182 replies to this topic

[Shadowrend]

Very well my red named friend. I will stop bothering you with GM critiques.

PS: You may remove this text and write "[CUT - Feature request]", or whatever you find more pleasing.

[Desert Dog]

Hi Mike, we at the GMC are a roudy&cranky lot, who are very demanding.
I've started a topic in the community forum:
http://gmc.yoyogames.com/index.php?showtopic=502491&pid=3716666&st=0&#entry3716666

I'm hoping that people could post their bug requests there, and we can work on the legit claims, and get a working example/demonstation up for you, which we'll then post here. We'll try avoid sending grumblings, feature requests, and/or vague bug reports.

I hope this idea suits you, if it doesn't, feel free to remove this post, and that topic I just started.

Thanks for your time mate,
~DD

Edited by Desert Dog, 14 March 2011 - 09:18 PM.

[Jakyl11]

dont know if this has been brought up but in windows7 when loading a game with no sound or with sounds, game maker stalls at loading sounds for like 30 seconds or more sometimes its quick sometimes its really slow and it gets rather annoying when you just want to test a small bit of code.

[Yal]

One bug I can remember:

If the image_xscale of an instance is zero during any drawing step, the game will crash. In previous versions of Game Maker, the game would flow on as normal, but the particular instance would be impossible to collide with and would of course not be visible on the screen. I suspect this bug has to do with rewriting drawing routines to make Game Maker draw PNGs instead of BMPs, or perhaps with the new collision mask system.

While it's generally a bad idea to let instances have an image_xscale of zero, having the game crash when you do doesn't feel like the best way of encouraging people to learn better coding habits.

I assume the bug also appears when image_yscale is zero, and when you use a draw_sprite_ext(), draw_sprite_part_ext() or draw_sprite_general() function call with one of the scale arguments being zero, but I haven't tested these.

[gnysek]

About drawing lines which was mentioned some posts before - nVidia cards draws lines, rectangles etc. in other way that some ATIs (ATI draws +1 pixel on x & y) - I reported it as bug in GM7 or GM8 long time ago, but Mark Overmars said that he didn't change it.

[Yal]

@Yal: Tried... didn't crash. If you can provide a simple GMK, I'll try and look at it.

It might be Lite-only. I haven't tried it since I got Pro, but some minor bugs in the Game Maker editors has stopped happening since - for instance the previously mentioned.
Never mind, it seems like I've been wrong. It was in the 21th post of the topic, and the reason the freeze happens is not what I thought.

As for the bug in this post, I might be able to shed some light. It's a lot more probable to have the Sprite Editor crash if you close it down quickly after entering it (either from closing the Image Editor, or by opening the Sprite Editor from the Sprite Form). It also crashes a lot more often in Lite than in Pro, while I had Lite it would crash on a regular basis, but since I've gotten Pro it has only crashed once.

Edited by Yal, 15 March 2011 - 11:00 AM.

[Davve]

When you scroll up/down really fast and then hit the top/bottom, the code window freezes for a second. The longer the script is, the longer is the freeze (about a second or two on 500 line scripts). Not sure if this is a problem to anyone else, but it can get really annoying.

[IceMetalPunk]

When you scroll up/down really fast and then hit the top/bottom, the code window freezes for a second. The longer the script is, the longer is the freeze (about a second or two on 500 line scripts). Not sure if this is a problem to anyone else, but it can get really annoying.

That might be either the code highlighting or the syntax checking working to catch up to the part of the code shown in the window as your scroll. Try disabling syntax checking, then try disabling code highlighting, then try disabling both. See if any of that makes a difference.

-IMP

[slayer 64]

i just got a sweet glitch today. i guess we can't draw models while making a primitive =(
if i comment out the "d3d_model_draw(oPlayer.mPlayer,ax,ay,az,-1)" then i can see the primitive lines in the world. if i draw the model i can't. why does it matter? the only reason i'm drawing this way is because i'm debugging. normally i would put all the lines in a model and display that.

d3d_primitive_begin(pr_linelist)
    
    for(i=0;i<ds_list_size(listNodes);i+=1)
    {
        node=ds_list_find_value(listNodes,i)
        ax=ds_list_find_value(node,X)*mapScalex+ mapScalex/2
        ay=ds_list_find_value(node,Y)*mapScaley+ mapScaley/2
        az=ds_list_find_value(node,Z)*mapScalez+ mapScalez/2
        neighbors=ds_list_find_value(node,N)
        
        d3d_model_draw(oPlayer.mPlayer,ax,ay,az,-1)

        for(j=0;j<ds_list_size(neighbors);j+=1)
        {
            neighbor=ds_list_find_value(neighbors,j)
            bx=ds_list_find_value(neighbor,X)*mapScalex+ mapScalex/2
            by=ds_list_find_value(neighbor,Y)*mapScaley+ mapScaley/2
            bz=ds_list_find_value(neighbor,Z)*mapScalez+ mapScalez/2
            
            bx=ax+(bx-ax)*.4
            by=ay+(by-ay)*.4
            bz=az+(bz-az)*.4
            
            d3d_vertex(ax,ay,az)
            d3d_vertex(bx,by,bz)
        }
    }
    
    d3d_primitive_end()

Edited by slayer 64, 15 March 2011 - 09:18 PM.

[Yourself]

why does it matter?

Because models draw primitives. Would you expect to be able to draw a triangle primitive 'inside' of a line primitive? While this may not seem obvious, it definitely seems like intended behavior to me.

[amd42]

why does it matter?

The model functions are just shortcuts for drawing groups of primitives. d3d_model_draw() probably just loops through the model data and calls the d3d_primitive functions. Very inefficient, I know, but it's probably too hard to fix at this point.

Edited by amd42, 16 March 2011 - 02:17 AM.

[Camman]

[CUT - Feature request ]

(has been addressed in the C++ runner...)

[My bad, I considered it a bug because it caused the game to perform contrary to how it would on other systems]

Edited by Camman, 16 March 2011 - 04:42 PM.

[Jakyl11]

The thing I hate the most is how game take forever to load (at least on Windows 7 (64 Bit)). When games load, they stall at around 80% (in debug mode it says something about loading sounds even though I don't have any included sometimes). It is quite annoying and eats away at time I could be spending doing other stuff.

That is my only gripe

read up and +1 me

[Smarty]

Okay - I don't like to bring forward the bug I've mentioned before, but since I've seen no dismissal or acknowledgement of its existance anywhere in the thread (including the first post) and since it has seen a decent vote count, I'm afraid I'll just have to risk being called out for pushy behavior... So Mike, would you please consider:

http://gmc.yoyogames...dpost&p=3712786

I'd at least like to know what the status of it would be, or if you need any additional information - even though there is a link to the bug report in the tracker, with example source code. It is of considerable impact on games that use bounding box collisions in mirrored sprites.

Edited by Smarty, 16 March 2011 - 12:26 PM.