A bit of blind coding, last night.

My monthly calendar reminder popped up, last night, to remind me to do all of my "full" backups.

I started the rewrite of the controller code, last night. Digging deep, I realised I could probably reuse quite a lot of the old code, since the engine already grabs Joystick data each frame, as well as using a bunch of good-old fashioned numbers to do a lot of the legwork.

The framework isn't quite running properly on MacOS, so today will probably be spent trying to get that going. For some reason, since the recent MacOS Security Update, I don't seem to be able to Remote-Desktop into the Mac via the WinLaptop. Hopefully it's just some weird thing, requiring a reset or something, and not a hole that's been closed, making it unusable.. Hmm.. :\

After a day spent trying to track down the issue(s), I've finally got WinSock to return exactly what I expect it to return! I can't rest yet, as I need to foolproof the values being returned, and also then get it working on things other than Windows... WinSock is strictly for Windows, so .. It's going to be a system-by-system rewrite to get everything on the same level. But I'm about 50% confident that I'm able to send requests from the Windows editions of the games. *sigh* T...

It's definitely picking up the controller..

Time for another test. If you're bored, are running Windows, have a controller or two at hand, and want to help out, please give this test a run. Download the Test Here (Windows only, right now. Linux is.. ... Being Linux!)

Mouse, and of course, touchscreen. They both took a bit of a tumble, last night. The overscan seemed to be working, but I was only testing it with controllers, since the overscan is typically only used on Android TV versions of the games. Yesterday, however, I chose to reuse the overscan functions on iOS.

<- 4:3 | iPhone -> The menu seems to be scaling fairly well, which is nice. Poor single-pixel Platdude outline looks a bit gnatty, but that's "GL_Nearest" sprite scaling for you :\ I'm confident that, once on an actual device, the resolution increase will help to fix that. Either way, the screen seems to fit all my test screen resolutions well enough to move ahead with the next task. Gud gud!!

So, after having read up on URL access for a couple of hours, I managed to get about 0% through it, yesterday. I might actually end up trying to find a proper book on the subject, and having a good focused read-through. .. Although I imagine any book would probably focus on Win/Linux, and not the cross-platform usage I'm after. I've really coded myself into a pickle, here!

Today I've been working through what will and won't be part of the "Framework". That is, what values and settings and functions should be "Framework", which should be "Engine" and which should be "The Game".

I posted Benchmark 2 over on Socoder, yesterday. You can Download the Benchmark here.

OK, a fair number of tweaks in the past few hours (and a nice big nap, too!) I'm now running Ubuntu32Bit, and have recompiled the Benchmark test with (hopefully!) all the important libraries Statically Linked, thus apparently making for a more compatible ".exe" You can grab the Benchmark Test here. If it runs, let me know. More importantly, if it doesn't, let me know! Thanks, again, for any and all testing.

Seems the crash in the Linux code was happening when it tried to stream a .wav file, so I had to rejig a whole bunch of stuff to get it working again. After all of that rejigging, the Benchmark test now runs 100% on Linux, which is nice. If you'd like to give the benchmark test a try, You can Download the test here as a .tar.gz. The test has been compiled using GLFW3 and OpenAL, under Ubuntu 16.04 LTS, with the g++ command, because that's all I could currently get it working with!! You'...

Developments

The Benchmark thing from the other day starts running well on the Linux setup, but seems to crash after about 2 minutes..

Developments

I finally got around to shifting the framework over to the Mac, last night. Hitting the compile button swiftly planted around about 1,000,000,000 red "WARNING:ERROR!" flags all over the code.

Most folk seem to be giving me successful results with yesterday's benchmark test program. If you haven't already, Do so here!

If you can, please give this benchmarker a run. It's a simple little tool. Unzip it, run the .exe, and watch as three "Bloke" sprites circle around for a little while. Don't touch anything. The benchmark tool will do it all for you. Running through a bunch of sprite-pushing tests, blending methods, and audio streaming, too. It'll also flip to fullscreen to check there, too, so be warned if you're watching any other media on your PC at the time!!! Once it's done (about a minute...

This seems to be functional as well as can be expected. An .ogg file is playing in the background. Feel free to replace the ogg with your own ogg, should you have any! As with the previous audio test, up and down will (*should!) pitch the music up and down, but now it'll glide more smoothly, since the pitch is being handled by the main loop, not the re-buffer part of the code. Other than that, nothing of note has changed, I don't think. You can download the Ogg Playing test here.

OK, I think I managed to get this working!! Run the game, and you should hear the "Sheep Goes Right - Level Complete" sound playing over and over and over. This is in fact an audio stream. This first test is for .wav only, so I left the "Data/550/TestStream.wav" as just a small little sample, but if you want to try switching it out for some actual music, simply replace that file. Audio must be in .wav format, though. Be sure to let me know if it does/doesn't work for you. You can Download St...

I haven't got around to tackling Streams, yet, and have spent all morning working on audio loading functionality. Mainly that the thing kept breaking when trying to load too many files. Typical memory leaks and the like were fairly rampant, too. But I've persevered, and I think it's more or less stable, now.

Developments

With my little "Waahoo" sound playing, I'm fairly confident that the first step of the audio chaos is finally working the way it should. .. I think..

First step, trying to get OpenAL to compile.

Today's been about neatening the engine in places, and also adding a Random Number Generator.