Network packet loss due to audio usage, Win7

[Ian (PassMark)]

The behavior occured with the Sound test and also without the sound test. It may be possible that the sound test makes it more likely, but my testing didn't suggest this.

[TomH]

Nearly 2 years since the last post, I'm wondering if there is a conclusion/fix/work-around of this byte drop issue?

I am not using BurnIn test but rather testing UDP in PerformanceTest 7.0-Advanced-Networking. Two units in server/client pair with one CAT5 network cable in between, TCP test always pass but UDP always has byte drop.

Any suggestion you may share? Thanks

[Ian (PassMark)]

This post referred to an issue where constant UDP broadcasting could impact the BurnInTest V5/V6 standard ICMP network test. Changes were made to the V7.0 BurnInTest standard network test that I beleive resolves this issue.

I don't beleive this is in anyway related to the behavior you are seeing with the PerformanceTest Advanced network test. If you still have the problem, please email us or post in the PerformanceTest forum.

[Comark Corp]

This post referred to an issue where constant UDP broadcasting could impact the BurnInTest V5/V6 standard ICMP network test. Changes were made to the V7.0 BurnInTest standard network test that I beleive resolves this issue.

I don't beleive this is in anyway related to the behavior you are seeing with the PerformanceTest Advanced network test. If you still have the problem, please email us or post in the PerformanceTest forum.

Hi Ian,

I can confirm that the problem still exists with burnintest v7.0 build 1005.
FYI: We are still using version 5.0 for production, but if we were to consider moving to Bit 7, this would probably keep us away. If you still have the bitbreak.exe executable, you can see for yourself.

Jay

[Ian (PassMark)]

Jay,

I just re-tested with the bitbreak exe and it does seem the issue where 3rd party software packets on the test network can still cause false timeout errors. From our testing we do however believe that this is reduced in V7.0 when multiple network ports are tested. I will add this back on my to do list to investigate further.

Thanks.
Ian

[Comark Corp]

Jay,

I just re-tested with the bitbreak exe and it does seem the issue where 3rd party software packets on the test network can still cause false timeout errors. From our testing we do however believe that this is reduced in V7.0 when multiple network ports are tested. I will add this back on my to do list to investigate further.

Thanks.
Ian

Hi Ian,
I think this issue got forgotten, and it still causes us grief from time to time. Now that we purchased a site license, we'd really like to get this resolved. Anything you could do for us would be greatly appreciated.
Jay

[Ian (PassMark)]

Hi Jay,

I did do some work on this a while back. I have had another look at it today and have a new debug build with some minor changes and additional debug logging. I will send you an email with the link and to see if I can get the source code for bitbreak.exe from our past discussions so I can test again with it here.

Thanks.
Ian