[ingenieur]

Originally Posted by Commanderwiggin

I have a 09/2012 X5D.

The X5D and E90 335d for the US use the same SGBD file - so yes you are set. I do not think the ELM327 will work.

[pheno]

Originally Posted by ingenieur

The X5D and E90 335d for the US use the same SGBD file - so yes you are set. I do not think the ELM327 will work.

Hey, thanks for the confirmation.

Generally speaking are there many BMW owners that have ELM327 ? If there's enough demand for that I believe it should be possible to make it work with TestO also. Benefit ? :-)

[DWR]

Originally Posted by pheno

Hey, thanks for the confirmation.

Generally speaking are there many BMW owners that have ELM327 ? If there's enough demand for that I believe it should be possible to make it work with TestO also. Benefit ? :-)

Pheno, many people here use other tools that use the ELM327, such as Torque. It would be fanatastic to be able to run Testo from the ELM327. The cable is one of the reasons Testo is a little less convenient than other logging applications (but Testo is still much more powerful).

[pheno]

Originally Posted by DWR

Pheno, many people here use other tools that use the ELM327, such as Torque. It would be fanatastic to be able to run Testo from the ELM327. The cable is one of the reasons Testo is a little less convenient than other logging applications (but Testo is still much more powerful).

ELM327 Wifi looks cool but it must add some extra latency. There's USB version also.Elm327 capabilities are quite impressive. I'll be ordering one.

[bvanlieu]

Originally Posted by pheno

ELM327 Wifi looks cool but it must add some extra latency. There's USB version also.Elm327 capabilities are quite impressive. I'll be ordering one.

I personally have found the ELM to be a very cheap slow product. OK for pulling things from a stop but cannot be run while operating my 2011 335d (bad things happen, like car turning off).

After mucking with the ELM, I invested in OBDLink, the open source module that has better components and samples faster: no issues using that Wifi with various apps and worth the extra $ IMHO.

- b

[pheno]

Originally Posted by bvanlieu

I personally have found the ELM to be a very cheap slow product. OK for pulling things from a stop but cannot be run while operating my 2011 335d (bad things happen, like car turning off).

After mucking with the ELM, I invested in OBDLink, the open source module that has better components and samples faster: no issues using that Wifi with various apps and worth the extra $ IMHO.

- b

Thanks, looks interesting.. it has to be ELM327 compatible I guess ?

[DWR]

Originally Posted by bvanlieu

I personally have found the ELM to be a very cheap slow product. OK for pulling things from a stop but cannot be run while operating my 2011 335d (bad things happen, like car turning off).

After mucking with the ELM, I invested in OBDLink, the open source module that has better components and samples faster: no issues using that Wifi with various apps and worth the extra $ IMHO.

- b

There are varying degrees of quality. Mine works fine, although I suspect the cable is faster, just less convenient.

[pheno]

Generally speaking would it be nice to use Tool32/ISTA/INPA/Ediabas/ with ELM327 USB/WiFi ?
Hmm.. did I just ask a rhetorical question ?

[TDIwyse]

Originally Posted by pheno

Generally speaking would it be nice to use Tool32/ISTA/INPA/Ediabas/ with ELM327 USB/WiFi ?
Hmm.. did I just ask a rhetorical question ?

Mmmm... how about with bluetooth and an Android app for TestO? Like Torque/BMWhat ... only better :-)

[pheno]

Originally Posted by TDIwyse

Mmmm... how about with bluetooth and an Android app for TestO? Like Torque/BMWhat ... only better :-)

Would be nice yes Actually I had plans to create an Android app over a year ago with USB/OBD2 interface but it never took off.

I am adding possibility to run custom jobs in "serial" mode to TestO which should make logging faster than now with Ediabas. And then DWR can start working on Road Dyno :-) maybe after that is time for some Android.

[TDIwyse]

Originally Posted by pheno

...tests show that with Ediabas avg speed for 4 params is 100ms per job and with 'serial' it is 70ms per job...

Some feedback from my Windows 8 notepad TestO setup from today. It doesn't appear the serial job is going to be any faster with my particular setup and configuration.

Testing the 4 parameter serial job showed an average timestep of ~96mS, whereas the "normal" mode with 7 parameters showed an average timestep of ~100mS.

