ChronoSoft

lr100 wrote:

CommanderData wrote:
Scroll of armor protection:
I really like this idea, and may change the functionality. I suppose I should warn everyone though just to be fair...

CD

No000!!! Its hard enough to get good armor as it is!

See, that's why I need to air these ideas out in public We'll hold off on that change for now... I can agree that armor going "poof" on you would be really painful. Maybe more so than weaponry!

He definitely has a lot of good little points, between you (lr100) and Wizzard I think we'll actually button up a lot of these little holes. I've been having some great success at finding and fixing things today so far. An update may be submitted tonight or tomorrow... don't expect a whole lot of "new" stuff, a smoother game will be the biggest draw

Speaking of armour....

If you had one of every armour type (except elfin and shozu), and read enough armour enchants so they all had an equal ac, are there any pros/cons between them and if so what?

jhawk wrote:Speaking of armour....

If you had one of every armour type (except elfin and shozu), and read enough armour enchants so they all had an equal ac, are there any pros/cons between them and if so what?

At the moment they'd be "mostly equal"... Leather obviously would not rust (neither would elfin or shozoku). The spartan cuirass has some beneficial properties and detriments too (I'll leave it as a research project- you can find the details in the *huge* 1.3 thread over at touchArcade.com, we'll put up the details here soon )

We are actually going to make heavier armors have some side effects in the future!

And magic armours right...?

ARRRRRRRRRRRRRRGGGGGHHHHHH!

I think I finally know what's been causing the game to crash occasionally when you go down the stairs!!! But fixing it may not be easy despite knowing the cause. For you programmer types (probably just lr100, but if anyone else is a dev here send me a PM and say hi ) we seem to have a problem with memory being overwritten in our World Structure.

This data overflow happening in the generation of a new level (when I define the rooms and hallways), and could stomp player variables and the players name, as that data is stored in the structure immediately after the problem area... This theory explains every crash log I've seen on the subject. I just need to find the root of it and eliminate it.

Hell, this might even cover the dark hallway problem too!

C programming gives you plenty of rope to hang yourself with!

I'll try to keep everyone updated. We're probably looking at 1.42 submission in the next 24 hours, with a ton of little nagging bugs fixed and a dash of new content for everyone to enjoy.

playing in my normal desperate search for id scrolls, even more so then usual...i happen to find one and after i read it and trying to guess which out of 10 to identify i get a phone call, dude im playing a game leave me alone..... reopen and no scroll and nothing identified

call came in the selecting popup screen

i am now re-exploring an entire level for the fourth time because once i take a couple steps into the room with the stairs the game crashes to home screen......not to happy

apparently seventh times the charm....

CommanderData wrote:ARRRRRRRRRRRRRRGGGGGHHHHHH!

I think I finally know what's been causing the game to crash occasionally when you go down the stairs!!! But fixing it may not be easy despite knowing the cause. For you programmer types (probably just lr100, but if anyone else is a dev here send me a PM and say hi ) we seem to have a problem with memory being overwritten in our World Structure.

This data overflow happening in the generation of a new level (when I define the rooms and hallways), and could stomp player variables and the players name, as that data is stored in the structure immediately after the problem area... This theory explains every crash log I've seen on the subject. I just need to find the root of it and eliminate it.

Hell, this might even cover the dark hallway problem too!

C programming gives you plenty of rope to hang yourself with!

I'll try to keep everyone updated. We're probably looking at 1.42 submission in the next 24 hours, with a ton of little nagging bugs fixed and a dash of new content for everyone to enjoy.

Good luck! Enable Guard Malloc is your friend!

jhawk wrote:i am now re-exploring an entire level for the fourth time because once i take a couple steps into the room with the stairs the game crashes to home screen......not to happy

apparently seventh times the charm....

My post a few spots up is all about that... I think I have it sorted! Update will be submitted today

lr100 wrote: Good luck! Enable Guard Malloc is your friend!

Yeah, I turned that on per your suggestion... The FUNNY thing is I still was not able to make it happen for me while watching. It was a totally unrelated thing... just sitting there reviewing all the ways that the player's object could be stomped or released in code... only to determine that it was IMPOSSIBLE- no code released it, or wrote to it inappropriately!

I went in and started looking over my game world structure because I wanted to take a break and add some other little tidbit in (we wont talk about that here) and then it hit me all at once! The structure! When the level is generated it builds a couple of arrays containing room and hallway data. Those are in the structure. The character name is next in line. I have seen crash logs due to invalid pointers to the character name during development. Check. Below that are your currently worn/wielded items. I had a report in TouchArcade last week of a guy wearing "Food". Double Check. What's next after that? The player object! The root of your crash logs- assigning his starting room once the next floor down is done being generated! Big F'in Check!

Basically something in the level generation is overstepping the end of the passage array space and spilling into the elements below it in the world structure. When this happens, it also could explain incorrect details about hallways and why some sections remain black- because the rest of the descriptive details have spilled out of the normal array. This morning I am reviewing the passage generation code. There has to be something wrong in there, but it's pretty complicated