I've attached some information from an aJile contact.  He also
mentioned that Dave Hardin, their chief technical officer, will
be at SigAda.


--- begin e-mail excerpt ---
> Has anyone run the J code output from an Ada compiler on this

Yes.  So far I have only run a very simple app (e.g. hello, world) using
JGNAT as the Ada compiler, which produces .class files as its output.
The .class files then are processed by our static linker and formatted
into a native JEM executable.

> or a similar chip?

There are only weak imitations of the JEM ;-)  Seriously, the JEM chip
has several technical competitive advantages as well as the low power
consumption advantages.

> If so, how does it compare with a standard Ada executable on a
> Pentium or a Sparc?

I have not run any significant apps with JGNAT as of yet.  However,
ACT is interested in having someone at aJile who is using JGNAT,
so now might be a good time for me to put together some benchmark

If you would like to point me to some "standard" Ada benchmarks,
I would be willing to run them on our aJ-PC104 evaluation board and
produce some timing information.

I would be willing to carry on email correspondence with you and/or
Robert Leif, but at the moment I don't have the time to get
entangled in a Team-Ada discussion ... if I'm going to spend time
on this, I would like to come up with some results that would be
useful to aJile, ACT, the Ada community, and the Java community.