Messages

Is there any chance of getting a copy of the database?

There isn't enough information being sent to uniquely identify an installation, so either the numbers are significantly higher (if .NET version is incorporated in making a unique ID) or it is tracking HTTP data to try and make unique IDs. I was only reading value at the bottom of the page (just the total), but it is impossible to say where this number has come from without knowing what the web server is doing.

Perhaps it is easiest to just ask for a copy of the database.

In fact, if JPEG works for mask importing, that could be considered a bug.

You mean - that it is included in a file filter?
I think AGS uses same file filter for all image-related open/save operations.

Yes, the file selector isn't filtering anything out, but there is a check after loading that should stop incompatible formats being used.

Code: csharp

if (bmp.PixelFormat != PixelFormat.Format8bppIndexed)
{{
    // don't use bmp
}}

Here are the stats of unique users per month (excluding the months which returned 0).

Spoiler

2008-06 to 2008-07: 40
2008-07 to 2008-08: 75
2008-08 to 2008-09: 88
2008-09 to 2008-10: 98
2008-10 to 2008-11: 1454
2008-11 to 2008-12: 3538
2008-12 to 2009-01: 4383
2009-01 to 2009-02: 4577
2009-02 to 2009-03: 5008
2009-03 to 2009-04: 5229
2009-04 to 2009-05: 5065
2009-05 to 2009-06: 4972
2009-06 to 2009-07: 5210
2009-07 to 2009-08: 5248
2009-08 to 2009-09: 4765
2009-09 to 2009-10: 4596
2009-10 to 2009-11: 4722
2009-11 to 2009-12: 4754
2009-12 to 2010-01: 5332
2010-01 to 2010-02: 5614
2010-02 to 2010-03: 5364
2010-03 to 2010-04: 5016
2010-04 to 2010-05: 4774
2010-05 to 2010-06: 4814
2010-06 to 2010-07: 4807
2010-07 to 2010-08: 4826
2010-08 to 2010-09: 4703
2010-09 to 2010-10: 4566
2010-10 to 2010-11: 4460
2010-11 to 2010-12: 4506
2010-12 to 2011-01: 4832
2011-01 to 2011-02: 4840
2011-02 to 2011-03: 4729
2011-03 to 2011-04: 4826
2011-04 to 2011-05: 4872
2011-05 to 2011-06: 4742
2011-06 to 2011-07: 4649
2011-07 to 2011-08: 4680
2011-08 to 2011-09: 4466
2011-09 to 2011-10: 4447
2011-10 to 2011-11: 4332
2011-11 to 2011-12: 4198
2011-12 to 2012-01: 4706
2012-01 to 2012-02: 5127
2012-02 to 2012-03: 4860
2012-03 to 2012-04: 4706
2012-04 to 2012-05: 2463
2013-05 to 2013-06: 21
2013-06 to 2013-07: 56
2013-07 to 2013-08: 66
2013-08 to 2013-09: 66
2013-09 to 2013-10: 82
2013-10 to 2013-11: 92
2013-11 to 2013-12: 91
2013-12 to 2014-01: 108
2014-01 to 2014-02: 934
2014-02 to 2014-03: 2574
2014-03 to 2014-04: 3588
2014-04 to 2014-05: 3832
2014-05 to 2014-06: 3800
2014-06 to 2014-07: 3779
2014-07 to 2014-08: 3789
2014-08 to 2014-09: 3583
2014-09 to 2014-10: 3568
2014-10 to 2014-11: 3611
2014-11 to 2014-12: 3576
2014-12 to 2015-01: 3908
2015-01 to 2015-02: 4024
2015-02 to 2015-03: 3877
2015-03 to 2015-04: 3813
2015-04 to 2015-05: 3885
2015-05 to 2015-06: 3905
2015-06 to 2015-07: 3865
2015-07 to 2015-08: 4047
2015-08 to 2015-09: 3934
2015-09 to 2015-10: 3737
2015-10 to 2015-11: 3677
2015-11 to 2015-12: 3621
2015-12 to 2016-01: 3820
2016-01 to 2016-02: 3928
2016-02 to 2016-03: 3968
2016-03 to 2016-04: 4000
2016-04 to 2016-05: 3852
2016-05 to 2016-06: 3817
2016-06 to 2016-07: 3893
2016-07 to 2016-08: 3925
2016-08 to 2016-09: 3866
2016-09 to 2016-10: 3752
2016-10 to 2016-11: 3663
2016-11 to 2016-12: 3542
2016-12 to 2017-01: 3643
2017-01 to 2017-02: 3605
2017-02 to 2017-03: 3620
2017-03 to 2017-04: 3906
2017-04 to 2017-05: 3848
2017-05 to 2017-06: 3506
2017-06 to 2017-07: 3483
2017-07 to 2017-08: 3537
2017-08 to 2017-09: 3489
2017-09 to 2017-10: 3533
2017-10 to 2017-11: 3434
2017-11 to 2017-12: 3226
2017-12 to 2018-01: 3346
2018-01 to 2018-02: 3331
2018-02 to 2018-03: 3225
2018-03 to 2018-04: 2104

[close]

Presumably there was a technical issue for releases in 2012/2013, but otherwise still peaking around 4000.

