| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: IPhone Boinc Client port |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 44
|
|
| Author |
|
|
andyarmstrong
Cruncher Joined: Jun 25, 2008 Post Count: 22 Status: Offline Project Badges: 
|
Hi!
I am quite an able programmer - in 4th year of a honours graduate degree at St-Andrews for computer science. I just got the linux version with no gui of the boinc client to work on my linux dedi box and something struck me. A large amount of people with jailbroken iphones have the ability to run unsigned iphone applications - and even run signed applications through apple's appstore. Essentially the iphone is a darwin system - does anybody have any ideas what the dependancies to make the linux distro work on other platforms - i mean you can access a terminal from the iphone and im wondering if its a simple as running the ./boinc --attach_project <url> <key> --daemon command on the phone when say plugged into a pc so battery is not exhausted to use its brilliant processing power - afterall, every little helps? Any thoughts? Andy |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
People with more time than (ahem)
Yes, people have experimented with stuff like this. I can hook you up with them if you like. But for practical purposes - it's not an effective way of computing. Every little doesn't help when you take things like cost into account. |
||
|
|
andyarmstrong
Cruncher Joined: Jun 25, 2008 Post Count: 22 Status: Offline Project Badges:
|
Well tbh i cannot see a cost if you do it when ur phones plugged in etc - there is no cost especially if you set it up as a hobby?
I just tried sticking the executable for linux on the phone - says cannot execute binary file :'( any ideas why this might be. Would be great to get in touch with people who have experiance or tried this kinda stuff! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The costs are electricity and the lifespan of your phone.
If you want to just have a bit of fun trying to make it work, that's fine. It's going to be a lot harder than you expect, but you may get somewhere. But if you seriously want to use your phone this way, then just drop the idea. |
||
|
|
andyarmstrong
Cruncher Joined: Jun 25, 2008 Post Count: 22 Status: Offline Project Badges:
|
i dont like that kind of attitude!
I will make it work - i think ill basically use the optimized version of the open source boinc client and compile it on the phone itself That way it should work. The thing is about these things - a lot of people will use it even if its not very often - and even if it makes little to no impact on the results returned, even if it returns one result every year from all users who are running the phone software - its making a difference and thats the kind of attitude im thinking. Especially when mobile phones are becomming increasingly powerful - iphone specs are crazy - and using top on the iphone states its cpu is never really above 35% - so stickin the rosetta on it at a cpu cap of 5% when charging - will defintately, definately make a difference and the phone itself would hardly know the difference. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I'll ask Nicolás to help you.
----------------------------------------I'm curious: what are the claimed specs? even if it returns one result every year from all users who are running the phone software - its making a difference and thats the kind of attitude im thinking. The deadlines plus the fact that iPhone results would have to be paired with other iPhone results raise the bar for contributing. The real problem, though, is I don't think you will be able to run the World Community Grid applications at all.[Edit 1 times, last edit by Former Member at Jan 4, 2009 1:01:12 AM] |
||
|
|
andyarmstrong
Cruncher Joined: Jun 25, 2008 Post Count: 22 Status: Offline Project Badges:
|
From a website:
* ARM1176JZF chip with TrustZone (enables trusted computing environment for media, apps, network, OS, etc. -- very bad for hackers) * Can vary in clock speed up to 700MHz or more, depending on implementation (thanks, Nigma) * ARM Intelligent Energy Manager (claimed to reduce power consumption 25-50% in portables) * 16K / 16K cache * Features vector floating point coprocessor ("for embedded 3D-graphics") * ARM Jazelle enabled for embedded Java execution (hmm...) * SIMD, high perf integer CPU (8-stage pipeline, 675 Dhrystone, 2.1 MIPS) * 0.45 mW/MHz power draw (with cache) |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I believe this is the same chip:
http://img237.imageshack.us/my.php?image=img0002qd5.jpg Even if it were twice as fast, it still would struggle. |
||
|
|
andyarmstrong
Cruncher Joined: Jun 25, 2008 Post Count: 22 Status: Offline Project Badges:
|
very possibly - where did that shot come from - who got the software running on their phone?
Its more a hypothesis - i would like to have an application offer people the choice to do something good, even if the impact of it would be low - and it sets the standard for the future - gets more people involved and thinking about grid computing perhaps leading to more desktops becomming attached etc. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
As I said earlier: "The deadlines plus the fact that iPhone results would have to be paired with other iPhone results raise the bar for contributing. The real problem, though, is I don't think you will be able to run the World Community Grid applications at all."
Given that even my pathetically slow computer is over ten times faster than the iPhone, and that an iPhone would have to be permanently plugged in and switched on to even hope of completing a task by the deadline.... Desktops and laptops are so cheap now, that it's very hard to make an argument for grid computing on a hand-held device. I can understand the attraction of even a small contribution - but if the technical limitations reduce that contribution to zero, then it's worth nobody's time. I can see what you're thinking: "Phones will get faster, we'll find a way to exploit the graphics coprocessor, too." But desktops get faster, too - and research projects are always searching for more power to exploit. This means the minimum requirements increase, too. Yes, I'm a killjoy. But a practical one, I think. |
||
|
|
|