isposure v1.3 flawed?

Forums › New Zealand Broadband › isposure v1.3 flawed?

ZollyMonsta

2628 posts

Uber Geek
+1 received by user: 218

Trusted

Topic # 24205 17-Jul-2008 20:19

I've been using isposure since it was opened to the public and it seems to be flawed.

A perfect example tonight is it reporting download speeds of 0.74mbps.
speedtest.net via the napier speedtest server reports 9006kb/s

Upload speeds reported at the same time via isposure reports 0.80mbps
speedtest.net via the napier speedtest server reports 1955kb/s

Which makes me wonder, where is the epitiro server based? And why does the results seem to far off the mark?
I've emailed about this to their support team but are still to recieve a response.

Surely incorrect results like this skew results for the ISP's?

speedtest.net

cokemaster

Nate wants an iphone
3900 posts

Uber Geek
+1 received by user: 28

Mod Emeritus

Trusted

Lifetime subscriber

Reply # 148997 17-Jul-2008 20:46

I've noticed similar trends here, on multiple broadband connections...

webhosting |New Zealand connections | geekzone IRC chat
Loose lips may sink ships - Be smart - Don't post internal/commercially sensitive or confidential information!

Filterer

484 posts

Ultimate Geek
+1 received by user: 5

Reply # 149026 17-Jul-2008 22:31

I get a wierd problem on a server I rdp on to running it. It makes multiple tray-icons and it depends to seem related to the time between logins. If I leave it for a couple of days I literally have 20 tray icons for isposure

pɐǝɥ sıɥ uo ƃuıpuɐʇs

Try Wrike: fast, easy, and efficient project collaboration software

Detruire

1238 posts

Uber Geek
+1 received by user: 58

Reply # 149205 18-Jul-2008 16:50

I log in and get a couple of tray icons, plus I can't stop it starting on startup though msconfig.

rm *
Sick of your ISP? Get a $20 credit when you try Bigpipe via this link.

manhinli

2483 posts

Uber Geek
+1 received by user: 4

Trusted

Reply # 149213 18-Jul-2008 17:10

Detruire: I log in and get a couple of tray icons, plus I can't stop it starting on startup though msconfig.

isposure also installs itself as a service. Goto Control Panel > Administrator Tools > Services, find the IsposureAgent service, right click and enter Properties, and select manually.

You can then start or stop the service at anytime you like. I personally like to make two shortcuts, one to start and one to stop:

net start IsposureAgent
and
net stop IsposureAgent

Find me on Twitter!

I posted 1, 2 x 10^3 times!

Detruire

1238 posts

Uber Geek
+1 received by user: 58

Reply # 149230 18-Jul-2008 17:50

manhinli: isposure also installs itself as a service.

Thanks for that, I'd never have though to look for it as a service.

rm *
Sick of your ISP? Get a $20 credit when you try Bigpipe via this link.

rugrat

1169 posts

Uber Geek
+1 received by user: 157

Reply # 151498 26-Jul-2008 03:24

I'm going to uninstall it, as it seems to be useless with dymanic ip address's. When ip address changes it thinks logging in from different location and wants all data provided again, plus history is lost.

ZollyMonsta

2628 posts

Uber Geek
+1 received by user: 218

Trusted

Reply # 151499 26-Jul-2008 05:10

Yes I did the same (uninstalled) and informed them of it's faults.

It seemed to always report less than 1mbps connection when the PC started up.. Of course this will skew any results they gather.

Bung

2357 posts

Uber Geek
+1 received by user: 255

Reply # 151511 26-Jul-2008 08:33

rugrat: I'm going to uninstall it, as it seems to be useless with dymanic ip address's. When ip address changes it thinks logging in from different location and wants all data provided again, plus history is lost.

I've been trying to compare 2 modems on the same line. I don't know if the ip address was changing with the changeover of modem but the history only seems to record at hourly intervals. Any other speed tests do seem to influence the ISP average for your exchange.

Has anyone had a reply from Epitero? They popped into the first few Geekzone threads and now seem very absent.

