RaymondHooks's Forum Posts

  • Hello. I'm assuming that you basically have no experience in programming, from your comments above. If that's the case, you may have figured that you're a bit of a hard case to help. Given that you have figured out ways to make things work, such as using an object that frankly scares me (timeline), then don't give up.

    While Scirra has attempted to make make it less daunting, knowledge of programming is still needed, because setting up event sheets is, indeed, programming.

    No, no, no. That part in bold is not true at all. Speaking as a long time C programmer who has worked on commercial games, having knowledge of C is fine but certainly not a requirement in the least bit to learning Construct or any event-based system for that matter. It wouldn't make a difference if you were the best C programmer who ever lived, it doesn't mean you're going to do well with this program and there are a lot of C guys who are frustrated to no end with these types of programs. The bottom line is, his question is just another thing that should have a focused tutorial and there isn't one, that's the problem... the user is not the problem... we have to stop blaming the user... the documentation isn't good enough or non-existent, that's the problem and why they're asking the questions to being with... besides that, programming a global variable in C is not the same as in Construct, so you can't say because they ask the question about Construct that means they've never programmed, lol.

    I think he's displayed that he has a general enough knowledge of what he's doing, but he probably needs a code example and that's part of the issue with a lot of the help section answers. We need to be giving code examples rather than just answers of a general direction... everyone doesn't learn just by general direction, and typically a true visual example is universally more apt to be understood.

    All of that said, if he says he's already figured out a different way to do what he wants, I don't see the point of even extending this. The Timeline Object works fine, and it may work fine for what he wants to do. Without knowing all the specifics of his game you're just assuming what works best for it without actually knowing. He may not even need to use global variables for what he's trying to do, perhaps he only needs an operation for one layout. Only he would know that.

  • First, this doesn't have really anything to do with whether Guyon is an expert not. He is a newer user of Construct so you could argue that he is not an expert (no offense to Guyon meant .

    I didn't isolate that to him being an expert. I said that he already knows how to use the product and maybe at an expert level, and that would be a big reason as to why a person in that position wouldn't be much concerned with comprehensive tutorials and instructional manuals. If you already know what you're doing, you're not spending any time reading manuals.

    Were we not all new users at some point? How do you think Guyon gained his experience? He didn't magically become skilled in Construct one day and he didn't have somebody spoon feed him all the answers. He worked hard and learned how to use the tool.

    Yes that's true, but that's beside the main issue which is lack of adequate instruction. It's very commonplace for these types of things to have better documentation that can be learned from. The reason for it is pretty simple really... it's just not the best idea to expect people to learn without it. A lot of the people here are smarter than they probably give themselves credit for, so they think the rest of the world can do what they're doing with the same amount of effort (I get that feeling from your post as well), but this is not the case at all. I know some very good C coders who gave up on MMF back when it didn't have much instruction, and have given up on trying to learn Construct for the same reason. So if those types would give up, consider what the average guy who has never made a game or has no programming discipline is doing when they keep getting stuck in Construct?

    I code in C myself, mostly C# (had to learn through instruction), but in order for me to learn MMF I had to get the resources and learn through instruction just the same. For MMF, they have books and guides and instructions. I read those, I learned it. Then when moving over to Construct the concept of event-based programming made sense because of what I learned in MMF. But, if it weren't for what I learned through MMF's instructional resources, it would've taken me much, MUCH longer to pick it up and I probably wouldn't have started using Construct.

    The best resource for Construct right now is the search function, but that at the same time also creates its own issue as I've pointed out in a previous post.

    We can all agree that more wiki documentation needs to be added. Since you seem passionate about this issue why don't you then do something to contribute? It doesn't take an expert to document a behavior, object, expression, etc.. I have added several wiki entries (within two months of finding Construct) not even knowing about how the object worked before I first started out. I scanned the forums, tried out examples, built my own examples, and just experimented with what the object did. I then added an entry and even examples in many cases. You and any other Construct user could do the same.

    I may do that, but it's not my responsibility or any end-users responsibility. It's something done out of courtesy from an end-user standpoint. But if the same courtesy approach is used from those who actually create a program, it really doesn't make logical sense. Good instructions are just something that should never be looked at as totally separate from the development of development software, because you can never assume how well end-users will pick up on what you've developed. The search engine gives definitive evidence of there being an issue, so I don't think there is a valid argument against that. At this point you have 2 situations:

    1. Wait for end users to contribute enough to the resource to eliminate the issue, which is more than likely not going to happen.

