Adamw

Pulsedamper in the fuelsystem? Or is it deleted?
Also you can try to change the lenght of the vacuumtube to the fuelpressure regulator.
Have not looked at the log yet. This is just based on experience of myself and others

yes one of those will be close to correct (-84 or 276).  I cant tell you which one though, it is just normally easiest just to try one, if it doesnt start or backfires, use the other one.
All of your other settings sound ok. 

Open the "set base timing" screen, that is where you enter the offset.  Remember to hit the enter key after changing the offset.

It looks to me like the DASH2Pro CAN stream sends "non-driven" wheel speed so your setup above should have worked ok.  No harm in trying what Simon suggests though and assign both driven and non driven to the same wheel.
Also check in the Dash2Pro config that the speed display is looking for "non-GPS" speed:

 
In the Dash2Pro CAN config, make sure the ECU speed channel is enabled and also make sure you dont have more than 15 channels enabled:

 

Our generic dash stream sends out all 4 wheel speeds, but your car only has one VSS connected so only one of those wheel speeds will work, the other 3 will read zero.I just had a look at Gaugeart's documentation but it doesnt state anywhere which wheel speed channel their device is actually looking for so you are going to have to experiment a bit in PCLink.
Go to your DI3 settings, change its function to try LF,LR,RF,RR, hopefully one of those will work.
 

MGP should roughly match your gauge. (MGP = Manifold Gauge Pressure). -21" is -10.3 psi so while its not dead on, its in the same ballpark. Also, -7.4psi MGP sounds like a reasonable number for idle vacuum. When the engine is off, but ECU on, does it read roughly 14.7psi/101kpa? If this checks out, i'd say your MAP sensor is reasonably healthy.
Do any other sensor values jump out at you as wrong? hit f12 and have a look at the various runtime value and if anything looks completely out of line with reality (eg when its cold with ECU on and engine off, most temps will be within 5deg of ambient).
When its running badly with everything plugged in, is your AFR high or low? can you post a log and your pclr please?
 
 

With aggressive ignition cut limiting on a engine with hydraulic lifters the large bangs in the exhaust port/manifold can lift the valves slightly and cause the lifters to "pump up".  This can play out in different ways depending on the valve train design.  SR20's for instance are almost guaranteed to through a rocker off.  Some other engines pump up enough for the valves to hit the pistons.  I dont think the 4G63 usually has much drama in this respect but I personally still wouldnt use prolonged ignition cut limiter if I had hydraulic lifters.   

You will need a wideband connected if you want to measure/display/use lambda. A narrowband sensor is not capable of giving a valid lambda measurement, it only gives a rich/lean signal.

We would need to see a log to confirm but I'm highly suspicious that your issue is trigger error.  
The Nissan 360 trigger works ok on relatively stock engines when all mechanical parts are in good condition, however they become much less useful in heavily modified engines, or when parts of the CAS drive become worn.  Heavier valve springs, more aggressive cam shafts and extended/higher RPM means there is much more torsional vibration on the camshaft drive and since the "slots" are only 1 deg wide it doesnt take much of a "shake" to cause trigger error.
If you want to run an RB engine to 9000RPM you should have a proper crank trigger.  You may get away with one of the aftermarket "24 slot" discs but really they are still not ideal.

The G4+ Mini Plugin does have built in knock hardware and it is connected to the stock knocksensor/wiring.

With laptop connected to ECU, press F12 key to open up the Runtimes screen.  Move to the triggers/limits tab.  Crank the engine and have a look at the 3 parameters shown below.  Both triggers should be green - Yes, and RPM should show 150-200.

To give a little more info...
There are many different scenario's or reasons for a piggy back type install but typically it is done where the aftermarket ECU cannot properly control or duplicate some function that the OEM ecu has.  This would most commonly be features such as automatic transmission control, cruise control/DBW, or CAN comms to other modules in the vehicle such as the instrument cluster.  
The OEM ecu would typically only be left with the bare minimum function it needs to control, all other engine control will be done by the aftermarket ECU.  The problem with many modern OEM ECU's is they have many saftey and diagnostic strategies built in that detect when for instance the injectors are disconnected.  This means the installer needs to be very familiar with how the OEM system works so he can work out a method to "fool it" into thinking everything is connected and happy.  Piggyback installs are often very difficult to get right and should only be attempted by a very competent installer.

I thought the explanation in the help was already clear enough but I will give a couple of examples to hopefully help (note my examples will ignore deadime and any other compensations).  The main thing you are missing from your example above is that your fuel equation load source will apply a multiplier as well.