Lurch

1054 posts

Uber Geek
+1 received by user: 71

Reply # 151710 27-Jul-2008 06:44

The website loads very slowly, some of the pages dont display properly. The tool itself reports slower speeds than if I use speedtest.net.

Actually my browser crashed when I first attempted to download the app.... Ahwell have uninstalled it. Looks like a waste of time to me..

Bung

2357 posts

Uber Geek
+1 received by user: 255

Reply # 151716 27-Jul-2008 08:26

Lurch: The tool itself reports slower speeds than if I use speedtest.net.

For this Telecom user the isposure tool reports the same speeds as Telecom's own speedtest so it's not as simple as isposure alone is broken. Speedtest.net also shows faster speeds.

Lurch

1054 posts

Uber Geek
+1 received by user: 71

Reply # 151779 27-Jul-2008 12:50

Speedtest reports 8500 kb/s where isposure reports 4000 kb/s... If you are implying I'm with Telecom I'm not :-)

Bung

2357 posts

Uber Geek
+1 received by user: 255

Reply # 151781 27-Jul-2008 13:04

I'm not implying anything.

I've been trying to compare a D-Link 502T with a Thomson ST536 so I've been using isposure and Telecom's test. Both consistently give similar results, both show the much faster Thomson modem as having much lower throughput than the D-Link. Confusing the matter is Speedtest.net giving X2 better results for the Thomson. Now capped so further testing will have to wait.

mcranna

3 posts

Wannabe Geek

Reply # 154451 5-Aug-2008 20:26

RESPONSE FROM EPITIRO

Hi All; apologies for the delay in responding to your queries. With over 7500 agents now out there in NZ alone (our target was around 2100), we have been fairly busy ensuring there is enough bandwidth through APE and telco, not too many delays in establishing sessions, and trying to answer as many questions / comments as possible that come through the "contact us" function.

We can't comment on individual test results; there could be a range of reasons why your performance is as it is, but I can tell you what we actually test. The question that keeps getting asked is regarding the line speed test, so here are some basics to get the discussion started.

The line speed test is
- a TCP test
- run with multiple sessions until we max out the line's capacity
- to an endpoint server in NZ

This is a reasonably accurate line speed test; and it does cost us quite a bit to establish. Because of the NZ peering situation, we need to hire two pipes (APE and telco) and big at that (to ensure we don't measure contention elsewhere), and we also need to limit the amount of test sessions per second for the same reason. The enpoint server is placed in a maximised location in NZ, so we do not measure contention in the offshore Internet; just what your ISP is providing (notwithstanding some local vagaries).

A couple of other things;

- it's not really designed for commercial type use; it's a home user tool. However, we are working on a version for businesses.
- Manhinli is right - the best way to turn it off is via the control panel. The idea of stop/start shortcuts is bloody good. Nice one.
- There's a bug in IE that causes the occasional "grey box of death", when a result graph won't load. We've sweated blood reducing this to just a few cases now, but please let us know if it's still happening.
- Lurch: Give us a chance! Nothing is perfect first off. Not even Microsoft got it right first time.

- We're planning enhanced functionality down the track, including tests of your CPE, so please consider keeping it installed.
- Can I suggest responding through this thread with any issues, rather than through Isposure, as I will get queries through here faster. I'll do my best to get back to you, or get someone to get back to you, within 24 hours.
- Really appreciate the time you guys are taking to comment on this; I can guarantee you your feedback goes straight back to the Dev team.

Mike Cranna

Bung

2357 posts

Uber Geek
+1 received by user: 255

Reply # 154497 5-Aug-2008 22:46

Bung:
I've been trying to compare a D-Link 502T with a Thomson ST536 so I've been using isposure and Telecom's test. Both consistently give similar results, both show the much faster Thomson modem as having much lower throughput than the D-Link. Confusing the matter is Speedtest.net giving X2 better results for the Thomson. Now capped so further testing will have to wait.

Thomson problem fixed with a reset to factory specs, previously had let install CD run. Why this should result in any difference is anyone's guess. Now Isposure and Speedtest.net also give similar results.