Strange values in v$sga_target_advice for Database Version 12.2.0.1

I noticed that the view v$sga_target_advice shows in my opinion illogical values for the lowest sga_size after we migrated our databases from 11.2.0.4 to 12.2.0.1. Can others confirm this observation? Is this a known bug?

I executed this query on several databases wiht an uptime between 7 days and 20 days:

SELECT

   sga_size,

   sga_size_factor,

   estd_db_time_factor,

   estd_physical_reads

FROM v$sga_target_advice

ORDER BY sga_size ASC;

and the results are:

  SGA_SIZE SGA_SIZE_FACTOR ESTD_DB_TIME_FACTOR ESTD_PHYSICAL_READS

---------- --------------- ------------------- -------------------

      3080           0.125              0.9461            32391375

      6160           0.250              1.5866           189440957

      9240           0.375              1.2671           104011944

     12320           0.500              1.0836            54734945

     15400           0.625              1.0514            46271079

     18480           0.750              1.0239            38801628

     21560           0.875              1.0047            33528312

     24640           1.000              1.0000            32391375

     27720           1.125              0.9952            31037416

     30800           1.250              0.9921            30392827

     33880           1.375              0.9888            29388695

     36960           1.500              0.9864            28643693

     40040           1.625              0.9855            28643693

     43120           1.750              0.9852            27416060

     46200           1.875              0.9852            26227296

     49280           2.000              0.9852            26227296

  SGA_SIZE SGA_SIZE_FACTOR ESTD_DB_TIME_FACTOR ESTD_PHYSICAL_READS