Tuesday, May 13, 2025

Re: IEEE 754 mode for Java

> Are you able to instrument the simulation at some kind of checkpoint

Good call Colin, thanks.  Managed to create a simple(ish) simulation that fails, and it tracks the game frame and cars position.  There is a clear reproducable point where the JVM starts to differ from JavaScript.

Screenshot 2025-05-14 111708.png

Still diving into the code to make sense of why, but progress!  Thanks again.

On Wednesday, 14 May 2025 at 5:26:28 am UTC+10 Jens wrote:
The head scratching continues ...
 
As a last resort: Add -XX:+PrintFlagsFinal to your JVM parameters in IntelliJ and in Maven. It shows every flag the JVM has and its final value used by the JVM. Save both outputs to a file and make a diff to see any differences. Maybe some flags with regard to optimization are different and may affect double calculations.

-- J.

--
You received this message because you are subscribed to the Google Groups "GWT Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to google-web-toolkit+unsubscribe@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/google-web-toolkit/c4591cd5-9c64-4ca8-a26d-bf2d693d2ef0n%40googlegroups.com.

No comments:

Post a Comment