NetBeans 6 delivers great updates to the Matisse GUI builder. Spend a few minutes with Roman Strobl and get an expert briefing on what's new and what has changed. (sponsored)
In this, the third and final installation of Andres' Introduction to Groovy series, you learn about how Groovy handles variable numbers of arguments, named parameters, currying, and more about Groovy operators. Including, some new operators.
Swing Fuse (actually just Fuse), is a framework designed to make it easier to create your own custom desktop components. In this article, Daniel Spiewak shows you how to get started and provides sample source code you can download.
Willam Louth shows how he uses JXInsight Probes to investigate probable performance issues with code bases that he is not familiar with. He also highlights possible pitfalls in creating a benchmark, as well as in the analysis of results.
GNU Classpath version 0.92 has been released. Among the highlights of this release a are full 2D graphics implementation based on Cairo that is enabled by default. This enables the use of applications like JEdit, FlickrBackup and JFreeChart out of the box.
Also new in this release is the inclusion of an applet viewer and plugin that can be embedded in web browsers or other applications.
Great to see progress on classpath ... classpath is really great. I hope they get the speed-problems of the cairo based implementation fixed, however it continues to become more and more useful.
However running binary apps with it reminds me a lot at the times when I was using wine
> Great to see progress on classpath ... classpath is
> really great. I hope they get the speed-problems of
> the cairo based implementation fixed, however it
> continues to become more and more useful.
The fact that more and more real applications are starting to work well is really important. I've had success with applications like StatCVS, FindBugs, Checkstyle, JFreeChart (of course!) and JEdit, and as more people are drawn in to try out these applications, we'll get more bug reports and the snowball will start to roll faster...
> However running binary apps with it reminds me a lot
> at the times when I was using wine ;)
I'm pretty sure you have filed bug reports already...but please keep them coming in. It helps the GNU Classpath developers to know what is being used and what needs fixing. Lately it seems the majority of patches on the mailing list are bug fixes and performance enhancements - a lot of work is being done to improve the quality of GNU Classpath, and I think it shows in this latest release.
Is this a full Java2D or Graphics2D implementation?
How can you tell which API's/Classes are implemented (or supposed to)?
Graphics2D is obviously a much smaller portion than the full Java2D that includes really complex elements such as font kerning, shapes, transforms and quite a few other features.
Is there any place I can check this out?
Also is there some guide on how to try classpath's latest version for Windows users?
Shai Almog
vPrise Software makers of vPrise Workgroup
http://wg.vprise.com/ founder of bean-properties the leading OSS properties implementation in Java https://bean-properties.dev.java.net/
I recall that he made a slight modification to the code (to bypass something that wasn't working in the startup), but I'm struggling to find the reference right now (it will be online somewhere). When I find it I'll post an update, probably on Monday when I get back to my office. I did run Roman's modified version and remember being quite impressed by the output. Performance wise, it was usable but still some way behind your (Sun's) implementation.
I think the only benchmark in the suite that can't be run against GNU Classpath right now is the 'batik' one. The 'chart' benchmark is working, I ran it a few weeks back. Results are on a development machine at my office, so I'll have to post them later (my memory isn't that great, but I think GNU Classpath ran at about half the speed of the latest Sun JRE for the chart benchmark).
> Alternatively, are there figures for typical
> operations on JFreeChart?
See the DaCapo benchmark suite above. In addition, I plan to write some benchmarks for JFreeChart soon so I can work on some performance improvements (in JFreeChart) that are backed up by real (and hopefully reasonable) statistics.
Re: Is this a full Java2D or Graphics2D implementation?
> How can you tell which API's/Classes are implemented
> (or supposed to)?
> Graphics2D is obviously a much smaller portion than
> the full Java2D that includes really complex elements
> such as font kerning, shapes, transforms and quite a
> few other features.
I don't know about font kerning, but all the shapes and transforms (AffineTransform) are implemented, stroking, compositing and anti-aliasing are done, and GradientPaint works. I think TexturePaint may be missing still, and there are some known bugs in the Area class.
> Is there any place I can check this out?
I guess the Java2Demo mentioned in another thread will give you a good idea of what works and what doesn't.
> Also is there some guide on how to try classpath's
> latest version for Windows users?
Hmmm...there are not many GNU Classpath developers working with Windows, so I'm not sure if this is possible or not. Probably the place to ask is on the Kaffe mailing list - Kaffe uses GNU Classpath and is or was runnable on Windows (I don't know the current state).
Re: GNU Classpath 0.92 Released With Full 2D Graphics
> It would be interesting to have comparative
> benchmarks between ClassPath and Mustang.
Well Classpath itself is only the class-library so an additional factor is the used jvm.
