TiAm's Forum Posts

  • ludei

    Got my invite today. Interface is very, very slick; definitely a big step up from the old compiler. Love that I could search for cordova plugins by keyword and automatically add them.

    However, still working on getting a project compiled; getting 'Error Android' for android. Probably being dense here, but where do we go to see the log of build errors? iOS built fine.

    Since you place so much emphasis on aesthetics, here's one little faux pas: the color of the cursor/selection overlay in the config.xml editor is way too pale; the cursor is basically invisible. I just copied in/out of notepad++ to edit it, but it would be nice to forgo that extra step.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • This threw me too. Here what works:

      1. Start new project. 2. Import your code base. 3. Set 'Use Cordova Plugins' to 'Yes' 4. Set 'Game Project' to 'No'. Unintuitive, but there it is. 5. Import, go from there.

    DON'T chose 'blank canvas game'; it doesn't work. Instead of your game you get some kind of test app.

  • Some things are not easily done (efficiently) in c2. However, I really enjoy putting this sort of thing together because it's so quick to do .

    Whoa, gory. Wasn't expecting that...

    On a lighter note, that rope/falling animation looks great. Keep it up!

  • ludei

    Entered my email into the beta, looking forward to testing it out. Few questions incoming:

    Is there any chance of ever seeing WebAudio working alongside Canvas+? This is the one feature I really miss when deploying with Canvas+.

    Have you compared your new Webview+ solution to Intel XDK/Crosswalk export on any low-mid range devices? The old webview+ was always slower than crosswalk for me, even when comparing with the borked post-v7 cw's.

    What version of crosswalk is WV+ based on, and do you plan to update it with every new version of crosswalk, or follow a custom schedule?

    Just to clarify: we will be able to export from C2 with the stock 'Cordova' option, using whatever atomic plugs we wish, and still compile to Canvas+?

    . . .

    Off Topic: You should tweak the copy in your beta email, because it's really confusing:

    [quote:30m6hdi8]

    "As we are receiving tons of requests, we preferred to keep the service in public beta for a bit longer. "

    I assume you mean private?

  • R0J0hound

    Thanks for replying. The flag idea does sound like it would incur a hefty perf overhead, but...well, I think you know more about this than me.

    I'll try capturing the canvas multiple times and see if that works.

    Shame there isn't a good, fast way to capture the canvas built into the webgl spec, seems a bit of an oversight.

  • TiAm Somebody

    I'm not having much progress with getting the canvas pasting to work reliably. It's currently kind of random if it works or not, so I may end up having to go a slower route of making an off screen texture and rendering everything to that.

    Hey, any chance of you working your rojo-mojo on this?

    I use this functionality for the pause menu in my current project to do a 'background blur' of all layers but the top menu layer. I just grab the canvas when pausing, paste it onto the paster object, and use an animated blur effect on it.

    Problem is...at some point 'load texture from canvas' stopped working on my mobile. Used to work in cw7 and chrome mobile, iirc. Works fine on desktop, except in IE (and it never worked there).

    The only other way of doing this (other than the painfully slow CanvasSnapshot function) is to apply the blur to all layers except the top, which is too slow on mobile.

  • Crowdsource it.

    If a user has x amount of rep, or a badge, that person can flag any dubious first post, the suspected spammer and its post/s get sent to the corner.

    The newb gets limited to a hidden forum, and mods give a thumbs down later.

    That could backfire because rep can be gamed up reasonably high by a bot account (for example, the rep bonuses for views to profile), and if there are no post limits, and only low rep users can be 'crowd-modded', bots could very quickly escape 'rep hell' by gaming views/posting rapidly.

    If this was done, the rep level req for users to ban posts would have to be very high (at least 5000, maybe 10,000) to curtail abuse.

    I think it's a bad idea because if there is an exploit in the rep/badge system anywhere, spammers could start banning legit topics.

    Elliott

    This. Tom, where oh where are you...

  • ...the captcha breaker is probably the result of a website that sends captchas to users and pays them a few pennies for each correct answer. It has been done before.

    Really? That's brilliant...in an evil genius sort of way.

    Ashley

    Seriously though, we need some sort of temporary restriction on new members making more than 1 new topic in a day. Or maybe a really low rep limit, like 150 or 200 rep.

