Hi.. I just want to Share/Update my progress..
I did solve lots of problems.. But I still didn't get into a state where it is easy to make a game with My Editor.
It must be done the right way, or it will be just another Editor.
I have a new brilliant Idea, and I want things to be as visible as possible.
though honestly, it's rather frustrating that I still didn't get where I wanted to by now. but that's how it is when you invent things as you go
And it's important to keep Trying..
More updates Coming Soon..
Also, It is better to wait in patient then to release something that is not exactly what you think ok
One More Step..
Just a little tiny demo i did with my editor.. dont expect too much.. I just wanted to release something..
and show some basic movements..
I'm planning to make a demo for various types of games, all in one Demo, but still lots of things to solve..
Like for example Collision Issues..
(https://i.imgur.com/f8nvTit.jpg)
Download: https://drive.google.com/open?id=1vwkCYVfyIYQ4wCY1Buvv40VqvDFtSK7Y
I'll take a look at it when I have a better internet connection. :)
It takes like 10 minutes to make what i did with this demo using the editor. (when you know what you doing it take 3 minutes)
I just spew this one out to take it out of my system,
The next thing will be far more impressive..
This one has like 2 out of 10 impression rate in my level scale.
now, back to the frustrating work of improving.
there is a lot of small detail work that improve the experience of using it
which only when all the details are coming all together, only than it can be valued for its true value.
whenever i improve something i instantly notice it. (Obviously if it works well)
Just gave it a spin, works well on my end. :)
Yea. it works, that's an important info.. Good enough for now
My old W7 box doesn't seem to have this at the mo:
got it from here
https://www.dll-files.com/download/20c0afa9589cfea14ea0ef8cba67f93b/msvcp110.dll.html?c=N0t4bTVpelNTR21mbWQ0SVNhd3U3QT09
needed this one to:
https://www.dll-files.com/download/4474fab985172f226d3308ade4c825cc/msvcr110.dll.html?c=N0t4bTVpelNTR21mbWQ0SVNhd3U3QT09
i know this issue with those dll's
i download them manually and put them either on the xors folder on in system32 folder
Hardcoal, I think you could just include them in the root of your executable folder and it would work for all and sundry. Good work by the way ;)
I thought i have, Guess I was wrong :p
Update Progress..
Ok i've made ReloadMap work..
I know it means nothing to people here , but its is another step forward for me
More Improvements have been Added..
??? ::) I think ive said it all
Ok, After a long break, I continue working on this project..
Hope to show some results soon..
good work :)
Thanks..
Ok, now that Ive passed this memory issue I had. I can go back and work on the important stuff..
now i have a question..
In unreal when you run a map and than press stop, the map loads pretty fast..
can anyone know or can explain how does exactly they do it?
how can they go to previous condition without loading all the 3d objects again.
I have my own ideas but i want to hear other opinions
Things are going not bad with the editor..
I have problem with the Blueprint which is that you cant zoom out on it, so its hard to work when everything is so big..
I yet to find a solution for this. and its pretty important.
other than that Im rather pleased with what i achieved so far, it start showing signs of a future.
If ill keep going on.
Maybe, just maybe, soon , more people will show interest and join.
The whole deal that this is based on xors3d is irrelevant.. engines can always be replaced..
The deal is, is with the mechanism, and if people like it or support it.
im working very hard to simplify it, day by day.
But im still not quite certain about all this.
This blueprint concept is not better than programming
(https://i.imgur.com/1ml2Co4l.jpg)
For programmers (or even wannabes like me) it's probably not better. But I am sure there are a lot of people who would prefer it over pure code.
I had a similar discussion with a friend about visual shader editors. I always say they look more confusing than the pure code would and actually make the work harder and more obfuscated.
But he (as a non-programmer) basically said "yes, maybe, but the chopped up and layed out nature of it helps me to visualize what I am doing and how everything strings together step by step. I wouldn't be able to follow the pure code at all."
He does have a point. Especially with all the shortcuts, tricks and abbreviations often used in code.
I guess it really depends on who your target audience is. Some people will love and happily use it.
I like how your implementation looks. Nice and clean.
I didn't know how hard blueprint will be when i began.
Im actually upset with it and with the time i invested on it
I could have already made 10 times based on code.
Im kinda stuck. And thinking what or if to do.
It all looks nice and working, but its inconvenient compared to programing..
Im absolutely thinking moving on to code.
Or mixing them both.
