FourG

And what navigation software would you put on it? There are few Navi systems that run on Linux (i.e. TomTom's PNAs), none that I know of are for sale to run on third party hardware. And without navigation, AVIC isn't exactly useful.

There aren't many as polished as iGo, Tom Tom or Garmin, but there are open-source nav programs that exist:

http://navit.sourceforge.net/ (I like this one, uses gpsd for location and speechd to announce directions)

http://sourceforge.net/projects/viking/

http://www.gpsdrive.de/

http://roadmap.sourceforge.net/

http://roadnav.sourceforge.net/

OK, the remainder of this will be a small rant, feel free to stop reading and go about your day/night. I'm about to get touchy-feely, consider yourself warned.

For teksimian, the whole point of considering Linux would be to use a set of apps of his choosing, contribute fixes and enhacements back to the apps he uses or if he so chooses write his own from scratch. In theory, he could do the same on a Win CE platform. Yes, if he chooses Linux he would likely also include writing his own drivers to duplicate the ones already done for Win CE for our device. Yes, Win CE is probably the platform that's targeted the most because of market share and a healthy support ecosystem. But something to consider is that the platform is not completely open and doesn't have the lowest capital cost of entry (I'd cite the non-free dev tools required as the biggest obstacle). This might have caused teksimian to seek an alternative. Other alternatives could have easily have included QNX, FreeBSD or any other OS that can run on an ARM architecture.

I respect that many of us have firmly held opinions regarding alternative OSes on our AVICs, but you have to recognize there are individuals in our AVIC community interested in running them. In this case, all teksimian wanted to know was what hardware the system uses. This request was for facts about the hardware and not soliciting our preferred choice of OS. Of course, this is my own opinion and guides how I interact with teksimian or anyone else on the forum. Just something I hope will be considered by others and hopefully make them pause before typing a flippant response, since the forum has recently driven away some great contributors like IDT and might scare off potential contributors like teksimian with adversarial responses. Remember, we have the Wiki available to use to start collecting FAQs with their answers, and can guide new people towards them and hopefully cut down on the frustrating repetition of n00b questions.

Just some food meant for thought. I hope I haven't offended anyone, but chances are I did anyway.

I could have sworn that I saw the icon of MSN Direct on my F700 shortly after I upgraded to 3.0, but can't find it anymore...

 

After searching, I've read I can't access MSN Direct on my F700 - is this true? Did I see a ghost? If not, how do I access MSN Direct. I could have sworn it was on the bottom of some screen, just can't remember where.

 

I have searched for this here, but can't find out *how* to access MSN Direct.

 

Cheers!

What you might have seen is a temporary glitch in the vBeta variable in iGo. During the boot up process, the Navi reads what board it is based off some resistor values on the board, if that value gets mis-read it might accidentally set itself to the value used by the F90/F900BT. If that happens and vBeta gets set wrong, the menu with the MSN button on it might get selected for display (the menus for all models are inside the data.zip, it just depends on a variable to select which one to show).

Technically, the F700BT should not have the MSN Tuner chip or the chip that holds the MSN firmware installed on its Navi board for cost savings, but you never know...

What hardware comprises the avic700bt?

 

I'm wondering if there's linux drivers for it, and whether linux would run on it.

 

An alternative OS would be so nice. Pioneer's OS sucks imo, and now they want money for upgrades? not cool.

OK, opinions aside (and yes, I agree with BorisM's point that if we could just cook a better WinCE image and applications we'd be in a quicker starting point), here are some of the facts about the hardware:

CPU is a SiRF (who bought Centrality) Titan CPU with an ARM1136J core and a VPU coprocessor (hence the "dual core" claim).

256 MB of DDR1 RAM (unknown device manufacturer/model) @ 133MHz.

Bootloader is stored on a Spansion S29AL004D70BF102 (256 MB).

NAND Flash chip is a SAMSUNG K9GAG08U0M-PIB0 (2Gb).

MSN firmware (F900/F90BT only) is an SST 34HF1641J.

AC97 Audio Codec (unknown vendor) provides audio out and touch panel 4-wire position input data.

Front Panel (buttons) interface via ADC chip.

CPU is connected via USP to:

- Parrot bluetooth handsfree chip (only provides handsfree and serial port profiles, no A2DP)

CPU is connected via UARTS to:

- MSN module (UART0)

- Serial connector (not populated, probably UART1 for debug messages on WinCE console)

- uCOM system FPGA (UART2, sounds like it might be a Xilinx from one of Carver's posts)

CPU is connected via I2C to:

- Video Decoder switch (I2C0)

- Dead reckoning ADC (I2C1) - Provides Gyro and G sensor data

CPU is connected via LVDS to:

- 5.8" TFT LCD touch panel (unknown vendor/model)

So far we haven't found Linux drivers for any of this hardware floating around in the Internet. What's especially important to consider is if there is bootloader code in the Navi board to make the CPU kickstart the uCOM over the UART, since that device controls everything on the AV board (e-Volume, audio and video source switching, IP Bus communication with XM/HD modules).

Lots of block diagrams and schematics can be found in the AVIC-F90BT/XS/UC service manual, and teardown data can be found in carver's posts (search the Hacks/Mods F-Series forum for his username).

Really? That solved it? I was told that those lines did not really apply to our units and were more for the standard igo units. I will add them in permanently if it fixes it!

It doesn't really apply if you have done the hardware hack (i.e. swap the Mute wire and ground the parking brake). But some of us haven't done it for one reason or another (me, I was lazy, knew of the software version and thought it would have the same effect as grounding PB, then decided to fix next time I had the dash open which was never because the replacement HD tuner died on the bench instead of making it into the car). Now I just add the statement to any button in the screens I use the most, and live without it in the lesser used screens.

According to GPSPassion and other forums, one could make disable it globally via sys.txt, but I guess it makes the interface much slower:

[interface]

;use_separated_safety_mode=1 ;(0) makes it slow

;use_separated_safety_in_mainview=1 ;(0) makes it slow

I have my mods on two different platforms. I have a user that says they cannot touch anything on the screen if the parking brake is not up at least a little on my newest set of mods. On my older mods from back in September this person can use everything just fine. The parking brake variable is set the same on both mods. Is there anything else that might affect this? I am searching through files not but coming up with much since this is really system level.

 

Thanks,

 

Alex

I think I've added " USE_SEPARATED_SAFETY_MODE=0" to pretty much every button in navigatemap_800_480.ui in your latest mod. Couldn't mute the Nav voice or cancel a route without the handbrake applied (I'm in a manual, have the softmod applied but not the hardwire PB to GND and move mute hack).

You would still need some Navi software (and even with broadband Google Maps isn't exactly a solution) and I am not aware of too many Navi packages running on Linux. TomTom's PNAs use it but it is for their Linux and their hardware...

How is Google Maps not exactly a solution? Check out the new Motorola Droid... it uses Google Maps Turn-By-Turn Navigation.

It also assumes you've got a data connection at all times. If you don't have a cached source of data/images, it's less than useful.

Would love to port Android to our AVIC but like BorisM pointed out we don't yet have enough information about how the Navi board controls the other boards in the system (protocol to talk to AV uCOM ASIC to change audio/video sources and control volume, commands to get data from the MSNDirect chip, control the Parrot handsfree chip, etc). Just because you can boot the alternate OS doesn't mean it's useful without a large amount of effort to re-create the drivers and support libraries for the custom peripherals that were written and tested on the old OS.

If the forum had a "like" button, I'd click it! Looks good.

Thanks bud, i found there only 2.000 so i have installed it. Can you give me 2.006 or testmode for 2.000??

Sorry Azux, but that URL has firmware that's newer than 2.000 under the "Firmware" section. Try a few clicks next time. One last hand-holding effort:

2.009 for F900BT/F700BT (http://www.pioneer.eu/files/support/avi ... E_6175.zip)

2.007 for F500BT (http://www.pioneer.eu/files/support/avi ... E_6050.zip)

3.010100 for F910BT/F710BT (http://www.pioneer.eu/files/support/avi ... E_6219.zip)

1.0001 for F310BT (http://www.pioneer.eu/files/support/avi ... _10001.zip)

Testmode 2.3 should work just fine with 2.009, IIRC, as long as you're using an SD card. Good luck if you were trying to use a USB key.

Here buds listen i have got update 2.006 but its says that is only work with models bought in U.S can some one give me 1 for eu?

Try Pioneer's Europe site: http://www.pioneer.eu/eur/content/suppo ... 700BT.html

It's got 2.009 for EU models there.

Azux, have you tried upgrading to 2.006 first? That's a free upgrade path, and Testmode 2.3 works with that version.

Ditto on the big icons. Small targets are just too tough to hit when you're driving (example: I've modded copy of your mod to make the next turn icon to be the nav mute instead, since it's big and closest to me and what I use the most in the map). Mind you, most of the quick menus shouldn't be used while driving period, but smaller targets make for less accuracy when stopped at the light.

Does the clock read an hour ahead in all screens, or is the time correct in the Map and wrong in all the audio control screens (FM/AM/HD/XM/iPod)? If the hour is correct in the Map and wrong in the audio screens even when the Timezone's set manually, you're not the only one with this problem. Myself and several others are experiencing it as well.

I wonder if something like this http://reviews.cnet.com/8301-31066_7-10390928-267.html?part=rss&tag=feed&subj=Crave could be ported to our unit?

Nope, our units don't have a full Bluetooth stack, just a neutered one custom made by Parrot that only provides hands-free and serial port profiles. If we can figure out how to hack the USB to accept one of the USB bluetooth dongles w/ a CSR radio and manage to get a configuration interface working, general devices like these might be an option...

Anyway, this still makes me wonder wether Leetlauncher can boot completely from a USB stick. This could make possible to insert the USB stick everytime I wanted to use leetlauncher (as most people in this forum have been doing with the SD card).

If you look through some of Leet's MortScript files and replace any reference to "My Flash Disk" with "USB Disk" you might be off to a good start. The only other issue might be if he's hard-coded references to "My Flash Disk" into some of his custom programs (like the main launcher, say for sourcing images).

I'm pretty sure you only get Atlasmgr if you've got an RDK for one of the SiRF chips, and that usually comes with a nice NDA to sign before you get your hands on one. There's lots of references to AtlasMgr on the QNX Foundry site that talks about setting up the Titan RDK with their board support package.

Thanks for your suggestion, I've already tried couple of times to hard reset my iPod but it didnt help. When AVIC interface language is changed, avic asks if its ok to reboot and if clicked "yes", it reboots without losing audio settings, so it should be somehow possible

If you look at Diaftia's mod, he has the ability to launch executables from iGo under Settings->Navi->Run (I have mine launching explorer.exe, for example). Testmode 2.3 includes a "restart" app that would let you soft-reset the unit, you could probably modify Diaftia's mod to launch "restart" from that screen (you'd have to copy "restart" to somewhere on "\My Flash Disk" so it's available if you don't have the Testmode SD card in).

Does anyone have the schematics of the machine…I remember seen it one on site but don’t remember where I wonder if adding another circuit will resolve it or copying the existing one re partitioning it unsoldering it and replacing it with higher capacity module will work and the same to the flash ram will do I can be very handy with the iron if I have to…

We have the schematics to the AV board and others via the Service Manual, but it doesn't have a schematic for the Navi board where the RAM and Flash ROM live. carver's done some work on hardware hacking on the NN746 Navi board (see this post and this thread (Upgrade NAND Flash from 2GB to 8GB?) to see some of his efforts).

One thing to note is even if you succeed in swapping out the flash module, the eboot/BIOS of the board is probably hard-coded to initialize the smaller part, and would also likely advertise a smaller size to the OS when it hands off the boot process to it. So you might need a firmware hack after the swap for things to work as expected.

It would be nice if the units bluetooth could be paired with one of these small Logitech Bluetooth Keyboards...

http://www.logitech.com/index.cfm/keybo ... 8&cl=us,en

Sadly the Bluetooth in our AVICs isn't a full-featured one, it's a phone and serial port only device made by Parrot. It looks like the Navi board communicates with the Bluetooth module over a serial or USB connection and uses Parrot's DLL/driver to run the board, rather than a full Bluetooth stack that can talk to HID devices.

Checked the registry, the entries for time zones (under "HKLM\Time Zones\") seem to indicate the DST stuff mandated by the Energy Policy Act has been applied to the WinCE 5.0 Platform Builder that Pioneer's using, so much for hoping it was stale TZ rules and auto DST screwing up the hour.

Guess it's time to see if this works (set via sys.txt):

 

[timezone]
sync_os_timezone=0
reset_os_timezone=0
multi_timezone=1

These settings seemed to work for the dev system to address the clock +1hr issue, though it will sort of break the ability to set the time from iGo 8 since it won't show the same time in the form as the system time it sets. Using the "Down" or "Up" buttons in the Time Settings screen, I was able to move the system time's hour down and up, all the while the iGo clock remained at the same value. But iGo did leave the system time alone after each boot, so once the time was set right it stayed put. Seems to me like there's an iGo script bug somewhere that needs squishing, going to go hunting later.

