Re: [asterisk-users] building a phone
Tzafrir Cohen wrote: > Hi folks > > A common wisdom here is that one should use a proper hardware phone > rather that an extra software on the user's PC. Why is that such a big > issue? > > One thing that bothers me with the current crop of hardware SIP phones > is that they are hopelessly properitary. > no more so than a vcr or other consumer device. > So what would it take to build a fully-adaptable phone? > why bother when there are lots of < $100 perfectly fine phones already ? > Here are some of my thoughts. This is not anything I plan to do soon (if > at all), but I really find it strange that there aren't such phones > already. > > why not just take an existing device such as the n770 or nintendo ds. load a sip client of your choice on, and you are done - wireless phone, touchscreen, runs any other software you want, already low cost, no manufacturing. (both devices already have wifi, speaker and microphone as well as a colour touchscreen, so what else is missing ?) there are also iphone clones coming down into the same price range that are fully programmable. > == Small Quantities: > When you look at such systems it becomes aparant that you can get much > nicer prices if you buy large quanities. But this is something that will > be a problem. Not only for prototying. The fact that you're limited to a > strict hardware setting is very limiting. No mixing and matching like in > a standard PC. I'm not exactly sure how to overcome that. > > == Platforms: > There are many embedded platforms nowadays. I assume that the relevant > application requires some non-trivial CPU power. I would exclude e.g. a > 486-based systems. My target phone should be able to handle at least two > concurrent Speex calls. Preferrebly 6 speex calls and above. > > OTOH, I can't afford a monster CoreDuo. I need a quiet system with no > fan. Thus the target CPU may be higher end VIA or Atom. Not sure about > Geode. > > There are also some interesting ARM-based boards around. I'm completely > unfamiliar with them but I suspect that they may prove to be cheaper. > > == SIP Software: > Not really sure here. There must be something close to usable already, I > guess. > > == Micro Browser: > Hell no! > > The device should have an LCD display, and the content of that display > should be programmable. Programming it using a HTML renderred is a bad > design decision. > > The device should be a good phone. It should not attempt to be a web > browser, as it will be a lousy one. > > == Handset: > I suppose that an obvious starting point for a handset is "skype phones" > such as USB handsets from yealink. Far from an optimal design, but a > driver already exists. > > > == Ease of Use: > A phone must be usable. The target device must be something my mom can > use. However that does not mean it must be easy to program. It must be > programmable and hackable. But I can live with a complicated user > interface for that. If such phones become successful and useful, better > interfaces will eventually be written. > > > ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
Tzafrir Cohen wrote: > Hi folks > > A common wisdom here is that one should use a proper hardware phone > rather that an extra software on the user's PC. Why is that such a big > issue? > Marketability for one. People worldwide understand the telephone paradigm. You have a handset and a box with numbers. You pick it up and dial, talk through the handset, and listen in the other end. It's simple. It's an elegant design. And everyone from 1 year olds to my 97 year old grandfather can use it. Software phones? Not so much. In fact, not even close. The additional complexity of running software on a machine ALONE would keep my grandfather and that 1 year old from using it. Headsets? Seriously? Since when have those been user-friendly OR comfortably. In essence, adherence to a software phone paradigm breaks a century of design advancement in telephone ergonomics, psychology, and reliance, and replaces it with something that's clearly just a kludgy add-on to a product which was never originally designed for the task. > One thing that bothers me with the current crop of hardware SIP phones > is that they are hopelessly properitary. > > So what would it take to build a fully-adaptable phone? > > Here are some of my thoughts. This is not anything I plan to do soon (if > at all), but I really find it strange that there aren't such phones > already. > > > == Small Quantities: > When you look at such systems it becomes aparant that you can get much > nicer prices if you buy large quanities. But this is something that will > be a problem. Not only for prototying. The fact that you're limited to a > strict hardware setting is very limiting. No mixing and matching like in > a standard PC. I'm not exactly sure how to overcome that. > This is one of the biggest reasons all the hardware phones are proprietary -- they're each written for different basic hardware. > == Platforms: > There are many embedded platforms nowadays. I assume that the relevant > application requires some non-trivial CPU power. I would exclude e.g. a > 486-based systems. My target phone should be able to handle at least two > concurrent Speex calls. Preferrebly 6 speex calls and above. > > OTOH, I can't afford a monster CoreDuo. I need a quiet system with no > fan. Thus the target CPU may be higher end VIA or Atom. Not sure about > Geode. > > There are also some interesting ARM-based boards around. I'm completely > unfamiliar with them but I suspect that they may prove to be cheaper. > > == SIP Software: > Not really sure here. There must be something close to usable already, I > guess. > > == Micro Browser: > Hell no! > > The device should have an LCD display, and the content of that display > should be programmable. Programming it using a HTML renderred is a bad > design decision. > > The device should be a good phone. It should not attempt to be a web > browser, as it will be a lousy one. > > == Handset: > I suppose that an obvious starting point for a handset is "skype phones" > such as USB handsets from yealink. Far from an optimal design, but a > driver already exists. > > > == Ease of Use: > A phone must be usable. The target device must be something my mom can > use. However that does not mean it must be easy to program. It must be > programmable and hackable. But I can live with a complicated user > interface for that. If such phones become successful and useful, better > interfaces will eventually be written. > > > Just a note here -- a complicated user interface, though you personally may be able to live with it, will pretty much ensure that the phones never become successful enough for a better one to be written. UI design is about 10% code and 90% psychology (and so FEW people who call themselves UI 'programmers' understand that). Just having a UI that can get you from point A to point B without typing in commands is NOT a UI worth making, as it will never be a UI worth using. ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
It's a dream! It's since years that I'm thinking to have an open hardware project targeted to a SIP application. I'm thinking, for example, to have a modular system that can be targeted to different custom appliances like, for example, (video) door bell opener/intercom, or building/desktop music streamer, or SIP compliant actuators. I have a (very) little experience on electronic projects. Is there something I can do to help starting a similar project? Thank you and best regards. Marco Signorini Tzafrir Cohen wrote: > Hi folks > > A common wisdom here is that one should use a proper hardware phone > rather that an extra software on the user's PC. Why is that such a big > issue? > > One thing that bothers me with the current crop of hardware SIP phones > is that they are hopelessly properitary. > > So what would it take to build a fully-adaptable phone? > > Here are some of my thoughts. This is not anything I plan to do soon (if > at all), but I really find it strange that there aren't such phones > already. > > > == Small Quantities: > When you look at such systems it becomes aparant that you can get much > nicer prices if you buy large quanities. But this is something that will > be a problem. Not only for prototying. The fact that you're limited to a > strict hardware setting is very limiting. No mixing and matching like in > a standard PC. I'm not exactly sure how to overcome that. > > == Platforms: > There are many embedded platforms nowadays. I assume that the relevant > application requires some non-trivial CPU power. I would exclude e.g. a > 486-based systems. My target phone should be able to handle at least two > concurrent Speex calls. Preferrebly 6 speex calls and above. > > OTOH, I can't afford a monster CoreDuo. I need a quiet system with no > fan. Thus the target CPU may be higher end VIA or Atom. Not sure about > Geode. > > There are also some interesting ARM-based boards around. I'm completely > unfamiliar with them but I suspect that they may prove to be cheaper. > > == SIP Software: > Not really sure here. There must be something close to usable already, I > guess. > > == Micro Browser: > Hell no! > > The device should have an LCD display, and the content of that display > should be programmable. Programming it using a HTML renderred is a bad > design decision. > > The device should be a good phone. It should not attempt to be a web > browser, as it will be a lousy one. > > == Handset: > I suppose that an obvious starting point for a handset is "skype phones" > such as USB handsets from yealink. Far from an optimal design, but a > driver already exists. > > > == Ease of Use: > A phone must be usable. The target device must be something my mom can > use. However that does not mean it must be easy to program. It must be > programmable and hackable. But I can live with a complicated user > interface for that. If such phones become successful and useful, better > interfaces will eventually be written. > > > ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
