Adamw

You will need load = MAP if you intend to ever have an idle valve.
MSEL sell a nice Bosch 1Bar MAP.  

I dont see anywhere in the help for the AEM device that tells you to use the find device tool.  This tool is only for products that can be configured via CAN bus like CAN keypads and Link CAN lambdas.  The AEM device has no configurability.  

Same way you see any parameter, add them to a time plot or gauge etc.  

 
 
Provided you have idle speed control mode set to closed loop then the integral will be in the Idle actuator menu.

 
No, in G4X the idle position values are " percent open", so larger numbers should raise the idle speed.  If you have a 3wire valve then change your Aux 1 active state to the opposite of what it is now.  

I wouldnt bother with 60-2, but a 36-1 or the RX8 one would be noticeably better than the 12T S6.  The main difference is how much drift you get under high crank acceleration.  With some basic testing I done many years ago, gong from 2 crank teeth to 12T was a massive difference, going from 12T to 24T was a good improvement again.  Going from 24 to 36T I couldnt really tell any difference with the gear I was testing with.  

LH Exhaust cam sensor needs to be connected to trigger 2, the others can go to any of DI1-4.  Set the VVT to user defined mode, assign the correct DI and aux to each cam in the bank 1/2 settings.  Then run the engine at a fast idle (above the vvt RPM and ECT lockout) and set the cam angle test to "calibrate".  The ecu will automatically determine tooth count and populate the tooth offset tables on its own, the cam angle test will turn itself back off when done.  It should be functional after that.    

Yes.

Aux 1 has a high side drive chip fitted to it for the VVT, so this cant be used as a normal low side drive type aux. 
Aux 16 is connected internally to a relay type circuit for use as an E-throttle relay, so this is not available. 
Everything else will be ok.
For the trigger, the crank signal must go to pin 33 and the cam signal to pin 32.  

Yep, will do R32-R34

As per the bottom pic.  The ecu and dash have the terminating resistors inside so these need to be at each end of the bus, then any other CAN device should branch off the main trunk.  

Since you have the "lock timing to" set to 15, you should be adjusting the offset until the timing light shows the 15deg timing marks are lined up.  If the engine doesnt have 15deg marks then change the "lock timing to" to 10deg and adjust offset until the 10deg mark lines up.  
Note you have to hit the enter key after typing in a new offset and you will see the text turn blue to confirm it is applied.  When all adjusted, then close the set base timing window and do a store.  
I just looked at you log and noticed that you are doing it without the engine running - this is not as accurate as doing it with the engine running, so dont spend too much time on it at this stage - instead just get it close enough to start, get it started, make a few adjustments if needed so that it idles on its own, then check base timing again with it idling.  

Yep.
Yep, most likely will need to increase the master.  You may have to drop the idle base position a bit if it idles too high when it has enough fuel.  

Your ECU has the 5pin CAN ports so you would need the CANJST.  The CANJST4 is for the newer ecu's which have a 4 pin CAN port.
The advantage Vs the CANPCB + CANF is it is cheaper and it uses a more common DTM connector which is easier to work with (crimp terminals and can be disassembled).  The CANF is solder connection and can be painful to solder.  
Since you already have the CANPCB/CANF, carry on and use it.  You only need 1.  

This is the one I hear of most often:  https://all4swap.ru/product/is300
There is another one but I cant remember who does it.  

You need to click on the "set base timing" option.

The amount of retard you get is based on 3 factors.  "Knock level detected", "Ignition Retard Gain" & "Knock Trim Gain Table" - these are all multiplied together.  
"Knock Level Detected" is Knock level - Knock Threshold.  So in your first example on cyl 1 you have a knock level detected of 0.62 (1.62-1.00), your Ign retard gain is 1.00 and knock trim gain table is 1.  So you would get 0.62 x 1.0 x 1.0 = 0.6deg instantaneous retard.  
In the second example on cyl 3 you have a knock level detected of 0.18 x 1 x 1, = 0.18deg (truncated to 0.1deg).
With normalised mode a knock event will be a significant spike above the threshold so your retard would be more significant than your examples.  Looking through my maps with normalised mode I usually have a retard gain around 2.0-2.5 and my trim gain table is just 1.0 right across.
Below is an example pull from our subaru where I have put an extra 5 deg advance in a single cell in the ign table at 3500rpm to purposely make it knock (you can see where Ign angle bumps up as we pass through), Cyl 4 knocks, the ecu instantly pulls 4 deg out of cyl 4 and you see after that the cyl 4 noise level drops back below the others.   
Threshold was 2.8, knock level was 4.64, gain was 2.5 (I have the retard limit set at 4 in this car).

