DNS series head unit (new generation dual core RNS-510 alternative)

[wanxu]

Yep. There are some skins that make it not as terrible, though.

Did some more playing around. Found that iGo8.3.5 isn't a solution. It's ever so slightly more responsive when open, compared to Primo, but as soon as you start navigation/simulation, the same thing happens where everything just bogs down/slows down. Go back to the menu of iGo8.3.5, or stop the simulation, and it goes back to being responsive again.

So it's either that iGo8/Primo is being a resource hog during navigation/simulation, or the DNS just isn't optimised properly/fully, yet.

I don't understand why its so hard, when Primo/iGo8 works fine in many other units, many of which have 1/4 of the processing power that the DNS is supposed to have. Hell, the exact same copy of Primo works ****ing perfectly in my Eonon unit, even when simulating a 4000km trip, and that has a 500mhz processor, 128mb of ram, and essentially the same version of WinCE6.0.

Ugh.

It's easy: in this unit EVERYTHING is processed by the same processor, while in the other units there's a side to process the GPS and the other side process everything else.

Anyway, you can edit sys.txt in Igo to set it to use less memory. Check out these options, and don't use TTS voices:

[debug]
cache=2048
max_memory=143724405 ; 91MB
reserve_memory=396683
no_file_cache="1"
no_tmc=1 ;

[moda]

It's easy: in this unit EVERYTHING is processed by the same processor, while in the other units there's a side to process the GPS and the other side process everything else.

Anyway, you can edit sys.txt in Igo to set it to use less memory. Check out these options, and don't use TTS voices:

[debug]
cache=2048
max_memory=143724405 ; 91MB
reserve_memory=396683
no_file_cache="1"
no_tmc=1 ;

But it should be exactly the same as every other unit? and it looks to be the same hardware as the S600, which according to Bazzle, works fine. Very strange!

So I've given both 9.6.13.267029 and 9.8.0.377454 a go, and both work better than 9.6.5. They still lag and stall on track skip, when simulating a route. I'm beginning to think that its the Simulation thats the problem, may not be realistic. Might have to give it a go on a trip and see how it goes. Because they all work fine and are responsive when its just sitting there displaying a map, or even once you've calculated and started a route, but as soon as the simulation starts it bogs down. Not sure how it works once you actually start driving. Will see.

[moda]

So I've given both 9.6.13.267029 and 9.8.0.377454 a go, and both work better than 9.6.5. They still lag and stall on track skip, when simulating a route. I'm beginning to think that its the Simulation thats the problem, may not be realistic. Might have to give it a go on a trip and see how it goes. Because they all work fine and are responsive when its just sitting there displaying a map, or even once you've calculated and started a route, but as soon as the simulation starts it bogs down. Not sure how it works once you actually start driving. Will see.

Drove around a bit today with both 9.6.13.267029 and 9.8.0.377454, both perform about the same, but 9.8 is slightly faster in the menus etc, and is not compatible with any of the tweaks etc.

Still exhibit the same issues with SD playback. Maybe slightly better, i'm not sure, its really hard to tell. As soon as you stop/pull over, though, the lag goes away. Very strange!

To be honest, I really can't be bothered any more. There is little to no lag in FM radio or iPod playback, even during navigation, which is good enough for me.

I'm starting to think it may be an issue related to the CANBUS decoder (because the CANBUS stream seems to slow down to update every 2 seconds, then goes back to being normal speed when there is no lag etc) so i'm going to wait until we get the new chip before playing with it any more. Remember back when they first shipped, there were all sorts of horrible issues that were caused by the CANBUS decoder.

Otherwise, its still an optimisation issue, I guess.

[wanxu]

But it should be exactly the same as every other unit? and it looks to be the same hardware as the S600, which according to Bazzle, works fine. Very strange!

No, because I assume that S600 works like older units, with the CPU just working with the GPS, so GPS is independent of the other side. If Igo is getting 100% CPU power, the music playback will continue as if its getting 0%. DNS810 works with the same processor, so if Igo is using CPU, music will be affected.

Where is Bazzle talking about this new S600 unit?

