Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
November 1, 2014
arrowPress Releases
November 1, 2014
PR Newswire
View All
View All     Submit Event





If you enjoy reading this site, you might also want to check out these UBM Tech sites:


 
Current of Android-x86 PC OS interview with Chih-Wei Huang
by Ruthaniel van-den-Naar on 06/12/14 04:21:00 pm   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.

 

Disclaimer: Sorry for my English is crappy. // I would appreciate if someone would fix it.
Im man of thoughts, not fancy pictures, if you seek for pictures go elsewhere.

Introduction
It is not good to rely not only on a single supplier, because you become its hostage. Therefore, all who are interested in computers, we seek an alternative to Windows.
Despite to my extensive experience with Linux distributions and MacOS X, I have concluded that the final winner and Windows challenger could be Android-x86.

Why Android-x86(link)?

  • Runs on PC(x86).
  • Original Android runs on ARM.
  • Android - Teoreticaly could run on every present and future device, PS4 and Xbox one included (in pure theory).
  • Is everytime better develop for only one OS (even with some platform differences), that for bunch of them.
  • Mac OS or iOS are too closed and fixed on Apple device and its unlikely to change it.
  • Linux market is realy fragmented and its development is relatively slow.
  • Android - is without doubt OS with faster development speed and it counts.
  • Android is open and free.
  • Today, when i can buy good living room pc for 200$ i realy dont wan to pay next 100 bucks to Microsoft for crippled OEM licence, or 200$ for retail (HW independed licence)

So I decided to interview its main developer Chih-Wei Huang, see below.

> You are probably worlds biggest expert in Android-x86. How it happens, could you tell something about you and your history?
First of all, please use our (only) correct project name: Android-x86. We dislike androidx86 since there was a pirate site called androidx86.org that just copied our effort and ask for donation (though it seems to be down now). You can find more the history in the discussion of our forum.
About me, I may not be the biggest expert. But I may be the only one developer who worked for android-x86 since the past over 4 years. Before that, I have been a linux user for almost 20 years. As a free and open source fan, I have tried to contribute the FOSS community since 1997.I have worked for linux i18n/l10n, voip projects (GnuGK/Asterisk), KDE/Gnome and multimedia projects like GStreamer and bugfixes for many projects, including linux kernel.
You may find more info about me from the wiki page: http://en.wikipedia.org/wiki/Chih-Wei_Huang

> Were you see long-term real life usage of Androidx-86 devices? Do you develop it only for some geeks, for developers or for normal users?
I can't judge the long-term thing. At first I began the android-x86 project just because. I want to study the android platform and its porting, and I'm happy to share my results.
   After the project published I received more feedback and encouragement so I continue working on it. I even couldn't believe it has lasted for over 4 years.
It's unfair to say the project is just for geeks. But I agree it's not friendly enough for end-users. I either didn't spend much effort to make it closer to normal users because of my time and manpower (see the answer of the next question). If someone is going to improve it, I'm glad to see.

> How big team of developers working on this project?
There is no a real "team" of the developers. As said before, I'm the only one who works for the project since it began. The 4.4-RC1 porting is almost finished by me solely. There are some developers joined and  contributed something, but then they just disappeared. I'm not sure why. But it should be a normal situationn for an open source project.
Except me, the most important contributor of the project is Chia-I Wu (Olv) who ported the mesa project to android-x86 as the OpenGL ES implementation. It allows us to use the hardware acceleration via Intel and AMD's GPUs. But he is also busy on his job so I only bothered him if we encountered critical issuesthat I can't solve myself.
Kyle Evans (fuzzy7k) is another developer who joined since 2011 who contributed new targets and bugfixes. He continues to provided patches to the 4.4-RC1 release. There is a group (2-3 persons I think) from German(?) who helped to port android-x86 4.0 (ics-x86) to the WeTab tablet. They contributed several important improvements of the HALs. But I haven't heard from them for a long time.

> How many of users know you think that tried Android-x86, at least through live distro and how many installed it into physical computer? Do you have some stats about usage?
I don't have the stats and I don't really care about that. Just checked. The 4.4-RC1 release has 584,441 downloads and the 4.3-test release has 1,365,182 downloads.

   > Are you in contact with Google? Do you know why Google isn't developing Android-x86 for desktops itself? Is affair of Microsoft or something else? ChromeOS looks like duplicity.
I have no official contact window of Google. But I have many friends in Google including the android team. I'll also contact Google android engineers directly for issues clarification or patches submission. I don't know Google's plan. But as you said, ChromeOS is probably the right one for its desktop roadmap.
   This is a common guess.

