calibre PEX post-simulation problem, too

[lhlbluesky]

first, thanks everyone who give me help these days.

now, i still have some problem. in pre-simulation, my circuit can work well. but when i extracted PEX netlist in calibre, and do layout post-simulation with the extracted netlist (with spectre), i find that, some modules' performance has some degree decrease, for ex: accuracy from 11bit to 10bit; and some modules' performance has severe decrease, for ex: for a buffer, input of 1.5V relating output of only 800mV, why?

i want to know which nets or parastic R or C affected the circuit performace, but the PEX netlist is too complicated, and with too many parasitic resisitors and capacitors, and i find their value not very large, in general, cap < 1fF, res < 40ohm
, however, what is the reason for my post-simulation?

here are some warnings and notices;

for PEX netlist, there are the following warnings:
pin not connected for cell AMP_1 (30) local nn=18 net num=10 pin_layer=593 seed_layer=18;
like this, they exist in many places.
for post-simulation output log:
only one connection to the following 5 nodes:
I89.XI0/XI58/N0
I89.XI2/XI22/net037
I89.XI2/XT21/N0
I89.XI2/XI48/N0
I89.XI2/XI30/N0
however, i checkd these nddes in layout, they connect well, and lvs is ok. but what is the reason, can anyone give me some advice?
i doubt that the way i extract the netlist has some problem, but i'm not sure.

besides, for calibreview post-simulation, what's the related procedure, can anyone give me some related paper or books?

thanks all. expected for answers.

 

[lhlbluesky]

any advice?

 

[wpchan05]

Assume your schematic sim is running ok, can you use your layout netlist with no parasitic and redo the simulation? The layout simulation without parasitic should be identical to your schematic simulation.