[moda]

No, because I assume that S600 works like older units, with the CPU just working with the GPS, so GPS is independent of the other side. If Igo is getting 100% CPU power, the music playback will continue as if its getting 0%. DNS810 works with the same processor, so if Igo is using CPU, music will be affected.

Where is Bazzle talking about this new S600 unit?

How do we know that thats how it works, though?

Bazzle mentioned it in my thread on fixmygps, here: iGo Primo causing unit lag

Bit of back and forth with Audiosources, they reckon that i'm the only customer that has complained about this problem, and that they've passed it on to the engineers to rediagnose.

Surely not though. Or am I going crazy? Everyone else in this thread is seeing the same issues too, right?

[Goobz]

Yep. Mines worse than yours. It locks up all on its own without music playing.

[moda]

Yep. Mines worse than yours. It locks up all on its own without music playing.

Locks up when you're using the GPS, or just in general use?

Do you by chance have RDS turned on? FWIW, a few weeks ago I was playing around and turned RDS on, locked up straight away and had to do a fuse pull.

[asa572]

My nav is laggy no matter what!
I have had my unit freeze and have needed to pull the fuse while listening to the radio while navigating.
It's a waste of time trying to use it with USB or SD playing.

[moda]

My nav is laggy no matter what!
I have had my unit freeze and have needed to pull the fuse while listening to the radio while navigating.
It's a waste of time trying to use it with USB or SD playing.

Same question as above, do you have RDS enabled?

Mine has been rock solid stable in all other functions, this July firmware is particularly good. It's only laggy in the specific use case (SD playback + Navigating). Never had it lock up or had to do a fuse pull though. As I said, it's responsive. if i'm listening to iPod or FM Radio. Tiny bit of lag on changing volume, but otherwise fine.

SD Playback on its own actually works really well now, even with FLACs, very very small delay changing/loading tracks, basically as you would expect any other function to work etc.

Can't remember, you two have DNS610s though, yea? Which might explain it....

[asa572]

I haven't been able to find the RDS setting since the latest update.
I figured no use looking for it to hard because it has never worked properly any way.

[moda]

Fair enough. That's the only thing I could think of that's causing general unit instability, otherwise it might just be the DNS610 software is immature, which is strange because it should be the same :\ I think RDS is in the hidden menu now, maybe? (ie hold half way down on the left of the aux screen until it pops up).

I'm half hoping that this new decoder chip just magically solves everyones problems.... buuuut yea, probs not.

[Goobz]

I have the 810. I do have RDS on. Usually in using BT audio with Nav (so doubt its RDS causing the issue) and of I try to switch between screens to change something, bang... Locked. As you worked out though, from memory, its only when I have a route programmed. I've had it freeze using my gf's iPhone on the apple plug (streaming Pandora) with Nav running too.

[lovsund]

Hi all,


Been following this thread since a couple of months ago, and I have the DNS-610 installed in my VW Polo 6R since March/April.


The software has from my point of view gone from worst possible to actual a pretty decent unit. I new about the problems when I ordered the unit, and it is really nice to see frequent firmware updates from Audiosources. There are still some bugs to be solved (bluetooth connectivity and random freezing), but performance wise, using Igo Primo and other features at the same time can be a big mess. (Causing my girlfriend to more or less stab the unit with the car keys when it wont turn off.)


However, I´m done some research regarding this issue, and it seems that it has come down to the very old OS (windows ce 6.0) that can not easily been solved by just fiddeling with memory optimization and misc settings. Please correct me if I´m wrong, I´m not an expert in the field, but not totally novice when it comes to this sort of hardware/software.