 

I think the problem is we have a unnamed padding column in the export so there is no character between the quotation marks, MLVHD doesnt seem to like that.  If I put any character where I have the two pink lines below then it opens fine.  Im not sure if that column needs to be there for some reason or if it is just an oversight.  Ill ask Vaughan to have a look when he gets a minute.  

Your gain is too high - it varies a lot but from engine to engine but a more typical gain is around 1-2.  Adjust the gain so that at medium RPM/medium load your average knock level is around 100-300, and say no more than about 600 at WOT & redline.  
You will likely get less random noise using one of the narrowband frequency channels (I havent tuned a 1JZ so cant offer a suggestion from experience, just try the calculated value as a test).  I would also increase  your window start to 15 so you are only "listening" for knock after the TDC piston slap has occured and shorten the window length to 50deg.  
Once you have less erratic noise and a knock level profile that trends with RPM then you will want to adjust the individual cyl gains so that the knock levels of each cylinder on average are as similar as you can get. 

The amount of retard you get is based on 3 factors.  "Knock level detected", "Ignition Retard Gain" & "Knock Trim Gain Table" - these are all multiplied together.  
"Knock Level Detected" is Knock level - Knock Threshold.  So in your first example on cyl 1 you have a knock level detected of 0.62 (1.62-1.00), your Ign retard gain is 1.00 and knock trim gain table is 1.  So you would get 0.62 x 1.0 x 1.0 = 0.6deg instantaneous retard.  
In the second example on cyl 3 you have a knock level detected of 0.18 x 1 x 1, = 0.18deg (truncated to 0.1deg).
With normalised mode a knock event will be a significant spike above the threshold so your retard would be more significant than your examples.  Looking through my maps with normalised mode I usually have a retard gain around 2.0-2.5 and my trim gain table is just 1.0 right across.
Below is an example pull from our subaru where I have put an extra 5 deg advance in a single cell in the ign table at 3500rpm to purposely make it knock (you can see where Ign angle bumps up as we pass through), Cyl 4 knocks, the ecu instantly pulls 4 deg out of cyl 4 and you see after that the cyl 4 noise level drops back below the others.   
Threshold was 2.8, knock level was 4.64, gain was 2.5 (I have the retard limit set at 4 in this car).

 

No, G4+ Xtreme has built-in E-throttle.

Import the fuel table out of our Monsoon base map, that is quite generic and will run near anything.  Charge cooling of 20 will be a starting point.  

The fuel pressure sensor is typically measuring rail pressure so the analog input it is connected to has no awareness of manifold pressure.  The error setting on the fuel pressure sensor analog input will make the rail pressure appear to be fixed at 3bar with no increase with boost to the ecu.  The engine will go rich because the ecu will think the differential pressure is dropping so will increase inj PW.  

In many ECU's in batch fire mode the injector timing is locked at a fixed point.  In G4X it does exactly what you tell it to do.  You are right though, it would generally make little difference to how the engine runs in multipoint group mode.  it may hint that your injector spray pattern is poor so timing makes a big difference to how much vapor Vs liquid fuel gets into the cylinder. 
It doesnt, it only knows position within 360degs.  So Inj timing of 390BTDC is the same as inj timing set to 30BTDC.  
In the captures below, trigger offset is 0 and ign timing is zero, ign mode is wasted spark, so 0BTDC is the first falling edge after the trig 1 missing teeth, you can see in the first pic the end of injection and spark both occur in the same place.  
Inj timing 0BTDC:

 
Inj timing 390BTDC (30BTDC would give same result):

 
Inj timing 550BTDC (190BTDC would give same result):

 

The modification works out very similar cost wise, possibly a little cheaper depending on shipping.  I know some of our europe offices are set up to do this work  but I dont think our US office is yet doing this service, check with tech support.   The bonus with the modification compared to the external module is you will gain a couple of aux outputs and they will usually add a couple of extra analog inputs at the same time on ECU's that dont have a lot of spares.  

In G4X you do the setup in the function that you want to use rather than on the pin like you did in our older ecu's.  So for example if you want aux 5 to control the fuel pump, you dont go to aux 5 and assign fuel pump to that pin - you instead go to the fuel pump function and set aux 5 as the output there.  This is to allow things like CAN or math functions etc to be used and inputs and outputs.  
For a switched cam this is done as a GP output, so go to Aux outputs>GP output>GP output 1, set your output to inj 6 and your activation conditions.  

Yes we have been working on CAN support for the 2015+ STI.  I have most of it working, so far I have only tested on 3 local cars 2015-2018 and I think they are all NZDM models. On two of those cars everything seems to work correctly,  on the 3rd car the boost gauge seems to read incorrectly sometimes so I still need to go back and do more investigation on that.   
I will send you some test software to try for us as it would be nice to confirm that it works in a JDM model.  I will just get the software assembled into a more user friendly installer package and will private message you when I have it.