops!

    Currently in C2 my workflow to replace an animation is to create another animation with a slightly different name, delete the old one and then rename the new one to the old name. I can't even marquee select the animation to delete frames, so that I can then reimport into the same animation. The ideal way is to just be referencing a folder of my choosing, so that when I change the animation frames in the source folder it updates it C2. Or have the option to reference it in another location.

    Likewise changing audio is a nightmare. The game I am working on has some serious audio fx going on with 9 sfx variations for each effect (it is a musical game) so when I import these 9 wav files into C2 I get 27 files (including ogg and ,mp3). Now to replace these I need to delete them , reimport and then relink in the actions. Actually I have to be careful to import the new files first, relink and then delete the old, since if I delete audio it will delete the actions! Would be very easily resolved by having the files referenced by C2 and just swap them out at source. Also, why is the audio list cluttered up with the ogg and mp3 files ? Obviously these will be used, but the user doesn't need them to be seen. Scirra should look at how it works in Unity where the Assets window is basically a mirror of your actual project structure. Currently it is all dumbed down and out the users control which is maybe by design for the beginner, but will only frustrate a serious user.

    It is a pity, since C2 is so nice to use, but the combination of no asset management and the woeful export options has mostly driven me away from C2. I am finishing off a long term project with it, and will keep up with development and the community but I have mostly moved on. It really needs some big changes if it wants to keep up with the big guns of Unity, GMS, UE4. I will always keep it around for pure HTML5 output as that is its real strength....funnily enough

  • andreyin ...looks like a nice game...congrats on getting your title on Steam! How did you get onto early access? Did you have to get Greenlit first?

  • If you want to do 2d and aren't scared of code, I would recommend GMS, Cocos-X or Godot engine(which is in beta but also does 3d). As much as I love C2, it has been limiting what I want to do and I have been moving projects to some other engines. Though I will still use C2 for html5 games for sure, nothing beats it there!

    If you are doing 3d then you can't go wrong with Unity or UE4, and I wouldn't really consider anything else at this stage, except maybe the new Shiva when it is released. I am actually finishing off my first 3d Android game using a small open source engine called ZgameEditor. http://www.zgameeditor.org/index.php

    ZGE is fairly basic, but quite similar to GMS in its structure and scripting language. No layout editor at the moment, but it is planned. The main thing I am loving about it is how easy it is to export and test your apk, especially considering the nightmare that it is with C2. As always...horses for courses...

  • Run, don't walk over to the Blackmagic site and download Davinci Resolve lite version. It is insanely good and a very high end app, as it is mostly a colour grading solution, but the video editing features are fine for basic use. The new version is out any day now, and will have major new features added to its editing toolset...so good in fact that it will take a lot of market share from Premiere, Finalcut and Vegas. I have been using Vegas professionally for many years and will be moving over to it , once it is released. The free version is not hobbled much at all!

    http://www.blackmagicdesign.com/au/products/davinciresolve

    Other free options are Lightworks, which has been used on hundreds of feature films and Blender which has a basic non-linear editor:

    http://www.lwks.com/

    http://blender.org

    The cut down version of Vegas is also quite nice along with Hitfilm express if you are looking for low priced basic editors.

  • As much as I hate Adobe as a company, there really is little to compare to After effects when it comes to motion graphics. I do mograph for a living, and AE and Cinema4D are the industry standard tools for good reason. There are better hardcore compositing apps that are nodal such as Nuke and Fusion, but AE is the go for motion graphics and game graphics, as you can do anything from cutout character animation to eyecandy fx very easily and just spit out a PNG sequence for your game sprite.

    Another aspect that is overlooked by many game artists is using animated shape layers/layer styles for game sprites. I am developing a couple of games that have a modern vector/hard edge look and the use of shape layers, point animation and glow/blur effects would be very difficult to achieve with any other tool.

  • why don't Scirra buy over this plugin an integrate it in C2 and build specific editor tools for it like they did with Spriter support?

    Yeah, it would be a good direction to add to the next generation of Construct (C3?) and would help differentiate itself from C2. Hopefully enough people support this project and the devs can really turn it into something special. Then Scirra may see the viability of adding 3d support, or least support for this plugin.

    I still have my doubts that the average C2 user will get a good handle on 3d, as it can get a lot more involved than 2d, and you do need to know at least some vector math and principles of 3d...normals, parent/child hierarchies, local vs world space etc. And that is before actually modelling , rigging, UVíng and texturing your assets.

  • > Been having a play with the plugin and while it is early days, I certainly can see some potential for it. The main issues are mainly to do with events and interactions...namely iterating through instances and collision detection. Also, some sort of layout editor (or use of the three.js online editor) will be a must for anything more complex than a procedurally spawned game .