Here are my findings (sorry for the tech gibberish):


The hardware, according to Audiosources is:
CPU: ARM Cortex A9, (ARMv7-A Cortex) dual core 1,2 ghz
System memory: 512 mb ram
System: WinCE6.0


The operating system, Windows CE 6.0 (anno 2006), is not able to utilize either the dual core functions, nor fully use the modern instructions features that the A9 CPU holds. Therefore, the system and Audiosources software is only capable using one crippled core. The newer OS Windows Embedded Compact 7 (anno 2011) is fully able to use multiple cores and ARMv7 new instruction settings (source: Windows CE - Wikipedia, the free encyclopedia)

Below whitepaper from Adeneo Embedded, as a Microsoft Windows Embedded Gold partner states: (http://community.embeddedswstore.com...on%20final.pdf)

"Windows Embedded Compact 7, the latest release of the Windows CE Operating system, brings significant improvements such as an ARMv7 compiler and NEON or Multicore support. These improvements provide dramatic increase of OS performance, especially on high-end ARM architecture such as Cortex-A-based processors, and are key advantages when developing advanced multimedia
embedded devices.
We measured, for example, average performance gain on a Cortex-A based processor between code built with the ARMv4i compiler (the legacy compiler of Windows Embedded CE 6.0) and the ARMv7 compiler (provided with Windows Embedded compact 7). We found an average performance gain of about 8%, with a 23% improvement in integer operations and a 19% improvement in floating point
operations."

and continues:

"Previous versions of Windows CE (read Windows CE 6.0) supported only ARMv4 or ARMv4i, so software compiled for more
advanced CPU cores (such as the Cortex-A did not support the new instructions and would perform slower than expected on high-end CPUs."

My conclusion is that there is a really decent dual core CPU in the DNS610/DNS810! But since the old software is not able to fully take advantage of its power, we will never see the full potential of the until Audiosources may update the OS to Windows Compact 7 or later. This may also be the cause why some of you have ANS/EONON/Dynavin etc units with older single core CPUs that is performing better that the DNS series.

[moda]

I have the 810. I do have RDS on. Usually in using BT audio with Nav (so doubt its RDS causing the issue) and of I try to switch between screens to change something, bang... Locked. As you worked out though, from memory, its only when I have a route programmed. I've had it freeze using my gf's iPhone on the apple plug (streaming Pandora) with Nav running too.

I'd still turn off RDS anyways, but that's just me.

Strange, because with the latest software, i've never had it crash/freeze. Maybe a factory refresh might help?

The software has from my point of view gone from worst possible to actual a pretty decent unit. I new about the problems when I ordered the unit, and it is really nice to see frequent firmware updates from Audiosources. There are still some bugs to be solved (bluetooth connectivity and random freezing), but performance wise, using Igo Primo and other features at the same time can be a big mess. (Causing my girlfriend to more or less stab the unit with the car keys when it wont turn off.)

However, I´m done some research regarding this issue, and it seems that it has come down to the very old OS (windows ce 6.0) that can not easily been solved by just fiddeling with memory optimization and misc settings. Please correct me if I´m wrong, I´m not an expert in the field, but not totally novice when it comes to this sort of hardware/software.

...

My conclusion is that there is a really decent dual core CPU in the DNS610/DNS810! But since the old software is not able to fully take advantage of its power, we will never see the full potential of the until Audiosources may update the OS to Windows Compact 7 or later. This may also be the cause why some of you have ANS/EONON/Dynavin etc units with older single core CPUs that is performing better that the DNS series.

Yep pretty much. I'm fairly sure we're not going to be able to get AS to update it to WinCE7.0 though.

Almost all the new chinese units out run some version of Android, which would be a better platform to run on anyways. Not sure AS are interested in doing that either

It's still strange since, though the S600 is running the same hardware, it works fine WRT performance, according to Bazzle anyways.

Not sure what the answer is at the moment, though :|

[woweriux]

Gotta agree with the others, since the July firmware my unit has been very stable with no crashes or lagging (other than RDS and SD/NAV). Bluetooth works great, connecting automatically and maintaining the connection throughout. Sound quality is also very good whether it's radio, cd, sd or bluetooth. Of course there are still minor issues to sort out (misspellings, missing features, etc.) but for day to day usability the 810 is very solid.

For the people that are having issues, I'm starting to suspect "user error" on firmware flashing in some instances.

Regarding the WinCE 7.0 theory, Sandy would it be possible for AS to try that as a solution? I assume it would be a similar firmware flash, correct?