On Thu, Nov 11, 2010 at 7:22 PM, Casper Bang <[email protected]> wrote:

> > Ah, you've put it much better than I did!
> > That's it exactly.  Why target Oracle's Java optimised JVM or
> > Microsoft's C# optimised CLR?  (Apart from "because they're and
> > everywhere".)  Isn't there an open VM suitable for hosting these open
> > languages?
>
> Arguable the CLR is less optimized for C# (and interpretation), than
> the JVM is for Java. The JVM  has 4x overloaded bytecode operators for
> multiplication, division, addition, subtraction, negation and
> modulus... one for each primitive Java type (int, long, float and
> double). This is a result of the initial design to favor Java
> interpretation over compilation. Yet both the JVM and CLR are tied to
> the objective paradigm and are stack-based designs.
>
>
Having different byte codes for different types of an operation as in the
JVM while the CLR uses associated meta to determine types is irrelevant. At
some stage of verification/loading the host is quite capable of knowing
exactly what types appear in what slots etc. Stack based designs are a
lowest common denominator. From that start, one can more easily extrapolate
or improve given a particular host cpu architecutre and quirks.

-- 
You received this message because you are subscribed to the Google Groups "The 
Java Posse" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/javaposse?hl=en.

Reply via email to