Waldemar wrote:

> Actually, the reality is Java is plenty slow in many applications.

I don't know about that. I have seen some nasty, klunky Java programs, but my experience is that they are the exception rather than the rule. Netbeans and Eclipse both started out as klunkers, but neither is today.


> There is JNI for a reason.

JNI was developed for compatibility with legacy code. It was used in the early days as a performance optimization tool, but these days not so much. Even for Java games, JNI is used only to access native APIs like OpenGL, OpenAL, or DirectInput. The old Dirty Java article at Gamasutra lost its relevance with the release of Java 1.4.

> Never mind cases where Java is not even considered (fast servers,
> OS internals, communication, graphics, driveres, embedded, etc, etc.)  As soon as
> Java reaches C/C++ speed, C++ will disappear.  Not too worry, won't happen any
> time soon.
> 

I don't expect people to write device drivers or any other performance critical code in Java :) That's not what Java is intended for. It *is* quite prevalent in embedded systems -- your cell phone is most likely equipped with JME, and you may even have some form of Java in your DVD player. The PS3 is equipped with a limited version of Java.

Java isn't a bare-metal systems language, but it isn't *slow*. The reality is somewhere in the middle. It's slower than C++ in most cases, yes. But there are plenty of applications that can be written in Java where a 5%, 10% or even 20% performance difference isn't going to matter. With my interest in games, I love to point out Java games like Tribal Trouble, Puzzle Pirates, and Bang! Howdy, all of which are 3D games, or the shooters developed by Puppy Games which are 2D on OpenGL. Since the JVM is bundled with each of them, most people don't know or don't care that they are written in Java. Performance is just fine in each of them.

The preceding paragraph is the typical response you are going to get from a Java programmer when you say Java is slow. There's a lot going on behind the scenes in the JVM, as Walter mentioned, and Java programmers understand that. There's a trade-off between safety/security and speed. They accept that. Do you think anyone at all would choose Java if it truly was a snail at everything? They like the ease of writing concurrent applications in Java compared to other languages. They like Java's inherent networking capabilities. They like the advanced development tools available for the language, such as the free profilers, full-featured IDEs, and the ability to choose between and fine-tune for several different GC implementations to get the best performance possible. Some hardcore Java programmers might be willing to write a video or audio codec in Java, but most are more grounded and understand that Java isn't suited for that.

So my point is that Java programmers use Java for a reason. Ask any of them and they will surely point out faults they find in the language (Generics being a popular one). But very few who choose to use Java will cite performance as a negative for *their purposes*. So to entice a Java programmer to come to D, the speed argument just isn't going to fly.

To court Java programmers, you have to convince them that D's features are shinier than Java's. You have to show them an advanced tool chain that lets them monitor every aspect of an application's performance. They want to be able to dynamically instantiate objects via a class name string. They want easy use of web services, support for web applications, and not to worry about porting issues across different platforms.

No matter your opinion of Java, or Walter's, people who use it on a regular basis see it in an entirely different light.

Mike Parker wrote:
> 
> So my point is that Java programmers use Java for a reason. Ask any of them and they will surely point out faults they find in the language (Generics being a popular one). But very few who choose to use Java will cite performance as a negative for *their purposes*. So to entice a Java programmer to come to D, the speed argument just isn't going to fly.

Agreed.  Programmers who are really serious about performance over other considerations will likely not be using Java, so this isn't a convincing argument here.  My personal problem with Java is that I hate the syntax.  Although it's fairly similar to C++ and even D, I find the nuances that are different to make it tremendously unwieldy.  And the deal-breaker is that Generics are complete garbage--I simply refuse to go without a decent way to express algorithms in a generic manner after having used C++ for so long.

> To court Java programmers, you have to convince them that D's features are shinier than Java's. You have to show them an advanced tool chain that lets them monitor every aspect of an application's performance. They want to be able to dynamically instantiate objects via a class name string. They want easy use of web services, support for web applications, and not to worry about porting issues across different platforms.

If I were to use Java, these would be the reasons.  And I think D has a long way to go here.  There is some library work that seems quite promising re: dynamic libraries (DDL), and Mango is fairly capable for server/web programming, but D is extremely weak in the toolchain department.  Windbg may work with D in Win32, but it's complete crap.  I have yet to use it without my debug session ending because windbg crashed.  The situation is so bad that I've considered developing exclusively on Linux in hopes that GDB support is better.  But fortunately, the times I want to use a debugger are so few and far between that this remains an idle fantasy.  And as for performance, DMD has profiling and code coverage tools built-in, which is a huge bonus in that regard.


