Page 2 of 2

Posted: 09 Jun 2012, 23:30
by Soarer
tobo wrote:Sorry, haven't checked this thread in a while :P . I've added the descirptor as attachment for those that requested it. I checked the above microsoft page and the keyboard seems to register up to 21 keys.
Ah, lovely, thank-you!

Right, so... it's got two interfaces, the first of which is a completely normal 6KRO (and modifiers) 'boot mode' keyboard.

The second is nominally a mouse, but it can send 4 different reports:
  • standard mouse
  • power control keys
  • multimedia keys
  • 15 extra keyboard keys
Since it's a low-speed device, this is a fairly sensible way to do things. The report for the extra keys only needs to be sent when more than 6 keys are pressed (not including modifiers). However, when more than 6 are pressed it will take 2 polls to send both the standard and the extra report (because the extra report is twice the size that can be sent in a single transaction), which will take up to 16mS.

Looking for problems, I reckon it's all OK in practice. On a Mac, which will think of the two reports as two independant keyboards, this arrangement certainly won't cause any issues until more than 6 keys are pressed. Then, there may be confusion over which keys are repeating, or which keys are affected by modifiers. But that's hardly going to matter in games or whatever, where you might be wanting to press so many keys.

So overall, not bad. For a low-speed device. It's a shame that full-speed is still viewed as too expensive, because it removes the needs for these tricks, and is much faster as well. It's a chicken and egg situation I guess - full-speed controllers won't get cheap until keyboard makers start using them, and makers won't use them until they're cheap :evil:

Posted: 10 Jun 2012, 00:00
by tobo
Thanks for the explanation. But what advantages would a full-speed controller realistically have?

Posted: 10 Jun 2012, 00:02
by JaccoW
Nice, thank you Soarer for the technical information. Mind if I use this in a review I will be putting up somewhere later this month or perhaps July. (with proper references off course) :)

Posted: 10 Jun 2012, 00:10
by Soarer
tobo wrote:Thanks for the explanation. But what advantages would a full-speed controller realistically have?
The obvious one is speed - it can send data every 1mS, so there's less delay. But it can also send 64 bytes each time vs only 8 for low-speed. That means full NKRO is quite easy and tidy to do (with one bit per key), with no messing around with pretending to be multiple keyboards. So while NKRO might not be important, it comes with no downside :D
JaccoW wrote:Nice, thank you Soarer for the technical information. Mind if I use this in a review I will be putting up somewhere later this month or perhaps July. (with proper references off course) :)
Sure, no problem. I've just spotted an error I made about the timings though - it's 16mS not 24mS because it can start sending both at the same time. Will edit... done. :ugeek:

Posted: 10 Jun 2012, 00:37
by ripster
This is what Paul Dietz at Microsoft R&D did a full two years ago.

http://geekhack.org/showthread.php?9230 ... post165890

And others..

http://geekhack.org/showthread.php?1531 ... post304795

Posted: 10 Jun 2012, 00:59
by Soarer
ripster wrote:This is what Paul Dietz at Microsoft R&D did a full two years ago.

http://geekhack.org/showthread.php?9230 ... post165890

And others..

http://geekhack.org/showthread.php?1531 ... post304795
Yeah, shame they didn't make more of a fuss about it being full-speed - we might have more of them by now!

I still don't know why they used such convoluted descriptors though... it could be a hint that the X4 started out as a low-speed USB design... dunno.

Posted: 10 Jun 2012, 01:22
by webwit
ripster wrote:This is what Paul Dietz at Microsoft R&D did a full two years ago.

http://geekhack.org/showthread.php?9230 ... post165890

And others..

http://geekhack.org/showthread.php?1531 ... post304795
You read a lot of geekhack. Are you a member?

Posted: 10 Jun 2012, 01:24
by 002
He's like a punished child looking into the playground through the bars :(

Posted: 10 Jun 2012, 02:01
by ripster
Like a pedophile within 100 yards of the school I've been caught.

I will go put back my ankle bracelet on now.

And some metaphors are NOT roads and best not taken to the end.

Posted: 10 Jun 2012, 02:05
by ripster
002 wrote:He's like a punished child looking into the playground through the bars :(
Lol.

Look who is talking, troublemaker!
http://geekhack.org/showthread.php?3118 ... post610501

Posted: 10 Jun 2012, 02:11
by 002
Haha - I don't make trouble, friend!
Just pointing out to people why it happened to avoid an avalanche of "wtf's". It's geekhack though so I suspect people will ask anyway..

Posted: 10 Jun 2012, 02:17
by 002
Perhaps you would prefer this?
Like an ostracised shepherd who got too close to the sheep ;)

Posted: 10 Jun 2012, 02:53
by ripster
That is a simile, not a metaphor.

Baaaaaaaa...
Baaaa..
Ba,..


What is funny is the wikis are in lockdown despite 99% of the borked wikis being me simply doing WikiVault tests that iMav ASKED me to do.

What a piece of shit wiki software.

And open them up. If I had wanted to fuck them all up I would have done it LONG ago iMav!

But we are drifting off topic..............