Download for Windows Download for Linux Download for FreeBSD Download for Mac Manual Wiki Forum IRC Trac

Tuesday, October 27, 2009

Getting ready for 2.1.8

It's been a little over three months since we released 2.1.7 stable, and we are happy to announce that version 2.1.8 is getting closer every day.

After we released 2.1.7 we split Aegisub development into two main branches: The stable branch where we apply bugfixes for the last stable version, and the development trunk where we work on grand new features.

For all of you not using Windows we also have some good news. Aegisub 2.1.8 is well on the way to become the first version we can call stable on Linux, FreeBSD and other UNIX systems. There's only a few build-system issues remaining to fix. (Unfortunately it isn't as well for the Mac version, more on that later in this post.)

More on what you can expect to see in 2.1.8 after the jump.

The perhaps most important bug-fix in 2.1.8 for many people will be for the issue hitting everyone on Windows 7: "Could not lock buffer for filling" when trying to play audio. It turned out to be a really silly problem, but now it's fixed.

Another problem a lot of people have had is opening high-definition video, getting cryptic "Failed seeking video" errors. It turns out this is caused by some very limited OpenGL drivers shipped by Microsoft that can't handle texture images larger than 1024x1024 pixels. If you install a graphics driver directly from your chip manufacturer (usually NVidia, AMD ATI or Intel) you should get a better OpenGL driver, but it can be tricky and might not always be possible. We do have a solution for this in our unstable branch, and if it doesn't break anything else we should get it into the 2.1.8 release.

We also fixed some issues with slow seeking in MKV files and opening files with Unicode filenames in FFmpegSource2, and some bugs in Automation 4 Lua. Notably, the new "relayer" and "restyle" functions in Karaoke Templater didn't actually work, they have been fixed.

There's more minor problems most of you probably haven't bumped into, you can see an overview of everything on the bug tracker: Milestone 2.1.8 issues.

There are also some things we'd have loved to get in, but just aren't feasible. Some people have complained that the audio spectrum display is very slow, which is entirely true, but it turns out that fixing it causes an avalanche of other problems to pop up, and it's just such a major undertaking that it will never make it into the 2.1.x series of Aegisub. But you can be assured that there is a major rewrite of things under way, and it's very fast!

As mentioned at the beginning, it also seems like we won't be able to get a really stable version for Mac out in the 2.1.x line of Aegisub. The technical reason is that we use wxWidgets 2.8, which has some big problems on Mac in some core areas Aegisub depends on, and we can't switch Aegisub 2.1 to use wxWidgets 2.9 without causing a lot of new bugs to pop up. Our unstable development branch is using wxWidgets 2.9 however, and it's looking very good on Mac, but the changes required to support wxWidgets 2.9 in Aegisub were large enough that they aren't getting into Aegiub 2.1.


Tuesday, July 14, 2009

Aegisub 2.1.7 released


It's been almost a month since the 4 year anniversary and the promise of 2.1.7 "within a few days", but lots of stuff happened, people were travelling and otherwise not reachable. But I dare say it's not a bad thing, this version of Aegisub we are releasing now is certainly more polished than what we could have managed four weeks ago.

So far, this release is just for Windows. Use the Download button above to get the new version. Release notes and more on what to expect after the break.

A lot of things have changed since version 2.1.6, and sure, it has also been a long time: More than half a year! Here's a summary of the more important changes.

Update: We now also have a complete changelog since 2.1.6 available, courtesy of TheFluff/rhx.

  • DirectSound audio player has been completely rewritten (again) - this should give much better stability of audio playback. Note that the old DirectSound player is still available, you can change to it in Options. If you're using Windows 7, the old one seems to be more reliable.

  • PCM WAV audio provider actually works again, and with files of any size. (No more crashing with files bigger than 256 MB.)

  • Tip of the Day has been removed.

  • Loads of changes to FFmpegSource2 (FFMS2) giving better support for almost everything, making it more stable and so on.

  • Lots of memory leaks have been fixed, Aegisub should use less memory now.

  • Bug fixes in many file format readers and writers. Loading and saving subtitles to foreign formats should be more reliable now. Problems with frame-based/SMPTE timecode-based formats fixed.

