Alturion Professional Advanced GPS 6 - The New D3 ? Is There A SDK ? - Page 6

ccsnet · 07-22-2005, 05:05 PM

Quote: Originally Posted by rickster

Yeah I saw the pdf file. Could be complex. Its possible.

Well NaviPC does it... need I say more ?

My opinion is to create an engine in the same way FD and MM use D3 as an engine... but not as complex - every body else will make the front end then ( or adapt ).

This way ( and infomatics is the same ) it can be higly customisable and taylored to what the programmer wants.

Terran

RoyN · 07-22-2005, 09:44 PM

yeah, a sort of an "sdk" would be really good... although a open source frontend in itself would probably be better in terms of ttm (time-to-market)... If we need devers to set out to make an SDK and THEN have devers go out and make a frontend with all the features might pose some time problems and would also require a lot of work from the SDK developers (because the guys who make the front itself will be the ones requesting more features)...

stevieg · 07-23-2005, 11:01 AM

Quote: Originally Posted by RoyN

yeah, a sort of an "sdk" would be really good... although a open source frontend in itself would probably be better in terms of ttm (time-to-market)... If we need devers to set out to make an SDK and THEN have devers go out and make a frontend with all the features might pose some time problems and would also require a lot of work from the SDK developers (because the guys who make the front itself will be the ones requesting more features)...

If it's Open source then there is no point making an SDK / engine etc as developers will be able to use it as they wish... If someone needs the functions of an SDK to perhaps communicate with RR/FP/CF for example they can submit the features into the codebase for the navigation software and bam they have a programmable / exposed interface.

Being able to make a FD/RR type front end seems a waste of time, as something like this would realy for portability need to be written in C/C++ for portability and speed and a VB 'front end' would only serve to slow it down.

So yeah, everyone get down the library and get some books on cross-platform programming and C

ccsnet · 07-23-2005, 12:07 PM

Quote: Originally Posted by stevieg

If it's Open source then there is no point making an SDK / engine etc as developers will be able to use it as they wish... If someone needs the functions of an SDK to perhaps communicate with RR/FP/CF for example they can submit the features into the codebase for the navigation software and bam they have a programmable / exposed interface.

Being able to make a FD/RR type front end seems a waste of time, as something like this would realy for portability need to be written in C/C++ for portability and speed and a VB 'front end' would only serve to slow it down.

So yeah, everyone get down the library and get some books on cross-platform programming and C

SDK was probably the wrong choice of word... more what Steve said ie make a common way to read the open source stuff.... ish... I know what I mean

.

Any way - What he says...

Terran

RoyN · 07-23-2005, 04:17 PM

Yeah, as I said in my post, a cross-platform frontend that could read the dvds directly and be designed for in-car use would be far better than a "toolkit" or "open library" me thinks...

and of course with the complete gps app released Open Source, a developer could adapt just the engine for another app if he so wished

ccsnet · 07-23-2005, 08:04 PM

BTW - had a thtough - it may have to be a SDK any way - if the apps in C ( like Steveig suggested ) and most people on here programe in VB......

Terran

RoyN · 07-23-2005, 09:44 PM

Yeah but if someone is developing the app in C, or the SDK in C, then he can dev a fullfledge carpc app in C no? (that could include linux and mac!!)

ccsnet · 07-24-2005, 09:04 AM

Quote: Originally Posted by RoyN

Yeah but if someone is developing the app in C, or the SDK in C, then he can dev a fullfledge carpc app in C no? (that could include linux and mac!!)

TBH - No idea... not a programmer - just looking at the majority view.... and as the majority of apps round here I thought that even if a C SDAL prog was made a vb SDK could still be available.

Terran

Maximus · 07-24-2005, 10:16 AM

Quote: Originally Posted by RoyN

Yeah but if someone is developing the app in C, or the SDK in C, then he can dev a fullfledge carpc app in C no? (that could include linux and mac!!)

Sure thing !
I mean, the GPS stuff shouldn't take more than a weekend or two to finish, then why shouldn't the developer be able to spend a day extra to write a fully functional frontend/car-pc app as well ?

I could only imagine the hours that go into a GPS application, take Alturion for example, V6 isn't their first and they probably have more than one developer... yet they have bugs in their interface, flaws in the routing and a non-working SDK.... and that even though the interface and SDK should be the "easy" part.

