I was having making some graphical charts from the tube task charts located in the manuals. I started with the Mark VII since it was easier to follow than the listed stage names for each triode based on modes. Anyhow, after doing that I did create a spreadsheet of the Mark V90, Mark V:35 (same as the V:25), JP2c and the Mark VII. I will show that first. It is somewhat helpful but the graphical representation is easier to see how things get shifted. This also identified that the Mark V:25 and V:35 are a bit closer to the Mark VII but mostly in crunch mode. The Mark V90 uses a dedicated triode from V2 for CH2 which was probably needed to complete the Mark I circuit. Not like the Mark V:25 or V:35 as that triode stage has different plate and cathode resistors. The 25/35 actually inserts the overdrive stage in between V1A and V1B which is the same as the Mark VII crunch.
Here is the table or matrix of the task charts in one image.
Since things jump around a bit from what is next in line in the signal path, it is easier to see it in graphical form instead.
This is for the Mark V90: In this image, V5A is the overdrive gain stage since it is first (in red) and the V4B is the boosting stage. V3A is more or less used as a buffer stage with some attenuation circuitry and V6A is used to define the IIC+. IV and Extreme characteristics as the final gain stage.
The Mark V:25 and V:35 still make use of the extra buffer stage V3A and attenuate it before the final defining gain stage of V5B. Much like the V90, this is where the IIC+, IV and extreme modes are defined. What I found interesting is that the overdrive gain stage V2A was used in the crunch mode. Sort of makes me wonder was this sort of a new change in design that lead to the Mark VII?
Just for references. I may as well include the Mark VII graphical chart. I did not bother with the clean modes as I was more interested in the gain channels. The clean and fat modes would be the same as the top line with the V3B and V4A completely bypassed. this is also similar to the JP2C but the top line only.
Here is the table or matrix of the task charts in one image.
Since things jump around a bit from what is next in line in the signal path, it is easier to see it in graphical form instead.
This is for the Mark V90: In this image, V5A is the overdrive gain stage since it is first (in red) and the V4B is the boosting stage. V3A is more or less used as a buffer stage with some attenuation circuitry and V6A is used to define the IIC+. IV and Extreme characteristics as the final gain stage.
The Mark V:25 and V:35 still make use of the extra buffer stage V3A and attenuate it before the final defining gain stage of V5B. Much like the V90, this is where the IIC+, IV and extreme modes are defined. What I found interesting is that the overdrive gain stage V2A was used in the crunch mode. Sort of makes me wonder was this sort of a new change in design that lead to the Mark VII?
Just for references. I may as well include the Mark VII graphical chart. I did not bother with the clean modes as I was more interested in the gain channels. The clean and fat modes would be the same as the top line with the V3B and V4A completely bypassed. this is also similar to the JP2C but the top line only.
oops. I thought something was missing. It was the phase inverter tube I added. The Mark VII does not have the phase inverter tube shown. So with the Mark V in general (all models) the FX loop send is the output of the GEQ and the FX return is V6B for the V90 and V5B for the V:25 and V:35. 