  • The Kanji Timer function has been almost completely rewritten, squashing all known bugs and giving a prettier GUI. The use is the same.

  • OpenGL errors in the video display are no longer fatal. You will get an error message, and video won't open if Aegisub doesn't support your graphics card/driver, but it shouldn't crash. See below for a workaround if video doesn't work for you.

  • Some additions to Karaoke Templater, you can now create loops with variable iteration counts (including infinite!) making a lot of effects more feasible.

  • The "Local configuration" option was removed from Options. If you want a "portable" version of Aegisub (as it was intended for making) we now have a separate download package for that. It also actually works as advertised now. See the download page for details.

  • An innumerable amount of other minor, cosmetic changes that just makes everyday use more convenient and smoother.

Read this if video doesn't work:

One of the most common problems with Aegisub is video mode not working with a wide range of graphics cards and graphics drivers, this especially seems to hit users of ATI/AMD graphics cards hard.

Until we can make our usage of OpenGL more robust (or make an alternative implementation of the video display in software and/or Direct3D) we have a workaround that involves forcing Aegisub to use a software OpenGL implementation instead of the one supplied by your graphics card driver.

Before you try this, first try updating your graphics card driver. Get the latest driver directly from the maker of your graphics chip (usually NVidia, ATI/AMD or Intel) if possible, otherwise look at the support site for your computer vendor.

Download MesaGL 7.4.2 for Windows — Unpack this archive to your Aegisub installation folder, next to the aegisub32.exe file. Aegisub should then use software rendering for the video display. It will be very slow/sluggish, but should work in all cases.

If you have any questions about Aegisub, first see our manual, it covers most of the program. Lots of common problems have also already been covered on the forum, try searching the forum, there's a good chance someone has had your problem before.


Wednesday, June 17, 2009

Four years of Aegisub

Today it's just about four years since Aegisub was first conceived.

It was on the 17th of June 2005 in a private channel on the ChatSociety IRC network. We were discussing the various problems with making advanced typesetting in ASS subtitles and how poor the tools generally were.

The original name was Visual SSA, and it was originally just intended to be a tool to assist in doing typesetting. Yes, the visual typesetting tools we only got implemented somewhat in the more recent versions was the original goal of the program. General-purpose subtitle editing wasn't it.

It wasn't many days into development before video was shown.

At that point, showing video was all it did. As things went, basic loading, editing and saving of subtitles was added soon, and the interface started looking a lot like the current versions, but still not quite.

The important part here is what's above the subtitle editing area: Where the audio box sits today is space for a toolbox. The intention was to add visual typesetting tools there.

Of course, that never happened, and audio took over that place instead. In fact, the goal turned towards making a replacement for the subtitling program Medusa, which was popular then. (A few people are still using it today!) This is where the name of Aegisub comes from: Aegis is the shield of Athena, upon which Medusa's head was put after she was slain by Perseus.

Development went on, features were added, slowly we got more testers and went towards a relatively stable version. On December 27 2005, version "1.00 beta" was released, just a little more than 6 months after the first idea.

Development continued with regular "beta" releases.

We had the notorious version 1.06 with the "movax bug" in the installer: The installation script had a typo in it, which caused it to, instead of adding Aegisub to Windows' Add/Remove Programs list, clear most of the data for the list. After it was discovered, 1.06a was quickly released fixing just that problem, though several people had already been hit by it.

It was first at version 1.07 we went open source. At this point, development still happened on a private SVN server with no public access, and all bug reports and feature requests were taken on the forum. The source for both 1.07, 1.08 and 1.09 was distributed as part of the installer package, there was still no public repository.

However, soon after 1.09, we moved to a new host for source code repository, which allowed us to open up the repository for anonymous read access. It was also around this time we got a real bug tracker. The 1.10 beta was the first version released under fully open development, that happened on August 7 2006.

If you compare the screenshots of 1.00 and 1.10, you'll see not much had changed. There's a large number of changes you first notice when you dive into the menus, of course, but compared to what has changed since then, it's just tiny.

After 1.10, the plan first was to release 1.11, and that should be done before the end of the year. First, things went pretty smooth, but then new huge features started slipping in, they weren't stable, and thus the release slipped. Shortly after the new year passed, we decided the changes were big enough that we'd now call it 2.00 instead.

