diff --git a/TODO b/TODO
index b596c605..a75c867d 100644
--- a/TODO
+++ b/TODO
@@ -5,6 +5,7 @@
   - Performance check on Guido's reimplementation strategy
 
 * FIXME audit
+
 * const audit
 
 * Replace vset with a call to merge.; 
@@ -13,6 +14,8 @@
 
 * Strong test for norm2, conj and all primitive types. -- tests/Grid_simd.cc is almost there
 
+* Thread scaling tests Xeon, XeonPhi
+
 ================================================================
 *** New Functionality
 ================================================================
@@ -23,9 +26,6 @@
   - use protocol buffers? replace xmlReader/Writer ec..
   - Binary use htonll, htonl
 
-* Bug in SeedFixedIntegers gave same output on each site. -- Think I fixed but NOT checked for sure
-  Implement and use lattice IO to verify this.
-
 * Expression template engine: -- DONE
    -- Norm2(expression) problem: introduce norm2 unary op, or Introduce conversion automatic from expression to Lattice<vobj>
 
@@ -33,6 +33,7 @@
 
 ** Make the Tensor types and Complex etc... play more nicely.
   - TensorRemove is a hack, come up with a long term rationalised approach to Complex vs. Scalar<Scalar<Scalar<Complex > > >
+
   QDP forces use of "toDouble" to get back to non tensor scalar. This role is presently taken TensorRemove, but I
   want to introduce a syntax that does not require this.
   - Reductions that contract indices on a site should always demote the tensor structure.
@@ -71,6 +72,7 @@
    // Fourier transform equivalent.
 
 Actions -- coherent framework for implementing actions and their forces.
+
 * Fermion
   - Wilson
   - Clover
@@ -81,22 +83,24 @@ Actions -- coherent framework for implementing actions and their forces.
 * Gauge
   - Wilson, symanzik, iwasaki
 
-Algorithms
+Algorithms (lots of reuse/port from BFM)
 * LinearOperator
 * LinearSolver
 * Polynomial 
 * Eigen
+* CG
 * Pcg
 * Adef2
 * DeflCG
 * fPcg
 * MCR
 * HDCG
-* HMC, Heatbath
+* HMC, 
+* Heatbath
 * etc..
 
 ======================================================================================================
-FUNCTIONALITY: it pleases me to keep track of things I have done (keeps sane)
+FUNCTIONALITY: it pleases me to keep track of things I have done (keeps me arguably sane)
 ======================================================================================================
 
 * Command line args for geometry, simd, etc. layout. Is it necessary to have -- DONE
@@ -158,4 +162,6 @@ FUNCTIONALITY: it pleases me to keep track of things I have done (keeps sane)
 * Conformable test in Cshift routines. -- none needed ; there is only one 
 * Conformable testing in expression templates -- DONE (recursive)
 
+* Bug in SeedFixedIntegers gave same output on each site. -- DONE
+  Implement and use lattice IO to verify this.            -- cout for lattice types DONE
 
diff --git a/lib/Grid_simd.h b/lib/Grid_simd.h
index 39265896..7ad7367b 100644
--- a/lib/Grid_simd.h
+++ b/lib/Grid_simd.h
@@ -234,5 +234,60 @@ namespace Grid {
   typedef vRealF vReal;
   typedef vComplexF vComplex;
 #endif
+
+ 
+  inline std::ostream& operator<< (std::ostream& stream, const vComplexF &o){
+    int nn=vComplexF::Nsimd();
+    std::vector<ComplexF,alignedAllocator<ComplexF> > buf(nn);
+    vstore(o,&buf[0]);
+    stream<<"<";
+    for(int i=0;i<nn;i++){
+      stream<<buf[i];
+      if(i<nn-1) stream<<",";
+    }
+    stream<<">";
+    return stream;
+  }
+ 
+  inline std::ostream& operator<< (std::ostream& stream, const vComplexD &o){
+    int nn=vComplexD::Nsimd();
+    std::vector<ComplexD,alignedAllocator<ComplexD> > buf(nn);
+    vstore(o,&buf[0]);
+    stream<<"<";
+    for(int i=0;i<nn;i++){
+      stream<<buf[i];
+      if(i<nn-1) stream<<",";
+    }
+    stream<<">";
+    return stream;
+  }
+
+  inline std::ostream& operator<< (std::ostream& stream, const vRealF &o){
+    int nn=vRealF::Nsimd();
+    std::vector<RealF,alignedAllocator<RealF> > buf(nn);
+    vstore(o,&buf[0]);
+    stream<<"<";
+    for(int i=0;i<nn;i++){
+      stream<<buf[i];
+      if(i<nn-1) stream<<",";
+    }
+    stream<<">";
+    return stream;
+  }
+
+  inline std::ostream& operator<< (std::ostream& stream, const vRealD &o){
+    int nn=vRealD::Nsimd();
+    std::vector<RealD,alignedAllocator<RealD> > buf(nn);
+    vstore(o,&buf[0]);
+    stream<<"<";
+    for(int i=0;i<nn;i++){
+      stream<<buf[i];
+      if(i<nn-1) stream<<",";
+    }
+    stream<<">";
+    return stream;
+  }
+
+
 }
 #endif
diff --git a/lib/lattice/Grid_lattice_base.h b/lib/lattice/Grid_lattice_base.h
index ace7565d..915879cb 100644
--- a/lib/lattice/Grid_lattice_base.h
+++ b/lib/lattice/Grid_lattice_base.h
@@ -176,7 +176,27 @@ PARALLEL_FOR_LOOP
         }
         return ret;
     };
+
  }; // class Lattice
+
+  template<class vobj> inline std::ostream& operator<< (std::ostream& stream, const Lattice<vobj> &o){
+    std::vector<int> gcoor;
+    typedef typename vobj::scalar_object sobj;
+    sobj ss;
+    for(int g=0;g<o._grid->_gsites;g++){
+      o._grid->GlobalIndexToGlobalCoor(g,gcoor);
+      peekSite(ss,o,gcoor);
+      stream<<"[";
+      for(int d=0;d<gcoor.size();d++){
+	stream<<gcoor[d];
+	if(d!=gcoor.size()-1) stream<<",";
+      }
+      stream<<"]\t";
+      stream<<ss<<std::endl;
+    }
+    return stream;
+  }
+
 }
 
