The Wifi Problem Will Be Found To Be Caused By:

[torpor]

Ah right, I get it .. you're right, that would be pretty funny if we got tripped up. Thanks for pointing it out ..

 

[WizardStan]

Either way, the thing that matters is that a proper test harness is set up to use to get details about the performance of the driver .. I'll have a look at helping out with this later this evening.

Exactly. I was just pointing out so that no one is wondering why only a few bytes are being transferred. I remember stumbling on that years ago

"that's strange. /dev/zero seems to go much faster than /dev/random. Conclusion: zeros, being round, are easier to push through the intertubes. The ones that make up the random are obviously getting stuck."

 

[Sphinxter]

Whores spitting in pandora's milk.

 

[hobbyman II]

that's strange. /dev/zero seems to go much faster than /dev/random. Conclusion: zeros, being round, are easier to push through the intertubes. The ones that make up the random are obviously getting stuck.

don't forget to use a legible font , maybe the hardware can't read dingbats , use Monster cables, they have a good font that improves audio playback :rolleyes:

seriously tho, some things that I have found to be causing problems in the past, (in my limited experience)

device re-negotiating connection every time it passes data due to lease being 0 seconds

misconfiguration of 127.0.0.1

some device not connected but driver getting in the way

wrong revision of driver for device model number (eg chip xyz mk4 rev 1.1 is several orders of magnitude different to xyz mk4 rev 1.100001 for no good or obvious reason)

incorrect setup sequence, some devices insist on settings file being in correct order

invalid control code/s in config file

device falling back to crud default because of init error

endian error or including checksums on data in wrong sequence

and the classic "gets em every time"

I count from one, computer counts from zero, error, causes more hassle when doing bitwise stuff, just ONE typo that LOOKS like sense, can ruin your week.

 

[VRAndy]

"All the world's evil" and/or "hope".

 

[may88]

I vote for "None of the above!"

Now this may not go down with "open" purist but was there not talk of a working closed source driver used in the early days?
Implies the hardware was good at some stage.

 

[Tensuke]

...Wasn't there a Palm driver or did I just make that up/not see it mentioned?

 

[Poem58]

Now this may not go down with "open" purist but was there not talk of a working closed source driver used in the early days?
Implies the hardware was good at some stage.

Yeah, I was thinking the same thing. One of ED's videos where he posted on the forum from his dev board. Everything seemed to work then. Never heard of a problem until they moved to the "open" driver. I do remember them saying the new driver would take some work. (Keep in mind my next statement is based on having absolutely NO coding skills) I wonder if looking at the original closed driver might help shed some light on what's wrong with the current one? If that's even possible.

 

[Prometheus]

I seem to recall that the dev-board post was done with a USB Ethernet device, rather than wirelessly - but please do correct me if I'm wrong.

As for looking at the closed driver... Well, it's closed, isn't it?

 

[zevdawg]

I seem to recall that the dev-board post was done with a USB Ethernet device, rather than wirelessly - but please do correct me if I'm wrong.

As for looking at the closed driver... Well, it's closed, isn't it?

And people are too afraid to reverse-engineer stuff cause of legalities nowadays.

(but again, someone did it to the iPhone, and that same dude did it to the PS3 - consequently now has no OtherOS option in the new firmware cause of em')

 

[lulzfish]

Last I heard, the closed driver worked well, but only with an older kernel or something, so there was no way to have it alongside other required OS updates.
inb4 poll "would u rather have wifi or kernel", plz put in offtopic at least.

So they switched to an open one, and it's been awkward.

 

[WizardStan]

Didn't Notaz just recently say that the closed source driver didn't work any better? And that it was basically rewritten into the N900 (was it the N900? Or something else that I've forgotten) driver anyway, almost exactly the same? And that neither one of them are actually compatible with the Linux stack anyway, so required a lot of work to get them to actually get them to do anything at all, let alone anything useful?
I'm pretty sure he did.

 

[Exophase]

Didn't Notaz just recently say that the closed source driver didn't work any better? And that it was basically rewritten into the N900 (was it the N900? Or something else that I've forgotten) driver anyway, almost exactly the same? And that neither one of them are actually compatible with the Linux stack anyway, so required a lot of work to get them to actually get them to do anything at all, let alone anything useful?
I'm pretty sure he did.

Yes.

Hey, you did ask.

 

[crade]

that same dude did it to the PS3 - consequently now has no OtherOS option in the new firmware cause of em')

Yeah cuz he made Sony fraudulent.

 

[torpor]

Well it looks like for now we'll be doing fine with the USB WLAN dongles (see news on PandoraPress and so on) so thats a good stop-gap measure for us until we work out whats going on with the onboard hardware .. hopefully it won't be too difficult to debug. I will look at it as soon as I get my Pandora's in my lab, since I'm going to be doing driver work anyway (MIDI and Audio) .. hope we can get it resolved.

 

[notaz]

Last I heard, the closed driver worked well, but only with an older kernel or something, so there was no way to have it alongside other required OS updates.
inb4 poll "would u rather have wifi or kernel", plz put in offtopic at least.

So they switched to an open one, and it's been awkward.

Wrong, there is no such thing as "closed driver" now. Read this post again.

The only drivers are few variations of TI driver (mostly equivalent) and the new(ish) Nokia driver. TI one was closed until Android code drop, now it's open. All those were tested and have similar (bad) results.

 

[droneb]

Last I heard, the closed driver worked well, but only with an older kernel or something, so there was no way to have it alongside other required OS updates.
inb4 poll "would u rather have wifi or kernel", plz put in offtopic at least.

So they switched to an open one, and it's been awkward.

Wrong, there is no such thing as "closed driver" now. Read this post again.

The only drivers are few variations of TI driver (mostly equivalent) and the new(ish) Nokia driver. TI one was closed until Android code drop, now it's open. All those were tested and have similar (bad) results.

now i get it why all the fuss about the small posibility of it being a hardware bug

i've been looking long for the post that started all of this!.

 

[LeChuck87]

http://www.youtube.com/watch?v=lRL4zi35nVU&playnext_from=TL&videos=TPyGGQC5P2w&feature=sub

ED's wifi vid!


edit: mali, you were caught on tape! (or youtube..)

 

[mali]

^ He poked my post with his stick!

Nice video, mmenu looks great, too

 

[verm71]

^ He poked my post with his stick!

Nice video, mmenu looks great, too

Yes, but whose avatar was displayed first?