Thanks for clarifying.
What would be the initial start date for the data, if I wanted to go from the beginning?

I'm curious if you know of programmers who've succesfully made this work.

I think this is probably a good parallel for working on an engine:
https://www.patreon.com/monkey2/
...but working exclusively on a game would probably get a lower level of interest as people are only interested in using the final product, rather than supporting something they already use.

No idea if a 'Request a feature for AGS'-reward wouldn't backfire on the freeusers and devtime. Though these asked for features are probably really needed, so it should be a good thing. But if some more games like last & furious come out and maybe 2-3 months after the exclusive patreon release get released for everybody I'm fine.

You could potentially decentralise the work (to reduce pressure on individual people) with something like this:
https://opencollective.com/
I imagine that no-one would enjoy working on the current engine, but diverting attention to MonoAGS (as a replacement) and ScummVM port (replacement does not have to be backwards compatible) might generate some interest. I appreciate though, that this isn't really answering the question which is being asked here, but thought I should mention this as an option if the original engine is removed from the equation. Personally I'd donate some money in either direction, if there was potential for it to work.

This seems to be broken:
http://www.adventuregamestudio.co.uk/stats.php

Code: html

<html>
<head>
	<title>AGS Editor Version Stats</title>
</head>

<body>

<h2>Total Unique Users: 0</h2>
</body>

Are these still being stored somewhere, and are they publicly accessible?
(I would like to see them)

What I mean is that scripting API stays almost the same, from the looks of it. MonoAGS, on other hand is pretty different both in terms of scripting API and optimal coding approach.

I'm wondering if the workflow is similar enough that, even if the scripting API is not a directly comparable, the editing tools could be shared (or just modular enough so both sets work in the same environment). The benefits would be, easier migration to MonoAGS from AGS, a newer editor for AGS in the short term, a boost for MonoAGS in the long term.

It might be good to let the editor list and download compatible script modules (minimum version check). Involving a wiki just adds more resources that may be out-of-date.

One of the patches for Dustbowl was to fix issues with the map 'room' being too large, so the newer size (bottom of this post) is probably useful as a rough guide.

It was more for an idea of what he is testing. I imagine there is enough script functions for the mouse that you could probably make a game play itself, but since the ScummVM main event loop doesn't match the original, just checking the timing of functions and blocking actions are similar between the two would be useful. If XAGE also supports script modules, perhaps those sort of tests could be implemented in separate modules, or also used as examples in the manual? The majority of the audio functions should be working, so I'll try and make something to test those to start with.

I think some of the games I was testing were 16 bit, so that might explain why some were working. I'll have a look at it once I'm done with a couple of other things. I've not looked at Java in a long time, but there is a developer guide that uses GLES2 in a tutorial here and I've got one device that I could try it on.

I made another build, but it is not final work, just for the test: https://www.dropbox.com/s/f62oqlis9a7zgho/AGS-3.4.1-debug.apk?dl=0

I've tried this one and didn't have any colour issues, only problem was being unable to start games where graphics filter is 'undefined'. This didn't happen with every game though, so might be related to some settings not being read correctly?

Has tzachs got any test framework for MonoAGS?

Have there every been any test games, which just implement and test all the scripting functions? The reason I ask, I've probably spent more time in AGS writing scripts to test than in ScummVM actually testing them. I imagine test scripts could also be useful for the normal branch, the cleanup branch, and for XAGE. I'd like to suggest that the GitHub repository hold a test game along with the demo game (or potentially just make the demo game be a function showcase/tester, rather than an actual game).

I would think that if each new release was tagged on Github, and release binaries were also on Github, it would be possible to link everything so no-one has to be manually editing things when a new version is released. They have a REST API where you can query for the latest release and its download link, so worst case, you just query for the latest "stable" release and build static pages on the web server. The python requirement is for additional passes after conversion from LaTeX (so those changes could be done with sed or awk if you don't want python installed) but I imagine it wouldn't be too difficult to just make the intermediate HTML pages part of each release (so include a static HTML site as well as the CHM manual).

What is there is terms of server infrastructure, now that TeamCity is gone?
If there is nothing but the webserver, can it pick up tagged releases with git or is it more generic/restricted web hosting?

When that window is closed it is calling

Code: csharp

spriteController.Refresh();

...in case a sprite was added in the selector. I'll take a look at it.

[DBoyWheeler]

Concept: DBoyWheeler
Playability: Cassiebsg (and thank you for the encouragement)
Artistic Execution: Matti

Side note: I think Glenjamin's is the best combination of the three voting categories