deathwingA

Developement alone need help and Tips

Recommended Posts

Hello everyone,I'm new here,and i'm planning on starting a developement of a game similar to Psycho Break(Evil Within in English?) using VS2017 C++ with help of my school's custom engine(which i'll refer as IEX) that use DirectX 9 as its API(a bit old). 

Knowledge i have: basic C++, a bit of shader,a bit of design patterns, able to use autodesk softimage, basics of 2D game programming and 3D(maybe).

Experience I had with developement includes 4 2D games, 2 3D games(one of which is being done in Unity),all of them is team projects.Other than the one with Unity, everything is developed with help of school's engine(because it already has a lot of things from rendering to vertices based collision)

 

What I seek

1.Any tips or helpfull site(?)for developing games(I suck at searching for what i need for some reason)

2.A way to load and render fbx models because IEX can only render .imo,.iem,and .x files

3.A more smooth and lighter collision because mine check every vertices with the other models it supposed to collide with

Lastly,my english isn't that good so please be gentle.Thanks.

 

Share this post


Link to post
Share on other sites
8 hours ago, deathwingA said:

A way to load and render fbx models because IEX can only render .imo,.iem,and .x files

If your engine supports some other 3D file formats (not FBX), then I suggest you convert your FBX files into whatever file they support. It will be much easier then integrating another loading mechanism.

To convert files you can use 3DS-Max (if you have a license from your school) or Blender.

8 hours ago, deathwingA said:

.A more smooth and lighter collision because mine check every vertices with the other models it supposed to collide with

You said your school's engine already provides collision detection. You should probably use that. If it is not performant enough, then you can allways use different physics models, with less verticies. What kind of game are you making. For many games, using bounding boxes is enough

 

Edited by SillyCow

Share this post


Link to post
Share on other sites
10 hours ago, SillyCow said:

You said your school's engine already provides collision detection. You should probably use that. If it is not performant enough, then you can allways use different physics models, with less verticies. What kind of game are you making. For many games, using bounding boxes is enough.

Sorry for the late reply and thanks for your time.I'm not too experience with collision,but does bounding boxes does it for TPS-like game? Because it should be lighter for my game that way since school's engine causes lag if the game has lots of models to collide with.

Share this post


Link to post
Share on other sites

Depends on the accuracy of what you want to detect...

If you are just trying to detect if your character's walking path is bumping into walls and cars, then yes. It is also good enough for most shooting detection.

It is not good enough if you want "real" physics: Knocking over trash cans and such.

Share this post


Link to post
Share on other sites

If all you want, is to reduce the number of models that are being detected, just add a "spatial indexer" such as an "octree". Octrees are very well studied and there are implementations of octrees for every popular language. Find one, download it, and use it.

Integrating a physics engine is a very advanced subject. You should be prepared to do a lot of difficult work. That said, you are willing to put in *alot* of dev work, then by all means go ahead and do so. However, you sound like a beginner, so I hesitate to give you this advice. Now that we got the disclaimer out of the way, here are your options.

Here is a list of engines that you can use:

https://en.wikipedia.org/wiki/Physics_engine#Engines

Physx and Havok are industry standards.

But... like I said, this is alot of work. I could hardly justify something like this for a school project (even if you were a professional dev). Think twice before going down this path, and ask yourself if they really meant you to extend your school's engine :-) .

Share this post


Link to post
Share on other sites

Bullet Physics is pretty easy and a good place to start for object collision, using primitives or simplified convex shapes.  It's dead simple to integrate and get going with.  I'd recommend it as a place to start . 

Share this post


Link to post
Share on other sites
9 hours ago, SillyCow said:

If all you want, is to reduce the number of models that are being detected, just add a "spatial indexer" such as an "octree". Octrees are very well studied and there are implementations of octrees for every popular language. Find one, download it, and use it.

Integrating a physics engine is a very advanced subject. You should be prepared to do a lot of difficult work. That said, you are willing to put in *alot* of dev work, then by all means go ahead and do so. However, you sound like a beginner, so I hesitate to give you this advice. Now that we got the disclaimer out of the way, here are your options.

Octrees looks like it could help,depends on the graphic,i guess. I don't think I have time to integrate new physics engine except for easier ones

 

8 hours ago, Madcap said:

Bullet Physics is pretty easy and a good place to start for object collision, using primitives or simplified convex shapes.  It's dead simple to integrate and get going with.  I'd recommend it as a place to start . 

Looks interesting,will try this if things get too laggy. on my side(and if i have time). 

 

Thanks for the replies guys,its giving me more insight on the subject . 

Share this post


Link to post
Share on other sites
On 10/7/2017 at 7:16 AM, deathwingA said:

(I suck at searching for what i need for some reason)

This is going to sound rough, but this is something you'll need to work on.  A lot of software development basically boils down to problem solving and research.  That's not to say you shouldn't ask for advice, but sometimes you will need to solve things without the option of asking what the best approach is, and it will be on you to make that decision.

My second bit of advice is to reconsider the scope of your project.  If you are a student, it may not be a good idea to aim for a high budget game at this point.  Games like that take years to make, by teams of experienced devs, full time, funded by people with relatively deep pockets.  From what I've seen, this seems to be the most common advice handed out on this site:  Set reasonable/attainable goals.  A small team of students trying to make a AAA game is not reasonable/attainable.  Trying to make the next big thing in gaming on your own right out of school is not a reasonable goal.  Ask yourself what you're capable of right now, and aim a bit higher than that (you'll learn on the way), but know the difference between aiming a bit high and setting yourself up for failure.

Share this post


Link to post
Share on other sites
15 hours ago, trjh2k2 said:

This is going to sound rough, but this is something you'll need to work on.  A lot of software development basically boils down to problem solving and research.  That's not to say you shouldn't ask for advice, but sometimes you will need to solve things without the option of asking what the best approach is, and it will be on you to make that decision.

Yes ,in fact ,that is what i have been doing , tho' I'm so slow  I thought I might as well post somewhere while still researching on my own,to get more info,I guess.

 

15 hours ago, trjh2k2 said:

My second bit of advice is to reconsider the scope of your project.  If you are a student, it may not be a good idea to aim for a high budget game at this point. 

Valid point,I plan to do only simpler and doable replica game(with a bit of my own system) and only 1 stage of the game,only to get something to show to the companies I plan to apply for. 

However,I do find myself in situation where I tried to perfect everything in the game,which do confuse me and causes loss of time .........I'll remember it,thanks.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now