Jase00's Forum Posts

  • It's so fast and just makes sense. Rarely if not ever feel like the C3 gets in the way of what I'm trying to do.

    You could do something, a mistake happens, and 9 times out of 10, it's a random mistake you made, and if it's not, or it's a limitation, then you can often work around it with the huge variety of actions, conditions, plugins, etc.

    Yes every software has its weird parts, or could improve somewhere, but I simply don't encounter this.

    Multiplayer is a good example, it's my favourite plugin, I don't get all the buzz about it not being good enough or something. Even if the built-in sync stuff is not for you, then, you can do everything by hand anyway, we have the getbit/setbit actions, we have the ability send messages reliably or unreliably. Yes, again, always room for improvement, maybe room options or more properties for room info (such as name of host when displaying a room list), but guess what, you can workaround this and inject the name into the room name, and tokenise the room name when showing it to the player! Ahh this stuff is why I love game dev. Maybe to some it is convoluted, but to me it's adaptable.

    Major kudos to Scirra team for what they've achieved, but I want to point out Ashley, Diego, and other past devs, because - it's clear that almost every game developer HATES designing UI, can you imagine how it must be making C3... that... Is incomprehensible. Not to mention the rest of the behind the scenes stuff like runtime, bug fixes, backwards compatibility, but, cmon, UI is notoriously hated. The level of patience they must have is God tier.

    Also major kudos to the third party devs, unlocking even further usability within C3, effects, and things thought to be impossible or unlikely to see. I can't comprehend the work that goes in and will just default to "magic", undermining the the actual skill involved.

  • > On your other post, I shared a bug report link that has these exact steps you wrote!

    >

    > When you say C3 app, you mean in nwjs, or in Google Chrome?

    >

    > AFAIK it happens on latest Chrome (also happened about half a year ago, again a Chrome issue), and will likely be fixed around 20th Feb according to what Ashley found. Can use Chrome Dev, or Firefox, for now.

    Sorry, I think I must have missed it. Thanks! I've been requesting features that already exist, reporting bugs that have already been solved... and then forgetting I already did those things. ugh, I'd lose my head if it wasn't attached to me. I'm getting to be like that old person who complains about a problem that was solved 2 decades ago and keeps forgetting that it was solved. :(

    Haha no worries, I get that feeling too, our minds get a bit polluted with so many different things, hard to keep track of!

    At least this issue will be history soon.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • What I hope to see from Flowcharts, is wayy more simplified event sheet stuff, and more "do most of the work in the flowchart view".

    Events could be mostly "on node changed" etc., not as much tokenising and such. How? I don't know, just wishful thinking but wanted to share.

    I'd absolutely love to see a function dropdown list in flowcharts, which I know others have mentioned and may not be on the cards.

    The reason a function list would be amazing, is, in flowcharts current form (respectfully, it's early days), it is a feature that you could think "well, I could just make this kind of system with event sheets already".

    Things like hierarchies, 3D, custom actions, they all have either saved time, gained us performance, or unlocked something we could never do before.

    Flowcharts feels like a "you could use this, or could do it manually" type of situation, there'd be not much saved time (because of all the events and tokenising, which you'd be doing if you made your own flowchart system via events and JSON - although one time save would be not making some UI to link stuff together), no performance difference AFAIK if you use flowcharts or did it by hand, and it is indeed something we can do ourselves via events.

    This is why I feel a function list or something, would be incredible to see - it suddenly opens a brand new way of designing logic, could even adapt flowcharts beyond dialogue systems and more for AI, kinda like Sims had with it's developer software Edith for designing Sims AI logic. I mean, you could do AI, with current flowcharts or a manually created flowchart system in events, but just feels like it adds that extra flair that Construct is known for, visual rapid development, what you see is what you get.

    But, I get this may not be the intended vision and just needlessly complex. Worth sharing feedback even if it's echoing previous comments, I suppose if a lot of people say something, then that means something.

  • On your other post, I shared a bug report link that has these exact steps you wrote!

    When you say C3 app, you mean in nwjs, or in Google Chrome?

    AFAIK it happens on latest Chrome (also happened about half a year ago, again a Chrome issue), and will likely be fixed around 20th Feb according to what Ashley found. Can use Chrome Dev, or Firefox, for now.

  • To use return values, you will need to double click your function (as if you're about to rename your function) and you will have an option to set the "return type", either number or string.

    This will change how the function works, I.e. you can no longer call the function as an action anymore, and can only call the function as an expression (e.g. You can set a Text object's text to "function.block-left-right").

  • Iirc if you do this, this will give you 0 or 1, depending on if Variable1 is actually set to 22 or not (0 being no its not equal, 1 being yes it is equal).

  • Not sure about the sprite editor hangups, but the "oh snap" bug, sounds like it's chrome issue and is fixed in dev build of chrome (which will come to regular Chrome on Feb 20th).

    github.com/Scirra/Construct-bugs/issues/7796

  • Agreed!

    github.com/Scirra/Construct-feature-requests/issues/65

    Gimmie some thumbs, only got 1 at the moment!