My point is this: Even if you find 10 people who are willing to do this for free (beeing an open source project), in their spare time, it'll be a loooooooong time before you have anything remotly like Destinator.

Yeah, yeah, I know... if no one ever tried just because it wasn't "possible" then we wouldn't have any software to begin with.
It's just that many seem to forget the ammount of work required since it's "just a few lines of code".

/M

ccsnet · 07-24-2005, 01:00 PM

Quote: Originally Posted by Maximus

Sure thing !
I mean, the GPS stuff shouldn't take more than a weekend or two to finish, then why shouldn't the developer be able to spend a day extra to write a fully functional frontend/car-pc app as well ?

I could only imagine the hours that go into a GPS application, take Alturion for example, V6 isn't their first and they probably have more than one developer... yet they have bugs in their interface, flaws in the routing and a non-working SDK.... and that even though the interface and SDK should be the "easy" part.

My point is this: Even if you find 10 people who are willing to do this for free (beeing an open source project), in their spare time, it'll be a loooooooong time before you have anything remotly like Destinator.

Yeah, yeah, I know... if no one ever tried just because it wasn't "possible" then we wouldn't have any software to begin with.
It's just that many seem to forget the ammount of work required since it's "just a few lines of code".

/M

I certainly dont think its a few lines and I also belive that we should decide some kind of standard purley because its more than likley going to be a community project and most programe in VB... in fact I dont know any C programmes on here.

So... I propose 2 things for the mo...

1) Open another thred specifically about this ( cause lets face it we are heading off topic big style, 2) Place links to the docuemntation and open it up to the community at large - see what they've got to say... cos I bet most havent really seen this thred and I have seen in the past many a person looking at some thing simular - at least trying any way.

Terran

ShawJohn · 07-24-2005, 01:59 PM

I recon it would take a couple of years to get this up and running properly. It can take 3months just for ppl to do a skin, and David (not meaning to put him down) took nearly six months to 're-write' CF from 1.2b to 1.3/2.0b, and that time didn't include learning to program

I'd love to get into some heavy programing having previously done some VBA writing Company databases but simply haven't got the time to learn the skills needed for a full standlone App or even a Linux OS/Car PC App combined into a lightweight super fast system like tomtom go with a media player that would run on a pc.

Why not center in on the TomTom SDK if you want better mapping and results this decade

ccsnet · 07-24-2005, 03:44 PM

Right - as we are so off topic now... I have opened this one http://www.mp3car.com/vbulletin/show...219#post569219

Terran

s003apr · 07-24-2005, 11:31 PM

I have been working on a navigation database. I am implementing it by storing publically available information like tiger line road data into an SQL database. My results so far show that this is very promising. Using MySQL, my query times are very fast. Additionally, by sticking to a mainstream SQL database, APIs already exist whic allow you to access the data in a variety of programming languages.

Navteq data is great and I would gladly pay $30-$40 for a database and work on software for it because that is what the popular routing apps cost (Delorme, MS Streets & Trips), but $200 for the same data in an accessible form is too rich for my blood.

Anyone interested in checking out my project can go here:

http://sourceforge.net/projects/navdb/

ShawJohn · 07-26-2005, 01:08 PM

Well I've been playing about with alturion 6 DVD edition for PC & PPC and I can't say I'm overly impressed. It is fiddly, no good at 800x480, tricky to program - even on a 17" CRT, the maps look far worse than D3 2004 although they may be more accurate.

Only thing I was struck by is the POI database. It has far more POI's than D3.

Needless to say, back to FD

Until someone tames TomTom for PC use

Maximus · 07-29-2005, 03:54 AM

I sent Alturion an e-mail with some comments and three questions a while back...

Quote:

Anyway.... on to my questions:

1. Will the Skinning be supported on the desktop version ?

2. What are the chances of a SDK with more features, for example, a SDK
without the need for the whole Alturion GPS application to run.
and the possibility to list cities/addresses directly rather than having
the user enter an address and then check if it's valid.

3. Will we see a true "Mobile PC" version ?
(Wouldn't be neccesary if "we" would get a better SDK/skinning.

Got the reply the other day...

Quote:

Dear Sir,

Thank you for your interest, we will pass your feedback to devellopment,

Kind Regards,

Alturion Support

User Name		Remember Me?
Password