With many customers having speed problems. I was wondering if they could use different test? Maybe 12MB or 15MB.
This could help them save DATA.
This was discussed some time ago, and the chance that someone does a test with 12MB, if they are getting 30Mbps or higher, testmy.net will very often retest using a 200MB package, so we settled with the 25MB package as that's the lowest you can really go if you get anywhere near plan speeds to keep testmy.net from forwarding to a higher package.
That aside, the Hughesnet test it self uses large package sizes as well, testmy.net on average will use less data than even the Hughesnet test IMO.
I'm not talking about those that are getting 30Mps or higher. I'm talking about those getting between 1&5 Mbps or slower.
Unlikely the protocol will be changed even in those circumstances as they could be hit with burst speeds showing higher than what they are getting, and vice verse.
Just for clarification:
The main issue with a testmy test with an unspecified amount is the time it takes to complete. If testmy sees it completed too quickly, it bumps it up to get a sample of a longer duration.
Naturally, the problem with this is that anything reported here becomes an unstandardized mess and impossible to correlate.
You eliminate one of those variables by stating a firm sample size (25M down/3M up) and disable "forwarding" to keep it from advancing to a larger size. Usually selecting 25/3 will automatically disable forwarding.