  • You do not have permission to view this post

  • With all info given, it is indeed weird.

    I have 1 possible idea, very unlikely:

    Did you had addons, and you did NOT bundle them in your project, so each test you did on other computers and browsers, you had to install those addons, then open your project and get errors?

    If that is what you're doing, then, did you modify those addons or update them?

    I say all of this, because, I tinkered with addons before and this made projects unable to be opened (didn't check console when this happened).

    Beyond this, I may be wrong, but, might be worth emailing your project to Scirra's support email, as, it's not a bug, but it's a severe issue with your project. Just clarify in the email what you've tried, other computers, other browsers, older C3, older backups, etc.

    It may be impossible for them to solve, or may be a huge endeavour for Scirra to investigate especially if its a large project, but, well, if I were in your position, that would feel like my only hope with such a severe issue.

    Do you have other info you can think of - did you mess with containers a lot in your project? Did you only add containers recently? Have you had a look at your c3p file with 7zip/zip software, does it open? Is it blank? Is it corrupt when you try to open it and open a random file within? Does this happen with any other projects? If unsure, try making a new project or take an example project, save it, close, open.

    I even wonder if "container" is a red herring in the context of Construct, maybe it's not container-related but a weird generic error message referring to another type of "container" within JS/html.

  • I'd say it's reliable, but I believe the debugger costs a bit of performance itself (though not 100% sure, or, if it does cost performance, is that reflected in the cpu %?).

    It may be more cleaner to add a text/SpriteFont and set text to cpuutilisation, then preview without debugger and see what you get.

    For the FPS, I've found Chrome is always wayy more performant, whilst Firefox lags in same exact game.

    For the FPS dropping randomly, I've only ever seen that on mobile devices. My mobile is a 120hz display, and even when I force it to "gaming mode", it drops to 60fps when it is overwhelmed, then few seconds pass, then it jumps back up to 120fps. Not sure if any desktop computers also do this, but I wouldn't have thought so.

    Perhaps the whole "nearly 100% but on older device it barely reaches 15%" thing, my guess is, is it using webgl2, or is it using your gpu on the device that reaches 100%? The debugger will tell you if webgl2 is being used, and also you could create a text object that outputs gpuutilisation - if you get a value, it's using your gpu. If you get "NaN", then perhaps gpu stuff is being offloaded onto the cpu, causing 100% cpu usage.

    Just some thoughts, hope it points you in the right direction!

  • Curious! I wonder if AZERTY is a keyboard with its own region profile on a computer - as in, are the physical scancode values different comparing QWERTY and AZERTY, or, when you plug in an AZERTY keyboard, are you expected to change the keyboard region before it works correctly?

  • I don't know the answer but have some suggestions.

    What browser do you use? I think best to try Chromium.

    You may have already tried this, but have you tried different type of drivers, like open source and the vendor provided ones?

    In the past, I used Manjaro with a Nvidia GPU, I remember one of the drivers would just give black screen with only DOM elements appearing (which didn't react, like it crashed) for C3 games, the other driver was better and a 3D C3 game looked correct (but I recall there being refresh rate issues, and I couldn't solve it, so went back to maining Windows).

    Also, again I'm not super familiar, but could Wayland be the issue? I seem to see that pop up often in reddit for Linux gaming issues, although this may only be relevant for Nvidia or not relevant at all, just a guess!

  • ...it's normal it went fast during the first days because people posted their most wished/relevant suggestion from the old platform as there was a reset.

    Ah true, this went right past me.

    Github provide a bunch of way to filter and sort issues.

    Just sort them from the "most recent" to see the new ones, "recently updated" to see recent discussions, or use searching syntax to filter only last month.

    That helps, thanks for pointing that out! Particularly "recently updated" even if a random obscure old suggestion that got 0 votes, suddenly gets comment, then it's in the front of the list if sorting this way.

    Being able to add tags like "Behavior", "Editor" could help indeed but i suspect tags are reserved for Scirra...

    Yeah, for now it's doable to scroll through a few pages, hopefully some form of categorisation can be done long term if suggestions get into the hundreds.

    FWIW Godot, the most "democratic" game engine out there also use Github issues, doesn't limit the amount of suggestion per user and don't care about having 3955 opened issues (with about 7 new ones per day), given their incredible growth and speed of developement i would say it works quite well.

    Excellent point! Understandably Godot may have a significantly bigger team (I'm not sure) but the fact they use Github for suggestions shows that it is tried and tested, which gives me a lot of confidence! My opinion is a lot less mixed now.

  • Infact I counted manually, assume we did have only 1 suggestion per person.

    There's currently 64 suggestions, subtract 51 (so everyone had 1 suggestion), and there would only be 13 suggestions!

    I do agree it's more about gathering community opinion, but does make it look way more clean if it was just 1 per person. Very mixed, I don't know.