Sean

Walter Bright wrote:

> 
> I haven't worked with Java for over 10 years now. But in March I attended "Java Performance Myths" by Brian Goetz, who is very knowledgeable about getting the most out of Java. He indicated that Java  needed another 10 years before it would be able to consistently match C toe to toe. And this is happening despite years of massive investment in Java by lot of very smart people, just to approach what relatively simple C compilers can do. This to me indicates there's a fundamental problem with Java.

I've never heard Brian Goetz speak, but he's a prolific writer. One of his articles, Urban Performance Legends (http://www-128.ibm.com/developerworks/java/library/j-jtp04223.html), is one I often direct people to in Java performance discussions. But the question one must ask when evaluating any language is, what is this language suitable for? When you look at everything Java gives you, the trade-off of not matching performance of C is acceptable for a wide range of applications, but not everything under the sun (no pun intended).

Books like Code Complete and Writing Secure Code exist because languages like C and C++ have no built-in mechanisms to prevent severe errors like buffer overruns. How many times does this happen in server applications or even non-networking libraries? libpng had a patch recently, last year I believe, to fix a buffer overrun. That just shouldn't be happening itn modern software. Java's answer to those issues is certainly extreme, but a lot of people find that beneficial -- particularly for network software. I've written a lot of C code and a lot of Java code. It's much more pleasant to write Java because I can write a lot less of it. I don't mind a bit that it's not going to beat C in the performance arena because I don't need it to for what I use it for.

One of the things I like about D is that it incorporates some of the same mechanisms, but in a more flexible way (i.e. array bounds checking can be turned off). One of the things I miss about Java when I use D is the ability to instantiate any class by name. I've made heavy use of that for some component-based designs in Java. Designing D applications requires me to think differently, much as Java required a new way of thinking when coming from C.

> 
> I think the problem is that Java is just lacking in some needed features - like a full set of basic types, simple aggregates, out parameters, etc. The alternatives are computationally expensive, and so the optimizer has a lot of work to do to reverse engineer them and figure out that all the programmer was doing was a workaround for a POD stack aggregate.
> 
> Java also has a few millstones that make things difficult for optimizers - the required array bounds checking, the VM, etc.

I have no background at all in language design, or optimizers, but I'm guessing it's more than just a language feature issue. Most of Java's critical optimizations happen at runtime during the runtime analysis and compilation phases. Runtime compilation is still a comparatively young technology in practice. There are still a lot of kinks to be worked out in the process, a lot of progress to be made. I assume when Goetz said Java's performance would match C in 10 years, he was specifically referring to the Hotspot VM, or runtime compilation in general.

There are two VMs that ship with the SDK, referred to as the client and server VMs. The client VM has a shorter runtime analysis period than the server VM, so the server VM can make more aggressive optimizations after longer periods of analysis. I've read that it can eliminate array bounds checks in certain cases. Whatever it does, Java's runtime optimizations are built around the language's features. For example, in the latest 1.6 release they just laid the groundwork for escape analysis, which will eliminate some object allocations on the heap and move them to the stack. The full implementation should be available in 1.7. I've even heard rumors that a JSR for a struct type will make it into a future release, which could open the door for more optimizations and, possibly, features like out parameters.

> 
> C++, much promoted for its efficiency, has a subtle problem with efficiency. We all know that the biggest gains in efficiency come from algorithmic improvements. But C++'s strength is in optimizing not the algorithms, but the micro-optimization details of implementing them. C++ code tends to get so complex that once we get it to work, we dare not go tinkering with the overall algorithm, but just concentrate on twiddling with the micro-optimizations.
> 
> I discovered this when working on DMDScript. I had spent a great deal of time tinkering with the details of the C++ version trying to make it go faster. Then I translated it to D, and found myself instead of tinkering with the low level details, I was tinkering with the algorithms. With a few changes there, that would have been hard to retrofit into the C++ code, I had it running faster than the C++ one.
> 
> And that may be D's efficiency trump card over C++ - it's much easier to modify existing, working code in ways that are too risky to do with C++.

But that's a completely different world. When trying to win over C++ programmers, words like efficiency, optimization, and performance carry a great deal of weight. Speed is king. Not so in the Java community. It's a different mindset with different goals. If Java programmers were overly concerned about performance, they wouldn't be using Java. Sure, they want their software to be performant and a good Java programmer knows how to squeeze every ounce of performance out of his code, but performance alone isn't going to sway them.

I think that overall D is going to be an easier sell to C++ programmers than to Java types. They are two very different markets that will require very different strategies.

Andrei Alexandrescu (See Website For Email) wrote:
> Say you wanted to introduce a programmer friend to D. She might know one of D's sibling languages (Java, C, C++, or C#), but you shouldn't rely too much on that. What features of D would you describe first to hook that friend in?
> 
> Andrei

Mostly it is the fact that a programing can be made more reliable and maintainable.

