diff --git a/src/parallel b/src/parallel
index ea99c35e..ac8745a9 100755
--- a/src/parallel
+++ b/src/parallel
@@ -1741,8 +1741,22 @@ sub usleep {
 
 sub multiply_binary_prefix {
     # Evalualte numbers with binary prefix
+    # k=10^3, m=10^6, g=10^9, t=10^12, p=10^15, e=10^18, z=10^21, y=10^24
+    # K=2^10, M=2^20, G=2^30, T=2^40, P=2^50, E=2^70, Z=2^80, Y=2^80
+    # Ki=2^10, Mi=2^20, Gi=2^30, Ti=2^40, Pi=2^50, Ei=2^70, Zi=2^80, Yi=2^80
+    # ki=2^10, mi=2^20, gi=2^30, ti=2^40, pi=2^50, ei=2^70, zi=2^80, yi=2^80
     # 13G = 13*1024*1024*1024 = 13958643712
     my $s = shift;
+    $s =~ s/k/*1000/g;
+    $s =~ s/M/*1000*1000/g;
+    $s =~ s/G/*1000*1000*1000/g;
+    $s =~ s/T/*1000*1000*1000*1000/g;
+    $s =~ s/P/*1000*1000*1000*1000*1000/g;
+    $s =~ s/E/*1000*1000*1000*1000*1000*1000/g;
+    $s =~ s/Z/*1000*1000*1000*1000*1000*1000*1000/g;
+    $s =~ s/Y/*1000*1000*1000*1000*1000*1000*1000*1000/g;
+    $s =~ s/X/*1000*1000*1000*1000*1000*1000*1000*1000*1000/g;
+
     $s =~ s/Ki?/*1024/gi;
     $s =~ s/Mi?/*1024*1024/gi;
     $s =~ s/Gi?/*1024*1024*1024/gi;
diff --git a/src/parallel.pod b/src/parallel.pod
index b2347440..c5d90bdf 100644
--- a/src/parallel.pod
+++ b/src/parallel.pod
@@ -331,9 +331,10 @@ Implies B<--semaphore>.
 
 =item B<--block-size> I<size>
 
-Size of block in bytes. The size can be postfixed with K, M, G, T, or
-P which would multiply the size with 1024, 1048576, 1073741824,
-1099511627776 or 1125899906842624 respectively.
+Size of block in bytes. The size can be postfixed with K, M, G, T, P,
+k, m, g, t, or p which would multiply the size with 1024, 1048576,
+1073741824, 1099511627776, 1125899906842624, 1000, 1000000,
+1000000000, 1000000000000, or 1000000000000000 respectively.
 
 GNU B<parallel> tries to meet the block size but can be off by the
 length of one record.
@@ -799,7 +800,7 @@ B<tr ':' '\012' < /etc/passwd | parallel -N7 chown {1} {6}>
 Implies B<-X> unless B<-m> or B<--pipe> is set.
 
 When used with B<--pipe> B<-N> is the number of records to read. This
-is much slower than B<--blocksize> so avoid it if performance is
+is much slower than B<--block> so avoid it if performance is
 important.