Japanese conversion - where to start?

[mattylondon]

Originally Posted by mattsimis

The Sat Nav takes over the screen when Nav is pressed, its like running two Operating Systems. One view is standard Audi (and Radio etc) and the other is custom (Audi JP OEM) Nav stuff. They do it this way (I suspect) as Kanji cannot be read easily on the low res standard Nav input. Their implementation is direct to screen.

Yes, additionally you have these installed (in blue). I haven't found what module supplies the 2nd LVDS input, I would guess it's the Japanese nav unit. Search continues!

Edit: yes, the 2nd LVDS is from your Nav Unit.

[mcs]

Bloody good info there.

I'm happy to photograph and document my MMI system for you if you need to compare and remove all the jap interface stuff.

Mike

[Delboy]

If as you say the different text requires a better display it may be possible that the screens better.

Could possibly rework the switch over device so an external trigger switched the display to a car pc type of unit mmmm need to do more digging.

But can't till get the Mac repaired

[mattylondon]

Originally Posted by Delboy

If as you say the different text requires a better display it may be possible that the screens better.

That's exactly what I thought Del. I did lookup all of the MMI screens and the part number only referred to PR codes with or without aluminium trim. Where a PR code of 8UV (High-level radio for China) or 8DZ (Radio "HIGH" Japan) was listed it only referred to a different adapter cable loom, probably the longer loom as it goes to that switching box.

[Architex_mA8tey]

Originally Posted by Delboy

I beleive his mmi is 2g Adrian, nice info though and be nice if admin could drop it into tech info

added to D3 tech dump

[mattylondon]

Maybe the MMI control unit can only support one byte character encoding. Japanese needs to be encoded in double byte or multi-byte, therefore the most efficient way to provide navigation is with a different nav unit (Japan Audi part numbers are 4E0910887B and 4E0910887H) that can support double byte and the video is transmitted directly to the MMI screen, rather then via the MOST loop. Just a guess.

[Delboy]

And you've jsit taken the conversation into a new level, sadly well above my intellectual level

[mattylondon]

One byte character encoding is enough for our Latin alphabet (and Cyrillic alphabet because I turned my MMI to Russian once) as it has 256 values. Kanji has 6879 characters, so one byte or 256 values isn't enough. Two byte character encoding gives 256 x 256, or 65536 values - enough for Kanji, Latin and probably every other alphabet!

It makes me think that the MMI system is only 8 bit. 8 bits = 1 byte. Our Nav units must also be 8 bit, and the Chinese and Japanese Nav units must be 16bit.

[mattylondon]

Mattsimis, these are my thoughts and inclination.

Your MMI Control Unit will need to be replaced, there are few options here depending if you have/want TV reception and the rear view camera.

Replacement of your radio.

Replacement of your Nav Unit and DVD of NZ maps. You mentioned that when you press the NAV button it switches over to Japanese Navigation. The NAV button is controlled by module E380 on a LIN connection off the Convenience CAN BUS. When the new MMI Control Unit is installed it will acknowledge this key press and go to normal navigation without errors.

TV unit may only need coding, I've not looked into fully.

New cable loom from MMI screen direct to MMI interface box, this will make the LVDS switching box redundant and may be removed (would like to see more photos of it)

I don't know if any of the antennas will have to be upgraded.

As for part numbers, this will depend on your current equipment level, and Mike MCS has kindly offered to scan his Australian '8 to help.

All the best

[IT]

Originally Posted by mattylondon

One byte character encoding is enough for our Latin alphabet (and Cyrillic alphabet because I turned my MMI to Russian once) as it has 256 values. Kanji has 6879 characters, so one byte or 256 values isn't enough. Two byte character encoding gives 256 x 256, or 65536 values - enough for Kanji, Latin and probably every other alphabet!

It makes me think that the MMI system is only 8 bit. 8 bits = 1 byte. Our Nav units must also be 8 bit, and the Chinese and Japanese Nav units must be 16bit.

Interesting.... Unicode type stuff was largely in the minority back in the early noughties when presumably the MMi was being developed, so it sounds like a plausible theory.

I guess they found it was easier as the time to write a custom overlay to handle the character sets instead of re-writing the MMI encoding to handle the extra bits, which from experience is not a trivial exercise.

Its all very exciting if I'm sadly truthful. It does get a bid mundane knowing the insides out on this stuff so when something totally unseen turns up its quite refreshing...

Unless you're the poor sod trying to get it to work that is....