Hey, I've changed thats files in testmode, now testmode start automatically but leetlauncher no..! i don't know what is my porblem, for information, i've an italian (europe) f900bt, maybe some files is not correct for my device...However i have no error, now simply when i reboot my device, testmode starts..(windows ce)..

Sounds like you overwrote the files CopyTool.exe and CopyTool.mscr in the Testmode folder. You need the LeetLauncher copy of both files there in order for the correct script to be launched to load up LeetLauncher.

If you compare the *.mscr from both folders in a text editor (MortScript files, in case you're curious), you'll see they're like batch files that tell the script engine (CopyTool.exe, which is really MortScript.exe but renamed) to do certain tasks on our behalf. This is why you need to create a "hybrid" Testmode folder that has the contents of LeetLauncher 3.0 with only the missing key files from Testmode 2.3 added to it. If you are asked to overwrite any files when you're copying them over from Testmode 2.3, say no and I think you should get the desired result.

I don't know why it don't start automatically, however i've another question, what is the role of testmode folder ? When I reboot my unit does not start neither testmode nor leetlauncher, if instead, i replace the folder testmode with another testmode (the testmode 2.3 of this site), then windows ce starts..

th3-sh4rk, there are some files in the "testmode" folder with very long names made up of a mix of numbers and letters. Try copying the ones from testmode 2.3's "testmode" folder into the leetlauncher 3.0's "testmode" folder. These are a sort of key that the system looks for to decide if testmode is valid, and Leet's last release was before 2.006 and 3.0 came out and the required keys changed.

Now that still doesn't explain why, as soon as I see "Mobile phone has been connected" on my car's F90BT screen during the boot process that the clock will jump ahead an hour if by some miracle it started out correct. I need to re-pair my phone with my desk system to see if it exhibits the same behavior or not, it may just be coincidence that the phone pairing event is timed with the clock getting set ahead in the non-iGo apps.

Looks to be coincidence that I see the clock on the Suby's F90BT jump ahead when the phone pairing message shows up. The desk's F90BT will also have the time jump ahead if the GPS locks quickly, phone or no phone. Did notice that if I go into explorer and run the Date/Time adjustment control panel after everything's launched that the non-iGo apps will reflect the system time I set. Guess it's time to see if this works (set via sys.txt):

[timezone]
sync_os_timezone=0
reset_os_timezone=0

So why don't Washington and California have the same issues, or do they? A quick look at a map might have been in order for whomever wrote that part of the program.

Yeah, it's freakin' hilarious. WA users apparently don't have the problem, there was a user that saw his clock jump ahead an hour when he drove south across the bridge to Astoria from WA, and then fall back to the right hour driving back north across back into WA. Can't describe how annoyed I was to spend the $60 x2 (one for the car, one for my dev system) only to find out iGo 8.3 *still* has the problem. To be fair, my iPaq 310 that runs iGo 8.0 also has the same bug with the auto-correction, so the fault lies with NavNGo and not with Pioneer in this *rare* case. The autocorrection timezone bug was supposedly fixed in iGo 8 R3 (i.e. 8.3) so I'm not sure why our ezrider still sets things wrong. Maybe there's a data file it relies on that has the actual bug...?

Now that still doesn't explain why, as soon as I see "Mobile phone has been connected" on my car's F90BT screen during the boot process that the clock will jump ahead an hour if by some miracle it started out correct. I need to re-pair my phone with my desk system to see if it exhibits the same behavior or not, it may just be coincidence that the phone pairing event is timed with the clock getting set ahead in the non-iGo apps.

Okay, found the "fix" for the street names issue but the clock is still FUBAR.

 

Any help here?

So far both osubeaver and myself have the exact same problem, and we're both in Oregon. The "Auto sets me to Idaho" problem's been there since 1.0 days. I have noticed that if it takes the GPS a while to lock (like it does with the radio on my desk) that the time gets set correctly in the non-iG0 apps too. If the lock is quick (as it is with the radio in my car, which doesn't get constant power removed between uses) then the hour is ahead by 1 hour until I go into Settings->GPS & Time->Date & Time and turn auto on, then off or bump the hour up and then down by 1. Then when I go back to the non-iGo screens the time has been set correctly.

I'm interested to see what the behavior is after Nov. 1st when DST goes away. If things start getting set right again, it might be possible that the WinCE DST settings play a part.

Yeah, for the most part the code is maintained in a text editor if it's for the iGo 8.x mods (i.e. APL2\ezrider.exe). We still don't have a Board Support Package for the AVIC written that allows access to the DLLs/hardware/etc for apps other than ezrider.