Marco Signorini wrote: > It's a dream! > It's since years that I'm thinking to have an open hardware project > targeted to a SIP application. > there is already a project called openmoko - join it and buy some hardware. The phone is large and clunky - the idea is good, but not something you're ever going to carry in your pocket, and somewhat silly when there is already smaller hardware out there that runs linux at less cost than their device. > I'm thinking, for example, to have a modular system that can be targeted > to different custom appliances like, for example, (video) door bell > opener/intercom, or building/desktop music streamer, or SIP compliant > actuators. > > I have a (very) little experience on electronic projects. Is there > something I can do to help starting a similar project? > > Thank you and best regards. > Marco Signorini > > > > Tzafrir Cohen wrote: > >> Hi folks >> >> A common wisdom here is that one should use a proper hardware phone >> rather that an extra software on the user's PC. Why is that such a big >> issue? >> >> One thing that bothers me with the current crop of hardware SIP phones >> is that they are hopelessly properitary. >> >> So what would it take to build a fully-adaptable phone? >> >> Here are some of my thoughts. This is not anything I plan to do soon (if >> at all), but I really find it strange that there aren't such phones >> already. >> >> >> == Small Quantities: >> When you look at such systems it becomes aparant that you can get much >> nicer prices if you buy large quanities. But this is something that will >> be a problem. Not only for prototying. The fact that you're limited to a >> strict hardware setting is very limiting. No mixing and matching like in >> a standard PC. I'm not exactly sure how to overcome that. >> >> == Platforms: >> There are many embedded platforms nowadays. I assume that the relevant >> application requires some non-trivial CPU power. I would exclude e.g. a >> 486-based systems. My target phone should be able to handle at least two >> concurrent Speex calls. Preferrebly 6 speex calls and above. >> >> OTOH, I can't afford a monster CoreDuo. I need a quiet system with no >> fan. Thus the target CPU may be higher end VIA or Atom. Not sure about >> Geode. >> >> There are also some interesting ARM-based boards around. I'm completely >> unfamiliar with them but I suspect that they may prove to be cheaper. >> >> == SIP Software: >> Not really sure here. There must be something close to usable already, I >> guess. >> >> == Micro Browser: >> Hell no! >> >> The device should have an LCD display, and the content of that display >> should be programmable. Programming it using a HTML renderred is a bad >> design decision. >> >> The device should be a good phone. It should not attempt to be a web >> browser, as it will be a lousy one. >> >> == Handset: >> I suppose that an obvious starting point for a handset is "skype phones" >> such as USB handsets from yealink. Far from an optimal design, but a >> driver already exists. >> >> >> == Ease of Use: >> A phone must be usable. The target device must be something my mom can >> use. However that does not mean it must be easy to program. It must be >> programmable and hackable. But I can live with a complicated user >> interface for that. If such phones become successful and useful, better >> interfaces will eventually be written. >> >> >> >> > > > ___ > -- Bandwidth and Colocation Provided by http://www.api-digital.com -- > > asterisk-users mailing list > To UNSUBSCRIBE or update options visit: >http://lists.digium.com/mailman/listinfo/asterisk-users > > ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
Jon Pounder wrote: > Marco Signorini wrote: > >> It's a dream! >> It's since years that I'm thinking to have an open hardware project >> targeted to a SIP application. >> >> > > there is already a project called openmoko - join it and buy some hardware. > > The phone is large and clunky - the idea is good, but not something > you're ever going to carry in your pocket, and somewhat silly when there > is already smaller hardware out there that runs linux at less cost than > their device. > > Thank you Jon, Really interesting project! I'll follow it. Best regards, Marco Signorini ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
Thanks for your reply, On Fri, Feb 27, 2009 at 10:53:21AM -0500, SIP wrote: > Tzafrir Cohen wrote: > > Hi folks > > > > A common wisdom here is that one should use a proper hardware phone > > rather that an extra software on the user's PC. Why is that such a big > > issue? > > > > Marketability for one. People worldwide understand the telephone > paradigm. You have a handset and a box with numbers. You pick it up and > dial, talk through the handset, and listen in the other end. It's > simple. It's an elegant design. And everyone from 1 year olds to my 97 > year old grandfather can use it. > > Software phones? Not so much. In fact, not even close. The additional > complexity of running software on a machine ALONE would keep my > grandfather and that 1 year old from using it. Headsets? Seriously? > Since when have those been user-friendly OR comfortably. > > In essence, adherence to a software phone paradigm breaks a century of > design advancement in telephone ergonomics, psychology, and reliance, > and replaces it with something that's clearly just a kludgy add-on to a > product which was never originally designed for the task. > But imposes many stupid design limitations as well. A limitation of CPU power. A limitation of screen space. A limitation of a pointing device. A user has a keyboard to enter URLs. You can do that with a dialpad. Sort of. And you curse whoever invented that. How do you dial to an address pointed from the page you were browsing? > > == Small Quantities: > > When you look at such systems it becomes aparant that you can get much > > nicer prices if you buy large quanities. But this is something that will > > be a problem. Not only for prototying. The fact that you're limited to a > > strict hardware setting is very limiting. No mixing and matching like in > > a standard PC. I'm not exactly sure how to overcome that. > > > > This is one of the biggest reasons all the hardware phones are > proprietary -- they're each written for different basic hardware. That's no inherent reason for being proprietary. It's proprietary because that's how they can make money of it. I think that from selling a hardware for which there's a good programmable phone you can eventually make more money. But then that's pure speculation. > > == Ease of Use: > > A phone must be usable. The target device must be something my mom can > > use. However that does not mean it must be easy to program. It must be > > programmable and hackable. But I can live with a complicated user > > interface for that. If such phones become successful and useful, better > > interfaces will eventually be written. > > Just a note here -- a complicated user interface, though you personally > may be able to live with it, will pretty much ensure that the phones > never become successful enough for a better one to be written. UI design > is about 10% code and 90% psychology (and so FEW people who call > themselves UI 'programmers' understand that). Just having a UI that can > get you from point A to point B without typing in commands is NOT a UI > worth making, as it will never be a UI worth using. It's a UI worth making because I don't spend a year over it. And because I'm not a UI designer and hte phone is first and foremost for me :-) -- Tzafrir Cohen icq#16849755 jabber:tzafrir.co...@xorcom.com +972-50-7952406 mailto:tzafrir.co...@xorcom.com http://www.xorcom.com iax:gu...@local.xorcom.com/tzafrir ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
> I assume that the relevant application requires some non-trivial CPU power. I > would > exclude e.g. a 486-based systems. I'm not sure that's the case. The industry has gone in the direction of throwing lots of silicon at a problem, often as an excuse for poorly written code, sometimes in an interpreted language. There are a number of high integration CPUs out there that I suspect could do this sort of thing. I develop device controllers for a variety of industry needs. They tend to have Ethernet, RS-232, sometimes 1 Mb/s synchronous communication. G711, quarter VGA color LCD with touchscreen and control loops running at about a 1 ms rate. The entire code takes less than 256K in C. My choice of processor is the DStni Ex (made by Lantronix and sold by Grid Connect) which is a high integration, high speed 186 core with two 10/100 Ethernet Ports and 256K of RAM on it in addition to the usual assortment of other stuff. The above required platform adds three support chips (one being the LCD controller). The CPU can run over 100 MHz. Memory accesses take one clock and typical instructions take two or three. Cost is in the $10 to $20 range for the chip and power consumption is around 1 W (the LCD backlight takes more than that!) I'm sure there are several other comparable platforms out there, such as by Digi International. The Geode is a good candidate as are some VIA chips, if one wants to use protected mode x86. The biggest thing for this is don't even consider Intel. For most of their life they have not provided cutting edge solutions for embedded use. Most of their stuff consumes too much power. And most importantly, they are targeting the very volatile, short lived PC market. By the time you get an embedded design up and running and reach market penetration, you won't be able to buy the chip any more. Wilton ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
Wilton Helm wrote: > > I assume that the relevant application requires some non-trivial CPU > power. I would > > exclude e.g. a 486-based systems. > > I'm not sure that's the case. The industry has gone in the direction > of throwing lots of silicon at a problem, often as an excuse for > poorly written code, sometimes in an interpreted language. just look at windows for a good/bad example of that > There are a number of high integration CPUs out there that I suspect > could do this sort of thing. I develop device controllers for a > variety of industry needs. They tend to have Ethernet, RS-232, > sometimes 1 Mb/s synchronous communication. G711, quarter VGA color > LCD with touchscreen and control loops running at about a 1 ms rate. > The entire code takes less than 256K in C. My choice of processor is > the DStni Ex (made by Lantronix and sold by Grid Connect) which is a > high integration, high speed 186 core with two 10/100 Ethernet Ports > and 256K of RAM on it in addition to the usual assortment of other > stuff. The above required platform adds three support chips (one > being the LCD controller). The CPU can run over 100 MHz. Memory > accesses take one clock and typical instructions take two or three. > Cost is in the $10 to $20 range for the chip and power consumption is > around 1 W (the LCD backlight takes more than that!) These days if you are writing stuff in C, the hardware platform is really not of ultimate importance since you can cross compile with gcc from just about anything to anything. What matters more is there are drivers available for all the other bits besides the cpu itself. I was just saying the other day, how I used to write pretty involved programs in 256 bytes in the original basic stamps. Most people today could not even conceive how a program could do anything useful in 256bytes let alone 256k or 256mb. > > I'm sure there are several other comparable platforms out there, such > as by Digi International. The Geode is a good candidate as are some > VIA chips, if one wants to use protected mode x86. The biggest thing > for this is don't even consider Intel. For most of their life they > have not provided cutting edge solutions for embedded use. Most of > their stuff consumes too much power. And most importantly, they are > targeting the very volatile, short lived PC market. By the time you > get an embedded design up and running and reach market penetration, > you won't be able to buy the chip any more. > > Wilton > > > > ___ > -- Bandwidth and Colocation Provided by http://www.api-digital.com -- > > asterisk-users mailing list > To UNSUBSCRIBE or update options visit: >http://lists.digium.com/mailman/listinfo/asterisk-users ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
On Fri, 27 Feb 2009, Tzafrir Cohen wrote: > Hi folks > > A common wisdom here is that one should use a proper hardware phone > rather that an extra software on the user's PC. Why is that such a big > issue? Both my laptop and desktop PC have poor quality microphone inputs. I use a USB "phone" on my laptop when out and about. > One thing that bothers me with the current crop of hardware SIP phones > is that they are hopelessly properitary. SIP is an open standard... > So what would it take to build a fully-adaptable phone? Persuade an existing manufacturer to provide an SDK for their phone... And thinking about it, don't Snoms run Linux? anyone asked if an SDK is avalable? > Here are some of my thoughts. This is not anything I plan to do soon (if > at all), but I really find it strange that there aren't such phones > already. > > > == Small Quantities: > When you look at such systems it becomes aparant that you can get much > nicer prices if you buy large quanities. But this is something that will > be a problem. Not only for prototying. The fact that you're limited to a > strict hardware setting is very limiting. No mixing and matching like in > a standard PC. I'm not exactly sure how to overcome that. > > == Platforms: > There are many embedded platforms nowadays. I assume that the relevant > application requires some non-trivial CPU power. I would exclude e.g. a > 486-based systems. My target phone should be able to handle at least two > concurrent Speex calls. Preferrebly 6 speex calls and above. > > OTOH, I can't afford a monster CoreDuo. I need a quiet system with no > fan. Thus the target CPU may be higher end VIA or Atom. Not sure about > Geode. > > There are also some interesting ARM-based boards around. I'm completely > unfamiliar with them but I suspect that they may prove to be cheaper. Custom DSP or ARM. Don't forget power requirements too. PoE may well be able to supply 15W, but imagine a building with 100 x 15W phones... Everything I've plugged into my meter idles at about 2W (Grandstream, Snom, Siemens, ATL) > == SIP Software: > Not really sure here. There must be something close to usable already, I > guess. > > == Micro Browser: > Hell no! "Executives" want it... > The device should have an LCD display, and the content of that display > should be programmable. Programming it using a HTML renderred is a bad > design decision. > > The device should be a good phone. It should not attempt to be a web > browser, as it will be a lousy one. I've actually used the RSS reader in my Grandstream Video phones.. It's usable, but I take your point about the full web thing! > == Handset: > I suppose that an obvious starting point for a handset is "skype phones" > such as USB handsets from yealink. Far from an optimal design, but a > driver already exists. Seen these: http://www.fit-pc.co.uk/meet-fit-pc.html#tiny And of-course the power line thing mentioned here a few days earlier - eg. http://www.theinquirer.net/inquirer/news/143/1051143/pc-plug-coming-soon-wall-near if you're going to use a USB phone, these platforms are ideal, if a little pricey.. I did play with a yealink phone and Linux some time back - get the display and keyboard working and it'll be very functional... (I got it to almost work with Zoiper, so a custom app. would be easy) Alas, I gave it to a friend, then go anothe from the same source (tesco), thinking it would be identical - and it was - in packaging, but was a totally different phone )-: > == Ease of Use: > A phone must be usable. The target device must be something my mom can > use. However that does not mean it must be easy to program. It must be > programmable and hackable. But I can live with a complicated user > interface for that. If such phones become successful and useful, better > interfaces will eventually be written. Dial the number, push the green button, off you go... Gordon ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
2009/2/27 Wilton Helm > > I assume that the relevant application requires some non-trivial CPU > power. I would > > exclude e.g. a 486-based systems. > > I'm not sure that's the case. The industry has gone in the direction of > throwing lots of silicon at a problem, often as an excuse for poorly written > code, sometimes in an interpreted language. There are a number of high > integration CPUs out there that I suspect could do this sort of thing. I > develop device controllers for a variety of industry needs. They tend to > have Ethernet, RS-232, sometimes 1 Mb/s synchronous communication. G711, > quarter VGA color LCD with touchscreen and control loops running at about a > 1 ms rate. The entire code takes less than 256K in C. My choice of > processor is the DStni Ex (made by Lantronix and sold by Grid Connect) which > is a high integration, high speed 186 core with two 10/100 Ethernet Ports > and 256K of RAM on it in addition to the usual assortment of other stuff. > The above required platform adds three support chips (one being the LCD > controller). The CPU can run over 100 MHz. Memory accesses take one clock > and typical instructions take two or three. Cost is in the $10 to $20 range > for the chip and power consumption is around 1 W (the LCD backlight takes > more than that!) > > I'm sure there are several other comparable platforms out there, such as by > Digi International. The Geode is a good candidate as are some VIA chips, if > one wants to use protected mode x86. The biggest thing for this is don't > even consider Intel. For most of their life they have not provided cutting > edge solutions for embedded use. Most of their stuff consumes too much > power. And most importantly, they are targeting the very volatile, short > lived PC market. By the time you get an embedded design up and running and > reach market penetration, you won't be able to buy the chip any more. > > Wilton > > > ___ > -- Bandwidth and Colocation Provided by http://www.api-digital.com -- > > asterisk-users mailing list > To UNSUBSCRIBE or update options visit: > http://lists.digium.com/mailman/listinfo/asterisk-users > I wonder what kind of hardware snom use, they got linux, they got openvpn. I would be nice to have that, and yes i want a gui, maybe not embedded to reduce load, but something like an external config generator software would be nice. ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
On Fri, Feb 27, 2009 at 09:40:09AM -0700, Wilton Helm wrote: > > I assume that the relevant application requires some non-trivial CPU power. > > I would > > exclude e.g. a 486-based systems. > > I'm not sure that's the case. The industry has gone in the direction > of throwing lots of silicon at a problem, often as an excuse for > poorly written code, sometimes in an interpreted language. Actually I have stated a bit above that that I would like the system to support 2 SIP lines (and preferably 6 and more) and allow using Speex in them. Supposrt of 2 lines is a must for features such as forwarding. > There are a number of high integration CPUs out there that I suspect > could do this sort of thing. I develop device controllers for a > variety of industry needs. They tend to have Ethernet, RS-232, > sometimes 1 Mb/s synchronous communication. G711, G.711 indeed requires much less CPU power. It is good enough for a LAN-only setup. Though with a bit of forward-thinking, G.722 support would also be nice. > quarter VGA color LCD with touchscreen and control loops running at > about a 1 ms rate. The entire code takes less than 256K in C. My > choice of processor is the DStni Ex (made by Lantronix and sold by > Grid Connect) which is a high integration, high speed 186 core with > two 10/100 Ethernet Ports and 256K of RAM on it in addition to the > usual assortment of other stuff. The above required platform adds > three support chips (one being the LCD controller). The CPU can run > over 100 MHz. Memory accesses take one clock and typical instructions > take two or three. Cost is in the $10 to $20 range for the chip and > power consumption is around 1 W (the LCD backlight takes more than > that!) Again, the main reason for me to require a higher end CPU is audio compression. But I also want the system to be run by a standard OS. It needs to be easy to add your own application there. Here's a plug that would cost you 99$: http://www.globalscaletechnologies.com/t-sheevaplugdetails.aspx#extern It has a USB and Ethernet output. The core of the system is: http://www.marvell.com/products/embedded_processors/kirkwood/index.jsp (yes, also two TDM ports with slics. No idea how to use them) That plug supports running quite a number of Linux distributions (Debian, Fedora, Gentoo, and some others) What would it take me to port Asterisk or Yate to it? An existing web interface? Hackability and ease of development is a must. -- Tzafrir Cohen icq#16849755 jabber:tzafrir.co...@xorcom.com +972-50-7952406 mailto:tzafrir.co...@xorcom.com http://www.xorcom.com iax:gu...@local.xorcom.com/tzafrir ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
On Fri, 27 Feb 2009, Tzafrir Cohen wrote: > The core of the system is: > http://www.marvell.com/products/embedded_processors/kirkwood/index.jsp > (yes, also two TDM ports with slics. No idea how to use them) > > That plug supports running quite a number of Linux distributions > (Debian, Fedora, Gentoo, and some others) > > What would it take me to port Asterisk or Yate to it? An existing web > interface? You don't want a PBX in it - you want a command-line VoIP client. I was asking (here and elsewhere) if such a beast existed some time back. All I got back was "put asterisk on it and use the OSS/ALSA stuff". Talk about using a sledge hammer to crack a nut.. I had a quick look at libiax and ran out of time... > Hackability and ease of development is a must. Quite. Now, way back I have (still have) a Nokia 770 tablet - that runs Gizmo - but, again it's a GUI application. Get a basic OS on the box, get a command-line phone that uses USB audio, stick a mini web server for configuration and off you go... Gordon ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
On Fri, Feb 27, 2009 at 05:48:28PM +, Gordon Henderson wrote: > On Fri, 27 Feb 2009, Tzafrir Cohen wrote: > > > The core of the system is: > > http://www.marvell.com/products/embedded_processors/kirkwood/index.jsp > > (yes, also two TDM ports with slics. No idea how to use them) > > > > That plug supports running quite a number of Linux distributions > > (Debian, Fedora, Gentoo, and some others) > > > > What would it take me to port Asterisk or Yate to it? An existing web > > interface? > > You don't want a PBX in it - you want a command-line VoIP client. > > I was asking (here and elsewhere) if such a beast existed some time back. > All I got back was "put asterisk on it and use the OSS/ALSA stuff". Talk > about using a sledge hammer to crack a nut.. I had a quick look at libiax > and ran out of time... Yate is the only h32, IAX and SIP free software soft phone I know. I figure adding one extra interface wouldn't be that tough :-) Linphone has a command-line interface. Not sure about others. But then again, maybe the system will run X11. That can simplify some hings. -- Tzafrir Cohen icq#16849755 jabber:tzafrir.co...@xorcom.com +972-50-7952406 mailto:tzafrir.co...@xorcom.com http://www.xorcom.com iax:gu...@local.xorcom.com/tzafrir ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
Gordon Henderson wrote: > On Fri, 27 Feb 2009, Tzafrir Cohen wrote: > > >> The core of the system is: >> http://www.marvell.com/products/embedded_processors/kirkwood/index.jsp >> (yes, also two TDM ports with slics. No idea how to use them) >> >> That plug supports running quite a number of Linux distributions >> (Debian, Fedora, Gentoo, and some others) >> >> What would it take me to port Asterisk or Yate to it? An existing web >> interface? >> > > You don't want a PBX in it - you want a command-line VoIP client. > > I was asking (here and elsewhere) if such a beast existed some time back. > All I got back was "put asterisk on it and use the OSS/ALSA stuff". Talk > about using a sledge hammer to crack a nut.. I had a quick look at libiax > and ran out of time... > > this mentality mystifies me too - why would you want a pbx running on a handset ? ok so you can, but once that novelty wears off in 5secs what is the practical use ? The only reason would be if you don't have any other phones, and if you don't why do you need a pbx to start with ? >> Hackability and ease of development is a must. >> > > Quite. > > Now, way back I have (still have) a Nokia 770 tablet - that runs Gizmo - > but, again it's a GUI application. > > Get a basic OS on the box, get a command-line phone that uses USB audio, > stick a mini web server for configuration and off you go... > on this subject - any tips on getting gizmo to actually work ? I could get rings through but no audio or even connecting the call after ringing, and its on a network where ekiga and grandstream phones work just fine with no stun or anything like that (network is private ips but directly routed to the server) > Gordon > > ___ > -- Bandwidth and Colocation Provided by http://www.api-digital.com -- > > asterisk-users mailing list > To UNSUBSCRIBE or update options visit: >http://lists.digium.com/mailman/listinfo/asterisk-users > > ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
>Again, the main reason for me to require a higher end CPU is audio >compression. But I also want the system to be run by a standard OS. It >needs to be easy to add your own application there. Mutually exclusive. I don't know any standard OS that doesn't waste about 10 x as many CPU cycles as a SIP phone should ever need. The problem is generalization. A standard OS is designed to support a wide variety of devices, including a wide range of screen sizes. The abstraction layers that make this possible often consume more CPU resources than the application they are supporting. Most of that isn't needed for this application. Compatibility with WXVGA isn't required. Even a full blown file system is a luxury. Linux is about the closest thing because it can be pared down. But it takes someone with considerable experience to know how and what to trim. I supervised a system that used "busy box" to create a compact system that lived on a small flash card an some RAM. As an example, I have been a Palm owner for a number of years. I laughed when the Win CE stuff came out to compete. The Palm OS was written for the task at hand. I could go a week or more on a charge. The Win CE devices had to be recharged after 8 hours! Why? The OS required too much which required far more compute power, which ate batteries. The SIP phone you propose could be done with about 1 W of power plus a couple more for backlighting. An OS based version would start at 5 W + backlight and could easily go to 15 W or higher. Not the end of the world, I suppose on a desk (if there aren't a hundred of them and I'm not paying the electric bill) but a huge difference if it has to run on batteries. Wilton ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
Tzafrir Cohen wrote: > On Fri, Feb 27, 2009 at 05:48:28PM +, Gordon Henderson wrote: > >> On Fri, 27 Feb 2009, Tzafrir Cohen wrote: >> >> >>> The core of the system is: >>> http://www.marvell.com/products/embedded_processors/kirkwood/index.jsp >>> (yes, also two TDM ports with slics. No idea how to use them) >>> >>> That plug supports running quite a number of Linux distributions >>> (Debian, Fedora, Gentoo, and some others) >>> >>> What would it take me to port Asterisk or Yate to it? An existing web >>> interface? >>> >> You don't want a PBX in it - you want a command-line VoIP client. >> >> I was asking (here and elsewhere) if such a beast existed some time back. >> All I got back was "put asterisk on it and use the OSS/ALSA stuff". Talk >> about using a sledge hammer to crack a nut.. I had a quick look at libiax >> and ran out of time... >> > > Yate is the only h32, IAX and SIP free software soft phone I know. > I figure adding one extra interface wouldn't be that tough :-) > > Linphone has a command-line interface. Not sure about others. But then > again, maybe the system will run X11. That can simplify some hings. > > ekiga (formerly gnome-meeting) seems to be the most included in distros etc., but it has a lot of shortcomings and is buggy IMO - continuously reruns wizard for some reason on a mint distro, wants to bind to vpn ip when its up instead of sticking to the previous setting, needs -c in the command line for calls so can't easily handle a sip url from a browser with a standard setting for helpers, barfs when you run from command line like that if there is already an instance running (should - just tell the original instance to handle the call via some form of ipc and shut down the new instance) they all seem like pretty easy problems to fix but how did something get to this state of maturity with some major shortcomings ? anyone know anything better ? ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
Wilton Helm wrote: > >Again, the main reason for me to require a higher end CPU is audio > >compression. But I also want the system to be run by a standard OS. It > >needs to be easy to add your own application there. > > Mutually exclusive. I don't know any standard OS that doesn't waste > about 10 x as many CPU cycles as a SIP phone should ever need. The > problem is generalization. A standard OS is designed to support a > wide variety of devices, including a wide range of screen sizes. The > abstraction layers that make this possible often consume more CPU > resources than the application they are supporting. Most of that > isn't needed for this application. Compatibility with WXVGA isn't > required. Even a full blown file system is a luxury. This is not entirely true - many of the nokia phones use a java OS as a core, and you can load pretty much any java software you want on them, but all the points about power and battery use are still valid. (and whether you really consider that truly an OS is questionable, but its out there) > > Linux is about the closest thing because it can be pared down. But it > takes someone with considerable experience to know how and what to > trim. I supervised a system that used "busy box" to create a compact > system that lived on a small flash card an some RAM. > > As an example, I have been a Palm owner for a number of years. I > laughed when the Win CE stuff came out to compete. The Palm OS was > written for the task at hand. I could go a week or more on a charge. > The Win CE devices had to be recharged after 8 hours! Why? The OS > required too much which required far more compute power, which ate > batteries. > > The SIP phone you propose could be done with about 1 W of power plus a > couple more for backlighting. An OS based version would start at 5 W > + backlight and could easily go to 15 W or higher. Not the end of the > world, I suppose on a desk (if there aren't a hundred of them and I'm > not paying the electric bill) but a huge difference if it has to run > on batteries. > > Wilton > > > > ___ > -- Bandwidth and Colocation Provided by http://www.api-digital.com -- > > asterisk-users mailing list > To UNSUBSCRIBE or update options visit: >http://lists.digium.com/mailman/listinfo/asterisk-users ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
On Fri, Feb 27, 2009 at 01:07:57PM -0500, Jon Pounder wrote: > this mentality mystifies me too - why would you want a pbx running on a > handset ? ok so you can, but once that novelty wears off in 5secs what > is the practical use ? The only reason would be if you don't have any > other phones, and if you don't why do you need a pbx to start with ? Because almost all the free software phones I know are X11 :-) I would prefer a decent "phone". But then again, if a certain software is also a PBX in addition to being a phone, I don't really care. I would care if being a PBX makes it worse as a phone than some alternatives. -- Tzafrir Cohen icq#16849755 jabber:tzafrir.co...@xorcom.com +972-50-7952406 mailto:tzafrir.co...@xorcom.com http://www.xorcom.com iax:gu...@local.xorcom.com/tzafrir ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
On Fri, Feb 27, 2009 at 11:11:35AM -0700, Wilton Helm wrote: > >Again, the main reason for me to require a higher end CPU is audio > >compression. But I also want the system to be run by a standard OS. It > >needs to be easy to add your own application there. > > > Mutually exclusive. I don't know any standard OS that doesn't waste > about 10 x as many CPU cycles as a SIP phone should ever need. It's not about what is wasted. It's about what you're left with. Anyway, my new home PBX is an Alix (alix6b2) unit: http://www.pcengines.ch/alix6b2.htm It's running Debian Lenny. chao:~# cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 5 model : 10 model name : Geode(TM) Integrated Processor by AMD PCS stepping: 2 cpu MHz : 498.056 cache size : 128 KB fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu de pse tsc msr cx8 sep pge cmov clflush mmx mmxext 3dnowext 3dnow bogomips: 998.10 clflush size: 32 power management: chao:~# free total used free sharedbuffers cached Mem:256640 251900 4740 0 78488 145972 -/+ buffers/cache: 27440 229200 Swap: 224868 52 224816 It also has an openvpn link, an Astribank device and the network of my very small LAN. Its standard load avarage is, well, 0. I never even bothered optimizing it. > The > problem is generalization. A standard OS is designed to support a > wide variety of devices, including a wide range of screen sizes. > The abstraction layers that make this possible often consume more > CPU resources than the application they are supporting. Most of that > isn't needed for this application. Compatibility with WXVGA isn't > required. Even a full blown file system is a luxury. Ask yourself how all those small mp4 players even work. I suppose most of them run Linux. See also http://www.rockbox.org/ > > Linux is about the closest thing because it can be pared down. But it > takes someone with considerable experience to know how and what to > trim. I supervised a system that used "busy box" to create a compact > system that lived on a small flash card an some RAM. This trimming can also be automated, if I see that a standard distribution will not do. See e.g. http://astlinux.org/ . > > As an example, I have been a Palm owner for a number of years. I > laughed when the Win CE stuff came out to compete. The Palm OS was > written for the task at hand. I could go a week or more on a charge. > The Win CE devices had to be recharged after 8 hours! Why? The OS > required too much which required far more compute power, which ate > batteries. The device does not run on batteries. The plug I mentioned above is 5W. Here's what happens when someone starts optimizing the power consumption: http://www.rowetel.com/blog/?p=61 Anyway, a general-purpose OS consumes extra memory, but relatively little power. I expect that a SIP phone (Requiring transport of many small packet => many interrupts) is a large power consumer. > > The SIP phone you propose could be done with about 1 W of power plus > a couple more for backlighting. An OS based version would start at > 5 W + backlight And this is based on? > and could easily go to 15 W or higher. Not the end of the world, > I suppose on a desk (if there aren't a hundred of them and I'm not > paying the electric bill) but a huge difference if it has to run > on batteries. -- Tzafrir Cohen icq#16849755 jabber:tzafrir.co...@xorcom.com +972-50-7952406 mailto:tzafrir.co...@xorcom.com http://www.xorcom.com iax:gu...@local.xorcom.com/tzafrir ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
>This is not entirely true - many of the nokia phones use a java OS as a >core, and you can load pretty much any java software you want on them, >but all the points about power and battery use are still valid. (and >whether you really consider that truly an OS is questionable, but its >out there) Java is the worst offender. Its resource requirements often exceed those of the application it is running. Java is useful for things like displaying web pages that are not time critical and where its write once, run everywhere philosophy is valuable. But anyone trying to actually do things like I/O control, call setup, transcoding, etc. in Java are asking for every issue I raised. If WCE can get 8 hours of battery life, Java would be about 3. Wilton ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
Witness the fact that the old Pingtel phones ran Java, and they were incredibly lame. I think part of what this thread misses is that DSP is a god chunk of what SIP phones need. A general purpose CPU is not the right tool for the task. A cheap DSP is better suited to compression, transcoding, etc. OTOH, presuming that the snom phones are Linux on a suitable platform soomeone could develop a custom software load for them and OEM the hardware. Michael --Original Message Text--- From: Wilton Helm Date: Fri, 27 Feb 2009 12:28:40 -0700 >This is not entirely true - many of the nokia phones use a java OS as a >core, and you can load pretty much any java software you want on them, >but all the points about power and battery use are still valid. (and >whether you really consider that truly an OS is questionable, but its >out there) Java is the worst offender. Its resource requirements often exceed those of the application it is running. Java is useful for things like displaying web pages that are not time critical and where its write once, run everywhere philosophy is valuable. But anyone trying to actually do things like I/O control, call setup, transcoding, etc. in Java are asking for every issue I raised. If WCE can get 8 hours of battery life, Java would be about 3. Wilton -- Michael Graves mgravesmstvp.com http://blog.mgraves.org o713-861-4005 c713-201-1262 sip:mgra...@mstvp.onsip.com skype mjgraves fwd 54245 ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
On Fri, Feb 27, 2009 at 09:39:59PM -0600, Michael Graves wrote: > Witness the fact that the old Pingtel phones ran Java, and they were > incredibly lame. > > I think part of what this thread misses is that DSP is a god chunk of > what SIP phones need. A general purpose CPU is not the right tool for > the task. A cheap DSP is better suited to compression, transcoding, > etc. Maybe. But it is also a great limitation on the hackability and thus on the speed of development. I think that general-purpose processors are strong enough for that. -- Tzafrir Cohen icq#16849755 jabber:tzafrir.co...@xorcom.com +972-50-7952406 mailto:tzafrir.co...@xorcom.com http://www.xorcom.com iax:gu...@local.xorcom.com/tzafrir ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
Michael Graves wrote: > Witness the fact that the old Pingtel phones ran Java, and they were > incredibly lame. > > I think part of what this thread misses is that DSP is a god chunk of > what SIP phones need. A general purpose CPU is not the right tool for > the task. A cheap DSP is better suited to compression, transcoding, etc. > > OTOH, presuming that the snom phones are Linux on a suitable platform > soomeone could develop a custom software load for them and OEM the > hardware. I'm surprised that no-one has mentioned Astfin. Basically uClinux and asterisk running on an Analog Devices Blackfin DSP. There's also some 'open source' hardware that's available - the IP04 and friends. I'm using an Edgepbx FX08, and they also have a two-port version (FX02). Atcom has a single-port one, the IP01. Though if I were going to prototype an 'open' SIP phone, I'd probably start with a beagle board (TI OMAP3530 - dual-core ARM+DSP). It's a pretty powerful SOC - its brother (3430) powers the Palm Pre. Just another datapoint :) Paul Some related links, if anyone's curious: http://beagleboard.org/ http://www.astfin.org/ http://www.rowetel.com/ucasterisk/ http://www.atcom.cn/En_products_IP01.htm http://www.edgepbx.com/ ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
On Sat, 28 Feb 2009 14:59:23 -0800, Paul Chambers wrote: >Michael Graves wrote: >> Witness the fact that the old Pingtel phones ran Java, and they were >> incredibly lame. >> >> I think part of what this thread misses is that DSP is a god chunk of >> what SIP phones need. A general purpose CPU is not the right tool for >> the task. A cheap DSP is better suited to compression, transcoding, etc. >> >> OTOH, presuming that the snom phones are Linux on a suitable platform >> soomeone could develop a custom software load for them and OEM the >> hardware. >I'm surprised that no-one has mentioned Astfin. Basically uClinux and >asterisk running on an Analog Devices Blackfin DSP. There's also some >'open source' hardware that's available - the IP04 and friends. I'm >using an Edgepbx FX08, and they also have a two-port version (FX02). >Atcom has a single-port one, the IP01. > >Though if I were going to prototype an 'open' SIP phone, I'd probably >start with a beagle board (TI OMAP3530 - dual-core ARM+DSP). It's a >pretty powerful SOC - its brother (3430) powers the Palm Pre. > >Just another datapoint :) Yeah, that'd be great hardware to select. What I was thinking is that this thread seems to be driven by those of a software bent. For that group perhaps there's an opportunity to write code for something like a snom 820. It's a solid solid hardware basis for the project. Snom would be foolish not to sell it for such use, even price it attractively. That way the hardware work would be done, and the software geeks could work their magic. Michael -- Michael Graves mgravesmstvp.com http://blog.mgraves.org o713-861-4005 c713-201-1262 sip:mgra...@mstvp.onsip.com skype mjgraves fwd 54245 ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
