November updates

[kornalius]

As long as PocketHAL is not fixed to work with this device there is nothing I can do. I use PocketHAL as a third party graphic engine for PPL. I hope there will be an update soon that will fix this.

Btw, the raw performance of PPL should be quite amazing with this little beast!

[mmtbb]

I agree. Great machine. I posted to PocketHal forum. I hope someone can help me. Maybe not though.

[mervjoyce]

Hi Kornalius,

Thanks for 0.8 beta update 19-11-2004. Installed and everything here runs 100% OK again.

Many thanks for finding and correcting the memory leak. I was having no success at all, but was not looking for something as basic as "every proc/func call".

Kornalius ... You've saved the day again!

I'll get on with expanding my program functionality.


cheers

[kornalius]

Little update:

A lot of little bugs here and there have been fixed since the last update.

I worked on fixing quite a few things with the object-oriented features in the last few days.

Today I will focus on the debugger.

Tomorrow I will try to implement isometric sprite display (automatic sorting of sprites on screen) if I have time. The sprites will support a new height property for isometric collision detection and sorting.

On friday I hope to release another build!

[mmtbb]

how much does oop slow down ppl?

[kornalius]

If you use it, it will be just a little slower probably won't notice! If you don't use it, it doesn't slow down PPL.

[Guest]

If you use it, it will be just a little slower probably won't notice! If you don't use it, it doesn't slow down PPL.

i would be using it in a game and would want to have it as fast as possible

[kornalius]

It's not slow it's just a touch slower than regular PPL code. The slowdown only occurs when PPL is looking for an object variable. In a loop it will be a bit slower but in other situations you won't notice. Dynamic optimization cannot be done just like regular PPL code.

Object oriented is a higher level language, therefore cannot be as fast as a lower level code in most cases.

[kornalius]

Another little Update:

The debugger is now more stable. I couldn't come up with a completly new algorythm for it but I found the reasons why it wasn't behaving normally before.

I hope to find time today to start working on the isometric display.

Btw, the next release will include the inifiles.ppl library written by PointOfLight some time ago that I totally forgot about, sorry POL.

The next release will be delayed until early next week, I won't be available during the weekend and don't feel like putting out a version that I won't be able to support for a couple days after it's release day. That is just not me!

[kornalius]

Small update:

- New functions:
CopySurface() : Copy a surface to another one.
CloneSurface() : Create a copy a surface.
SurfaceEffect() : Apply, Blur, Negative, AlphaBlending or GreyScale to a surface.

- Can't use NULL parameter for main surface buffer anymore. A new variable is used. BUFFER%.

- New functions:
Classes() : Return a list of all defined classes.
Assign() : Assign a pointer to an object. Can also be done this way: &object$ = pointer$;

I should start work on the isometric features today, more on this later...

[PointOfLight]

This is basically off-topic, but since you were talking about surfaces... is there a way to load a new image into an already existing surface?

[kornalius]

No, do you really need it?

[PointOfLight]

Can't really think of a good reason. Just wondered if it existed.

[kornalius]

Actually it's not possible after checking the code.

[PointOfLight]

What? Something not possible for you? I don't believe my eyes Really, though, I didn't have a good reason for it anyway. I was just asking. So stop talking to me and get to work on the isometric stuff