cancel
Showing results for 
Search instead for 
Did you mean: 

Large packet sizes - proprietary software? Fact or fiction?

DZad
I've been here awhile

Hi everyone... I've been trying to find an answer online for this claim, but I have yet to discover any useful info, perhaps because the question may or may not make sense and I'm not versed in the understanding of network technology enough to have a real answer...

I have a third-party device that uses Rogers Wireless data. I was told by the provider that uses larger-than-normal packet sizes to transmit data, and because of this, speed tests will not give accurate results... This seems like an odd claim, because from both my understanding, and my research, the packet sizes shouldn't matter... In addition to this there is a limitation of MTU being 2132 bytes possible for lte/4g transmission, including the headers... I suppose it could speed things up a little if the connection is very stable, but I imagine that Rogers has specific settings that these units need to run on. Either way it would make sense that the speed test would read the headers and know how much data is being transmitted regardless...

I guess my main question is am I being mislead? Or does the above scenario actually make sense?

Is it possible that the speed test results are not accurate? Because the max I've been able to produce with this unit is a little over 4 Mbps for both upload and download speeds... This is nowhere near what I was led to believe I would get...

Please help!

1 ACCEPTED SOLUTION

Accepted Solutions

Re: Large packet sizes - proprietary software? Fact or fiction?

Gdkitty
Resident Expert
Resident Expert

While it might come into a LITTLE effect, it shouldnt effect speeds that much.

I have done cellular bases speed tests on phones, etc.. which are able to pull 40-50 mbps on the tests.

(so same data network)

Now, any speeds on wireless, can all be subjective to WHERE the test is done from.  Signal in these cases, can 100% make a huge difference.   standing 5 ft in a different direction could change the signal enough to get other speeds.
If you dont have a decent enough signal where you are, can adversely effect the speeds.

View solution in original post

2 REPLIES 2

Re: Large packet sizes - proprietary software? Fact or fiction?

Gdkitty
Resident Expert
Resident Expert

While it might come into a LITTLE effect, it shouldnt effect speeds that much.

I have done cellular bases speed tests on phones, etc.. which are able to pull 40-50 mbps on the tests.

(so same data network)

Now, any speeds on wireless, can all be subjective to WHERE the test is done from.  Signal in these cases, can 100% make a huge difference.   standing 5 ft in a different direction could change the signal enough to get other speeds.
If you dont have a decent enough signal where you are, can adversely effect the speeds.

Re: Large packet sizes - proprietary software? Fact or fiction?

DZad
I've been here awhile
Thanks @Gdkitty, I appreciate the reply.

Both myself and a client of mine have tried the speed test in several locations where we have good LTE coverage for our phones, but still get max just above 4 Mbps for speed on these units...

I'm just trying to understand if it's possible that the test results are innacurate... this client is obviously not happy with the results and basically accused me of lying to him... I just wanted to make sure I'm not.

I was told that these units use machine-to-machine data on a private Network, also transmitting larger data packets, which they claim makes for stronger and faster connection. The company told me they have a proprietary software which handles these large data packets, and because of the size, it makes standard speed testing inaccurate.

I'm just putting it out there to see if that makes sense to anyone because from everything I've looked at I've not found a single thing that would suggest that's the case.

Thanks again for your reply!
Topic Stats
  • 2 replies
  • 3183 views
  • 1 Like
  • 2 in conversation