Rainy day - Game project. A practical approach to complete it.

[SiegfriedCroes]

for the game itself I used the PNG format for all graphics I did reduce the colors of all of them, it doesn't really matter for performance, just to reduce file size  

 

[Ziz]

Do you stalk me Ziz?

No, but I remember, what different members post on this board. It is not as big as it seems.

I would not ask if it was explained  FOR THIS GAME HERE. You must learn that in a forum, the READER matters and he does not always search in different threads for the same answers. I only asked because I have no clue about the "engine" and how it can handle textures/tiles. if 32 bit are not an problem then OK, nice. If 8/16 bit is much faster then it would be worth to optimize the graphics towards this. Questions like that have to be solved before the actual work begins.

Every engine have to handle with the hardware. And you often assumed, that 8 bit should be faster on the Pandora, because the number is lower.And me and others often explained you, that this assumption is wrong. It is not faster on the Pandora, just because you reduce the bits per pixel.

 

[PokeParadox]

Penjin just uses what is default for the system (i.e. 16b for Pandora)

EDIT: PenjinTwo uses a config file and tries to set those values of resolution/colour depth.

I would say optimially the images would match the screen colour-depth then SDL doesn't have to do much conversion.

 

[fusion_power]

for the game itself I used the PNG format for all graphics I did reduce the colors of all of them, it doesn't really matter for performance, just to reduce file size

Thanks for the answer, it's even better if you already did  this and yes, file size still matters imho. (hello 50GB new Wolfenstein Mega-Texture-Crap  :rolleyes:  ).

 

[MarkoeZ]

MarkoeZ.SetLurk(FALSE);

Text.Render("Rainyday Activity: Yaaaaay!!!!", Screen);

MarkoeZ.SetLurk(TRUE); 

 

[Wrath Of Khan]

Yep. Ill keep tabs on progress.


It deserves to live again...

 

[PokeParadox]

A tiny update... I'm working through converting the differences from Penjin to PenjinTwo I am working on this and I'm not dead.

Still nothing interesting to show like a first working screenshot running under PenjinTwo but I have a ResourceManager done (needs testing and debugging)

And I need to finish the conversion. Hopefully a few more evenings work... then we can actually start planning where to go from here!

 

[Wrath Of Khan]

^^^^^ Good stuff!

 

[x1212]

Is there currently something I could help with? I might have some spare time the next few days (again).

 

[PokeParadox]

Hi and sorry, I need to get the last few bits done which is not a lot of work but I've been able to find the time (Had a bit of crunch at work, thankfully we're back to normal again) and now I have visitors so I can't really code...

Once I get it compiling with PenjinTwo I'll create the PenjinTwo repo and we will figure out where to go from there.

I'm a bit reluctant to push anything until it is compiling and running something. hopefully I can make some more progress this weekend.

So unfortunately I don't think there is anything that you could help with here as it's still things which require understanding of both Penjin and PenjinTwo. But if you think you could help with the conversion perhaps I should check in the current modified source?

 

[x1212]

Well I don't have experience with Penjin and PenjinTwo, but if you find the time to check it in I will try at least, maybe the changes you have already done will give me an idea of what to do ... (however, having looked at a diff of the sources of the two engines it really looks a bit complicated)

I think creating a new branch for this should be enough to make sure that anyone cloning the repo will get a working version by default as the master-branch won't be changed then.

 

[PokeParadox]

I will try and get this checked in somewhere. I did find a very little bit of time to progress a little more (Especially since the weather was crap) I wanted to get it compiling but I discovered another couple of functions that I have no alternatives for in PenjinTwo... But I can just lift them directly... I think.

I think we are getting closer but I hope that it's not completely broken once I reach the first milestone! h34r:

I'll try and do this tonight or failing that it will be Wednesday.

 

[PokeParadox]

Um... I think I created a branch but I can't seem to push to it >_> Getting a little blue in the face about this right now. :S

 

[x1212]

hmm ... I assume you are trying to push to my github-repository?

As I added you a few weeks ago to the list of collaborators, it should work if you use your github-account while pushing.

Did you get some kind of error?

Edit: ah ok I guess you mean pushing to your local repository on this new branch?

        Did you 'git checkout branch' already?

 

[PokeParadox]

I'll give it another look tonight. I think I was getting in a muddle with local branches and remote branches. I've not used branching with GIT before.

I'll try and checkout from the remote branch - diff my changes accross and then push.

There's probably a better way if you understand GIT better but I think this should work at least...

 

[levi]

In case it helps, Git is a distributed SCM, so there's no concept of remote and local branches in it. Just changes you have that another site doesn't have and vice versa.


Git has a concept of branches, but AFAIK they operate more like labels in other SCM. A branch as more general concept exists in git (and other distributed SCM like Mercurial) because changes can have one or more children. Once two people do separate work based on the same change revision, it generates a new child of that change. That'll happen if you do the work at the same time, or fail to pull down new changes and work on the head (either deliberately or accidentally). Often, those are then labelled with using 'git branch', but they don't have to be.


Changes can also have more than one parent (naturally!) - that's what happens when you merge in a branch to your head revision to collate the changes. It descends both from your branch and the other branch. I find thinking of the changes as a chain of dots where a completed branch looks like a loop hanging off the shorter branch helped my understanding of the concept of branches in git and mercurial.


Git automates things a bit more than mercurial, which IMO can hide the core concepts to a certain degree more than Mercurial does. Pull down changesets and it seems to resync your local view to the head revision automatically, for example, but perhaps it's just that I'm not too used to git yet that means it sometimes catches me out with the stuff it does differently to Mercuial. The core concepts seem to be the same though.

 

[PokeParadox]

Sorry, I haven't found a good amount of time to sit at my PC and figure this out. So for now I'm doing what I know will work and sharing my current progress as a 7z snapshot:

http://projectinfinity.org.uk/lib/exe/fetch.php?media=pokeparadox:tempenjintwo_rainyday_tue15july_2014.7z

It doesn't compile yet and still has some functionality to replace in PenjinTwo.