Thanks for the good words.
imo, if you would have let go your visual game editor idea a few years ago, and focus on coding some templates, procedures, systems, that you can reuse in different games, you would have already finished a few games (if your goal is really to make a game someday) just saying... ;)
beside, for such visual editor, unity and unreal already exist (and more advanced)
Remi, i could have made 1000 games in the time i invested.. And i does makes me sad.
But on the other side..
I do enjoy the fact i have my own editor.
My editor if it goes well will have stuff no one have because i decide what to do.
My main goal was to make a game, but there was no game editor back than, and i didn't know even how to save a map.
When i started with blitz3d i was in shock there is no save map..
I knew nothing..
I like complex code and see where it gets me. (hopefully not an asylum)
But thanks for your wise opinion i totally see how pointless this may seem to others
surprise us, just finish 1 simple game ;)
( i have the same "virus" of always wanting to add better functionalities/graphics/effects, so i know what you are going through, try to control the "beast", try to keep it simple, and make it happen )
Quote
surprise us, just finish 1 simple game ;)
Fine. But some people would find a simple game very boring to complete, and therefore want to achieve something more interesting.
Remi D
I can make a Game in a Second without graphics, But I prefer concentrating on Development rather than proving myself at this point.
I already made little test games with it btw.
ATM this editor is not their yet. hopefully soon.
If not. I will give it up
this was a benevolent suggestion, (that i already told you a few years ago...), maybe try to make something else, from scratch, something simple, it could be refreshing, and you can create "graphics" using 2D or 3D primitives, look at what stevie g manages to create... ( https://stevieg.itch.io/gunstar?secret=qpSAuh6TnXfdOw94mJ0e6czB9k ) (with the old deprecated Blitz3d, as you say...)
Youre right Remi D and I value your kind suggestion. Great Idea!
I started learning unreal and unity and tested other engines..
For now I conciser Unity as the best option.
It has this days exactly what I targeted when I wanted a perfect editor.
Also, I need a general rest. not moving to another project.
I love direct coding. I admit.
Its very hard for me, to just make a game.
I always tend to get attracted to complex things.
it gives me excitement.
I dont seek to prove myself..
you will have to take my word when i say, i can make awesome games when the time comes.
if you prefer to create all in code, and since you seem to have good experience with Blitzmax, i suggest to keep using it.
again, stevie g created several games with 3d models composed of 3d primitives...
another example, using only primitives (3d shooter prototype, not finished) : https://www.syntaxbomb.com/index.php/topic,4975.0.html post#5
or it could be in 2D, as you prefer... my point was that there is no need for perfection to create a fun playable game... and by trying another way to make a game, you will probably learn a few things. (and maybe realize that a visual editor will never allow to create all kinds of gameplay...)
Sorry i forgot to check the games you posted.
I will check them now or soon.
Thanks for the refresh, it really helps..
Ok now, i want to say something about my editor.
I've reached the point of trying to implement a code.
That's a very important and even final step.
Im gonna make a button on my editor that will open the game code and i want when i press it that it will load let's say blide but with the code already loaded.
How do i do a command line that does that?
Opening blide with a specific file i like?
Im not gonna endlessly break my head with this editor, i promise
check out pub.mod/freeprocess.mod
https://blitzmax.org/docs/en/api/pub/pub.freeprocess/
bye
Ron
Quote from: Hardcoal on November 07, 2019, 00:34:17I didn't know how hard blueprint will be when i began.
Coding a blueprint / visual scripting system is no easy task at all. It's not too far removed from coding your own language but with less syntax checking. On top of that you have the mammoth task of the visual side and then further decide if the system is going to be interpreted on the fly or create source code. I admire you for diving into doing such a system.
When I did my GUI & designer thing for AGK I toyed with the idea of a blueprint system for the logic side but quickly decided it was much better ( and quicker to work with ) to have the GUI output raw source code as then it's just a case of adding code to the potential events ( button clicks, scroll bar movement etc etc ).
Also a blueprint / visual scripting system is very slow going for coders as simple tasks take ages to draw up and doing any big logic becomes a huge mind bending visual puzzle. I'd say if that's all you're used to then it'd be fine but us mad coders can skim read over source code and figure out what it's all about much quicker than looking at boxes and connecting lines. It's also much quicker to write the source than visually create it.
Having said that I do think that a blueprint / visual scripting system has it's place. For example a 3D modeller wanting to see their model rotating / moving in a scene would find it easier to visually do it than in code. Granted, slower than coding it but easier to do visually.