    Or

    2. Write it as a first-party, which would be the most realistic step to eliminating the issue.

    This is a simple situation. How many of us here can fix bugs and write improvements to Construct code? Maybe 1-3. Ok, how many can add wiki entries? At least 100 - 200 users. So why have one of the only guys who can improve Construct waste time writing documentation when a couple hundred would do that job instead?

    We'll have to agree to disagree that it would be "wasting time". Having comprehensive instructions--something that should never be separate from the scheme anyway--is never a waste of time. It's necessary, very much so. And it can be tackled by the same people if it's broken up in pieces and done over a long period time. No one is asking for them to drop everything and just write a manual only, as there are certainly more rational ways to go about it. I should also point out that the people who created MMF also wrote books and resources on it in their spare time. It just comes down to how a developer views this aspect and how they're willing to allot time to tackling it. If a developer chooses not to make instruction an important priority, that doesn't in any way negate the importance or necessity of instruction... it just means the developer hasn't made instruction an imporant priority... nothing more than that, really.

    Construct is not C#. C# is a language standard that has had millions of dollars and hundreds of people working behind. You can't even begin to compare Construct (a free and community project) to a large language. We could hire a technical writer and

    solve all the documentation problems if that were so.

    See my previous response.

  • I would much rather see his continued efforts on C2

    Of course, because you already know how to use the program and probably quite well if not expertly. New users would (and should) have a different view.

    All that's going to happen is that you're going to keep putting out new versions that new users don't have adequate documentation to learn from, thus continuing to ask the same questions over and over again making the forum even more filled with repeated questions, only adding to the issue of searching through a huge archive that keep getting larger.

  • And to continue, I think the first step in addition to updating the wiki with more fleshed out info, would be to break down the basics of the basic types of games one would make with Construct and then start writing visual step-by-steps. This was done with XNA and I'd say that it helped the community for that program grow exponentially from where it would be without it.

    I'd start by doing something like break it all down into innovations and basics. The Innovations category would take things like faking Mode 7 and other 3D concepts, and posting a step-by-step with an included cap.

    The basics category would be the same with such things as step-by-steps for creating hit-damage, collision, frame animation setup and control, time-keeping, score-keeping and the marriage of all of those things. And along with all of that, take all of the available operations and write example code for each and explain what each does and why and when to use them... without the "why and when" it's not a tutorial that a new user can learn from.

  • I'd say that this single thing is needed more than Construct hitting 1, 2 and beyond. It's more important than any thing being done with this great program right now.

    While I find many of the things understandable, I think the amount of questions that get asked makes it evident that the lack of documentation is an issue in need of great attention. In reality, without making it a major focus, you'll only be bringing along the people who already know how to use it, which is no good if new people can't understand even the basics without asking a question on the forum, or searching through the forum... the forum itself is a great resource, but it's become too large for the average new user to wade through to find something minor or specific to a function of the most basic part of the typical platform game. I seen questions for things that you would find in most 2D games, those things should have clear and easy-to-follow step-by-step instructions.

    I couldn't imagine learning some of the programs I've used over the years let alone learning C# as I have without comprehensive step-by-step guides. I mean, even if you do manage to learn without, you'll always be missing some part of the discipline and it sets you up for future issues.

    I think more wiki additions would be nice provided that there are code examples that tell the user how it should look in order for it to run. I see that when people give advice on the forum, a lot of it is explaining what you need to do but not exactly how to do it, that's why you see a lot of follow-up questions asking "Yeah, but how do I write it?", and then when there is no response to it it's all a waste and no one benefits; the asker doesn't benefit because he still has no answer, and the new user searching doesn't benefit because when he gets to the thread to find the answer there is still no answer because the responder never gave example code. So now the new user has to make a new thread asking a question thats already been asked, this time hoping he'll get an actual example code of how it should look.

  • As I mentioned, holding alt and clicking crop will crop all of the frames in an animation.

    Yes, I know.

    The point is that all of it (including batch resize and batch export) should just be a selectable option in the image editor without work-arounds or using events. It should be as simple and readily available as the batch import functions are.

  • ^^^ In response to the previous two ideas, there seems to always be a way to make something work in this program, it's very versatile. But personally, I would rather something like this be made an option in the image editor rather than using events to get the result. You should be able to highlight all of your frames in the image editor and then select what command you want to apply to those frames, click the command, and then done.