Simple consistent syntax is one part of this. The "It's easy to program in" also is part.

My favorite thing right now is this: Things can be made "intent-ional" (think "funct-ional") sort of a #define on steroids. This comes from use of const and templates. The after doing things like parser generation and ASM-stype bignum with template code I could actually stand reading really makes me think there is something here. After showing the template parser generator to a prof of mine, he said "Quit tempting me" to start learning D, this from a guy that likes Prolog (nothing wrong with that BTW).






p.s. & OT: that might as well be the abstract for the current version of my, "intro to D: why it will take over the world some day" paper:

http://www.uidaho.edu/~shro8822/term_009.pdf

comments welcome.

== Quote from Mike Parker (aldacron71@yahoo.com)'s article
[snip]
> I think that overall D is going to be an easier sell to C++ programmers than to Java types. They are two very different markets that will require very different strategies.

D has no chance against java/C#.  Unless, and I am aware of the diminishing freqency of these cases, performance forces the use of another language.  D would be a candidate for that other language.  It is the only situation where pitching D to java/C# people is not a waste of time.

D is a low level language competing with C/C++/ASM.

It has a chance but I don't know how long the window of opportunity will last.

Mike Parker a écrit :
> Being a long time Java programmer, I strongly disagree with you and Waldemar both. Speed is only an issue for people who don't use Java, or for those who don't really know how to properly write software with it.

Note that the "people who don't know to write software" include even Sun: in Solaris9, Sun rewrote the administration tools in Java, the result?
Well, the tools were so slow, that I learned to use the command-line tools instead of the GUI, because they were so damn slow that each time I used these tool, I was really pissed-off by the slowness..

It's probably possible to write fast Java apps, but even Sun don't manage to do it, and all the Java apps I've used have annoyed me by being too slow, so it's definitely not trivial: I think that using D it's easier to get fast apps.

Regards,
renoX

PS: Of course with D, there is the library problem..
I'd really like to have a wrapper of Qt in D..

Waldemar wrote:
> == Quote from Mike Parker (aldacron71@yahoo.com)'s article
> [snip]
>> I think that overall D is going to be an easier sell to C++ programmers
>> than to Java types. They are two very different markets that will
>> require very different strategies.
> 
> D has no chance against java/C#.  Unless, and I am aware of the diminishing
> freqency of these cases, performance forces the use of another language.  

Also in cases needing ASM access or where use of a VM is not an option.

The first case won't go away and the second could come into play when, for an example, it takes longer to load the VM than it does to run the app, or when size is a BIG issue. I'll admit that D doesn't yet excel in either of these last two cases, but it will (I hope), I don't think Java/C# ever will.

> D would
> be a candidate for that other language.  It is the only situation where pitching D
> to java/C# people is not a waste of time.
> 
> D is a low level language competing with C/C++/ASM.
> 
> It has a chance but I don't know how long the window of opportunity will last.

> p.s. & OT: that might as well be the abstract for the current version of my, "intro to D: why it will take over the world some day" paper:
> 
> http://www.uidaho.edu/~shro8822/term_009.pdf
> 
> comments welcome.


BTW: that version is the current version that I had when I noticed that it might have some bearing on the topic. I'm sure spelling and grammar errors abound, so please ignore them.

renoX wrote:
> Mike Parker a écrit :
>> Being a long time Java programmer, I strongly disagree with you and Waldemar both. Speed is only an issue for people who don't use Java, or for those who don't really know how to properly write software with it.
> 
> Note that the "people who don't know to write software" include even Sun: in Solaris9, Sun rewrote the administration tools in Java, the result?
> Well, the tools were so slow, that I learned to use the command-line tools instead of the GUI, because they were so damn slow that each time I used these tool, I was really pissed-off by the slowness..

The new Sun debugger (SunStudio) is the same way.  It has a lot of nice features, but is so slow I end up doing most of my debugging in emacs with dbx.  One of these days I'll take the time to turn off the bits that are killing performance and give it another try.


Sean

Walter Bright wrote:
> I think the problem is that Java is just lacking in some needed features - like a full set of basic types, simple aggregates, out parameters, etc. The alternatives are computationally expensive, and so the optimizer has a lot of work to do to reverse engineer them and figure out that all the programmer was doing was a workaround for a POD stack aggregate.

Could you expand on the above items?  What basic types are missing? Does a "simple aggregate" refer to structs/tuples?  What does POD stand for, and what is a POD stack aggregate?  Could you show a code example in D that performs badly in Java?  Pardon my ignorance.

-Jeff