Thanks Derron! Great Example.
but i still dont understand how can i make Tprocess open lets say Test.bmx file with Blide.
I can open it with OpenURL but it wont be a process.. hmmm.
my main goal is to make my editor reach an Element Code, Edit it and Go back to Editor.
Qube.
My Blueprint is very fast.. there is no compile process if i dont code..
its just a method i developed to make it run without compile..
its almost as fast as normal code.
anyway.. now im going to implement the programming side which hopefully will make it perfect.
I would really love to hear more advises from people about how i should approach matter..
or any other ideas..
one more thing i have to explain which i used to mention in the past.
I have serious issues with screens..
it may sound like a joke, but when i stare on the screen my mind almost goes blank.
its very hard for me to think clearly in front of monitors..
it reduces my ability to something like 20% of what im really capable. (an assumption)
i dont like to talk about this too much. so i only mention it now.
im waiting for an oled screen or i thought maybe a screen with higher refresh rate will help.
atm im suffering while im programming, and many times take a head ache pill.
its amazing how easy it is for me when i dont stare on a monitor to think, compared to staring on a screen.
i use black themes for years when i program. it helps alot.. but its not perfect.
any white or bright color i try to turn off..
I recommend to buy on VA panel monitors..
I remember the stupidity when everyone was excited from IPS panels.. and I knew they are crap.
i didnt understand how come no one talks about VA panels when I already had only VA monitors.
the thing about VA panel is that you Get deep blacks.. and almost no color bleed..
when i look at a screen, i feel like it radiates into my brain. and i can feel its flickers.
its not like looking on a true image. we are staring onto a florescent light.
the blue light thing or glasses ive tried them all..
yea they all help abit.. but there is fundamental problem with screen..
and its about time it will be resolved for the sake of us all healthiness ..
we cant avoid screens today.. so we must make them better
ok, now im gonna check stevie g games ;)
multiple ways to do so:
- interprocess communication and interaction (DDE, OLE, Server-Client, ...) dunno if blide offers such a thing
- or simply create a temporary file, open that with blide ... and meanwhile observe changes to the file timestamp to react if someone saves in the editor/blide, if the started blide process closes, remove the temporary file
bye
Ron
as Yasha once said : most of your programming (logic) should be done in your mind or on paper/text editor (with words)
That's half right.
But debugging requires experimenting.
It's still no excuse for monitors to burn out our brain cells.
@RemiD
Writing problems out on paper before, during and after programming is what you get taught when in academia. It's a much better process and can help you be more productive during the design steps. Doing it during and after sitting at the machine is a part of documenting what you are doing.
@Hardcoal
I agree that in this 'modern' age there's really no excuse for super bright, retina destroying, screens :o
BTW I think you're editor is a great alternative. On the contrary of keeping things straight forward, I'd love to come up with a language that's 1 step up from assembly - or even a BASIC variant of x64 assembly - still working on this idea and in reality it may never happen :)
well i always code a procedure in words before coding it in the chosen language, and i have managed to make some nice/complex things, so this approach must work somehow.
@Hardcoal>>concerning your eyes problems, have you ever tried to wear this :
(few tiny holes, you can see but it is disturbing the sight)
(https://upload.wikimedia.org/wikipedia/commons/thumb/d/d0/Rasterbrille.jpg/333px-Rasterbrille.jpg)
or this : (more bigger holes, you can see very well and not disturb the sight)
(https://images-na.ssl-images-amazon.com/images/I/61-EfGlW5EL._SX466_.jpg)
(i wear these when doing manual works outdoor, for protection, quite good)
Interesting glasses ill check it..
I know this monitor problem is solvable.
The question is when, and how...
There are new technologies already of paper like colored monitors, but the companies don't rush to release them from their own stupid reasons, or who knows why.. (excuses or evil.. I don't know)
Until than ill have to manage.
And generally im ok, as long as i don't wake up and instantly jump on the computer.
My phone has oled screen. You can feel the difference.
Yet, I did had an oled tv 2 years ago, but it wasn't good for my eyes.
I now use 58 inch va panel tv and so far, thats the best i can get.
I also have a head scratching problem when im sitting on the computer which indicates my body really struggles with something.
It comes and goes in synchronization with my computer use.
About writing down on paper before coding..
I always think and plan before i code, but..
If you take me back in time, i wouldn't be building a game editor but an ide.
We are not supposed to use paper to plan in general.
We need our ide to be also our planing sheet.
A good modern ide should begin with a flowchart that you put remarks and even sketches with mouse stuff.
