diff --git a/NEWS b/NEWS
index f6b2d6c91..036df0076 100644
--- a/NEWS
+++ b/NEWS
@@ -1,7 +1,9 @@
 v0.5.1
 New : Optimal parsing => Very high compression modes, thanks to Przemyslaw Skibinski
 Changed : Dictionary builder integrated into libzstd and zstd cli
+Changed : zstd cli now uses "multiple input files" as default mode
 Fix : high compression modes for big-endian platforms
+New : zstd cli : `-t` | `--test` command
 
 v0.5.0
 New : dictionary builder utility
diff --git a/programs/playTests.sh b/programs/playTests.sh
index 014a568c1..25ba687e9 100755
--- a/programs/playTests.sh
+++ b/programs/playTests.sh
@@ -23,9 +23,19 @@ roundTripTest() {
 [ -n "$ZSTD" ] || die "ZSTD variable must be defined!"
 
 
-echo "\n**** simple test **** "
+echo "\n**** simple tests **** "
 ./datagen > tmp
 $ZSTD tmp
+$ZSTD tmp -c > tmpCompressed
+$ZSTD tmp --stdout > tmpCompressed
+$ZSTD -d tmpCompressed && die "wrong suffix error not detected!"
+$ZSTD -d tmpCompressed -c > tmpResult
+$ZSTD --decompress tmpCompressed -c > tmpResult
+$ZSTD --decompress tmpCompressed --stdout > tmpResult
+$ZSTD -q tmp && die "overwrite check failed!"
+$ZSTD -q -f tmp
+$ZSTD -q --force tmp
+
 
 echo "\n**** frame concatenation **** "
 
@@ -83,6 +93,7 @@ $ZSTD -f tmp1 notHere tmp2 && die "missing file not detected!"
 echo "\n**** integrity tests **** "
 echo "test one file (tmp1.zst) "
 $ZSTD -t tmp1.zst
+$ZSTD --test tmp1.zst
 echo "test multiple files (*.zst) "
 $ZSTD -t *.zst
 echo "test good and bad files (*) "
diff --git a/programs/zstd.1 b/programs/zstd.1
index 8d69c4ddf..27d607f5c 100644
--- a/programs/zstd.1
+++ b/programs/zstd.1
@@ -14,7 +14,7 @@
 
 .SH SYNOPSIS
 .TP 5
-\fBzstd\fR [\fBOPTIONS\fR] [-|INPUT-FILE] <OUTPUT-FILE>
+\fBzstd\fR [\fBOPTIONS\fR] [-|INPUT-FILE] [-o <OUTPUT-FILE>]
 .PP
 .B unzstd
 is equivalent to
@@ -28,15 +28,13 @@ is equivalent to
 .SH DESCRIPTION
 .PP
 \fBzstd\fR is a fast lossless compression algorithm.
-It is based on the \fBLZ77\fR family, with FSE & huff0 entropy stage.
-zstd offers compression speed > 200 MB/s per core.
-It also features a fast decoder, with speed > 500 MB/s per core.
+It is based on the \fBLZ77\fR family, with further FSE & huff0 entropy stages.
+\fBzstd\fR offers configurable compression speed, with fast modes at > 200 MB/s per core.
+It also features a very fast decoder, with speed > 500 MB/s per core.
 
 \fBzstd\fR command line is generally similar to gzip, but features the following differences :
  - Original files are preserved
- - By default, \fBzstd file1 file2\fR means : compress file1 \fBinto\fR file2.
-     Use \fB-m\fR command if you want : compress file1 into file1.zstd and file2 into file2.zst
- - By default, when compressing files, \fBzstd\fR displays advancement notification and result summary.
+ - By default, when compressing a single file, \fBzstd\fR displays progress notifications and result summary.
      Use \fB-q\fR to turn them off
 
