Hello FreeSurfer Developers,

my fsgd is this：

GroupDescriptorFile 1
Title post compane
Class d
Class m
Class h

RescaleFlag 1
DemeanFlag 1
Variables gender Age Edu
Input bd1 d 0 15 8
Input bd2 d 1 19 13
Input bd7 d 0 17 11
Input bd8 d 1 19 13
Input bd6 m 1 22 12
Input bd9 m 1 26 7
Input bd10 m 0 53 15
Input bd11 m 1 56 4
Input hc3 h 0 45 7
Input hc4 h 1 18 13
Input hc1 h 0 52 7
Input hc2 h 1 24 10

When I run Three Groups (One Factor/Three Levels), three Covariates,my MTX is

1 -1 0 0 0 0 0 0 0 0 0 0
1 0 -1 0 0 0 0 0 0 0 0 0
It’s wrong,What should I do
--------------------------------
ERROR: matrix is ill-conditioned or badly scaled, condno = 1e+08

--------------------------------
-------- ERROR: READ THIS -----------------
--vvvvvvvvvvvvvvvvvvvvvvvvvvvvvv-----
Possible problem with experimental design:
Check for duplicate entries and/or lack of range of
continuous variables within a class.
If you seek help with this problem, make sure to send:
1. Your command line:
mri_glmfit --y rh_teach66.mgh --fsgd glmteach66.fsgd --C glmteach66.mtx --glmdir group66.area_66sm.rh --fwhm 10 --surface fsaverage rh --eres-save
2. The terminal output of this program (ie, everything it prints to the screen)
3. The FSGD file (glmteach66.fsgd)
4. The design matrix group66.area_66sm.rh/Xg.dat
Attempting to diagnose further
WARNING: Class d has 4 members. With 3 variables and using DODS, This is the bare minimum which may cause problems with the design matrix.
WARNING: Class m has 4 members. With 3 variables and using DODS, This is the bare minimum which may cause problems with the design matrix.
WARNING: Class h has 4 members. With 3 variables and using DODS, This is the bare minimum which may cause problems with the design matrix.
SumSq: Min=1.491914 (col 10), Max=2.676059 (col 11)
... could not determine the cause of the problem
--^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^----

ERROR: cannot find residual group66.area_66sm.lh/eres needed with --perm-resid
when running mri_glmfit, make sure to include --eres-save
ERROR: cannot find residual group66.area_66sm.rh/eres needed with --perm-resid
when running mri_glmfit, make sure to include --eres-save
glm66.sh: line 14: --perm-resid: command not found
-----------------------