I guess comparative benchmarks would'nt be that useful, Mustang is very well tuned and classpath is not. Furthermore classpath relies on Cairo which seems to be (for now) a bottleneck. Drawing very primitive stuff like lines is about 15x slower on classpath compared to Mustang.
Re: GNU Classpath 0.92 Released With Full 2D Graphics
> I guess comparative benchmarks would'nt be that
> useful, Mustang is very well tuned and classpath is
> not.
I think the numbers would be interesting, all the same. I'd like to know how far behind Classpath is, in performance terms, and see how that changes over time.
Let me also add that I have a lot of respect for the Sun engineers that have managed to get pretty spectacular performance out of the Java runtime, all things considered. They've set the bar high, and it won't be easy getting Classpath up there (but of course we will try).
Re: Is this a full Java2D or Graphics2D implementation?
I haven't worked in the Windows port in a while, but I've got a stack of patches from shudo sitting in my mailbox for improving Kaffe on cygwin/win32. I'd assume that the most interesting combination on windows would be kaffe + cygdll + native qt4 peers, but I haven't tried it out myself yet.
Re: Is this a full Java2D or Graphics2D implementation?
> the most interesting combination on
> windows would be kaffe + cygdll + native qt4 peers,
"most interesting"? Does that mean "most likely to work" or "most likely to fail in interesting ways", or "most (interesting) work to configure", or what?
Andy Tripp, CTO and Founder Jazillian
- Legacy to 'natural' Java.
Re: GNU Classpath 0.92 Released With Full 2D Graphics
> Let me also add that I have a lot of respect for the
> Sun engineers that have managed to get pretty
> spectacular performance out of the Java runtime, all
> things considered.
Indeed, they've done some very impressive work over the years on the performance of their implementations.
GNU Classpath 0.92 Released With Full 2D Graphics
URL: GNU Classpath
At 3:52 PM on Aug 10, 2006, Michael Urban wrote:
Fresh Jobs for Developers Post a job opportunity
Also new in this release is the inclusion of an applet viewer and plugin that can be embedded in web browsers or other applications.
The full release announcement is available here
Also, screenshots of JFreeChart using gnuclasspath are available here.
31 replies so far (
Post your own)
Great :-)
Great to see progress on classpath ... classpath is really great. I hope they get the speed-problems of the cairo based implementation fixed, however it continues to become more and more useful.However running binary apps with it reminds me a lot at the times when I was using wine
lg Clemens
Re: Great :-)
> Great to see progress on classpath ... classpath is> really great. I hope they get the speed-problems of
> the cairo based implementation fixed, however it
> continues to become more and more useful.
The fact that more and more real applications are starting to work well is really important. I've had success with applications like StatCVS, FindBugs, Checkstyle, JFreeChart (of course!) and JEdit, and as more people are drawn in to try out these applications, we'll get more bug reports and the snowball will start to roll faster...
> However running binary apps with it reminds me a lot
> at the times when I was using wine ;)
I'm pretty sure you have filed bug reports already...but please keep them coming in. It helps the GNU Classpath developers to know what is being used and what needs fixing. Lately it seems the majority of patches on the mailing list are bug fixes and performance enhancements - a lot of work is being done to improve the quality of GNU Classpath, and I think it shows in this latest release.
http://www.jfree.org
Re: GNU Classpath 0.92 Released With Full 2D Graphics
Great to see such progress!How about the Java2Demo from the Sun's JDK?
Does it run? The demo covers a lot of Java2D
functionality.
Dmitri
Java2D Team
Re: GNU Classpath 0.92 Released With Full 2D Graphics
It would be interesting to have comparative benchmarks between ClassPath and Mustang.Does anyone know of any? Do the ClassPath people have some?
Alternatively, are there figures for typical operations on JFreeChart?
Ian
Is this a full Java2D or Graphics2D implementation?
How can you tell which API's/Classes are implemented (or supposed to)?Graphics2D is obviously a much smaller portion than the full Java2D that includes really complex elements such as font kerning, shapes, transforms and quite a few other features.
Is there any place I can check this out?
Also is there some guide on how to try classpath's latest version for Windows users?
Re: GNU Classpath 0.92 Released With Full 2D Graphics
> How about the Java2Demo from the Sun's JDK?> Does it run? The demo covers a lot of Java2D
> functionality.
Thanks for your interest, Dmitri!
Roman Kennke posted some screenshots of the Java2Demo:
http://kennke.org/blog/?p=9
I recall that he made a slight modification to the code (to bypass something that wasn't working in the startup), but I'm struggling to find the reference right now (it will be online somewhere). When I find it I'll post an update, probably on Monday when I get back to my office. I did run Roman's modified version and remember being quite impressed by the output. Performance wise, it was usable but still some way behind your (Sun's) implementation.
http://www.jfree.org
Re: GNU Classpath 0.92 Released With Full 2D Graphics
> It would be interesting to have comparative> benchmarks between ClassPath and Mustang.
