Superhub : Firmware R37

[thenry]

a bit like how long cloning was going on? better late than never i guess.

[General Maximus]

yeah, I really wish they would get that sorted. I need to live forever.

Who's to say that R37 isn't "broken"? Let's face it, the previous R's have all sucked. Even if the firmware is stable and largely bug-free will if fix the borked wireless range?

Originally Posted by heero_yuy

I doubt it will do much for the wireless range. As I understand it it's the design of the hardware, particularly the internal antenna(s), that's the real problem here.

Exactly heero, still Christmas is almost here and I am asking Santa for a high end router

[qasdfdsaq]

Originally Posted by jempalmer

Exactly heero, still Christmas is almost here and I am asking Santa for a high end router

I have a pile of them I don't know what to do with...

[Qtx]

Somewhere in an alternative universe someone is singing "I all want for christmas is R42ooooooooooo".

Got that out of an xmas cracker, honest! *cough*

E4200 perchance qas?

[Sephiroth]

Originally Posted by heero_yuy

I doubt it will do much for the wireless range. As I understand it it's the design of the hardware, particularly the internal antenna(s), that's the real problem here.

Correct.

[qasdfdsaq]

To be honest I'm yet to find anything actually wrong with the design of the internal antennas (or the rest of the wireless hardware for that matter). Every test I've run on it has come back with an unremarkable average or normal result. Course there is one significant variable that simulation suggests is way off, but I don't have the equipment to measure directly in practice.

I suggested a year ago that it's more likely a quality control issue than a design one, and frankly that's still the most likely thing I can think of today.

[Sephiroth]

You're letting your rep down, Qasi - saying stuff like that. If I recall correctly, the QC issue was a matter of glue. But that ultimate SH mod tells a different story.

Originally Posted by qasdfdsaq

but I don't have the equipment to measure directly in practice.

Oh no qas, you lost your hammer, again!

[qasdfdsaq]

Originally Posted by Sephiroth

You're letting your rep down, Qasi - saying stuff like that. If I recall correctly, the QC issue was a matter of glue. But that ultimate SH mod tells a different story.

I don't care about rep, only facts.

The measured transmit power of the SH in stock configuration is around 17.5dBm at the card, the antennas are basically a perfect 1/4 wavelength dipole, add the transmit power and the raw antenna gain (2.7dB) together and you've got 20.2dBm being radiated in a perfect sphere. That's exactly bang on the legal limit (20dBm), at least when considering measurement error.

In theory at least, the wireless hardware of the Superhub is as near enough "perfect" as theoretically possible - for coverage at least, not performance or reliability - you cannot get a stronger or more even wireless signal out of any device without breaking the law.

The only problem is the computed reflection coefficient, which if correct, indicates up to 80% of the signal reaching the antenna is being bounced back as a result of impedance mismatch.

Assuming that final point (which I cannot verify) is wrong, there's absolutely nothing wrong with the Superhub wireless, from a signal strength point of view. The rest of the problems must be firmware or quality control (or performance, since only one antenna is covering most areas whereas two is needed for optimal speeds). That and the big circuit board blocking off 1/4 of the wireless hardware...

[Sephiroth]

I think we may be glancing off each other. You're reporting your research at component level and I'm alluding to design in the context of effect.

Where we meet in terms of your reply, is the reflection coefficient. Presumably you've used industry data for the components involved - the RC (nor anything else) is not given in any SH/Netgear spec. If there is impedance mismatch to the extent you mention, then the complaints we see about wireless performance are explained.

I'd call that bad design - to pput anything in the field that has such severe characteristics.

[banjo]

Can we get back to the R37 Firmware , has anyone got it yet ?

[Sephiroth]

Originally Posted by banjo

Can we get back to the R37 Firmware , has anyone got it yet ?

The technical banter provides useful insights into how stuff works.

Also, inter-twined in this is whether or not R37 can help wireless range. If the hardware design is poor or there are issues of the type that Qasi has described, then R37 won't help.

So we are on track.

BTW, to answer your question, although R37 is slated for XXL100 to come down today, 2 reboots later, not yet.