    Importing images in a batch in Construct is as simple as clicking 'import frames' and then highlighting all the frames you want, then clicking 'open'. It should be just that simple to batch export, resize, and crop, as well. We're already 50% of the way there since the importing function already allows for batch commands.

  • Most of my stuff is drawn in another program, but I can see the need for batch processing because of the need for accurate animation testing of your changes. Having to switch back and forth between programs is a bit annoying.

    The more you can do all in one program, the better.

  • Haha, I like the title of this thread

    It sounds to me like a question about allowing the end user to put in his own custom soundtracks that trigger when an event happens. The best example I can think of is with a football game called NCAA Football 09 for XBOX 360. The way it worked was, the game gave you a series of events like:

    1st down

    3rd down

    Sacks

    Interceptions

    Touchdowns

    And then the end user would select one of the above events, then it would open their XBOX 360 hard drive where the end user could search for a music track that they wanted to play for the specified event. Lots of people have music on their hard drives (I think just about every PC comes with pre-loaded classical music), and of course there is free music all over the place on the net, so the end user finding material to use wouldn't be a problem at all. Thing is for the program creator, how the heck do you do it in Construct? That I don't know. It's not something I've ever considered doing with the program, I'd bet there is a way it can be done though.

    If someone knows, this is one of those things that should be wiki'ed, because custom soundtracks is a great feature.

  • Construct is a great program. It should be taught in Universities. The fact that it's free and works as well as it does should make it a no-brainer because schools wouldn't have to pay any licensing fees. Also, Construct Is deep enough and allows for so many types of games, that there appears to be few real limits except for the imaginations of the people who create with it.

    I've been writing in C# since the beginning and as much as I enjoy it, the bottom line is with the way technologies have advanced there is simply no need for 2D game creation to be so difficult and require so much writing just to get something to move a pixel to the right or left. That's why Construct is so valuable, it cuts through all of that, and it makes prototyping a breeze which can't be said for other programs or writing in C. If Construct had an XNA runtime, I wouldn't even touch the XNA framework again... I'd be building my games in Construct, exclusively.

  • I've had the OP's issue, except mine was with graphic images.

    For example, I had a set of 30 animation frames that were 800kb's each. It took forever to preview. But when I cut the images down to between 130kb's to 200kb's, the preview loaded much quicker. Of course, I realize that smaller images load faster than larger ones.

    However, I know it's not my computer that caused the slow load times prior to the cutdown. My CPU is not the best, but at 2 and a half Ghz, and 3 GB's of RAM, it should handle everything in Construct fine. But there are quite a few hiccups with loading and saving even smaller images and layouts. When I've used the same images in Visual Studio/XNA, they loaded quickly at even large memory sizes.

  • After having done a pretty good test (posted here) and then reading the issues of others, it's quite clear that there is a problem. It's not simply an issue of the way people are programming (an accusation I've seen while searching on this issue).

    When I ran the test, I used the exact same commands with both the keyboard object and the 360 controller object, starting with the controller object. The keyboard object responded to all of them correctly, the 360 controller object responded to only a few and even switched operations around to other parts of the controller. I'd tell the left thumb stick to command a jump, and it would make the right thumbstick do it... screwy stuff like that. Certain buttons just didn't operate no matter what I did.

    Personally, my 360 controller works fine on every game I have. Never had a single problem. It also works fine for me in XNA (I program in C# sometimes) and MMF, but in Construct it's a no-go.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • If you run twice is it just as slow the second time? The second time it should cache the images which are the slowest part, and run a lot quicker.

    Yep. In fact, no matter how many times I save or run the same layout even without making changes to it, I get the exact same lag.

    Also, when I run debug layout and attempt to to restart the layout in the debug commands, it does the same thing. Pretty much everything lags except opening the program or running a layout with no objects in it.

  • I'm on a mobile device right now, but will try to post later.

    What I can say at the moment though, is that I have one layout with a 30 frame animation that totals about 800 kb's per frame. That one loads the Run layout function slow and saves slow.

    I've also had a frame where there was just one picture totaling about 2 MB's, and that one did the same.

  • I'm getting a lot of lag when saving, and when selecting "Run layout" from the software.

    When I save, it takes a long time for the process to finish (like 2 minutes or a little more on average) and the windows message windows always say "not responding" while it's running the process. And when I select to Run a layout it basically does the same thing.

    I just got another CPU 2 weeks ago so now I'm running Vista with 1.34 GB's of vram, 3 GB's of memory, AMD Athlon Dual core 2.3 GHz.

    Is there anything I can do to get better performance?