So I get that you get throttled back once you use up your data, but this is rediculous... this is right after I revived my email

  • 1
  • Problem
  • Updated 1 year ago
  • Acknowledged
Photo of Kendra Miller

Kendra Miller

  • 1 Post
  • 0 Reply Likes

Posted 1 year ago

  • 1
Photo of VeteranSatUser

VeteranSatUser, Champion

  • 3497 Posts
  • 2030 Reply Likes
I don't know what speed test and or ISP you are running to generate that test, but that is not an accurate test if you are running on Exede. The latency is way, way, way too low.

So either it's an inaccurate speed test or that test wasn't done on Exede.
Photo of les

les

  • 87 Posts
  • 20 Reply Likes
sadly Ive run the test on excede a few times in the past and got those results with excede - so it is possible with excede.
 I ran the same test with testmyspeed - speedtest and download speed and they were all very close.
 I called excede tech support and they had a different result but slightly better, but not great, so they ran it again and got worse results - later they sent a guy out to realign the dish and he replaced all the cable as the direct installers had sliced it in various places.
 then after much more testing between him and the exced offices - they were able to get acceptable speeds.
 sadly it is possible, but indicates a problem that warrants a call to exced to see what can be done. [it might take several calls to reach the person that is willing to help you]
(Edited)
Photo of VeteranSatUser

VeteranSatUser, Champion

  • 3497 Posts
  • 2030 Reply Likes
Yes speeds could be that low, but latency will NEVER be that low on an accurate test. My point is since I know the latency is wrong, there is no guarantee the other part of the test is any more accurate.
(Edited)
Photo of les

les

  • 87 Posts
  • 20 Reply Likes
obviously you have had different results,
I know what I have gotten on my tests
I watched as the tech guy here got the same results and worse on their visits here.
 you might not believe it, but latency can be even lower.
there is something wrong when your tests and those of exced are this low and lower. it needs to be tested by you are excede and when it is so low? there is something wrong. [depending on your package(exced sez)]

 but it is possible [period]
(Edited)
Photo of Andy Schack

Andy Schack

  • 854 Posts
  • 753 Reply Likes
What Vet said.....52 ms is way too low...should be about 700.

Andy
Photo of Old Labs (VS1-329-L12FZ)

Old Labs (VS1-329-L12FZ)

  • 3900 Posts
  • 3977 Reply Likes
It's Google... it shows up at the top of Google search whenever anything resembling a speed test search is performed - hey let's direct them to a speed test that will be automatically questioned as to it's validity! Good work Google... they're just trying to gather some statistical data for analysis and who knows for what malevolent or benign purpose.      

Google partners with Measurement Lab (M-Lab) to run this test off of Google. Not even close to being accurate with the latency displayed.

It's currently showing me at 5 Mbps download, when both speedtest.net and testmy.net are in the 15-16 Mbps range. Latency at 59 ms per Google and as expected elsewhere. They may just be displaying minimum speed during the test as DSL Reports appears to do.

Then there's dslreports.com showing 5 Mbps also but without looking into the details of their methodology it appears they simply display the minimum achieved during the test but I do like their showing how re-transmits can affect speed.

All speed tests are not created equally and depends on what's actually being measured. I use them all (Except Google) because if you investigate the methodologies used each provides some distinct info that can be useful. Speedtest.net results appear tro be more pertinent to browsing, testmy.net more pertinent to streaming/downloading and dslreports.com grading system can sometimes yield a good overall grade despite a slow speed due to the quality (affected by congestion).   

There's even a disgruntled Exede user trying to get a class action suit funded based solely upon that very type of Google result - she's still awaiting her 1st donation after two months - apparently the photo of her laptop sitting on her couch and a close up photo of her screen isn't garnering much support as proof positive after all it is Google and they're never wrong are they ;)

She and I went at it publicly here about 2 months ago about quoting things out of context, copying snippets, and improperly attributing some of them to the wrong individuals over on Facebook to support her claims - she never did grasp what I was trying to say; I'd point you to the conversation but it's meaningless now - she was banned and it's only a one sided conversation as her posts are no longer there.

In short, try one of the others mentioned here, Kendra - but my recommendation is testmy.net. Looking at the published speedtest.net methodology, I do have some concerns about its accuracy - some of it leads me to believe it may be statistically (not intentionally) biased in favor of ISPs and/or the browsing experience. dslreport.coms appears accurate but you have to dig into it to find maximum speed attained.

Offhand however and looking at the time of the  screen capture above you appear to have been right in the prime time usage period and much slower speeds are to be expected depending on what plan you have and if you exhausted your priority data.
(Edited)
Photo of VeteranSatUser

VeteranSatUser, Champion

  • 3497 Posts
  • 2030 Reply Likes
I personally have found testmy.net the most accurate test measurement for me. I would be curious what her test results would be with that tool since we know the latency is off, how off are the speed test results.
Photo of Matt B

Matt B, Viasat Employee

  • 844 Posts
  • 433 Reply Likes
Yep!  As the champions said, the latency is waaay to low to be accurate for Exede service.

http://testmy.net/ is the site I recommend, as has several of the posts above.
Photo of Bev

Bev, Champion

  • 3076 Posts
  • 1291 Reply Likes
Just to show how much variance speed tests can have. All done today within minutes of each other. There are a lot of variable in any connection and, the results those cause can make a big difference on speed tests. 










Photo of Old Labs (VS1-329-L12FZ)

Old Labs (VS1-329-L12FZ)

  • 3900 Posts
  • 3977 Reply Likes
There can be dramatic speed variances within seconds during even the same test (remember everything you do involves a series of packets):

Here's a recent testmy.net test where where my measurement points displayed span from 5 Mbps to over 20 Mbps (and those are only the one's displayed there are many more collection points used than those). I've seen much wider variances - this isn't all that bad.
 


Differences between different speed test services/providrs can be also explained by the different statistical methodologies used.

If they are published (some don't), you need to dig into the details and methodology of each speed test provider and determine what's actually being displayed for the summary number (there's one I use that actually appears to display the minimum but provides a much more meaningful statistic to me % of re-transmissions) and each provider has a different opinion of what should be displayed (average, mean, etc.) Just understand that between providers you may be comparing apples to oranges.

Statistics can be used to achieve just about any result you desire ;)

Knowing re-transmission rates during the test is useful:

"The higher the packet loss / re-transmit percentage the more inefficient the connection is, and a very poor result may be indicative of congestion, inside wiring issues or other problems".

For example, DSL Reports provides a quality grade and it's possible to have a good speed grade and poor quality grade (and vice versa) - the grades appear to be granted on a bell curve of all of your particular ISP's users. I typically get a poor quality grade during peak congestion periods - the solution for severe congestion is to drop packets at the bottleneck (that's the way it works and  gotta do something with them or just bring everything to a halt waiting on some slow poke in the passing lane).

The moral is to look beyond the summary result for the potential cause, and remember ad blockers and other add-ins (even anti-virus) can possible impact the results (disable them if needed on your speed test sites but they do often reflect the speed you'll attain because of them and speed decreases because of them aren't ViaSat's fault). Finally know what the summary numbers being displayed actually represent.
 
(Edited)