Any feedback on isposure?

Forums › New Zealand Broadband › Any feedback on isposure?

adslgeek

50 posts

Geek

#31828 31-Mar-2009 19:25

Hi there,

I haven't heard of any complaining lately regarding isposure and I was keen to hear of recent feedback on the tool.

Especially some speed test results and how the results compared to the results from isposure. I have heard that they had made some upgrades to the software apparently?

In particular has anyone seen it capping at a top speed at all?

Cheers,
ADSL Geek

adslgeek

50 posts

Geek

#204528 31-Mar-2009 20:26

I have read the forum posts, and they seemed to have some real issues, but I highlighted them to Epitiro, and according to them, they have upgraded the software / fixed them.

I have some friends that have used it and they also fedback these issues (variety in speeds, differing speeds to speedtest.net), but what I was keen on help on was some recent speed test results actual comparisons.

In particular have you heard of it maxing out at a top speed?

Thanks heaps for any help you guys can help out with..

This was my results below:

I just tested my link using a number of different tests between 7:40 through to 7:59:

ISPosure (tested at 7:39pm): 24 MBps down (NOTE base on that = 192Mbps = clearly wrong)

I have a sync rate of 7616 cause it is only a ADSL1 modem I am testing.

Speedtest.net: 4.33Mbps down and 0.66Mbps up
nzdsl.co.nz: 4586 kbps and 621 kbps up

www.speedtest.maxnet.co.nz: Download Speed: 2.6 Mbps Upload Speed: 616.37 kbps
Telecom test

Your line speed is approximately 2229.1 Kbps or 273.2 KB/sec

( Where kb = kilobits and kB = kiloBytes )

ADSLgeek speedtest 5421Mbps (I know these test results are currently very rought)

My DSL modem attainable synch rate was:

Downstream Data Rate: 7616kbps

Upstream Data Rate: 736kbps

Cause I am currently testing a Binatone DSL modem that is only ADSL1.

I got this error at 7:56, but it didn't seem to stop the test conducting, despite the program clearly hanging my PC.

-----------

Microsoft Visual C++ Runtime Library

Runtime error!

Program C:\Program Files\isposure\IsposureAgent.exe

This application has requested the Runtime to terminate it in an unusal way/

Please contact the application's support team for more information

---------

I ran isposure again at 8:07

21.8KBps (again the B = bytes I am assuming means b=bits)

---------

I ran isposure again at 8:12
but the results locked on the old speedtest results and did not display new results after I had done more.

Keen to hear of more feedback if this is similar?

Also, in particular has anyone used it and it actually worked well and was accurate? I have not seen this posted
in forums so was curious why?

Cheers,

ADSL geek.

Hatch

Trade NZ and US shares and funds with Hatch (affiliate link).

Ilmarin

94 posts

Master Geek

#204536 31-Mar-2009 21:09

I had a second look at it recently and was disappointed not to find any noticeable improvements (I did send them some feedback after it first went live). The results are largely irrelevant, since international performance is not measured, peak/off-peak is not taken into consideration, 'gaming speed' is purely arbitrary (depends on target server and the game of choice), and 'browsing speed' seems pretty arbitrary as well (again the testing methodology doesn't seem to be stated anywhere). The results comparison is pretty much apples & oranges, or should I say apples and mixed fruit mush. Even if line speed (i.e. national performance) was relevant, I would want to compare ISP A average vs ISP B average, not MY result vs some ISP average. In theory, line speed average should be pretty much the same across all providers for ADSL anyway... it's only useful as a discriminator when looking at metrics like international performance (e.g. compare my results vs all results with line speed >= my line speed).

In my experience, speedtest.net gives the most useful and accurate results, even though they are still relatively inaccurate (international download file sizes too small, incorrect latency calculation, etc.). If I want a snapshot in time, that's what I use. If I want a database of results that I can use to compare ISPs, see how my ISP has performed historically both for me and for everyone else, compare international performance and peak vs off-peak, well... that niche simply hasn't been filled... yet.

adslgeek

50 posts

Geek

#204542 31-Mar-2009 21:37

Have you done any recent tests?

I was curious to see if my results were typical.

Cheers,
ADSL geek.

Ilmarin

94 posts

Master Geek

#204550 31-Mar-2009 22:07

I believe I was recently getting around 16 Mbps line speed as recorded by isposure (my sync rate is around 18 Mbps). I've never seen obviously erroneous results from it, although I suspect they have changed/modified the test server(s) and mechanism since they first went live, and last I checked it didn't do a very good job of determining when the connection was otherwise idle. A result of 24 Mbps (I don't recall them using MBps units) in your case certainly suggests a flaw in the test. If such results are repeatable... you could use a utility like Wireshark to try and look under the hood of their test and see if it's doing anything obviously incorrect.

jfnz

211 posts

Master Geek

Trusted

#204646 1-Apr-2009 11:55

I have only skip read this thread and missed the majority of it, but my feedback is that I have some concerns about the accuracy of the data returned.

Often I have customers approach me with isposure data in hand that appears to be somewhat skewed. When alternative testing is done (taking in to account the same times of day) it will show an entirely different side of the story often much more favourably than isopsure suggests.

Unfortunately if you're trying to prove the highest throughput on a line it's difficult to have a centralised server to run speed tests off due to the fact you have to take in to account routing delays and similar.

At the end of the day it should be used as a guide to represent a single practical real-world scenario as opposed to the be all and end all of Internet speed tests which I commonly see people reputing it as.

Please note: Any posts, comments, or contributions in this forum are posted by me as an individual acting in my own right and do not necessarily reflect the views of any company I work for, clients I've consulted for, or anyone else.

cbrpilot

945 posts

Ultimate Geek

Trusted

Spark NZ

#205042 2-Apr-2009 23:23

I have had quite the opposite problem with Isposure. I think it under-reports the performance of my Broadband.

My connect rate (line speed) according to my modem is >10meg.

According to Isposure - it's between 600kbit and 1.2mbit. All of the other metrics look pretty low to me.

Speed test to speedtest.net show I can get 4.5mbit to sites in NZ as well as about the same to their site in LA.

My views are my own, and may not necessarily represent those of my employer.

uktuatara

149 posts

Master Geek

#205581 6-Apr-2009 16:47

My experience has been dissapointing with isposure. When I punch in my location to the isposure tool, it lists me as being in Onehunga. Which has an upgraded exchange to ADSL2, however i know my location when compared to the exchange (Mangere Bridge) whcih is right at edge of the 2km radius which rules me out of getting the ADSL2 speeds I can theoretically obtain (Until my new cabinet goes in!)

So when I run my tests, my results are what is expected for when I'm this far from the exchange, about 1.5 Mbps. However if I run the comparision to other users in 'my area' since I am comparing to the Onehunga suburb, I can see that it's comparing me to users that are obtaining true ADSL2 speeds (the Exchange was upgraded some time in Feb 2008 according to Telecom Wholesale site), no doubt closer to the exchange, or even with a cabinet in their area.

So if I didn't know better, I'd think I should change to another provider! But when checking with my neighbours, we all get around the same speed, although we all use different providers.