Superhub : Superhub Firmware Beta Test

Originally Posted by zekeisaszekedoes

Basically you want to hide the superhub away in a cupboard because of the lights, )

yes your right, and now I don't need the wireless function of the Hub and just using it as a modem I can hide it away in a cupboard

I spent a fair amount of my later working life setting up and supporting fixed and wireless Networks, and just can't understand the patchy wireless performance from the Hub

from the limited information I have gleaned, I suspect the resistance on the antenna's change due to heat

if in fact they are mounted as silver strips on the circuit board, the power to them will be in milli watts, and the heat generated inside the Hub could affect their operation

This would go some way to explaining how the wireless performance can change several times in a day from "good" through to "no connection "and then back again

Aw well just a theory, but it does intrigue me,

[kwikbreaks]

Originally Posted by kenoliver

from the limited information I have gleaned, I suspect the resistance on the antenna's change due to heat

if in fact they are mounted as silver strips on the circuit board, the power to them will be in milli watts, and the heat generated inside the Hub could affect their operation

That's not it...

Originally Posted by qasdfdsaq

The antennas are not integral - as the wireless chip is on a mini PCI card, the antennas are mini PCB antennas connected via a cable and U.FL connector - pretty much the same configuration as any laptop really.

The resistivity and thermal expansion changes would be way too small to produce any noticeable change in performance imo.

[qasdfdsaq]

Sorry guys, a correction here - it's mini PCIe not mini PCI as I previously stated - I forgot to put in the 'e' earlier :-P

And indeed, temperature changes are unlikely to be the cause, as in these situations it seems like people's laptops can 'see' the Superhub but it doesn't respond when trying to connect to it - more likely a firmware bug.

[zekeisaszekedoes]

Originally Posted by qasdfdsaq

Sorry guys, a correction here - it's mini PCIe not mini PCI as I previously stated - I forgot to put in the 'e' earlier :-P

Not quite as primitive but still hardly the sort of non-integrated solution you'd want to put in a mass-manufacture device. It's not like you're ever going to change the mini PCIe wireless module for something else.

[gadge]

When should the final version be released? our superhubs reset itself twice this morning.

[kwikbreaks]

Originally Posted by zekeisaszekedoes

It's not like you're ever going to change the mini PCIe wireless module for something else.

It's not user changeable for sure but it gives VM/Netgear a relaively cheap upgrade path should they decide to fit something better. The cable modem/router isn't likely to change for quite some time but WiFi is still moving forward. Upgrading to something that actually works would be a good start...

Originally Posted by kwikbreaks

IUpgrading to something that actually works would be a good start...

"Amen" to that

[KenK]

Originally Posted by zekeisaszekedoes

I suppose the real question is, who sweet-talked VM into believing this was a good solution?

The bean-counters. It can only have been about price; not whether it works, meets customer needs, is upgradable ...

[zekeisaszekedoes]

Originally Posted by kenoliver

"Amen" to that

Indeed. Getting rid of the superhub should be a top priority. Standing behind the "well, it works for some users" isn't an adequate defense. Especially in cases where it only works because those particular users are placing VERY low demands on the device.

Originally Posted by KenK

The bean-counters. It can only have been about price; not whether it works, meets customer needs, is upgradable ...

Ah, so that's where the short-term planners from the banks went after they got fired. Mystery solved!

[Bullstein]

I dreaded getting the superhub but so far it's been excellent (to my amazement)

Speeds and ping are fine, wired to my ps3 is lag free gaming and I'm currently on my laptop in the back garden with the superhub (behind my tv) at the front side of the house 40 yards away. No problems

Originally Posted by Bullstein

I dreaded getting the superhub but so far it's been excellent (to my amazement)

Speeds and ping are fine, wired to my ps3 is lag free gaming and I'm currently on my laptop in the back garden with the superhub (behind my tv) at the front side of the house 40 yards away. No problems

I don't think anyone could fault the Hub on it's wired performance, it's solid and reliable

But in my case I'm not able to say the same about it's wireless performance

Maybe the beta testing of R28 and it's impending release will improve it

until then I continue to rely on my Time Capsule to take care of the wireless side of my network

Hopefully your positive experience will continue

[Phil-ntl]

The latest update from the firmware team is that during testing we've found three significant bugs in the modem mode functionality in the new firmware. We're currently working on fixing these bugs and producing an updated firmware version. We're continuing to test the firmware to make sure it's as robust as possible before we release it for testing.



So at the moment it's looking unlikely that we'll be able to start the beta test next week. As soon as I have more news on this I'll post an update.



Thanks

From the VM beta forum

[qasdfdsaq]

Woop woop. Another delay.

[Helix]

I don't get what that the business about them receiving the firmware last week was. Surely the firmware should be developed with VM engineers involved and testing as it goes along, rather than providing a download for VM to test and then they find bugs and the firmware developers have to fix it and then send another update to VM.

Seems to be a very bizarre way of developing this, but then the whole process has already shown that VM/Netgear know nothing about software development.

[KenK]

Originally Posted by Helix

I don't get what that the business about them receiving the firmware last week was. Surely the firmware should be developed with VM engineers involved and testing as it goes along, rather than providing a download for VM to test and then they find bugs and the firmware developers have to fix it and then send another update to VM.

Seems to be a very bizarre way of developing this, but then the whole process has already shown that VM/Netgear know nothing about software development.

It's fairly standard. Whoever's got the contract to develop the software provides occasional 'code drops' for the customer to try out and report back on. Maybe this hasn't happened often enough in this case; but why should VM spend staff time on a process that they're paying someone else to provide.