• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: Surely someone knows this ?! Jens knew
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Surely someone knows this ?! Jens knew


  • Subject: Re: Surely someone knows this ?! Jens knew
  • From: J P May <email@hidden>
  • Date: Tue, 30 Mar 2010 23:51:00 +0200

On Mar 30, 2010, at 1:52 AM, Hamish Allan wrote:

 I'd have thought it was an implementation detail and therefore subject
 to change. It's also premature optimisation until you've done some
 profiling!

+1. This is, AFAIK, a proprietary protocol (on top of Bluetooth) so there's not a lot of info about what goes over the wire.

Ah - thanks for the info, I did not know that, good one. Just saw your email now.


I've gotten emails from people who were trying to figure out the same thing and felt from testing it's at least "a good few bytes", so that's the answer really. there's no point (seems to be the consensus) worrying about shaving off the last byte if you're only sending a couple bytes.

(It would be interesting to talk to a bluetooth mouse engineer, how do they do it, so smooth!)




Anyway, latency is generally a bigger issue than bandwidth. Regardless of how much or little data can fit in a packet, it's better to send data in larger chunks to minimize the overhead of having to wait for acknowledgement of each message.

it's a remote control device, like a, err, Toyota accelerator pedal sending a channel (wirelessly) to the engine computer. sending GK's un-reliable mode, no acknowledgements.





But as Hamish said: premature optimization. Just implement it first by doing "the simplest thing that could possibly work", then try it out and see if it's adequate. If not, optimize and iterate.

The whole thing was completely finished and tested, I was just looking in to it after the fact


(You guys are "real" network programmers, you forget that gamekit is one routine :) )

if you want to try it yourself, just bring up a gamekit client and server in xcode, choose the unreliable mode, and install it on two iphones. keep making the message bigger 1 byte, 2, 3, and see when you reckon the byte size of the message is big enough it is slowing it down.


on another topic, it's funny about gamekit ... it is just so reliable and simple in a business sense, clients and so on will inevitably demand it even in situations not ideal, you know. it's the beginning of the end of He-man networking! :)



Cheers

-Jens

_______________________________________________ Do not post admin requests to the list. They will be ignored. Macnetworkprog mailing list (email@hidden) Help/Unsubscribe/Update your Subscription: This email sent to email@hidden
  • Follow-Ups:
    • Re: Surely someone knows this ?! Jens knew
      • From: Jens Alfke <email@hidden>
  • Prev by Date: Re: Surely someone knows this ?!
  • Next by Date: Re: Surely someone knows this ?! Jens knew
  • Previous by thread: Re: Surely someone knows this ?!
  • Next by thread: Re: Surely someone knows this ?! Jens knew
  • Index(es):
    • Date
    • Thread