I posted the following in the bug hunt forum:

Var and Var Array send bug

It would be really useful if someone, anyone could double check our findings.

You can go to the above thread for further instructions.

This won't take but a few minutes of your time and working these bugs out is of benefit to the whole 3DGS MP community.

Just go to the thread, download the code (less than a meg), run the server, run the client, drop the wait() time on the clients by pressing the down arrow key and thus speeding up the network send, and report back in.

See how many clients and how fast you can push the system before it crashes... and if it doesn't post that as well.

Please, just a few of you testing this could make all the difference for JCL and I!

Hi Ricardo,
WED freezes when I try to open the server or client, so I couldn't test it. Is this a really big level? I let it be for 10 minutes, but it doesn't seem to do anything.

WED crahes when I open the server or the client level...

nipx

Ok, I run the test after adding test_string

I can set the wait to 0, but on the server the recived packets keep on 0

Hi Fastlane,

I tried it as well. 3 clients all set to zero, all stable.

I'm not sure if I'm actually getting the throughput needed to break it. I don't have 3 computers to test with, all my clients are on one computer, server on another. Also tried all 4 on the same server. Still Stable.

Is it the reliable bps on the server that you are getting to go over 10k?

Thanks giorgi.

do me an extra favor. Press the arrow down key a few more times on each client. You should see the throughput increase and try to break the 10k barrier.

Gonna be testing it more this week. After SED broke down and gave me an error, I did a repair and couldn't get the breakdown to happen again. It was sporadic to begin with and it "could" do something with a broken version... like I said more testing is needed but if anyone can back me up with tests on of there own that would be great!