Now, the rest is almost history, but not quite.

Visual typesetting finally got added, the thing Aegisub (then Visual SSA) was first intended for, and pre-release upon pre-release was put out. In the end we decided there were far too many versions called "2.00 pre-release" to make releasing "2.00 final" feasible without confusing everyone, so we instead decided 2.2.0 would be the first version of Aegisub labelled as "stable", and we would release 2.1.x versions until then.

2.1.6 was released on November 26, 2008, which is soon 7 months ago. The good news: 2.1.7 is almost ready and you can expect it within a few days, and version 2.2.0 should be less than a month away!

The installer for Aegisub 1.00 was 2.82 MB, the installer for Aegisub 2.1.7 on Windows will be close to 24 MB. We've come a long way since that day in June 2005.


Tuesday, February 10, 2009

Kumaji explained

Kumaji is an advanced subtitle rendering engine, in development.

At the time of writing this, Kumaji does not render any subtitles whatsoever and is in general in a very early stage.

There are several reasons I started the Kumaji project, I will try to cover them in this post.

The name

First what does "Kumaji" even mean? It's derived from Japanese where it would be written クマジ. If you reverse that, you get ジマク, "jimaku" (字幕), which means subtitles. Also, Kumaji can be understood as 熊字, "bear" + "writing", hence Kumaji has a bear for a mascot! (Lots of people have suggested using Pedobear, this is wrong, I don't want to have that association.)

"Kumaji" should be reasonably easy to pronounce for most people and as far as I know it doesn't have the potential to offend people, like "libass" might have. The name is also format-agnostic like the actual renderer will be.

The goals

The key goals are:

  1. Portable code without sacrificing compatibility

  2. Maintainable and hackable code

  3. Speed

  4. Flexibility

Portability is the first and foremost goal. All current subtitle renderers have major problems with this. Those that do compile and work on multiple platforms (libass and the abandoned asa) are strongly tied to details of text and font handling on UNIX-like systems, which means they fail on Windows and Mac platforms because those have much different ways of handling fonts which FontConfig doesn't wrap properly or over-complicates. The result is very sub-optimal. VSFilter depends on not just Win32 (and Wine doesn't implement everything it requires yet) as well as MFC and COM. Perian's subtitle rendrer is Objective-C and entirely dependant on Cocoa text API's. Kumaji will achieve portability by plugging in platform-specific code where appropriate. The motto would be doing the right thing on each platform, whatever the cost.

Maintainable and hackable code is important. It must be possible to jump into the code without having a great understanding of the entire system beforehand, and it must be possible to learn good techniques from reading the code. The code must be well-commented or self-explanatory all around. (Portions of the code with poor, little or no explanations should be treated as bugs and reported.) VSFilter is a prime example of how I do not want the code to end up.

Speed is obviously important, to a certain degree. Reasonable subtitle scripts should render in realtime so Kumaji can be used for softsubbing. This may mean writing some critical routines in multiple versions optimised for different systems, using SIMD intrinsics or hand-optimised assembly code. However, good algorithms and data structures still take priority over SIMD and assembly tricks.

Finally, Kumaji must be flexible. It should be possible to implement support for new subtitle formats without providing more than a parser for them. If a format requires special rendering support not present, it must be possible to add that without taking the entire system apart and jeopardising the previous goals. It should also be possible to use Kumaji as a framework to write custom special-purpose renderers. For example, one can imagine creating a Lua interface for Kumaji's internal functions and use that for scripting advanced karaoke effects.

Help wanted!

Currently, Kumaji is pretty much my own pet project, but I would really like to have some help. What's needed right now is data structure design. If you want to help I expect you to have some knowledge of digital typography, the intricacies of Unicode complex scripts and the Bidi algorithm, OpenType, as well as general data structure and algorithm design. Or you should be have or be able to take an interest in those topics and read lots and lots about them! (It's interesting stuff, really!)

Kumaji is being written in C++ using just the STL (no TR1 libraries, boost or otherwise) so if you want to help reviewing or writing code you should be familiar with that.

The project is being hosted at under the name kumaji.