> Could Google somehow legally stop Android-x86 development or not? It could be like book burning in dark ages, couldn't?
Android is licensed under an open source license -- the Apache license. Nobody could stop anyone to modify and/or redistribute it, including Google itself.

> From my point of view, main features which are missing in long-term in current build for desktop usage are multi-monitor support, run app in window, Wine and Virtualization. Do you see it similar or is there another important features which are needed? (im here know speaking about driver - we will discuss this later)
The features you mentioned are somewhat useful to desktop usage. But whether they are necessary depends on the applications. Unfortunately due to lack of manpower, none of the features you said is in the todo list. Personally I'm interesting in supporting multi-monitor. But no time to study the necessary technical details.

> How hard is development of Android drivers, could you simply use and recompile Linux drivers or reuse some part of them?
The project is not aimed to develop android drivers. We use the existing drivers of the linux kernel. They usually just work.

> How is present state of support you all major GPU vendors (Intel, Nvidia, AMD)?
I have written an article about the supported and unsupported GPUs. Take a look:
https://groups.google.com/d/msg/android-x86/RaIP7qcVitw/rRXZw_o-ZxAJ

> How is present state of drivers for 3g modems?
There is a preliminary support for some Huawei USB 3G dongle in 4.0-r1. But it's not be integrated into 4.4 yet. There is no strong demand to support it.

> With my Android 4.4 and Intel mesa drivers, i had experienced lots of black screen or partly displayed object in games, do you know where is problem? Especially Unity3D games seem to be broken.
I guess you are using i915 GPU. Some GLES commands are not supported or buggy. Unfortunately both Intel and Mesa developers gave up that driver so don't expect any improvement in the future. For users of i965 or Radeon GPUs, they are supported much better.

> Other problem of current version of this OS is gestures support, is there way how to simulate multi finger gesture with physical mouse or touchpad? I was unable to do it in games.
No plan to simulate it. It's not difficult to support touchpad. One just has to provide a correct idc file.

> How new features we could expect in near releases?
Simply speaking, no. There are two major reasons. First, this is a porting project. Porting android to our target devices and make it work well is the first priority. Developing new features is not. Second, we have no developer, as you should have already known.
What I'll sustain to do is updating kernel, mesa and so on from upstream to ensure more bug fixes and supporting more devices.

> Could we donate, or somehow support this project?
Sure. See http://www.android-x86.org/donate

> Do you want add something at the end you this interview?
We need more developers. If you are interesting in the project, try to contribute back bug fixes and improvements.

Update: Day after releasing of this arcitle was announced ConsoleOS Android-x86 based project, i dont know too much, about this project. Target are similar like Android-x86, but i can tell you that first table on KS project page is full of lies.


Related Jobs

Twisted Pixel Games
Twisted Pixel Games — Austin, Texas, United States
[10.31.14]

Senior Graphics and Systems Engineer
Twisted Pixel Games
Twisted Pixel Games — Austin, Texas, United States
[10.31.14]

Mid-level Tools and Systems Engineer
Sega Networks Inc.
Sega Networks Inc. — Madison, Wisconsin, United States
[10.31.14]

Mobile Game Engineer
Forio
Forio — San Francisco, California, United States
[10.31.14]

Web Application Developer Team Lead






Comments


Kujel Selsuru
profile image
Your english is a little rough but not unmanagable, thank you for posting this was an interesting read.

I am myself in favor of open source and agree with the idea of a viable alternative to Windows but since Android applications are written in Java it may not be a great option. Unless Android-x86 supports C/C++ among other languages it will not be a sufficient contender. Even then it should not replace Windows only balance it and Windows do the same in turn.

Jean-Claude Cottier
profile image
There are already some Android devices based around x86. Intel is working hard to convince NDK (c++) developers to compile their apps for x86 too (fat binary: arm+x86 in the same APK). All my apps have a tiny layer of Java, but everything else is C++. They are all compile for ARM & x86. There is also a way to have 100% C++ project (but in reality, a tiny bit of Java will still handle the communication with the system, but it is hidden from the devs).

Kujel Selsuru
profile image
That is acceptable, Java is just a pain to work with.

Ruthaniel van-den-Naar
profile image
If you are using some advanced middle ware, you dont need to care about lowlevel API. Yeah, deep under hood of Android is Java like engine, but there is also support for C++. In case of bigger game middleware you are still using C++(Unreal 4) or C# (Unity3D).
Projects like Xcom for Android shows, that Android platform performance is good enough, so you dont need to care about Java limitation in general, i think that Google is still working on bigger platform effectivity.


none
 
Comment: