Active Topics
-
Firefox with Leste (10)
to Maemo 7 / Leste by teroyk - 2 days, 19 hrs ago - more...
 |
|
2013-08-28
, 05:36
|
|
Posts: 1,974 |
Thanked: 1,834 times |
Joined on Mar 2013
@ india
|
#142
|
Originally Posted by wicket
yes yes this makes me shiver
First of all thanks to joerg_rw and the Openmoko guys for making this a real possibility.
The Neo900 will be competing with the €399 Jolla device for peoples cash. I only have one SIM card and only carry one device on me so I will buy one or the other. Right now it looks like Jolla has the advantage with regards to price and hardware. This means that the Neo900 will need to set itself apart with the software.
When I have to decide which device to buy I am going to be looking to see which OS will be able to provide me with freedom and functionality closest to what I get out of a FOSS desktop or server OS.
Personally, I'm not too fussed about Fremantle compatibility. I still find Fremantle lacking in certain areas. I'd much rather have Fremantle rebased on top of full Debian.
As mentioned by others, the issue of RAM could also be a deciding factor. I understand that it's a matter of taking whatever you can get but I would urge you to do your best to get 1GB.
On a side-note, if a VHF/UHF transmitter is being considered as mentioned earlier in this thread, maybe we could forget about HDMI-out and transmit a DVB-T2 signal instead?
9 months na i think i am in buttt no infrared no nokiabot....
|
|
2013-08-28
, 06:17
|
|
Posts: 2,102 |
Thanked: 1,937 times |
Joined on Sep 2008
@ Berlin, Germany
|
#143
|
Originally Posted by wicket
If money is the most important factor, well, this project might not be exactly, what you are looking for.
The Neo900 will be competing with the €399 Jolla device for peoples cash.
SNIP
When I have to decide which device to buy I am going to be looking to see which OS will be able to provide me with freedom and functionality closest to what I get out of a FOSS desktop or server OS.
But you also state you would rather have an OSS-computer in your hands with phone capabilities than a pure phone, that is the N900 and, may be, if we are lucky, the Neo900 as well.
My expectation regarding openness of Jolla is low. They were Nokia people and their track record to bring a true OSS-phone to the market is obvious in the struggle to update Maemo to more modern Debian versions.
 |
|
2013-08-28
, 06:59
|
|
Posts: 1,789 |
Thanked: 1,699 times |
Joined on Mar 2010
|
#144
|
Originally Posted by GrimyHR
Not possible, simply because there's closed binaries, missing code, and usually undocumented. On top of that, the drivers aren't quite optimized for battery life, performance and stability.
what about building a phone around those cheap chinese soc-s that are around 10$ and have 1gb ddr3, around 1.5ghz proc and 4 core mali 400 gpu/dsp...?
But I just figured out a MUCH MUCH better Neo900.
Get a company-aided and community backed device. Basically a qwerty-based Chinese smartphone with the Allwinner A10 SoC.
That's got all the documentation available; porting possible.
And its got chops powerful enough to match the GTA04.
Better yet, its high volume and low cost... and available now.
__________________
I'm flattered
Originally Posted by mscion
The Following 5 Users Say Thank You to mscion For This Useful Post
I vote that Kangal replace Elop!
I'm flattered
| The Following User Says Thank You to Kangal For This Useful Post: | ||
|
|
2013-08-28
, 07:17
|
|
Posts: 1,203 |
Thanked: 3,027 times |
Joined on Dec 2010
|
#145
|
as much as i would like truly open hardware, i don't mind certain binary blobs (gpu etc) if it means a better end product. if that means nda's for a select few core devs or something, so be it.
to make the two devices compatible with existing freemantle, i foresee the need to push updates to the n900 to cater for hardware abstraction. this update will add another layer using tiny bit more memory and a small speed penalty.
i would probably take aapo's route and rebase off current or previous debian release. this would highlight issues with hardware abstraction and closed source packages, which could be replaced by open alternatives taking advantages of features from updated base packages. in the end we could end up with a sort of maemo5.5 image which with a little work could then be pushed to n900 as well.
to make the two devices compatible with existing freemantle, i foresee the need to push updates to the n900 to cater for hardware abstraction. this update will add another layer using tiny bit more memory and a small speed penalty.
i would probably take aapo's route and rebase off current or previous debian release. this would highlight issues with hardware abstraction and closed source packages, which could be replaced by open alternatives taking advantages of features from updated base packages. in the end we could end up with a sort of maemo5.5 image which with a little work could then be pushed to n900 as well.
