diff --git a/book/rv32/chapter.tex b/book/rv32/chapter.tex
index 708e5fd..434d223 100644
--- a/book/rv32/chapter.tex
+++ b/book/rv32/chapter.tex
@@ -209,7 +209,7 @@ The address of the current instruction minus a numeric offset.
 \subsection{pcrel\_13}
 \label{pcrel.13}
 
-An address that is within $[-4096..4095]$ of the current instruction location.
+An address that is within $[-4096..4094]$ $[$\verb@-0x1000..0x0ffe@$]$ of the current instruction location.
 These addresses are typically expressed in assembly source code by using labels.
 See \autoref{insnformat:btype} for examples.
 
@@ -217,7 +217,7 @@ See \autoref{insnformat:btype} for examples.
 \subsection{pcrel\_21}
 \label{pcrel.21}
 
-An address that is within $[-1048576..1048575]$ of the current instruction 
+An address that is within $[-1048576..1048574]$ $[$\verb@-0x100000..0x0ffffe@$]$ of the current instruction 
 location.  
 These addresses are typically expressed in assembly source code by using labels.
 See \autoref{insnformat:jtype} for an example.
@@ -414,8 +414,8 @@ arranged in a different order.
 
 Note that the \verb@imm_j@ value is 
 %expressed in the instruction as a target address that is converted to 
-a 21-bit value in the range of 
-$[-1048576..1048575]$ representing a \verb@pc@-relative offset to the 
+an even 21-bit value in the range of 
+$[-1048576..1048574]$ $[$\verb@-0x100000..0x0ffffe@$]$ representing a \verb@pc@-relative offset to the 
 target address. 
 
 %In the J-type format the 20 {\em imm} bits are arranged such 
@@ -995,8 +995,8 @@ to form the \verb@imm_b@ value.
 \end{figure}
 
 Note that \verb@imm_b@ is expressed in the instruction as a target 
-address that is converted to a 13-bit value in the range of 
-$[-4096..4095]$ representing a \verb@pc@-relative offset to the
+address that is converted to an even 13-bit value in the range of 
+$[-4096..4094]$ $[$\verb@-0x1000..0x0ffe@$]$ representing a \verb@pc@-relative offset to the
 target address. For example, consider the branch instructions in
 the following code: