Game Creation/Design Catch All

Fredrik_S wrote:

huh! Mobile shader development. That's really cool. Gonna grab and check it out.

I have no idea what I'm doing with it but it's really cool.

Heh, might be time to dig into Blender then. I have a Udemy course on Blender that apparently is getting updated to 2.8, so that might make for a good summer start to the learning.

Godzilla Blitz wrote:

Heh, might be time to dig into Blender then. I have a Udemy course on Blender that apparently is getting updated to 2.8, so that might make for a good summer start to the learning.

DO EEET!

muraii wrote:
Godzilla Blitz wrote:

Heh, might be time to dig into Blender then. I have a Udemy course on Blender that apparently is getting updated to 2.8, so that might make for a good summer start to the learning.

DO EEET!

The next couple of months are super busy, but that'll give them time to update the course to 2.8. By June I should be able to start this.

"Learn Blender" has been on my list for quite a while, so I'm excited to hear that 2.8 is a step in a positive direction.

Heh, might be time to dig into Blender then. I have a Udemy course on Blender that apparently is getting updated to 2.8, so that might make for a good summer start to the learning.

That would be cool but right now I am stoked about Probuilder that is now part of Unity.
I am going to test it out now. I hope it has some sort of CSG as I see it has the capability to merge objects. If it doesn't I am curious how it handles collision and z buffer fighting with intersecting objects and non solid object aligned with solid objects (read a C shaped object aligned with a D shaped object, is it going to cause collision problems or visual anomalies)

Slightly off-topic here but it sounds like some of you guys may have the know-how I need.

I have used Blender sparingly in the past. I've also used some other free 3D-design softwares. Ie. When it comes to basic shapes that I want, I can eventually get it done.

I do 3D Printing, now. What I want to know is how to take a 3D object and add a texture to it that will export into a file that I can 3D print ( .stl file ). I followed some tutorials a bit ago and was able to get cool wood-grain textures to render on screen, but I wasn't able to figure out how to get them to save as a 3D printable file with the texture intact.

Using free software, do you guys know how to do this? As an example - I would want to make a simple table for tabletop gaming and apply a wood texture to it that prints out and looks textured. This is an example of what I'm going for:
https://www.thingiverse.com/thing:27...
IMAGE(https://cdn.thingiverse.com/renders/e1/96/9d/2b/c2/4c9e4558d1c71ceccce54205dd254d9b_preview_featured.jpg)

Those textures above were probably sculpted by hand. I'd like to be able to do that manual texturing along with taking an image of a woodgrain or other texture and apply it to the model to generate the texture and get that exported as a printable model.

Unfortunately, the 'free' part is sort of important. This is just a fun side thing so I can't justify spending a lot of cash on the good software packages. I don't mind dropping a little $$$, but not much.

-BEP

Blender has sculpting, is free and you can probably google some wood heightmap textures that you can us a "stamp" tool to add the wood grain.

A heightmap is a gray scale image that the sculpting program can convert into depth information.

Thanks, Fang. I did use that sculpting tool at one point when I was trying to figure it out, but I think I failed to figure out how to export it as a 3D Printable File. It's been quite a while since I did it and I was trying different software packages at the time, so my memory is a little hazy.

Man, I wish I had the know-how/skill/talent to sculpt the horn on that dino, let alone the whole thing.

I'll update to the latest Blender and see what I can figure out.

-BEP

You should inquire on some 3d printing boards or with some expert. Without having tried it I dont think anyone can say that a heightmap texture will be understood by the software ingesting a model for print. There could also be printer specific peculiarities to that.

If a printer can't understand the normal map/bump map/height map/displacement map/whatever the kids are mapping these days, then the only thing that will come out to the printed model will be actual model changes, which you might be able to do in something like Zbrush, as that program works a bit more like sculpting clay.

No no no... the heightmap is used to sculpt actual detail on the mesh. (look at the video) You can't transfer texture map detail into something printable unless you have a bump map or displacement map and convert it to actual displacement. Also for 3D printing, you want the highest resolution to capture as much detail as you can and to remove faceting.

I think I found my answer. Thanks for the info, all.