[DWR]

I think Pheno is seeing faster. I am seeing about the same time.
Also, I seem to get variation depending upon which parameters I choose to log - don't have a clue why that would be. It would be good if you could test the same parameters as are in the serial job. I'd be interested to see those results.

What is really concerning me is the stability of the sampling rate is worse with serial in your data. Also, are the Testo.ini files the same for both? Look especially at the 'pause' parameters.

[pheno]

Originally Posted by TDIwyse

Some feedback from my Windows 8 notepad TestO setup from today. It doesn't appear the serial job is going to be any faster with my particular setup and configuration.

Testing the 4 parameter serial job showed an average timestep of ~96mS, whereas the "normal" mode with 7 parameters showed an average timestep of ~100mS.

Thanks for testing!

[serial]
port=COM1
pause_before=0
pause_read=0
pause_after=20
syncwrite=1

pause_after is 20 by default but your ecu does not need that. It means waiting time (in ms) after data has been read from ecu. So now it adds 20ms after each job. It's there because my M60B40 with DME 3.3.8 needs that or it will hang up pretty soon.

So please try to set that setting to 0 or 1 and you should see faster times.

[TDIwyse]

Thanks for the feedback. Will try to work some of those suggestions later this week or weekend.

[DWR]

Originally Posted by TDIwyse

Thanks for the feedback. Will try to work some of those suggestions later this week or weekend.

Same here.
Really appreciate your contributions.

[TDIwyse]

Looks like the "pause" value change helped sampling rate for the serial job. I changed it from 20 to 1.

This is with the car not running just to make things simpler... I also used the same parameters for these tests per DWR's request.

Sampling speed improved from an average of ~107 ms to ~80 ms.

Question 1: What is STAT_LUFTMASSE_PRO_HUB_WERT measuring? Is it air mass proportional to "something" instead of just total air mass?

Question 2: Why would STAT_EINSPRITZMENGE_AKTUELL_WERT give such dramatically different values between serial vs normal measurements? This was back to back measurements with the car not running in both conditions ...

[DWR]

Originally Posted by TDIwyse

Looks like the "pause" value change helped sampling rate for the serial job. I changed it from 20 to 1.

This is with the car not running just to make things simpler... I also used the same parameters for these tests per DWR's request.

Sampling speed improved from an average of ~107 ms to ~80 ms.

Question 1: What is STAT_LUFTMASSE_PRO_HUB_WERT measuring? Is it air mass proportional to "something" instead of just total air mass?

Question 2: Why would STAT_EINSPRITZMENGE_AKTUELL_WERT give such dramatically different values between serial vs normal measurements? This was back to back measurements with the car not running in both conditions ...

OK, so apples to apples we are seeing a reduction in sampling time. Good.

1) Air mass per cycle, rpm normalized. I used that to see how volumetric efficiency changes with rpm. Unfortunately, it maxes out, so it turned out to be not that useful. Now, I just do the math, using total air flow.

2) Funny, I never looked at that. Was just concentrating on the sampling rate. Haven't got a clue why there are not the same. Pheno, any ideas?

Thanks for the feedback!

[pheno]

Thanks for testing and feedback.

2) Will be fixed in next version. There's an error in the code during calculations.

LUFMASS_PRO_HUB means air mass per cylinder.

[Makeclick]

Hello friends

I just joined here. This sounds so nice and i started to search maps from the flash file (A7569997) and i did find something. 105 maps or possible maps. Maybe i am not even close, but maybe i can learn and help something.

[DWR]

[QUOTE=Makeclick;18010063]Hello friends

I just joined here. This sounds so nice and i started to search maps from the flash file (A7569997) and i did find something. 105 maps or possible maps. Maybe i am not even close, but maybe i can learn and help something.

I believe you just posted in the wrong thread - welcome aboard!

Wasn't that meant for the "Transmission remap - Let's do it ourselves" thread?

WinOLS is not a great map finder. Hundreds of maps in the trans cal file. But, you have a real one there. Good work.

[pheno]

Getting back on this slowly.. It looks like there's more work than I thought to make these serial version "custom jobs" work. But I've decided to start the work nevertheless. I think the speed increase is noticeable - although I was hoping for even faster communication with the direct way.

[_TB_]

Pheno great work! It is really _really_ cool that the community just goes that extra bit to make logging superior!

Two questions;
- The version posted in OP - is that the latest version? And will it run standalone?
- Where can I get a prg file for 325d (DDE6.2) ?