It is already fixed. New build will be released on Friday. The divider was 1024 not a 1000.
I have built my entire map around lambda readings coming from V3 studio… Does this mean they are all incorrect? Or is this only when using the AFR View, rather than Lambda View? I really hope that’s the case 
If you read data via User CAN from exteranal controller, the values was a little bit off (2,5%).
@sourcerer did you ever find a resolution to this? Sounds like the only problem addressed was with the CAN calculation, but not the different readings between ECUs…
For such comparisons u all should measure and compare lambda and not afr.
Sensor is measuring lambda, but afr then is calculated by what u give the ECU as stoichiometric AFR. And this is not always 14,7. For E10 fuel used in Germany or USA it is 14,1. If u set this value not correct or different in different ECUs u will have differences in afr but lambda is the same. AFR just creates confusion cause many people have wrong fuel settings or the ECUs are intransparent which ratio they use.
Lambda is transparent and for all ECUs the same.
All Gauge which just show AFR without be able to set stoichiometric AFR u can throw away cause they assume stoichiometric AFR 14,7 which is not the standard anymore. Even Super+ in germany has 5% ethanol content, which results in 14,6.
Useing AFR is confusing and useless and creats such confusion like discussed in this thread here.
Not a scientific test, but we compared the values with the oem ecu installed on a vag tsi engine. We checked the lambda readings using vag com and the Aem wideband. The readings were very close to the oem ecu readings, so we went with the AEM wideband controller data.
If i remember correctly The latest software revision fixed some issues with the lambda readings. I haven’t tested it yet, but I will test it in the future.
If you use extranal controller via can there was a bug that instead of 1000, the divder was 1024. It was fixed in 3.068.
I understand. The issue is that nothing was changed except for unplugging one emu black and then plugging in a different emu black. And the wideband readings were different between the two. Was it a software setting?
I had problems with AEM trough analog to EMU Classic. Reading was way offf no matter what I did. Only CAN version seems to be correctly sent to emu. You can look on the internet about AEM reading some offset. (in my case offset was random each day so I could autotune endlessly)