And slowly you replace your sketches with code..
That's how i would do that.
I would also support the ability to work on a tablet that synch with your computer.
So you sit in bed designing your stuff and transmitting it to your computer..
If one day ill have a budget ill fulfill lots of what i come up with.
One more example of my approach..
I want to make let's say a spaceship auto creator
That means an app that in one press creates random spaceship in 3d which than can be modified as you please.
Than you can export it even as a png animation.
So it's good for 2d and 3d..
Same goes for running characters.
This allows people with no graphical skills and saves lots of time as well.
On final word, i want to recommend on my favorite daw software which i lately found.
It's called Mixcraft 8.1. An almost perfect music editor to my opinion.
Ok, that was my longest post ever, and im just making it longer by writing this
Ok I started to work on Game Coding in my editor, and it goes well..
Just wanted to update.
Soon ill be able to make a pacman
Ok, Game Coding on Editor, Faze one. Operational.
Faze 2. More thinking needed..
I did it quite smartly.. soon it will be clear why.
Take a look at my amazing Site! http://www.hcgameeditor.dx.am/
I put lots of hard work to it for it to be looking that great
Its shining mate!
That must of taken you, phewwww, hard to guess really, but if I had to guess, I reckon at least a good couple of minutes! ;)
Dabz
Ok Seriously.. it will start looking better, that was a joke ;)
Ok, I solved more problems after a long week.
there is a faze in developing that i dont like so much, its when you are not sure what you gonna do. so you make and destroy code, etc.. its no fun. I like to know what im gonna do. thats why it takes more time. This is a big waste of time comparing to when you know what youre gonna do.
Anyway.. thats enough for me for now. Im tired as it is from all this project.
Ill probably now try to work on the Gif thing ;)
More Progress..
I do not post unless I really made a major Improvement.
By posting, I feel like a landmark.
So its sort of a work log for me, because I know I only post when a Faze is over
I came to the conclusion it will be very helpful if i simply learn other engines like Unity.
so I can reflect on my own editor.
I should have done it a long time ago.
Genearlly.. I did a lot of similar things without even knowing it similar, but i could have being in a better place today, If i looked at it earlier
Just a Snapshot
The orange line you see in the picture is a physical collision that can be modified according to the shape you desire
(https://i.imgur.com/RwwlA0L.jpg)
Oo, that looks very swishy, great work. You should make a YouTube video of it in action :)
Great stuff! :)
Thanks guys
Its too early to make a video..
There is still a pretty much work to do
I did so much and still so much to be done..
But I see some rays of hopes.
Quote from: Hardcoal on December 30, 2019, 03:24:47
Thanks guys
Its too early to make a video..
Sometimes it feels like you're coding forever and not making much progress and then other times it all just comes together. Keep plodding on as it looks like you've done heaps of work so far 8)
I want to say that along with trying to develop my current Editor..
Im also experimenting with a side editor. with another concept..
Just trying to see a different approach to the one I have taken.
The approach Ive taken with my current Editor, I would not take this day.. Relying on Blueprint..
It is very time consuming.. (But will see..)
Besides.. it so fun making a new start. when everything flows so fast.. You Code fast..
Since its just a prototype
You should always work parallel Just in order to make sure youre on the right Track
Very impressive looking UI, intriguing work - Happy New Year!
very clean and nice looking UI..nice work..
Thanks Alien.. I will keep improving it :)
Things are going rather well with the Editor. although Ive taken a Break.. for some reasons..
The thing is.. Even though it goes well, I find little energy to continue ATM..
Im also Experimenting on another Editor I started, So it takes some of my Energies..
I just managed to make that when I load my Editor it will only load one map. and each time i press another map that is not loaded it will load it only than.
This saves the loading time of the whole game in one go.
Now I need to work on Fast PlayMap to ReplayMap System.. Shouldn't be too hard..
Speed Is essential For a comfortable Game Editing.. You cant make Discounts about it.
Sometimes when I do stuff. Im kinda amazed it actually works..
I cant believe something so complex as this game editor.. can Handle the whole complexity of itself.
Well. Up till now, when I Did Play Map, And Stop Map, There was a slow process of loading the map again.
I found out a way.. to turn 10 seconds of loading to 0.1 second of loading.
Even in unity they dont have that, and Thats funny. such a big company cant think of a way to make Reload map faster..
The best solution for this in my opinion (which i didn't fully applied yet) is To clone each action you do to an identical map that is in the memory.