Michael Graves wrote: > On Sat, 28 Feb 2009 14:59:23 -0800, Paul Chambers wrote: > >> Michael Graves wrote: >> >>> Witness the fact that the old Pingtel phones ran Java, and they were >>> incredibly lame. >>> >>> I think part of what this thread misses is that DSP is a god chunk of >>> what SIP phones need. A general purpose CPU is not the right tool for >>> the task. A cheap DSP is better suited to compression, transcoding, etc. >>> >>> OTOH, presuming that the snom phones are Linux on a suitable platform >>> soomeone could develop a custom software load for them and OEM the >>> hardware. >>> >> I'm surprised that no-one has mentioned Astfin. Basically uClinux and >> asterisk running on an Analog Devices Blackfin DSP. There's also some >> 'open source' hardware that's available - the IP04 and friends. I'm >> using an Edgepbx FX08, and they also have a two-port version (FX02). >> Atcom has a single-port one, the IP01. >> >> Though if I were going to prototype an 'open' SIP phone, I'd probably >> start with a beagle board (TI OMAP3530 - dual-core ARM+DSP). It's a >> pretty powerful SOC - its brother (3430) powers the Palm Pre. >> >> Just another datapoint :) >> > > Yeah, that'd be great hardware to select. > > What I was thinking is that this thread seems to be driven by those of > a software bent. For that group perhaps there's an opportunity to write > code for something like a snom 820. It's a solid solid hardware basis > for the project. Snom would be foolish not to sell it for such use, > even price it attractively. That way the hardware work would be done, > and the software geeks could work their magic. > I'm a card-carrying (embedded linux) software geek, and I know I'd be interested :) Anyone got some influencial contacts inside Snom? or Aastra, for that matter, their hardware also seems good quality from what people have said. Another possibility is talking to Atcom (or other VoIP ODMs), they seem to have done pretty well from the IP04 and derivatives. They've experienced the benefits of an open development model, perhaps they'd be interested. Not sure what the quality of their existing handset hardware is like. Anyone on the list have the contacts to get the ball rolling? Paul ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
I have influential contacts inside snom... CS -Ursprüngliche Nachricht- Von: asterisk-users-boun...@lists.digium.com [mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Paul Chambers Gesendet: Sonntag, 1. März 2009 01:30 An: Asterisk Users Mailing List - Non-Commercial Discussion Betreff: Re: [asterisk-users] building a phone Michael Graves wrote: > On Sat, 28 Feb 2009 14:59:23 -0800, Paul Chambers wrote: > >> Michael Graves wrote: >> >>> Witness the fact that the old Pingtel phones ran Java, and they were >>> incredibly lame. >>> >>> I think part of what this thread misses is that DSP is a god chunk of >>> what SIP phones need. A general purpose CPU is not the right tool for >>> the task. A cheap DSP is better suited to compression, transcoding, etc. >>> >>> OTOH, presuming that the snom phones are Linux on a suitable platform >>> soomeone could develop a custom software load for them and OEM the >>> hardware. >>> >> I'm surprised that no-one has mentioned Astfin. Basically uClinux and >> asterisk running on an Analog Devices Blackfin DSP. There's also some >> 'open source' hardware that's available - the IP04 and friends. I'm >> using an Edgepbx FX08, and they also have a two-port version (FX02). >> Atcom has a single-port one, the IP01. >> >> Though if I were going to prototype an 'open' SIP phone, I'd probably >> start with a beagle board (TI OMAP3530 - dual-core ARM+DSP). It's a >> pretty powerful SOC - its brother (3430) powers the Palm Pre. >> >> Just another datapoint :) >> > > Yeah, that'd be great hardware to select. > > What I was thinking is that this thread seems to be driven by those of > a software bent. For that group perhaps there's an opportunity to write > code for something like a snom 820. It's a solid solid hardware basis > for the project. Snom would be foolish not to sell it for such use, > even price it attractively. That way the hardware work would be done, > and the software geeks could work their magic. > I'm a card-carrying (embedded linux) software geek, and I know I'd be interested :) Anyone got some influencial contacts inside Snom? or Aastra, for that matter, their hardware also seems good quality from what people have said. Another possibility is talking to Atcom (or other VoIP ODMs), they seem to have done pretty well from the IP04 and derivatives. They've experienced the benefits of an open development model, perhaps they'd be interested. Not sure what the quality of their existing handset hardware is like. Anyone on the list have the contacts to get the ball rolling? Paul ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
On Sun, 1 Mar 2009 18:14:18 +0100, Christian Stredicke wrote: >I have influential contacts inside snom... > >CS So you do! What do you think? Would snom be interested in selling hardware into a group of users who would be loading community developed application firmware? It makes me wonder how many little routers Cisco sells that actually get loaded with WRT-DD and the like? Michael >-Ursprüngliche Nachricht- >Von: asterisk-users-boun...@lists.digium.com >[mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Paul Chambers >Gesendet: Sonntag, 1. März 2009 01:30 >An: Asterisk Users Mailing List - Non-Commercial Discussion >Betreff: Re: [asterisk-users] building a phone > >Michael Graves wrote: >> On Sat, 28 Feb 2009 14:59:23 -0800, Paul Chambers wrote: >> >>> Michael Graves wrote: >>> >>>> Witness the fact that the old Pingtel phones ran Java, and they were >>>> incredibly lame. >>>> >>>> I think part of what this thread misses is that DSP is a god chunk of >>>> what SIP phones need. A general purpose CPU is not the right tool for >>>> the task. A cheap DSP is better suited to compression, transcoding, etc. >>>> >>>> OTOH, presuming that the snom phones are Linux on a suitable platform >>>> soomeone could develop a custom software load for them and OEM the >>>> hardware. >>>> >>> I'm surprised that no-one has mentioned Astfin. Basically uClinux and >>> asterisk running on an Analog Devices Blackfin DSP. There's also some >>> 'open source' hardware that's available - the IP04 and friends. I'm >>> using an Edgepbx FX08, and they also have a two-port version (FX02). >>> Atcom has a single-port one, the IP01. >>> >>> Though if I were going to prototype an 'open' SIP phone, I'd probably >>> start with a beagle board (TI OMAP3530 - dual-core ARM+DSP). It's a >>> pretty powerful SOC - its brother (3430) powers the Palm Pre. >>> >>> Just another datapoint :) >>> >> >> Yeah, that'd be great hardware to select. >> >> What I was thinking is that this thread seems to be driven by those of >> a software bent. For that group perhaps there's an opportunity to write >> code for something like a snom 820. It's a solid solid hardware basis >> for the project. Snom would be foolish not to sell it for such use, >> even price it attractively. That way the hardware work would be done, >> and the software geeks could work their magic. >> >I'm a card-carrying (embedded linux) software geek, and I know I'd be >interested :) > >Anyone got some influencial contacts inside Snom? or Aastra, for that >matter, their hardware also seems good quality from what people have said. > >Another possibility is talking to Atcom (or other VoIP ODMs), they seem >to have done pretty well from the IP04 and derivatives. They've >experienced the benefits of an open development model, perhaps they'd be >interested. Not sure what the quality of their existing handset hardware >is like. > >Anyone on the list have the contacts to get the ball rolling? > >Paul > >___ >-- Bandwidth and Colocation Provided by http://www.api-digital.com -- > >asterisk-users mailing list >To UNSUBSCRIBE or update options visit: > http://lists.digium.com/mailman/listinfo/asterisk-users > > > >___ >-- Bandwidth and Colocation Provided by http://www.api-digital.com -- > >asterisk-users mailing list >To UNSUBSCRIBE or update options visit: > http://lists.digium.com/mailman/listinfo/asterisk-users > -- Michael Graves mgravesmstvp.com http://blog.mgraves.org o713-861-4005 c713-201-1262 sip:mgra...@mstvp.onsip.com skype mjgraves fwd 54245 ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