 
diff --git a/lib/lattice/Grid_lattice_peekpoke.h b/lib/lattice/Grid_lattice_peekpoke.h
index ad7dfc50..44d2a71f 100644
--- a/lib/lattice/Grid_lattice_peekpoke.h
+++ b/lib/lattice/Grid_lattice_peekpoke.h
@@ -110,7 +110,7 @@ PARALLEL_FOR_LOOP
     // Peek a scalar object from the SIMD array
     //////////////////////////////////////////////////////////
     template<class vobj,class sobj>
-      void peekSite(sobj &s,Lattice<vobj> &l,std::vector<int> &site){
+      void peekSite(sobj &s,const Lattice<vobj> &l,std::vector<int> &site){
         
       GridBase *grid=l._grid;
 
diff --git a/lib/math/Grid_math_tensors.h b/lib/math/Grid_math_tensors.h
index 0c1d67d1..8b3e4b1b 100644
--- a/lib/math/Grid_math_tensors.h
+++ b/lib/math/Grid_math_tensors.h
@@ -95,6 +95,10 @@ public:
       return *this;
     }
 
+    friend std::ostream& operator<< (std::ostream& stream, const iScalar<vtype> &o){
+      stream<< "S {"<<o._internal<<"}";
+      return stream;
+    };
 };
 ///////////////////////////////////////////////////////////
 // Allows to turn scalar<scalar<scalar<double>>>> back to double.
@@ -170,6 +174,15 @@ public:
     inline const vtype & operator ()(int i) const {
       return _internal[i];
     }
+    friend std::ostream& operator<< (std::ostream& stream, const iVector<vtype,N> &o){
+      stream<< "V<"<<N<<">{";
+      for(int i=0;i<N;i++) {
+	stream<<o._internal[i];
+	if (i<N-1)	stream<<",";
+      }
+      stream<<"}";
+      return stream;
+    };
     //    inline vtype && operator ()(int i) {
     //      return _internal[i];
     //    }
@@ -262,6 +275,19 @@ public:
   inline const vtype & operator ()(int i,int j) const {
     return _internal[i][j];
   }
+  friend std::ostream& operator<< (std::ostream& stream, const iMatrix<vtype,N> &o){
+    stream<< "M<"<<N<<">{";
+    for(int i=0;i<N;i++) {
+      stream<< "{";
+      for(int j=0;j<N;j++) {
+	stream<<o._internal[i][j];
+	if (i<N-1)	stream<<",";
+      }
+      stream<<"}\n\t\t";
+    }
+    stream<<"}";
+    return stream;
+  };
 
   //  inline vtype && operator ()(int i,int j) {
   //    return _internal[i][j];
diff --git a/tests/Makefile.am b/tests/Makefile.am
index b129d3d9..6af1ce3b 100644
--- a/tests/Makefile.am
+++ b/tests/Makefile.am
@@ -5,11 +5,14 @@ AM_LDFLAGS = -L$(top_srcdir)/lib
 #
 # Test code
 #
-bin_PROGRAMS = Grid_main Grid_stencil Grid_nersc_io Grid_cshift Grid_gamma  Grid_simd
+bin_PROGRAMS = Grid_main Grid_stencil Grid_nersc_io Grid_cshift Grid_gamma  Grid_simd Grid_rng
 
 Grid_main_SOURCES = Grid_main.cc
 Grid_main_LDADD = -lGrid
 
+Grid_rng_SOURCES = Grid_rng.cc
+Grid_rng_LDADD = -lGrid
+
 Grid_nersc_io_SOURCES = Grid_nersc_io.cc
 Grid_nersc_io_LDADD = -lGrid