Jump to content

  • Log In with Google      Sign In   
  • Create Account

Banner advertising on our site currently available from just $5!

1. Learn about the promo. 2. Sign up for GDNet+. 3. Set up your advert!


Member Since 19 Jul 2006
Offline Last Active Today, 04:09 AM

Posts I've Made

In Topic: 256 or 1024 color palettes/rotation (with 32bit color)

Today, 02:11 AM

Wow, that's brilliant. I'm gonna try it. :D

In Topic: How to draw a texture out of bounds of the RenderControl?

27 July 2015 - 12:17 PM

I don't know if it's a bug, but If I understand correctly:

I believe the display resolution of the backbuffer must be a valid device screen resolution, so it would use the closest matching resolution to what you set and I think SetViewport determines what portion of that will be rendered. If you need to render into a 4096x4096 space, myself, I would create a separate render-target and move the position of the render-target(as a texture to draw to backbuffer with spritebatch) around based on the scroll values -- so it pans around inside the display window.

In Topic: Are there any limitations to the DirectX Toolkit?

21 July 2015 - 09:06 PM

My personal experience with it - it has a lot of useful helpful functions that are highly efficient IMO and memory safe if used properly so I really like it. I use MonoGame/XNA but am also expanding a DX11 game engine at the same time, so I would say it has some things that are similar to XNA(which I likes), but you have the source code too so you can see how it works and understand how to make your own functions that do similar things. For example, I found the memory alignment helper(AlignedNew.h) very useful (inherit into your classes/structs) for 16 byte alignments (optimization). So since you can choose what you do or do not want to use and you can see the source code, I think it is useful in helping to learn DX11+ plus it does not bring any limitations and since it is very popularly used, I personally would assume it is here for the long run. You could even use it minimally if you wanted and just use the parts most useful to you.

In Topic: Monogame vs LibGDX for 3D development

18 July 2015 - 03:49 AM

I would probably refer to them rather as frameworks (altho I don't know much about LibGDX)

I know that anything you can do in DX11/OpenGL4.5, you can do with MonoGame (XNA was DX9 - where this is more like DX11+/OpenGL) - like a wrapper -- I've done 3D animated character stuff with it before (and used new shaders). Yeah, it's odd that not very many ppl use them for 3D - although 3D xna tutorials for example, will work exactly the same way (except it's faster now and you can use new shader models).

I don't know much about LibGDX so someone else should probably speak for that one. I've heard it is also very good but I don't know if there are any limitations.

I would be VERY suspicious of anyone saying there are performance issues with MonoGame - at least I've done my own benchmarking and found it compares extremely close to DX11 in C++ on desktop. FAST. Maybe the old versions had some issues? I don't know - that could be it.

The only way I could see that happening is not knowing how to program c# for performance(pools, garbage collection, locals, etc) or how to minimize locks and changes when accessing the video card. It isn't essential to use the provided methods either - as you can make your own primitive rendering stuff. I've heard you can use bullet for 3D physics too if you didn't already make your own. I think I'll go play with libGDX just for fun now too.. laugh.png

In Topic: Moving from MonoGame to SharpDX/SlimDX

15 July 2015 - 04:45 PM

I believe MonoGame uses SharpDX to some degree(cross platform) and basically is thereby DX11+ and you may notice that if you obtain the SharpDX toolkit version, you can set it up so that it is initially virtually identical to MonoGame/XNA. With SharpDX though, you'll probably want to use .cso's for compiled shaders where with Mono you'd use the pipeline(although optional). I tried using both SharpDX and MonoGame in multiple projects and noticed no performance differences or restrictions so far. I had to resort to checking DirectX 11 documentation frequently to decipher how to mimic a typical DX11 setting in c# -- and one may need to watch for managed and unmanaged stuff - and make proper dispose methods. Last I checked SharpDX version for DX12 did not yet have a lot of the toolkit stuff I was hoping for and I'm not aware if those are being worked on by anyone yet.