To be honest: I am not very optimistic regarding this project. The WRT is really a case where you essentially use stuff that is already available and which is very very stable (e.g. Linux). There is nothing really special for the WRT. For a phone, the picture looks different. There are so many components necessary that are either not available or not very stable. There is a tremendous risk of ending up with a project that has a lot of loose ends. But if someone wants to give it a try, sure. We have nothing to lose! Those who know embedded Linux will easily feel like home on the phone once they are logged in. Definitively an interesting topic for our Asterisk developer meeting that we want to run this month in Berlin. Maybe for starters we just compile an Asterisk and run it on the phone. That will be fun! CS -Ursprüngliche Nachricht- Von: asterisk-users-boun...@lists.digium.com [mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Michael Graves Gesendet: Sonntag, 1. März 2009 18:30 An: Asterisk Users Mailing List - Non-Commercial Discussion Betreff: Re: [asterisk-users] building a phone On Sun, 1 Mar 2009 18:14:18 +0100, Christian Stredicke wrote: >I have influential contacts inside snom... > >CS So you do! What do you think? Would snom be interested in selling hardware into a group of users who would be loading community developed application firmware? It makes me wonder how many little routers Cisco sells that actually get loaded with WRT-DD and the like? Michael >-Ursprüngliche Nachricht- >Von: asterisk-users-boun...@lists.digium.com >[mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Paul Chambers >Gesendet: Sonntag, 1. März 2009 01:30 >An: Asterisk Users Mailing List - Non-Commercial Discussion >Betreff: Re: [asterisk-users] building a phone > >Michael Graves wrote: >> On Sat, 28 Feb 2009 14:59:23 -0800, Paul Chambers wrote: >> >>> Michael Graves wrote: >>> >>>> Witness the fact that the old Pingtel phones ran Java, and they were >>>> incredibly lame. >>>> >>>> I think part of what this thread misses is that DSP is a god chunk of >>>> what SIP phones need. A general purpose CPU is not the right tool for >>>> the task. A cheap DSP is better suited to compression, transcoding, etc. >>>> >>>> OTOH, presuming that the snom phones are Linux on a suitable platform >>>> soomeone could develop a custom software load for them and OEM the >>>> hardware. >>>> >>> I'm surprised that no-one has mentioned Astfin. Basically uClinux and >>> asterisk running on an Analog Devices Blackfin DSP. There's also some >>> 'open source' hardware that's available - the IP04 and friends. I'm >>> using an Edgepbx FX08, and they also have a two-port version (FX02). >>> Atcom has a single-port one, the IP01. >>> >>> Though if I were going to prototype an 'open' SIP phone, I'd probably >>> start with a beagle board (TI OMAP3530 - dual-core ARM+DSP). It's a >>> pretty powerful SOC - its brother (3430) powers the Palm Pre. >>> >>> Just another datapoint :) >>> >> >> Yeah, that'd be great hardware to select. >> >> What I was thinking is that this thread seems to be driven by those of >> a software bent. For that group perhaps there's an opportunity to write >> code for something like a snom 820. It's a solid solid hardware basis >> for the project. Snom would be foolish not to sell it for such use, >> even price it attractively. That way the hardware work would be done, >> and the software geeks could work their magic. >> >I'm a card-carrying (embedded linux) software geek, and I know I'd be >interested :) > >Anyone got some influencial contacts inside Snom? or Aastra, for that >matter, their hardware also seems good quality from what people have said. > >Another possibility is talking to Atcom (or other VoIP ODMs), they seem >to have done pretty well from the IP04 and derivatives. They've >experienced the benefits of an open development model, perhaps they'd be >interested. Not sure what the quality of their existing handset hardware >is like. > >Anyone on the list have the contacts to get the ball rolling? > >Paul > >___ >-- Bandwidth and Colocation Provided by http://www.api-digital.com -- > >asterisk-users mailing list >To UNSUBSCRIBE or update options visit: > http://lists.digium.com/mailman/listinfo/asterisk-users > > > >
Re: [asterisk-users] building a phone
On Sun, Mar 01, 2009 at 07:04:42PM +0100, Christian Stredicke wrote: > To be honest: I am not very optimistic regarding this project. > > The WRT is really a case where you essentially use stuff that is > already available and which is very very stable (e.g. Linux). There is > nothing really special for the WRT. > > For a phone, the picture looks different. There are so many components > necessary that are either not available or not very stable. Hence my focus on small quantities and hackability. -- Tzafrir Cohen icq#16849755 jabber:tzafrir.co...@xorcom.com +972-50-7952406 mailto:tzafrir.co...@xorcom.com http://www.xorcom.com iax:gu...@local.xorcom.com/tzafrir ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
On Sun, 1 Mar 2009 20:24:55 +0200, Tzafrir Cohen wrote: >On Sun, Mar 01, 2009 at 07:04:42PM +0100, Christian Stredicke wrote: >> To be honest: I am not very optimistic regarding this project. >> >> The WRT is really a case where you essentially use stuff that is >> already available and which is very very stable (e.g. Linux). There is >> nothing really special for the WRT. >> >> For a phone, the picture looks different. There are so many components >> necessary that are either not available or not very stable. > >Hence my focus on small quantities and hackability. OK, but there is the basis for something interesting here. If there's sufficient documentation to expose the hardware functionality then the device could be a suitable platform. That implies a certain level of cooperation from the manufacturer. It might be a lot of work, but that just means that someone with the right skillset and determination would need to be the project lead. Or it could end up lame or incomplete, like most of the soft phones that I've tried. Michael -- Michael Graves mgravesmstvp.com http://blog.mgraves.org o713-861-4005 c713-201-1262 sip:mgra...@mstvp.onsip.com skype mjgraves fwd 54245 ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [asterisk-users] building a phone
It may not be necessary to replace Snom's firmware to add interesting functionality to the product. Though that was not the original poster's premise, admittedly. As to the 'loose ends', they usually remain so until someone is motivated to drive them to closure. Absence of a suitable hardware platform is guaranteed to perpetuate that situation :) The big question for me is whether Snom would provide some documentation to those prepared to invest their time. With GPL'd software likely to be part of the mix, such information couldn't be covered by a non-disclosure or some restrictive developer agreement. One of the things that helps to kick-start a developer community is to sell 'developer kits' (like Digium did). Single-unit quantities with a 'not-for-resale' provision, perhaps with membership of some developer program. Paul Christian Stredicke wrote: > To be honest: I am not very optimistic regarding this project. > > The WRT is really a case where you essentially use stuff that is already > available and which is very very stable (e.g. Linux). There is nothing really > special for the WRT. > > For a phone, the picture looks different. There are so many components > necessary that are either not available or not very stable. There is a > tremendous risk of ending up with a project that has a lot of loose ends. > > But if someone wants to give it a try, sure. We have nothing to lose! Those > who know embedded Linux will easily feel like home on the phone once they are > logged in. > > Definitively an interesting topic for our Asterisk developer meeting that we > want to run this month in Berlin. > > Maybe for starters we just compile an Asterisk and run it on the phone. That > will be fun! > > CS > > -Ursprüngliche Nachricht- > Von: asterisk-users-boun...@lists.digium.com > [mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Michael Graves > Gesendet: Sonntag, 1. März 2009 18:30 > An: Asterisk Users Mailing List - Non-Commercial Discussion > Betreff: Re: [asterisk-users] building a phone > > On Sun, 1 Mar 2009 18:14:18 +0100, Christian Stredicke wrote: > > >> I have influential contacts inside snom... >> >> CS >> > > So you do! What do you think? Would snom be interested in selling > hardware into a group of users who would be loading community developed > application firmware? > > It makes me wonder how many little routers Cisco sells that actually > get loaded with WRT-DD and the like? > > Michael > > >> -Ursprüngliche Nachricht- >> Von: asterisk-users-boun...@lists.digium.com >> [mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Paul Chambers >> Gesendet: Sonntag, 1. März 2009 01:30 >> An: Asterisk Users Mailing List - Non-Commercial Discussion >> Betreff: Re: [asterisk-users] building a phone >> >> Michael Graves wrote: >> >>> On Sat, 28 Feb 2009 14:59:23 -0800, Paul Chambers wrote: >>> >>>> Michael Graves wrote: >>>> >>>>> Witness the fact that the old Pingtel phones ran Java, and they were >>>>> incredibly lame. >>>>> >>>>> I think part of what this thread misses is that DSP is a god chunk of >>>>> what SIP phones need. A general purpose CPU is not the right tool for >>>>> the task. A cheap DSP is better suited to compression, transcoding, etc. >>>>> >>>>> OTOH, presuming that the snom phones are Linux on a suitable platform >>>>> soomeone could develop a custom software load for them and OEM the >>>>> hardware. >>>>> >>>> I'm surprised that no-one has mentioned Astfin. Basically uClinux and >>>> asterisk running on an Analog Devices Blackfin DSP. There's also some >>>> 'open source' hardware that's available - the IP04 and friends. I'm >>>> using an Edgepbx FX08, and they also have a two-port version (FX02). >>>> Atcom has a single-port one, the IP01. >>>> >>>> Though if I were going to prototype an 'open' SIP phone, I'd probably >>>> start with a beagle board (TI OMAP3530 - dual-core ARM+DSP). It's a >>>> pretty powerful SOC - its brother (3430) powers the Palm Pre. >>>> >>>> Just another datapoint :) >>>> >>> Yeah, that'd be great hardware to select. >>> >>> What I was thinking is that this thread seems to be driven by those of >>> a software bent. For that group perhaps there&
Re: [asterisk-users] building a phone
