Search Results
Searched for posts by fejoa in all forums

Showing results 71 - 80 out of 983 total
Modify your search
Posted by fejoa, May 26, 2019 at 8:45 pm
Pickles outside of jars, could they be naked?
Posted by fejoa, May 21, 2019 at 10:11 pm
Ok folks, after some bug hunting today I can confirm that this problem is to do with SDL2, and it originates here:

EDIT: Here's a PR that should solve the problem:
Posted by fejoa, Apr 22, 2019 at 10:06 pm
daniel wrote
CVS installation works fine.
What is the difference between that one and the one on github?

That's good that the CVS compiles and runs on RPi.

The github version has quite a few more dependencies than before. New things like libpng, zlib, plus a couple of home baked things like xbrz scaling, fantasyname generator, music and sound, and possibly some other things I have failed to mention. It's not guaranteed that the additional libraries and bespoke stuff will work with the RPi compiler toolchain, or even if it compiles the application may not be compatible with either the chipset architecture or device drivers.

One cool thing to do would be to clone the git repository and progressively work backwards (or forwards) to find the last version of IVAN that compiles and runs on RPi. That should give an indication of what new features definitely break the game on RPi.
Posted by fejoa, Apr 20, 2019 at 6:16 am
I'd be happy to put some money in the hat for this endeavor.

Sorry for putting you on the spot RustKnight, we understand
I think it would be more cost effective and easier on you to try for an online interview, either audio or video. Got to try with small steps first.

All that said, the first step would indeed be to approach original three. I'd say J_Kahvi would be easiest to contact. Just send him a PM to his account on this site, theoretically it should end up in his email inbox.
Posted by fejoa, Apr 19, 2019 at 1:34 am
Have we still got money left over from the forum fundraiser?
Posted by fejoa, Apr 17, 2019 at 11:13 pm
I think this is a great initiative and I think your questions are good. I share Cap's sentiments about the chances of success however. Strangely I feel like you might be most successful with a face to face meeting. Go ask Ernomouse if he will let you use his place as a base of operations in Finland

I don't know how you would find the original devs, but their names are in the AUTHORS file:

You can find J_Kahvi (TV) here: and on the forum, he's into philosophy. I think hexi (HS) is in politics now, but there seems to be no sign of holybanana (TK). I think holybanana was last seen writing philosophy (imaginary phenomenology) with J_Kahvi in 2008/2009, but that was ten years ago...
Either J_Kahvi or hexi will be your best bet to help you track down holybanana.
Posted by fejoa, Apr 10, 2019 at 3:55 pm
For me (windows 8.1, 64-bit) doing alt-tab and also setting Frameskip to 1 seems to help restore things to similar performance as before this bug.

The alt-tab thingee is interesting. I think it must have something to do with IVAN partially losing its context. I'll try compile/testing at various points in the Git repo to see when this error might have crept in.
Posted by fejoa, Apr 7, 2019 at 8:54 pm
Awesome project!

It might have something to do with the ARM chipset being functionally different somewhere. Possibly need to modify the source code to make it fully compatible. Not an impossible task, but I personally wouldn't know where to begin.

What operating system are you running on your RPi? What compiler toolchain are you using to build for RPi?

Are you compiling the GitHub source? That has quite a few additional dependencies Try compiling IVAN CVS instead first to see if that works:
IVAN CVS uses SDL1.2, so use the 'libsdl1.2-dev' library on e.g. Raspbian.
Posted by fejoa, Mar 28, 2019 at 6:30 pm
Let's try fix the first thing first. I actually remember seeing this error, but I don't remember how the hell I fixed it. The message about '>>' operators might be related to the size of int64 a la:
You need to tell Visual Studio the whole code needs to be handled as 32 bit.

Did you follow these steps?

1. Added _USE_32BIT_TIME_T to "Preprocessor Definitions" which I found under Alt-F7 "Configuration properties -> C/C++ -> Preprocessor"
2. Added WIN32 and _WIN32 to Preprocessor Definitions (this seemed not ti fix anything, but I did this for superstition)

It could be that step 2 helped me overcome the compile error after all. Here are my settings:

Under preprocessor definitions (Configuration->C/C++->Preprocessor) for Main, I have:


and for FeLib, I have:

Have a look in the attached pictures. Let me know if it helps or not
Posted by fejoa, Mar 8, 2019 at 5:52 am
So the Windows "go slow" bug could be related to the numlock state eh? Thanks for the tip Izzy, I'll have to take a closer look at that when I get a chance.