 
So assuming your fuel equation load source is set to MAP and you have a master fuel of 8ms:
If I have a value of 100 in my fuel table and the map sensor is measuring 100Kpa, then the approx pulsewidth at the injector will be 8ms.
If I have a value of 100 in my fuel table and the map sensor is measuring 300Kpa, then the approx pulsewidth at the injector will be 24ms.
If I have a value of 50 in my fuel table and the map sensor is measuring 100Kpa, then the approx pulsewidth at the injector will be 4ms.
If I have a value of 50 in my fuel table and the map sensor is measuring 50Kpa, then the approx pulsewidth at the injector will be 2ms.
 
 

This is really a limit of using modeled mode with TP as the load reference - it simply cant account for extra air entering the engine from any source but the TP.  If you have a reasonably stable MAP at idle then you will probably get a better result by connecting that, otherwise it will need a change to traditional mode to allow use of the extra compensation tables.
The reason these tables arent available in modeled mode is the whole idea is to base fuel delivery on calculated/predicted airflow.  If you have no way for the ECU to "see" this increased airflow (such as MAP) then modeled mode isnt going to work.

Fudge the cal 4 calibration to make the ecu read the same as the gauge. What you are seeing is ground offset. Maybe the ecu and the gauge are grounded at different places? This is one reason why getting a CAN based lambda unit is such great idea.

it looks like you haven't got the pull-up resistors turned on?

Yes it will plug straight in.  Note there are also two suggested modifications: 1, fit a TPS Sensor.  2, fit a IAT sensor.

Hi Dan,
Turbo drag racers normally use "launch control" rather than antilag.  Some refer to it as "two step".
Usually it is activated either by a clutch pedal switch or the transbrake button.  When sitting on the line with the transbrake on or clutch down the launch strategy is activated.  The launch strategy will normally consist of a lower than normal rev limiter (usually ignition cut), with the remaining sparks retarded and sometimes a little extra fuel added.  The limiter and retard combined increase exhaust energy and create boost.
Some basic settings:

Leave fuel trim off to start with, it is normally not needed.  Only add fuel if you cant get enough boost with reasonable retard levels.
 
There are many different ways to do your retard table but probably something like below is the most common:

You can use either the Siemens or Continental ECS, the continential is slightly easier to wire so is the preferred one.
%Ethanol is not normally transmitted in the generic dash stream but I could do you a custom CAN stream to get that working.

Ok attached are a couple of files.  The .lcs is a configurable version of our "generic dash stream", it has been modified a bit (the 8 knock channels have been removed).  I have added DI2 status to the end of it so hopefully you can see how it is done. The .xc1 file is a matching custom CAN config for race studio 3.
The Link end I think is easy enough to work out by trial and error.  The AIM end is not quite so intuitive so I will post a couple of pics to explain the basics:
Click the custom CAN button, then import, then browse for the .xc1 file:

 
Once that is imported you can right click on it and select "open this custom can for editing" and you will get a screen like below with the whole message format showing.  If you look down at the last frame you will see the DI2 status that I have added.  If you double click on those DI2 bytes it will bring up an edit screen which is where you set the start bit, length, scaling etc.  

 
Have a play yourself, if you cant get it to work then reply here with the channels you need and I will try to set something up for you.
Configurable Generic Dash - modified with DI2.lcs
LINK_G4+ Generic + DI_@20170524_021339_007952.xc1

You can't modify the preconfigured streams, normally you have to start from scratch. I do already have a configurable version of the generic dash stream that I can post here later when I'm at work. I think I also have some modified MXS configs that you can copy from.

You will also need to set up the individual cyl knock level gains and cyl allocations on the "cyl setup" page, so that all cylinders report a similar background noise level.
HP Academy have quite a good webinar on tuning the G4+ Knock control system: https://www.hpacademy.com/previous-webinars/049-setting-up-knock-control-link-g4-vipec/
You will need a ~$20 membership to view their webinars.

Yes the Thunder DI's were designed specifically to take ABS speed sensor type inputs so you will well and truly be fine. 
 
The momentary and toggle settings are mostly just user preference or to allow you to take advantage of some existing switch/button that is already in the car/dashboard/steering wheel/etc.
The Toggle setting is used where you have a switch that has two distinct positions, in this case the switch takes care of "latching".  The momentary setting is for when you have a spring loaded button/switch that returns to the off position by itself, with this setting the firmware takes care of the latching (i.e one press of on, another press for off)
Most functions you can use either type of switch/setting, the exception is Cruise control which only works with momentary switches at present.  I believe it is intended to make the cruise switch type more configurable soon.
 
Either is fine for a DI switch. 

I said set it to falling earlier because at that time you had nothing connected to the DI and you wanted it to be turned off.
When your switch is turned "off" the DI will be at 0V, we call this "low", when you turn the switch/button "on" the DI will see 12V, this is called "high".  So, we set active edge to "rising" because we want the launch to turn on when our DI "rises" from low voltage to high voltage.

KnockBlock G4 Manual.pdf