What if i make all the images in my game monochromatic and then color them in game? Would that improve load times?
I highly doubt that load times of a few images will be of any concern to you
fewer colours -> less data -> shorter time to read
I’m making a large game though so trust me… It concerns me quite highly.
I’ve had projects with a few thousand of images, and image loading time really wasn’t a problem. But Danny02 is right, the less information there is in an image, the better it compresses and the less data there is to read from disk.
I’d say, to make it looking good is more important than to save 5% image loading time.
Probably they load faster, but, wouldn’t it take some time to add color to each image?
I think you should do it the normal way just to keep it simple, plus, any modern day device will load textures pretty fast i think.
im sure it would take longer to color the images in any meaningful/appropriate way than it would to load already colored images.
You quite simply aren’t making a large enough game where this could matter at all, trust me. Don’t underestimate the power of the modern day CPU/GPU.
technically speaking, loading a 8 bit, grayscale or any image that is smaller (disk usage wise) will load faster, but unless you’re loading thousands upon thousands of images pretty much any modern day computer will handle it.
Here’s an example, I am also working on a resource pack for minecraft. It currently has over 4,000 textures (uses a mod called MCPatcher, that allows me to have multiple textures per block). Using the logic that MC is programmed in java and with LWJGL, i’ll assume we’d get similar results via LibGDX or Slick.
The results:
4,000 full 24 Bit Color PNG files - 10 seconds load time.
4,000 compressed 8 bit PNG files - 8 seconds.
So yes, it does matter. But with 4,000 64x64 textures, it only changed the load time 2 whole seconds. You could argue it was a “20% increase” but really, images aren’t a big deal to load. I’d just stick with full uncompressed images, most of your load time will be from running tons of game logic, not images. At least, that’s been my experience.
Welp, whole post erased by accident…
Basically what everyone is saying: “Don’t worry about it.”
Any (minuscule, as in microseconds per image) time saved in loading will be dwarfed by the time it takes to ‘color them in game,’ what ever you meant by that.
Besides, there are other more effective ways of reducing loading time, such as only loading what you need when you need it, and pre-loading stuff in the background when you know you’re about to need it.
One of the main rules of optimization is: “Don’t fix it if it ain’t broken.” Unless your loading times are unbearable (really), this isn’t yet an issue and is not something to worry about.
If this is a performance issue for you, you have serious problems somewhere else.
And if this is such a “large game”, shouldn’t you be a bit more experienced first?
:-\
haha well the game isn’t uk WoW big. And really I have no issue with loading (YET!) I’m trying to think ahead.
There’s two versions of that: Planning, and Premature Optimization. One is a useful skill, the other is pure evil.
Make sure you choose the correct one.
Think it’s a little bit of both. 
…I do not get it. Googling would have been faster then starting a new thread. For the love of God google first. You have made 14 topics in what a month? Not trying to make you afraid of asking questions but really google before posting.
If you want to “plan ahead” fine. So you have a bagillion images…demand load and unload them. If you truly are building a “big” game then you’ll need to be able to do this with various resources anyway.
Although I agree that Googling before posting is usually a good idea but now this post will show up in Google and give people an example such as :
It keeps the forum active imo, better than no one posting just because Google answered it, although some things should never be posted…such as the infamous “Which 2D library” thread that pops up once or twice a week.
It depends on the image format.
Logically less data should mean faster load times. In practice, however, in this case it makes no difference.
To really try and benefit from doing something like this you should use the correct image format (or even make your own) and a suitable compression algorithm.
To answer your question: No.
I wish there was more talk about game design (as in, how to make it fun, create ideas, transforum ideas into a game), and less about how to program java. Currrently we have 99% programming questions to 1% design questions.
Sorry for derailing here but I agree with Varkas. Programming is big obviously but there have been so many new topics that have literally been answer over and over again and are very much basic java type questions. Doesn’t matter how fast, nice looking, or technical your game is, if it isn’t fun it isn’t a very good game.
I would rather no one posting then posting for the sake of posting. Those posts become useless noise.