There has been some micro-benchmarking done on the Java2D code:
http://developer.classpath.org/pipermail/classpath/2006-June/001083.html
which shows that GNU Classpath has some catching up to do.
> Does anyone know of any? Do the ClassPath people have
> some?
The DaCapo benchmark suite looks interesting:
http://www-ali.cs.umass.edu/DaCapo/gcbm.html
I think the only benchmark in the suite that can't be run against GNU Classpath right now is the 'batik' one. The 'chart' benchmark is working, I ran it a few weeks back. Results are on a development machine at my office, so I'll have to post them later (my memory isn't that great, but I think GNU Classpath ran at about half the speed of the latest Sun JRE for the chart benchmark).
> Alternatively, are there figures for typical
> operations on JFreeChart?
See the DaCapo benchmark suite above. In addition, I plan to write some benchmarks for JFreeChart soon so I can work on some performance improvements (in JFreeChart) that are backed up by real (and hopefully reasonable) statistics.
http://www.jfree.org
Re: Is this a full Java2D or Graphics2D implementation?
> How can you tell which API's/Classes are implemented> (or supposed to)?
> Graphics2D is obviously a much smaller portion than
> the full Java2D that includes really complex elements
> such as font kerning, shapes, transforms and quite a
> few other features.
I don't know about font kerning, but all the shapes and transforms (AffineTransform) are implemented, stroking, compositing and anti-aliasing are done, and GradientPaint works. I think TexturePaint may be missing still, and there are some known bugs in the Area class.
> Is there any place I can check this out?
I guess the Java2Demo mentioned in another thread will give you a good idea of what works and what doesn't.
> Also is there some guide on how to try classpath's
> latest version for Windows users?
Hmmm...there are not many GNU Classpath developers working with Windows, so I'm not sure if this is possible or not. Probably the place to ask is on the Kaffe mailing list - Kaffe uses GNU Classpath and is or was runnable on Windows (I don't know the current state).
http://www.jfree.org
Re: GNU Classpath 0.92 Released With Full 2D Graphics
> It would be interesting to have comparative> benchmarks between ClassPath and Mustang.
Well Classpath itself is only the class-library so an additional factor is the used jvm.
I guess comparative benchmarks would'nt be that useful, Mustang is very well tuned and classpath is not. Furthermore classpath relies on Cairo which seems to be (for now) a bottleneck. Drawing very primitive stuff like lines is about 15x slower on classpath compared to Mustang.
Mfg Clemens
Re: GNU Classpath 0.92 Released With Full 2D Graphics
> I guess comparative benchmarks would'nt be that> useful, Mustang is very well tuned and classpath is
> not.
I think the numbers would be interesting, all the same. I'd like to know how far behind Classpath is, in performance terms, and see how that changes over time.
Let me also add that I have a lot of respect for the Sun engineers that have managed to get pretty spectacular performance out of the Java runtime, all things considered. They've set the bar high, and it won't be easy getting Classpath up there (but of course we will try).
http://www.jfree.org
Re: GNU Classpath 0.92 Released With Full 2D Graphics
I wonder what is the impact on JNode (a full java OS that uses classpath)? Does this mean I can run jedit on JNode?Re: Is this a full Java2D or Graphics2D implementation?
I haven't worked in the Windows port in a while, but I've got a stack of patches from shudo sitting in my mailbox for improving Kaffe on cygwin/win32. I'd assume that the most interesting combination on windows would be kaffe + cygdll + native qt4 peers, but I haven't tried it out myself yet.cheers,
dalibor topic
GNU Classpath - Core Libraries
IRC: irc://irc.freenode.org/#classpath | irc://irc.freenode.org/#kaffe
Re: Is this a full Java2D or Graphics2D implementation?
> the most interesting combination on> windows would be kaffe + cygdll + native qt4 peers,
"most interesting"? Does that mean "most likely to work" or "most likely to fail in interesting ways", or "most (interesting) work to configure", or what?
Re: GNU Classpath 0.92 Released With Full 2D Graphics
> Let me also add that I have a lot of respect for the> Sun engineers that have managed to get pretty
> spectacular performance out of the Java runtime, all
> things considered.
Indeed, they've done some very impressive work over the years on the performance of their implementations.
cheers,
dalibor topic
GNU Classpath - Core Libraries
IRC: irc://irc.freenode.org/#classpath | irc://irc.freenode.org/#kaffe