-BEP

I am a genius ;P
Heh, so I have gotten reasonably far into my prototype in Unity.
I have to say that I like it but it certainly is quirky. I mean things like it has built in pathfinding but in order to do click move or have the camera follow your player, you need to write a script...
The scripting so far seems easy enough to parse but you really don't need to get too complicated with scripts so things could get hard later on. Luckily I am keeping the scope of my project very conservative. I really plan on have my levels be interactive and give you the sense of being alive with atmospheric animations. So I plan on doing 2-3 things well and then expanding on them rather than try and do 15 things that don't evolve more than one step further.
I am also going to overlay the exploration with dialogue and exposition rather than have dialogue trees or walls of text.
I am enjoying it so far. I am always too hesitant but when I buckle down, I surprise myself quite a bit.

Awesome, glad to hear it’s going well! How about some screen shots!

I hope this works. Be prepared to be completely bedazzled by the stunning cutting edge graphics!
I am bummed it doesn't show the mouse cursor or clicks but it is click to move movement and camera follow on a true 3d test level.

https://imgur.com/kvAYjh3

fangblackbone wrote:

I hope this works. Be prepared to be completely bedazzled by the stunning cutting edge graphics!
I am bummed it doesn't show the mouse cursor or clicks but it is click to move movement and camera follow on a true 3d test level.

https://imgur.com/kvAYjh3

Looks smooth and functional, nice work!

There are 2 lessons learned in my brief delve back into game design:
Even with such seemingly simple accomplishment, I feel vindicated. I feel like this is what I should be doing. I feel like it is a multidisciplinary field but also that each discipline (coding, designing, art) are each multidisciplinary. When I am doing art I feel like to be successful I have to consider the code and design. The same goes for coding and design where you have to consider the art + design and code + art respectively. When it works, it is euphoric.

Next, game development is extremely taxing. I feel like I can burst for an hour and 45 minutes or so and then have to veg or practically nap for the next 30-60 minutes. It takes at least that long for your mind to wind down if at all possible. It can become borderline obsessive and intimidating to start back up again

fangblackbone wrote:

There are 2 lessons learned in my brief delve back into game design:
Even with such seemingly simple accomplishment, I feel vindicated. I feel like this is what I should be doing. I feel like it is a multidisciplinary field but also that each discipline (coding, designing, art) are each multidisciplinary. When I am doing art I feel like to be successful I have to consider the code and design. The same goes for coding and design where you have to consider the art + design and code + art respectively. When it works, it is euphoric.

Next, game development is extremely taxing. I feel like I can burst for an hour and 45 minutes or so and then have to veg or practically nap for the next 30-60 minutes. It takes at least that long for your mind to wind down if at all possible. It can become borderline obsessive and intimidating to start back up again :)

That's why I like it so much :). I'm responsible for the whole enchilada, not just the sauce. I'm responsible for how it looks, how it looks on the plate, how all the ingredients look when placed on it and in it as well as the perfect crunchiness so it makes the right sounds during eating, along with the sauce and how it's built on the inside.

In other news, I'm about 25% into prototyping an ARPG. Had some good talks with coworkers on their current problems with the genre which actually reinforced my design. Unreal has a really nice top-down, mouse click movement, template so I'm using that. Humanoids coming to save everything from the Pyramids, Sphereoids and Cubes!

Yogurt coming to save everything from the Pyramids, Sphereoids and Cubes!

Fixed! (a Love Death and Robots reference...)

I am glad I am not the only one with the game dev itch. I have tried several times to get into Unreal and its quirks just bounce off me really hard. I wish there were modern updates to the Source engine SDK. I have a ton of experience with the Hammer (Worldcraft) editor that would have me so much farther along in even as little as 3-4 days. I think Unreal is along the lines as Hammer that the differences are more off putting. Unity is drastically different so it is easy to stop myself from saying "I wish they wouldn't have done it that way".

Anyone has any experience with Stingray (Autodesk's engine)? The last I looked it seemed way to tied into Maya... (If I am going to do it my way, then it is going to be with the tools I already have strength in. I am certainly not going to pay a subscription for something I don't prefer as awesome as it seems)

fangblackbone wrote:
Yogurt coming to save everything from the Pyramids, Sphereoids and Cubes!

Fixed! (a Love Death and Robots reference...)

Hahahahaha

fangblackbone wrote:

I am glad I am not the only one with the game dev itch.

This will be my second prototype this year I did 5 last year, nothing I wanted to publish or finish out however. This concept is interesting, so I'm running it through it's paces to see if it's something I want to finish or not.

fangblackbone wrote:

I have tried several times to get into Unreal and its quirks just bounce off me really hard. I wish there were modern updates to the Source engine SDK. I have a ton of experience with the Hammer (Worldcraft) editor that would have me so much farther along in even as little as 3-4 days. I think Unreal is along the lines as Hammer that the differences are more off putting. Unity is drastically different so it is easy to stop myself from saying "I wish they wouldn't have done it that way".

Unreal is just quick for me to set up and hit go. It has a lot of templates which set up the barebones of a concept (inputs, viewing plane, basic animations and so on). I don't know that Unreal is going to be what I use - it was more to get something on the screen and playable fast so I can see it in action outside of my head.

I still bounce off of Unity hard - I can never seem to get anything done in it, or it goes about it in a non-intuitive way (creating an invisible entity at the middle of the map to load the main code for a scene is....uhh...not the best for my brain or how I work). I've looked at Godot (which I apparently need to look at again) and GameMaker - I'm almost tempted to just do everything in Love2D but I know that will be a death knell due to how much I'd have to write in boilerplate.

Love the conversation!

As someone delving into it more as a hobby than looking to change careers, I’ve found it a bottomless bucket of fun stuff to learn.

This morning I started on this City Builder course with Zenva. It was one of the courses that came with one of those Game Design bundles on Humble or something. Looks like a quick course, but I think I’ll pick up a few things.

I was also looking at this RPG tutorial series, but decided to do something shorter first.

Gamemaker would be so much simpler but it using it completely ignores my strongest strength: 3D. With competition so fierce in the space, I can't help but think the 3D version would be lightyears more appealing. You can do atmosphere and style in 2D for sure. But with lights, depth, animations, events and particles in 3D have an order of magnitude more potential with the extra axis even if the game's perspective (top down, 3/4) is the same.

And now I have spent the last hour looking at the PBR tutorials on the Unreal website ;P
For my kickstarter I am tempted to also sell it as a game dev walkthrough to see if that will foster community involvement and add value for wider interest/support.

I started making a lot more progress once I settled on one engine and stuck with it. There are so many good game design engines that it's easy to get drawn into a grass is greener on the other side kind of scenario.

Not that I didn't spend a lot of time dabbling in different engines and playing around with stuff, but after a couple months of that I felt like it was time to pick one and run with it.

Godzilla Blitz wrote:

I started making a lot more progress once I settled on one engine and stuck with it. There are so many good game design engines that it's easy to get drawn into a grass is greener on the other side kind of scenario.

Not that I didn't spend a lot of time dabbling in different engines and playing around with stuff, but after a couple months of that I felt like it was time to pick one and run with it.

That's pretty much where I'm at with Unreal. I can handle the warts a lot better than with Unity, and it seems to be easy enough to prototype in Blueprints then convert things to C++ if I feel the need to. Godot and others are more curious, but I can't actually use them freely other than Godot since my primary dev machine at home is a Mac, and I'm not willing to shell out the money to see if GameMaker will work for me or not.

athros wrote:

and I'm not willing to shell out the money to see if GameMaker will work for me or not.

I dont have any feelings about GameMaker whatsoever, so Im genuinely curious. GameMaker seems to be on Steam, so wouldnt it fall under the normal Steam refund policy?

athros wrote:

I still bounce off of Unity hard - I can never seem to get anything done in it, or it goes about it in a non-intuitive way (creating an invisible entity at the middle of the map to load the main code for a scene is....uhh...not the best for my brain or how I work).

I totally had that issue at first, but you can attach scripts to anything you want including the camera. You can also attach multiple scripts to one object so instead of attaching it to an empty object you could attach it to the player object or the game board or the UI or even a piece of level geometry. Having to have it attached to something in order to create an instance of that script in that scene is the part that takes a bit to get used to though. I came to Unity from XNA which while also using C# at least had the concept of a Main.cs class at it's core, which in many ways is a holdover from procedural style coding I guess.

I actually had a lot of fun playing around with the topDown template in Unreal Engine 4.
Unfortunately I hit a hard pass when trying to import an FBX file and it showed up black and wouldn't light. Well in full disclosure, when I built the lighting it looked like it was working properly for about 3 seconds, then it turned black again and was black when I ran the level. I adjusted the lighting to no avail. I created new brushes with the editor and they lit fine.

I am sure it is something to do with the UV's or lightmap and some such but creating and dealing with UV's is something I am gutting from my project because it will save a lot of time and I don't want to deal with them. I am going to push procedural shaders as far as I can take them.

By contrast, I went hog wild with importing an FBX level into Unity. And now um I made a level that is too mazy even for its designer ;P I wanted to stretch what I had done previously and I succeeded. I wonder if I am going to have to create a minimap and/or make it more linear. I could pull back the camera to make more of the level visible but I like the ratio I have of the player size to the path and the player size to the camera.

I really enjoyed the technique I used. I made a box that was 50x50 quads and started drawing selections that would become the paths textured in green and then rooms in orange. I then broke off pieces of the path into tiers that would be placed at different heights. I did the same with the rooms for each tier. And then I extruded the purple "walls" to corresponding heights around the tiers. When I was done with that, I selected parts of the path that would connect each tier and dragged one edge down to make a ramp connecting the tiers. The purple wall are placeholder for the actual level decoration that is going to be more detailed later. They also serve as bounding boxes to give me the scope of how much geometry I have to fill for each section. The paths and rooms with serve as the actualy player collision bounds and walkable area. I am hoping if I make it transparent, I will be able to use finer detailed decoration underneath it. Here is a screenshot of the level in Lightwave:
IMAGE(http://i68.tinypic.com/vdg7r9.jpg)

fangblackbone wrote:

I actually had a lot of fun playing around with the topDown template in Unreal Engine 4.
Unfortunately I hit a hard pass when trying to import an FBX file and it showed up black and wouldn't light. Well in full disclosure, when I built the lighting it looked like it was working properly for about 3 seconds, then it turned black again and was black when I ran the level. I adjusted the lighting to no avail. I created new brushes with the editor and they lit fine.

Just a heads up if other people run into this problem. A completely black model usually indicates that the normals of the model are flipped. You just need to select all the faces of the model and flip the normals (usually under UV / Flip normals in your 3d program).

polypusher wrote:
athros wrote:

and I'm not willing to shell out the money to see if GameMaker will work for me or not.

I dont have any feelings about GameMaker whatsoever, so Im genuinely curious. GameMaker seems to be on Steam, so wouldnt it fall under the normal Steam refund policy?

I would assume, but there's also a free trial from yoyogames official site.
I'm quite liking it if you just want to make games faster without having to really "become" a programmer. A lot of great games have been made using it as well.

I'd also recommend taking a look at Godot 3.1, open source from MIT.

So came across this old bit of code which is kind of fascinating. This was written by John Carmack to do a quick and dirty inverse square root approximation for the real time lighting they used in Quake 3:

float Q_rsqrt( float number ) { long i; float x2, y; const float threehalfs = 1.5F; x2 = number * 0.5F; y = number; i = * ( long * ) &y; // evil floating point bit level hacking i = 0x5f3759df - ( i >> 1 ); // what the f*ck? y = * ( float * ) &i; y = y * ( threehalfs - ( x2 * y * y ) ); // 1st iteration // y = y * ( threehalfs - ( x2 * y * y ) ); // 2nd iteration, this can be removed return y; }

So for back story this game was being worked on back when x86 processors were terrible at floating point operations and using a real square root function for calculating real time lighting took way too long. Luckily they didn't need it to be perfectly accurate since it was just for lighting so John was able to hack together this function that ran much faster than a true square root function would have at the time (thanks to SSE optimizations in modern CPUs they can run true square root functions about 40 times faster than they can run this code today).

So the two things he did that are somewhat magic are the evil bit level hacking and the wtf lines.

The evil line turned the float point number into a long int to setup for the bit level hacking which causes the binary value of the number to be shifted one place to the right. Basically it changes 0110 (6) into 0011 (3) which is much faster than doing a division by 2 operation especially on a floating point value, but renders the same result.

The wtf line is something of a mystery to this day. 0x5f3759df (1,597,463,007) is known as a magic number. Nobody knows how Carmack came up with it (he actually doesn't take credit for it either), but it worked. The - (i >> 1) part is actually the bit level hacking part.

The 1st and 2nd iteration part is running Newton's Method on the value of y which helps take the approximate value and makes it more accurate to what the value should be if this had been done correctly. This takes the margin of error down to less than 1 percent after only a single iteration, which is why they decided to not do any other iterations as this was good enough for their needs.

Basically this is all an example of very bad code that worked very well. Without this they could not have done real time lighting in Quake 3 as the proper way of doing this was too slow on the hardware of the time. Nobody knows where that magic number came from either. Carmack gives possible credit to other people who say they got it from someone else who say they got it from someone else etc, etc.