Hello,
#1Will it benefit me any to use multiple instances of the same mesh for memory considerations? For example, I have noticed that if I use the same mesh with different skins that there isn't as large of a nexus size needed. So is it my imagination or is this the way to go?

#2As for resolution size, am I saving resources by using the lowest resolution possible?

#3Should I try to make my textures smaller to save resources.

Thanks in advance

mythic's ai demo, has 1024x1024 textures and it takes along time to load, perhaps it just me, someone wise could probably tell you.

Nitro777,

Can't answer #1. Sorry. Now you have me curious .

#2 - When you are talking agout resolution, are you talking about the "screen" resolution (such as in 640 x 480 or 1024 x 768? If so, higher screen resolutions can be more demanding on the system. Run you level at different resolutions and then check your frame rate. The results you get will depend on several factors (video card, CPU, number of polys on screen, textures, etc.).

When designing a game/RT3D application it is good to pick a target machine (CPU, video card, etc.) to develop for.

#3 - Larger textures/skins will consume more video memory. If the video mememory is filled up, then things can crawl. I try to use larger textures only in areas where a player will really notice. Other areas (particularly ceilings, which most people are not looking at most of the time) I will use as small of a texture as I can get away with.

-Thanks for the replies.

@Dan, yes I was talking about screen resolution. I thought that if I aimed at lower resolutions I might be able to target more machines and get better performance.

quote:

---

I try to use larger textures only in areas where a player will really notice.

---

-sounds like a good idea, thanks.

As for 1) Sure, because the mesh, frame (, bones) informations must be stored only once.

ah yes, that was my thought, thanks for the confirmation

One consideration though if video memory is getting tight for your project.

Each skin will generally take up more memory than the mesh alone. So, if you have a mesh that over the course of the game uses 5 skins, and you combine them all on the single model you might want to think twice. I'll give an exampe of why. In level one, the mesh uses skins 1, and 2. However 3, 4, and 5 are still loaded into the video memory. In level 2 it uses skins 3 and 5. But you still use memory for skins 1, 2, and 4. In level 3 you get the pay off, it uses all the skins.

Part of designing the game is weighing the ups and downs of each technique to maximize your performance. While it would increase the overall file size, you would be better off creating a model for each level that only contains the skins which would be used within that level so as to keep as much video memory free as possible. Granted in A6 pro, you can load the skins seperate from the models, so this is not really an issue at all.

Hi Myrlyn,

I understand what you are saying, however, I don't think I worded my question #1 correctly...what I am suggesting is that instead of making three different skins and three different models, I would make one model with a banker skin, a baker skin and a candlestick maker skin, just duplicated instances of the mesh. Meaning that each model would only have one skin in the game...I wouldn't switch them in game, and I wouldnt give them more than 1 skin in the game...thats different, right?