Did a quick custom race with multiple classes as a test. The classes taking part were as Follows:
Gr.3 x4 (RH tyres)
Gr.4 x3(RH tyres)
600pp x5(SH, SM, CS)
550pp x2 (SH, CS)
500pp.x2(CS, CM)
16 cars in total
Let me just say first and foremost, that SOPHY, the new intelligent A.I. alleged to be included into GT7 couldn't come sooner and here's why.....
The track was Tsukuba 12 laps in the dry. 3 tire degradation and 6 fuel consumption.
I competed in the 600pp class in a race spec GT0 with SM tires 390hp 1410kg and finished in - wait for it - 2nd place...... 8 seconds behind the leader....
I chose Tsukuba because I wanted to see how the fast and slower cars managed traffic, unsurprisingly the faster cars (Particularly Gr.3's) couldn't intelligently get past the slower cars efficiently enough on such a tight track. On the 2nd lap the Gr.3 cars were flying and left everything. But when they caught the 500pp class their lap times tumbled, as did the Gr.4's
Then there was sand bagging going on for some reason, let me explain. Getting 1st in my class was a quick affair and after a thrilling 2-3 lap chase the Gr.4 cars slowed down even before they hit traffic for some reason, enabling me to pass them with ease. It was a bit of a downer actually because I was enjoying watching their battle, all the while expecting them to keep pulling away. But that didn't happen. They even let off the gass to let me pass...Disapointing.
I don't remember A.I. being this defective, even in GT5's Tsukuba enduro.
What I was baffled about the most, was the A.I. selected the wrong tires for 600pp category and downwards. I had originally intended all cars in the 600pp to be on SS tyres but forgot to switch them from a 600pp SM tyre build. Instead of being on SM tyres though, all but one of the A.I.(excluding myself) in the 600pp class were on SH's. One of the most powerful cars (a C4 with 402hp) in that category was even on CS's so that messed up it's performance in the race by a lot.
The 550pp tyres went from SH to CS. Then the 500pp cars went from CS to CM.(I forgot to change 1 of those too so that was on me) The two cars in the 550PP category in particular should have both been on SH, but all this is rough for experience sake.
I did a bit of thinking and thought maybe the computer is putting all cars on their stock tyres (before tuning). I even started doubting if the cars were even using the tuning sheets I gave them, the Gr classes included.
(A.I. was particularly clumsy around Tsukuba I noticed)
Not sure, but further tests will be done. But there is one argument that conflicts with the former. In the Gr.4 class I had placed a self-made Gr.4 car in there, the new Alpine A110. That car was successfully wearing RH's. It trailing behind it’s Gr.4 rivals at the end of the race as the A.I. spent 1 lap and a half trying to figure out how to get past a 550pp category car…
Maybe the tyre choice was too vast for the computer and it scrambled itself? IDk.
I'm going to try this again on one of the Sardegna layouts next. This time I'll make all classes from 600pp downwards wear the same tyres, likely SS. If it's still scrambled then I'll see if switching every class to slicks does the trick. Perhaps also there were just too many varying speeds for the some of the A.I. to manage around Tsukuba.

But it's moments like these I wanted to create, A GTR GT3 working around 600pp who in them selves are overtaking a 500PP category(silvia) Will definitely use a bigger track next time however
All in all I got some good data from this test, but the incompetent A.I. only augments my hope for SOPHY's arrival to the series.