    This is getting bad, when it hits hard the forums become unreadable...and like you say, you and Tom can't always be awake.

  • Whole bunch of IPs definitely suggests a botnet. We are still getting hit periodically.

    This is kind of an interesting read (especially the part about Bayesian Poisoning):

    http://www.embergaming.us/index.php?thr ... tempt.153/

  • Thanks a lot for all the effects you have been uploading; there's some great stuff here.

  • This is definitely the worst spam I've seen; the forum is swamped.

    Spoke too soon ! Perhaps low rep accounts should be limited to only be able to start one thread per day and limit the rate of posts they can make to one per 10 minutes, or something similar? I dunno - I don't have a monopoly on good ideas - but this just reacting to the spam makes this forum seem a bit behind the times compared to many others...

    Sounds like a good idea. Somehow hackers always find a way though.

  • WOW.

    :O

    This is literally perfect for a space shooter I've had on hold.

    Thanks!

  • — (and everyone)

    Crosswalk Export: TLDR edition:

    I asked Intel. They said "ask Google".

    Asked Google. They said "won't fix, ask Crosswalk":

    https://code.google.com/p/chromium/issu ... 1430679226

    Asked Crosswalk. They said "won't fix, roll your own crosswalk":

    https://crosswalk-project.org/jira/browse/XWALK-4158

    I asked myself...where do I get the time to do this?

    In other news: A fix is on the way...but who knows when:

    crosswalk-performance-mega-thread_p913870?#p913870

    Meanwhile: Google is sending out conflicting emails to devs with cw7 apps on google play:

    intel-xdk-openssl-fixed_t130231

    crosswalk-intel-xdk-experiences_p914746?#p914746

    Your game is fine. Your game will be taken down.

    ...Please clarify?

    Sorry Dave, we can't do that.

    . . . <img src="{SMILIES_PATH}/icon_mad.gif" alt=":x" title="Mad">

    This just in: Deploying C2 games to low-end android devices is a freaking nightmare. Always has been, always will be.

    I'm throwing in the towel and targeting powerful devices only; CW12 is good enough for that.

    Next step: iOS...and start learning Unity. <img src="{SMILIES_PATH}/icon_e_geek.gif" alt=":geek:" title="Geek">

  • [quote:2doytqvl]Recently we sent you a notification that one or more of your apps should be upgraded to more recent version of OpenSSL, due to security vulnerabilities. The notification was sent in error, and we thank you for previously making the necessary changes to your app.

    We apologize for any confusion this may have caused.

    Regards,

    Google Play Team

    ©2015 Google Inc. 1600 Amphitheatre Parkway, Mountain View, CA 94043

    Email preferences: You have received this mandatory email service announcement to update you about important changes to your Google Play Developer account.

    Now I got this from Google. Has anyone else recieved this from them? I haven't changed a thing.

    This is way too vague...does this mean we can upload new cw7 apps to google play? That existing apps get a pass?

    Also, this part strikes me as odd: "... we thank you for previously making the necessary changes to your app..."

    What changes?

    Methinks the issue is still clouded...

  • Unfortunately, there has yet to be a resolution to this.

    I tried to get this resolved by contacting Google, but they basically said 'no, we won't fix it because it's really old'. I tried to point out that this was a unique case, and that the fix was elementary, but in the end it felt like I was trying to reason with the Daleks. Eventually they locked the bug, took up their plungers, and went home.

    Another avenue of appeal is to the crosswalk project itself. I suspect they may be a little more open to reason, but I haven't had time to write up a bug yet and submit it to them.

    Intel is being strange about this, in that they acted like they wanted to get this resolved, made a blog post about it...and then just dropped the ball. I'm glad they are being active in the forum, and a recent alpha of the latest crosswalk build looks to be a big improvement, but part of providing a reliable method of deployment is supporting the only freakin' version of the exporter that works well on less-than-bleeding-edge devices.

    tl;dr: Why the hell does it come down to users to resolve an issue like this? And what happens when a more significant bug crops up...?

    /rant

    P.S.

    No hard feelings at Scirra, I get that Ashley is crazy overloaded, and this really isn't their responsibility in the first place.