    >

    > What is planned as far as collisions? Will it be auto cubic bounding boxes that rotate with a model or just a spherical, non rotating box? Of course options are always a good thing. The ability to use another mesh as bounding box would be great, though I am not familiar with the capabilities of the Three.js engine.

    >

    > Currently I am noodling with a prototype of a basic 2.5d shooter to learn the plugin. I have basic player movement and camera down fine. I can fire one bullet, but the issue becomes how to control instances. Of course this is all handled under the hood with normal C2. How would I check for its position to destroy offscreen? I can create multiple bullets but they all have the same name and only one responds to actions. I have tried using object>create>create by cloning but can't work it out...

    >

    > cheers...

    >

    the sub plugin will make these things easier, but right now it's easy to get things to work like regular construct by linking them to a sprite dummy, and associating the id of the 3D object to the sprite by saving q3d.idlast in a variable when you create the object. from there you can do whatever you want to make them "link". i don't know if this is clear but ill make an example if you want some help.

    There'll be options for the colliders in the sub plugin, thats what im trying to get working right now. Tiny tank isn't procedurally spawned so it's not really right to say its the only type of game you can make atm. You just need to use sprites to mock things up and spawn associated graphics.

    offscreen checks are difficult if you're using perspective projection, just give them a timer.

    Thanks for the reply...sounds good. Would love to have a basic example of the sprite dummy technique you have mentioned! Is this similar to what Russpuppy has done with his platform example? An example that shows how to use instanced objects , like ammo being fired would be very useful. For offscreen checks I was just trying to destroy them once they reached past a certain y axis position.

    cheers...

  • Been having a play with the plugin and while it is early days, I certainly can see some potential for it. The main issues are mainly to do with events and interactions...namely iterating through instances and collision detection. Also, some sort of layout editor (or use of the three.js online editor) will be a must for anything more complex than a procedurally spawned game .

    What is planned as far as collisions? Will it be auto cubic bounding boxes that rotate with a model or just a spherical, non rotating box? Of course options are always a good thing. The ability to use another mesh as bounding box would be great, though I am not familiar with the capabilities of the Three.js engine.

    Currently I am noodling with a prototype of a basic 2.5d shooter to learn the plugin. I have basic player movement and camera down fine. I can fire one bullet, but the issue becomes how to control instances. Of course this is all handled under the hood with normal C2. How would I check for its position to destroy offscreen? I can create multiple bullets but they all have the same name and only one responds to actions. I have tried using object>create>create by cloning but can't work it out...

    cheers...

  • Wowsers, colour me very impressed One try of your tank demo and I got out the wallet....$15 is a no brainer. I know C2 like the back of my hand and am an experienced 3d artist, so this has me very excited by the possibilities. I won't be giving up on Unity and Godot Engine in a hurry, but for simpler games, this could be brilliant.

    I haven't even downloaded it yet, so I am sure I will have feedback soon.

    Thanks Russpuppy, that example looks like a great starting point!

  • Thanks guys, Inno Setup sounds like the winner!

  • OK, thanks Ashley for clarifying that. I intend to distribute via itch.io , Desura etc so I am not really interested in web publishing at this stage. If someone pays for a game , they probably won't have an issue in running the dxwebsetup. My concern was more for anyone trying a demo of it (potential customers or press/reviewers) and getting turned off by lack of hardware acceleration.

    Are there any installers that you or others would recommend to package up Node Webkit with the dxwebsetup?

  • Try Construct 3

    Develop games in your browser. Powerful, performant & highly capable.

    Try Now Construct 3 users don't see these ads
  • I am getting close to releasing a desktop game using Nodewebkit and C2. Most machines I have tried, play the game terribly until I have installed dxwebkit.exe.

    Before I start sending around pre-release demos of my game, I am wondering how common it is likely that an end user will need to use this dxwebkit.exe file. I was under the impression that Nodewebkit is just chromium using Webgl/opengl, so why the need for this DirectX download? My game works fine in Chrome, but when exported to NWK will perform terribly unless the dxwebkit download/update is performed.

    Very confused as to what is going on here, and would appreciate some enlightenment Just concerned that when I send my game to press/reviewers etc, that they won't want to download this update to get the game working , and first impressions will be that it is unplayable at such a low fps.

    Just packaging up my game to send to the Intel Level Up competition (deadline today) and have been working on this game for months, previewing it fine in Chrome. Didn't realise these issues till now. Yeah, I know....I should have done some tests exports before now

  • I would be more concerned that Nintendo might be calling to ask you for a loan

    http://www.theverge.com/2014/5/7/5689878/nintendo-earnings-fy-2013