Then After stop, you just switch the played map with the cloned map.
This way you get instant Play and Stop..
I think In unreal that might be what they done.. because there is no loading time their. But im only assuming.
Im doing many things to make processes faster..
For example, when I load a game to the editor. I only load one map to memory..
The Rest of the maps are only loaded when you try to switch to them.
Saving Editor Loading Time.
That cloning approach has drawbacks:
- events need to be re-registered (as new object instances replace existing ones)
- old objects need to get properly cleaned up (unregister from events, unregister from globals, ...)
- you will clean a lot of stuff from memory after doing so (mem cleanup)
- while cloning you duplicate required memory (not a biggy for small projects - but think bigger, more ressources, ...)
I do such a think in my game for "savegames" - and it is prone to lead to segfaults/errors/... once you need to tackle memory leaks :)
Memory leaks? Yes ... if you forget to unregister an object properly, then it might be hold by some entity - and that other object holds another object (which should be removed too) - at the end you might have a chain of objects leading to "old map" not being removed at all and staying in memory.
To check this out: keep up the process explorer / task manager / "top" (on linux) and maybe even print out the amount of GCMemAllocated(). Brucey also added some stuff to BCC/BMK/BRL so you could build your individual BMK/BCC which counts alive objects. That way I managed to track down a lot of mem leaks happening because of forgotten properly done cleanups of certain objects.
Once you added display of GCMemAlloced() you should reload your map ... reload your map ... reload your map (and start "play" each time). If memory increases there, then more and more "managed" (so "cleanable") objects are alive. If that value stays pretty much the same - but Task Manager / top /... show an memory increase, then you are possibly leaking memory.
I was not aware of this earlier - as on Linux the memory increase was lower than on Windows - but I leaked 11 MB per Savegame load. 11 MB of "objects", not graphics... that was pretty much. On Windows with "legacy BlitzMax" and Threads it got even worse - a user reported the memory usage climbing from 200MB to over 2GB after some savegame deserialisation. But even after fixing stuff for NG it still happened in legacy - and only with "threads". So the legacy GC has some trouble with whatever I am doing (NG's GC can even tackle circular references - which I should not have that much as I avoid them like hell).
Hope it does not concern you - but better check that out (memory leaking) and also make sure that the core function of your editor works: running the compiler etc.
bye
Ron
I was testing memory increase.. all clean..
I had no errors what so ever..
if you program your stuff correctly it will work correctly..
Im not their yet but I think I got it under control.
> if you program your stuff correctly it will work correctly..
In my case it was hand made positional 3d audio which hold reference to an entity which else would clean up. That entity was kept - and with it some references it hold - which referenced other stuff ...
So at the end some audio stuff was keeping my game code from cleaning up properly.
Just want to say that as if the audio module was a "third party" thing you might have considered stuff in there to be "working no matter what I do".
Also "decoupling" (eg. through "event systems") can create issues (assume you send out an event with an object as parameter - and the parameter is stored by some listener as ":object", it is hard to find the reference then when cleaning up). Pay attention when coding such stuff.
Maybe a reason why some prefer "IDs" over "references" (albeit this might lead to other issues then ...).
bye
Ron
Thanks for the Tips Derron.
I dont use external Audio module Atm
What Ive done to check for undeleted types is a counter for each type..
Every time i create a new type I add 1 and when i delete one I subtract 1.
So I can compare the results when reloading a Map in my Case
also in my editor there is the memory display. so I can see if the memory keeps being stable.
I also run test where i create and delete object constantly to see if there is a memory accumulation
Cheers
Im really working hard to make the use of this Editor as convenient as possible (to me at least)
Sometimes I get stuck on boring stuff.. and it seems there is no progress externally.. But this Fazes must be face.
Working on cosmetics is the easiest part..
But its worth nothing, if the Editor is not what you want it to be
Although im not touching this editor for a while.. its not a lost project..
I decided change of strategy make it a 2D game editor only..
This simplify things.. way way more
I should have the brains to do so from the beginning.. than i would sure had a functional game editor years ago :(
It will have 3d objects.. but the Editing will be 2D only with depth only ability of movement..
that was my initial plan before my Ego made me divert to full 3D which made me waste years of time for something wrong!
what a shame..
Yet the Editor is like 90% ready.. so why waste it
since i work very organized .. i have no problem getting back to any project no matter how many years have passed and get back to the code almost instantly
Back for working..
Its Taking Shape.
Lots of details to make it work properly