 
@@ -45,22 +43,20 @@ It also features a fast decoder, with speed > 500 MB/s per core.
 .SH OPTIONS
 .TP
 .B \-#
- # compression level [1-19](default:1)
+ # compression level [1-21] (default:1)
 .TP
-.B \-d
+.BR \-d ", " --decompress
  decompression
 .TP
-.B \-f
+.B \-D file
+ use `file` as Dictionary to compress or decompress FILE(s)
+.TP
+.B \-o file
+ save result into `file` (only possible with a single input FILE)
+.TP
+.BR \-f ", " --force
  overwrite output without prompting
 .TP
-.BR \-m ", " --multiple
- multiple files mode
- In this mode, multiple files on the command line means compression or decompression of each named file
- Notifications are also turned off by default
-.TP
-.B \-D
- Use next file as dictionary content for compress / decompression
-.TP
 .BR \-h/\-H ", " --help
  display help/long help and exit
 .TP
@@ -73,17 +69,47 @@ It also features a fast decoder, with speed > 500 MB/s per core.
 .BR \-q ", " --quiet
  suppress warnings and notifications; specify twice to suppress errors too
 .TP
-.B \-c 
+.BR \-c ", " --stdout
  force write to standard output, even if it is the console
+
+.SH DICTIONARY
+.PP
+\fBzstd\fR offers \fIdictionary\fR compression, useful for very small files and messages.
+It's possible to train \fBzstd\fR with some samples, the result of which is saved into a file called `dictionary`.
+Then during compression and decompression, make reference to the same dictionary.
+It will improve compression ratio of small files.
+Typical gains range from ~10% (at 64KB) to x5 better (at <1KB).
 .TP
-.B \-z
- force compression
+.B \--train FILEs
+ use FILEs as training set to create a dictionary.
+ The training set should contain a lot of small files (> 100).
+ and weight typically 100x the target dictionary size
+ (for example, 10 MB for a 100 KB dictionary)
+.TP
+.B \-o file
+ dictionary saved into `file` (default: dictionary)
+.TP
+.B \--maxdict #
+ limit dictionary to specified size (default : 112640) 
+.TP
+.B \-s#
+ dictionary selectivity level (default: 9)
+ the smaller the value, the denser the dictionary, improving its efficiency but reducing its possible maximum size.
+
+.SH BENCHMARK
 .TP
 .B \-b#
  benchmark file(s) using compression level #
 .TP
 .B \-i#
  iteration loops [1-9](default : 3), benchmark mode only
+.TP
+.B \-B#
+ cut file into independent blocks of size # (default: no block)
+.TP
+.B \-r#
+ test all compression levels from 1 to # (default: disabled)
+
 
 .SH BUGS
 Report bugs at:- https://github.com/Cyan4973/zstd/issues
diff --git a/programs/zstdcli.c b/programs/zstdcli.c
index 7a6dae99c..5fe1869a9 100644
--- a/programs/zstdcli.c
+++ b/programs/zstdcli.c
@@ -142,8 +142,8 @@ static int usage_advanced(const char* programName)
 #ifndef ZSTD_NOBENCH
     DISPLAY( "Benchmark arguments :\n");
     DISPLAY( " -b#    : benchmark file(s), using # compression level (default : 1) \n");
-    DISPLAY( " -B#    : cut file into independent blocks of size # (default: no block)\n");
     DISPLAY( " -i#    : iteration loops [1-9](default : 3)\n");
+    DISPLAY( " -B#    : cut file into independent blocks of size # (default: no block)\n");
     DISPLAY( " -r#    : test all compression levels from 1 to # (default: disabled)\n");
 #endif
     return 0;
@@ -208,12 +208,17 @@ int main(int argCount, const char** argv)
         if(!argument) continue;   /* Protection if argument empty */
 
         /* long commands (--long-word) */
+        if (!strcmp(argument, "--decompress")) { decode=1; continue; }
+        if (!strcmp(argument, "--force")) {  FIO_overwriteMode(); continue; }
         if (!strcmp(argument, "--version")) { displayOut=stdout; DISPLAY(WELCOME_MESSAGE); return 0; }
         if (!strcmp(argument, "--help")) { displayOut=stdout; return usage_advanced(programName); }
         if (!strcmp(argument, "--verbose")) { displayLevel=4; continue; }
         if (!strcmp(argument, "--quiet")) { displayLevel--; continue; }
+        if (!strcmp(argument, "--stdout")) { forceStdout=1; outFileName=stdoutmark; displayLevel=1; continue; }
+        if (!strcmp(argument, "--test")) { decode=1; outFileName=nulmark; FIO_overwriteMode(); continue; }
         if (!strcmp(argument, "--train")) { dictBuild=1; outFileName=g_defaultDictName; continue; }
         if (!strcmp(argument, "--maxdict")) { nextArgumentIsMaxDict=1; continue; }
+        if (!strcmp(argument, "--keep")) { continue; }   /* does nothing, since preserving input is default; for gzip/xz compatibility */
 
         /* '-' means stdin/stdout */
         if (!strcmp(argument, "-")){