| The Following 2 Users Say Thank You to Android_808 For This Useful Post: | ||
 |
|
2013-08-28
, 08:32
|
|
Posts: 439 |
Thanked: 282 times |
Joined on Oct 2012
|
#146
|
i have a allview tablet , dual core arm a9 1,5 Ghz , 1 gb ddr3 , wi-fi b/g/n and mali400mp2 the price was 100 euro , this hardware on maemo would simply rock , just saying
__________________
N900 Beast featuring :
8GB OPTFS@ext4(*performance mount options)
overclock: cpu[125 - 1,150(*VDD1)+Dsp overclock
CSSU-T latest version
N900 Beast featuring :8GB OPTFS@ext4(*performance mount options)
overclock: cpu[125 - 1,150(*VDD1)+Dsp overclock
CSSU-T latest version
| The Following User Says Thank You to Alecsandru For This Useful Post: | ||
|
|
2013-08-28
, 09:09
|
|
Posts: 18 |
Thanked: 9 times |
Joined on May 2013
|
#148
|
Originally Posted by biketool
Among many other very interesting and helpful suggestions being posted, this one really appeals to me and I would love to see this in any device. Also, considering the current state of affairs in the world, people would clamor to pay for a device such as this, be it Neo900 or not. The above mentioned features (if added to the Neo900) could *easily* get it incredible media attention as well as huge support for large scale backing, allowing for far greater hardware/software possibilities.
This is very cool news, I had considered ordering a GTA04 for my old openmoko but the N900 form factor is so much better.
This effective phone community design has all of the features found in a N900 and adds the add-ons for the N9 including NFC, real working no-hack USB OTG, three axis magnetometer, air pressure altimeter, FM tx/rx, etc.
I suggest that a crypto signed ID be added to the board as the Chinese already make cheap N900 refurbs from repaired boards which sell as new but with no QC and a high failure rate. It would be nice to know that the QC is community/EU standard rather than roll the dice Chinese pirate(even if it is an open to copy design).
In this age of Snowden and Arab Spring we must consider what hardware including crypto coprocessors, mesh networking, and any other long range off network comm methods can be added at this critical stage in the project where the design can still be changed.
I think there is still time now to add a UHF/VHF transmitter preferably a software defined radio since there is more room on a N900 board. This feature is already implemented on the AndroidOS Earl backcountry tablet. http://www.meetearl.com/ This is key to communicating and networking in the absence of cellular telephone service or when used in a protest movement with a hostile telephony infrastructure. The UV-3R group is working on an open firmware so there are devs there who can help with the micro size SDR components and controls required. It would also bring in substantial interest and RF expertise from the amateur radio community as there is currently not yet a smart radio/phone (and only one junk VHF/UHF/GSM phone) even if they just want to flash it to Android.
http://www.liorelazary.com/index.php...uv5r&Itemid=17
http://groups.yahoo.com/group/UV-3R/message/8158
The SDR even if only tuned to FRS, PMR446, MURS is useful for walkie-talkie chat and sending environmental telemetry and GPS position between devices. If tuned to amateur radio bands the possibilities for packet radio become vast even including free access to the amateur radio satellite galaxy for those with a license.
Additionally if it can transmit on 122.5mhz AM(FM will work with mild distortion) it can communicate on the aircraft emergency band which is listened for even in the most remote areas by commercial aircraft with hundreds of km line-of-sight even if only as a beacon, when combined with a digital signal including GPS coordinates on 406mhz a global network of search and rescue satellites with parallel receivers and finely tuned DSPs are listening for distress signals and can have a rescue activated in around ten minutes, if properly formatted the SARSAT system will accept and activate rescue for unregistered EPIRB signals.
https://en.wikipedia.org/wiki/EPIRB
A software defined radio if given a properly formatted signal can also probably clone a SPOT communicator enabling satellite SMS type messaging with a cheap service plan http://talk.maemo.org/showthread.php...t+communicator
Infact, what I would love to see is phones with 'docking' ports like workstation class laptops do (lenovo w520, dell M precision series etc).
Is it possible for the jolla phone to have the 'other half' as a dockable solution with the above mentioned tech?
|
|
2013-08-28
, 09:27
|
|
Posts: 6 |
Thanked: 14 times |
Joined on Aug 2013
|
#149
|
Looking through the posts I can see that there are two goals to achieve:
1. To keep maemo/fre(e)mantle alive
2. To develop (promote) an open hardware platform
In he long run it is impossible to achieve the first goal without achieving the second - so the second goal should be the primary one.
If this open device is to succeed, there is a need to deliver it in grate quantities.
Operating N900 are currently in order of 10k of units now (I've heard about 30k in January). We'd need almost all current users to commit to this idea to make the price and the hardware appealing to all - according to dos1:
But in order to achieve this volume I think that there is a need to allay with all possible projects that might be interested in developing open HW phone and maybe not only a phone.
First I thought about open pandora project - since it has quite similar specs to GTA04 (they also might want to refresh their hw specs).
Additionally I think that companies like jolla and canonical (ubuntu) might be interested in supporting a community/open phone. None of them has qwerty phone in their portfolio and supporting the community sends good PR message. Both are SW companies, so having some open HW at hands would be good to them.
In terms of specs I'd like to see something similar to http://arstechnica.com/gadgets/2013/...n-the-droid-5/ .