I agree, the intersting part is adding what is not included in the standard firmware. Regarding documentation... On the one hand the phone is running a "regular" embedded Linux, I think that does not require additional documentation. The API to the phone is a different topic. It will really depend what content we are talking about. Many applications can be done using the mini-browser. The software does not even have to run on the phone for that. Maybe a concrete example of an application that has to run locally on the phone would be useful. CS -Ursprüngliche Nachricht- Von: asterisk-users-boun...@lists.digium.com [mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Paul Chambers Gesendet: Mittwoch, 4. März 2009 05:15 An: Asterisk Users Mailing List - Non-Commercial Discussion Betreff: Re: [asterisk-users] building a phone It may not be necessary to replace Snom's firmware to add interesting functionality to the product. Though that was not the original poster's premise, admittedly. As to the 'loose ends', they usually remain so until someone is motivated to drive them to closure. Absence of a suitable hardware platform is guaranteed to perpetuate that situation :) The big question for me is whether Snom would provide some documentation to those prepared to invest their time. With GPL'd software likely to be part of the mix, such information couldn't be covered by a non-disclosure or some restrictive developer agreement. One of the things that helps to kick-start a developer community is to sell 'developer kits' (like Digium did). Single-unit quantities with a 'not-for-resale' provision, perhaps with membership of some developer program. Paul Christian Stredicke wrote: > To be honest: I am not very optimistic regarding this project. > > The WRT is really a case where you essentially use stuff that is already > available and which is very very stable (e.g. Linux). There is nothing really > special for the WRT. > > For a phone, the picture looks different. There are so many components > necessary that are either not available or not very stable. There is a > tremendous risk of ending up with a project that has a lot of loose ends. > > But if someone wants to give it a try, sure. We have nothing to lose! Those > who know embedded Linux will easily feel like home on the phone once they are > logged in. > > Definitively an interesting topic for our Asterisk developer meeting that we > want to run this month in Berlin. > > Maybe for starters we just compile an Asterisk and run it on the phone. That > will be fun! > > CS > > -Ursprüngliche Nachricht- > Von: asterisk-users-boun...@lists.digium.com > [mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Michael Graves > Gesendet: Sonntag, 1. März 2009 18:30 > An: Asterisk Users Mailing List - Non-Commercial Discussion > Betreff: Re: [asterisk-users] building a phone > > On Sun, 1 Mar 2009 18:14:18 +0100, Christian Stredicke wrote: > > >> I have influential contacts inside snom... >> >> CS >> > > So you do! What do you think? Would snom be interested in selling > hardware into a group of users who would be loading community developed > application firmware? > > It makes me wonder how many little routers Cisco sells that actually > get loaded with WRT-DD and the like? > > Michael > > >> -Ursprüngliche Nachricht- >> Von: asterisk-users-boun...@lists.digium.com >> [mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Paul Chambers >> Gesendet: Sonntag, 1. März 2009 01:30 >> An: Asterisk Users Mailing List - Non-Commercial Discussion >> Betreff: Re: [asterisk-users] building a phone >> >> Michael Graves wrote: >> >>> On Sat, 28 Feb 2009 14:59:23 -0800, Paul Chambers wrote: >>> >>>> Michael Graves wrote: >>>> >>>>> Witness the fact that the old Pingtel phones ran Java, and they were >>>>> incredibly lame. >>>>> >>>>> I think part of what this thread misses is that DSP is a god chunk of >>>>> what SIP phones need. A general purpose CPU is not the right tool for >>>>> the task. A cheap DSP is better suited to compression, transcoding, etc. >>>>> >>>>> OTOH, presuming that the snom phones are Linux on a suitable platform >>>>> soomeone could develop a custom software load for them and OEM the >>>>> hardware. >>>>> >>>> I'm surprised that no-one has mentioned Astfin. Basically uClinux and >>>> asterisk running on an Analog Devices Bla
Re: [asterisk-users] building a phone
On Thu, 5 Mar 2009, Christian Stredicke wrote: I agree, the intersting part is adding what is not included in the standard firmware. Regarding documentation... On the one hand the phone is running a "regular" embedded Linux, I think that does not require additional documentation. The API to the phone is a different topic. It will really depend what content we are talking about. Many applications can be done using the mini-browser. The software does not even have to run on the phone for that. Maybe a concrete example of an application that has to run locally on the phone would be useful. How about an IAX client? It would eliminate all the stupid NAT firewalls with broken SIP ALGs instantly. Gordon CS -Ursprüngliche Nachricht- Von: asterisk-users-boun...@lists.digium.com [mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Paul Chambers Gesendet: Mittwoch, 4. März 2009 05:15 An: Asterisk Users Mailing List - Non-Commercial Discussion Betreff: Re: [asterisk-users] building a phone It may not be necessary to replace Snom's firmware to add interesting functionality to the product. Though that was not the original poster's premise, admittedly. As to the 'loose ends', they usually remain so until someone is motivated to drive them to closure. Absence of a suitable hardware platform is guaranteed to perpetuate that situation :) The big question for me is whether Snom would provide some documentation to those prepared to invest their time. With GPL'd software likely to be part of the mix, such information couldn't be covered by a non-disclosure or some restrictive developer agreement. One of the things that helps to kick-start a developer community is to sell 'developer kits' (like Digium did). Single-unit quantities with a 'not-for-resale' provision, perhaps with membership of some developer program. Paul Christian Stredicke wrote: To be honest: I am not very optimistic regarding this project. The WRT is really a case where you essentially use stuff that is already available and which is very very stable (e.g. Linux). There is nothing really special for the WRT. For a phone, the picture looks different. There are so many components necessary that are either not available or not very stable. There is a tremendous risk of ending up with a project that has a lot of loose ends. But if someone wants to give it a try, sure. We have nothing to lose! Those who know embedded Linux will easily feel like home on the phone once they are logged in. Definitively an interesting topic for our Asterisk developer meeting that we want to run this month in Berlin. Maybe for starters we just compile an Asterisk and run it on the phone. That will be fun! CS -Ursprüngliche Nachricht- Von: asterisk-users-boun...@lists.digium.com [mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Michael Graves Gesendet: Sonntag, 1. März 2009 18:30 An: Asterisk Users Mailing List - Non-Commercial Discussion Betreff: Re: [asterisk-users] building a phone On Sun, 1 Mar 2009 18:14:18 +0100, Christian Stredicke wrote: I have influential contacts inside snom... CS So you do! What do you think? Would snom be interested in selling hardware into a group of users who would be loading community developed application firmware? It makes me wonder how many little routers Cisco sells that actually get loaded with WRT-DD and the like? Michael -Ursprüngliche Nachricht- Von: asterisk-users-boun...@lists.digium.com [mailto:asterisk-users-boun...@lists.digium.com] Im Auftrag von Paul Chambers Gesendet: Sonntag, 1. März 2009 01:30 An: Asterisk Users Mailing List - Non-Commercial Discussion Betreff: Re: [asterisk-users] building a phone Michael Graves wrote: On Sat, 28 Feb 2009 14:59:23 -0800, Paul Chambers wrote: Michael Graves wrote: Witness the fact that the old Pingtel phones ran Java, and they were incredibly lame. I think part of what this thread misses is that DSP is a god chunk of what SIP phones need. A general purpose CPU is not the right tool for the task. A cheap DSP is better suited to compression, transcoding, etc. OTOH, presuming that the snom phones are Linux on a suitable platform soomeone could develop a custom software load for them and OEM the hardware. I'm surprised that no-one has mentioned Astfin. Basically uClinux and asterisk running on an Analog Devices Blackfin DSP. There's also some 'open source' hardware that's available - the IP04 and friends. I'm using an Edgepbx FX08, and they also have a two-port version (FX02). Atcom has a single-port one, the IP01. Though if I were going to prototype an 'open' SIP phone, I'd probably start with a beagle board (TI OMAP3530 - dual-core ARM+DSP). It's a pretty powerful SOC - its brother (3430) powers the Palm Pre. Just another datapoint :) Yeah, that'd be great hardware
Re: [asterisk-users] building a phone
Christian Stredicke wrote: > I agree, the interesting part is adding what is not included in the > standard firmware. > > Regarding documentation... On the one hand the phone is running a > "regular" embedded Linux, I think that does not require additional > documentation. The API to the phone is a different topic. It will > really depend what content we are talking about. May not be a need for much documentation on their embedded linux distro, but I'd be surprised if none is needed. While it's probably possible to figure out everything if you have the source for everything (including things like bootloaders) I wouldn't make the assumption Snom would make all the source available. Besides, some notes to get us started would save some time, even for experienced embedded linux developers. Though I do agree the bulk of documentation should be for the 'non-regular' parts. It'd be nice to have a developer section on their wiki, and be able to extend/update/correct developer documentation that they seed it with. That way the external developers can share what they learn, and each individual doesn't need to figure the same things out themselves. > Many applications can be done using the mini-browser. The software > does not even have to run on the phone for that. Maybe a concrete > example of an application that has to run locally on the phone > would be useful. OpenVPN, for example. I could see interesting things being enabled by having Avahi onboard too. It's not really true 'applications' which I think would be interesting to embed on the phone. I've long wanted to have bluetooth presence detection in desk phones, to be able to indicate the approximate physical location of an employee in the building(s). Kinda hard to do that with a mini-browser :) Mini-browsers certainly have their uses, don't get me wrong, but not everything can/should be done 'at arm's length'. A bit of local intelligence can go a long way, even if the bulk of the work is done centrally. Paul ___ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