Maybe some parts of Edge design can be incorporated (canonical claims it has negotiated better prices on them).
It might be good to seek some support from some open/free software/hardware foundations/associations.
1. To keep maemo/fre(e)mantle alive
2. To develop (promote) an open hardware platform
In he long run it is impossible to achieve the first goal without achieving the second - so the second goal should be the primary one.
If this open device is to succeed, there is a need to deliver it in grate quantities.
Operating N900 are currently in order of 10k of units now (I've heard about 30k in January). We'd need almost all current users to commit to this idea to make the price and the hardware appealing to all - according to dos1:
> 1000 buyers: 490 EUR (here it would include a better CPU, e.g. OMAP4)
> 10k buyers: 390 EUR (here it would include a new plastic case)
But in order to achieve this volume I think that there is a need to allay with all possible projects that might be interested in developing open HW phone and maybe not only a phone.
First I thought about open pandora project - since it has quite similar specs to GTA04 (they also might want to refresh their hw specs).
Additionally I think that companies like jolla and canonical (ubuntu) might be interested in supporting a community/open phone. None of them has qwerty phone in their portfolio and supporting the community sends good PR message. Both are SW companies, so having some open HW at hands would be good to them.
In terms of specs I'd like to see something similar to http://arstechnica.com/gadgets/2013/...n-the-droid-5/ .
Maybe some parts of Edge design can be incorporated (canonical claims it has negotiated better prices on them).
It might be good to seek some support from some open/free software/hardware foundations/associations.
|
|
2013-08-28
, 09:58
|
|
Posts: 1,548 |
Thanked: 7,510 times |
Joined on Apr 2010
@ Czech Republic
|
#150
|
Sounds interesting, count me in !
Just from top of my head, a few things I think the Neo900 hardware will make possible:
Just from top of my head, a few things I think the Neo900 hardware will make possible:
- Augmented reality
- The N900 is missing the magnetic compass, so making AR apps is not possible (without some crazy tricks with camera tracking as in Nokia Panorama)
- the Neo900 should have a magnetic compass and even accelerometers, so AR apps should be possible (well, once written
- Improved navigation support
- navigation apps can use the magnetic compass to show where north is, automatically reorient the map, show in which direction and how distant a POI is in regards to current device orientation
- the pressure sensor might enable better elevation estimation (GPS sucks at reporting elevation)
- all the new sensors combined (magnetic compass, accelerometers, pressure sensor, gyroscopes & magnetometer) might even enable inertial navigation, at least for shorter distances, such as tunnels or short indoors stretches (probably wouldn't be easy, but very, very handy if working)
- overall better navigation performance due to more computing power to render the navigation interface, do elaborate map rendering, cache more stuff in RAM, maybe even better voice rendering (but that's probably not CPU bound)
- Improved mapping
- N900 is ideal for OpenStreetMap mapping due to its accurate resistive screen
- and due to Neo900 being Fremantle compatible, the super useful OSM2GO OpenStreetMap editor should work without modifications, but faster
- Improved drawing
- N900 with its unique resistive screen is already often used for digital drawing
- Neo900 will have the same screen, so drawing on it should be faster due to better CPU and more RAM
- also due to proper USB OTG, artists might easily connect a USB keyboard and use keyboard shortcuts
- Improved game emulation support
- the N900 is already a superb emulator for multiple consoles and computer systems, mainly due to the similarity to OpenPandora and the GameGripper
- the improved hardware in the Neo900 should speed things up and make emulation of more demanding platforms, such as the PSP, possible
- V-synced output due to newer GPU & drivers
- Wide range of USB peripherals usable due to proper USB OTG & more CPU power
- 3D printers
- software defined radios
- TV tuners
__________________
modRana: a flexible GPS navigation system
Mieru: a flexible manga and comic book reader
Universal Components - a solution for native looking yet component set independent QML appliactions (QtQuick Controls 2 & Silica supported as backends)
modRana: a flexible GPS navigation system
Mieru: a flexible manga and comic book reader
Universal Components - a solution for native looking yet component set independent QML appliactions (QtQuick Controls 2 & Silica supported as backends)
The Neo900 will be competing with the €399 Jolla device for peoples cash. I only have one SIM card and only carry one device on me so I will buy one or the other. Right now it looks like Jolla has the advantage with regards to price and hardware. This means that the Neo900 will need to set itself apart with the software.
When I have to decide which device to buy I am going to be looking to see which OS will be able to provide me with freedom and functionality closest to what I get out of a FOSS desktop or server OS.
Personally, I'm not too fussed about Fremantle compatibility. I still find Fremantle lacking in certain areas. I'd much rather have Fremantle rebased on top of full Debian.
As mentioned by others, the issue of RAM could also be a deciding factor. I understand that it's a matter of taking whatever you can get but I would urge you to do your best to get 1GB.
On a side-note, if a VHF/UHF transmitter is being considered as mentioned earlier in this thread, maybe we could forget about HDMI-out and transmit a DVB-T2 signal instead?