diff --git a/libraries/gpu/src/gpu/Batch.cpp b/libraries/gpu/src/gpu/Batch.cpp
index 30e591cc60..5604fa21c7 100644
--- a/libraries/gpu/src/gpu/Batch.cpp
+++ b/libraries/gpu/src/gpu/Batch.cpp
@@ -107,6 +107,19 @@ void Batch::setInputFormat(const Stream::FormatPointer& format) {
     _params.push_back(_streamFormats.cache(format));
 }
 
+void Batch::setInputBuffer(Slot channel, const BufferPointer& buffer, Offset offset, Offset stride) {
+    ADD_COMMAND(setInputBuffer);
+
+    _params.push_back(stride);
+    _params.push_back(offset);
+    _params.push_back(_buffers.cache(buffer));
+    _params.push_back(channel);
+}
+
+void Batch::setInputBuffer(Slot channel, const BufferView& view) {
+    setInputBuffer(channel, view._buffer, view._offset, Offset(view._stride));
+}
+
 void Batch::setInputStream(Slot startChannel, const BufferStream& stream) {
     if (stream.getNumBuffers()) {
         const Buffers& buffers = stream.getBuffers();
@@ -118,15 +131,6 @@ void Batch::setInputStream(Slot startChannel, const BufferStream& stream) {
     }
 }
 
-void Batch::setInputBuffer(Slot channel, const BufferPointer& buffer, Offset offset, Offset stride) {
-    ADD_COMMAND(setInputBuffer);
-
-    _params.push_back(stride);
-    _params.push_back(offset);
-    _params.push_back(_buffers.cache(buffer));
-    _params.push_back(channel);
-}
-
 void Batch::setIndexBuffer(Type type, const BufferPointer& buffer, Offset offset) {
     ADD_COMMAND(setIndexBuffer);
 
@@ -153,3 +157,17 @@ void Batch::setProjectionTransform(const Transform& proj) {
     _params.push_back(_transforms.cache(proj));
 }
 
+void Batch::setUniformBuffer(uint32 slot, const BufferPointer& buffer, Offset offset, Offset size) {
+    ADD_COMMAND(setUniformBuffer);
+
+    _params.push_back(size);
+    _params.push_back(offset);
+    _params.push_back(_buffers.cache(buffer));
+    _params.push_back(slot);
+}
+
+void Batch::setUniformBuffer(uint32 slot, const BufferView& view) {
+    setUniformBuffer(slot, view._buffer, view._offset, view._size);
+}
+
+
diff --git a/libraries/gpu/src/gpu/Batch.h b/libraries/gpu/src/gpu/Batch.h
index 601ae63a77..c3a87c8f19 100644
--- a/libraries/gpu/src/gpu/Batch.h
+++ b/libraries/gpu/src/gpu/Batch.h
@@ -72,8 +72,9 @@ public:
     // IndexBuffer
     void setInputFormat(const Stream::FormatPointer& format);
 
-    void setInputStream(Slot startChannel, const BufferStream& stream); // not a command, just unroll into a loop of setInputBuffer
     void setInputBuffer(Slot channel, const BufferPointer& buffer, Offset offset, Offset stride);
+    void setInputBuffer(Slot channel, const BufferView& buffer); // not a command, just a shortcut from a BufferView
+    void setInputStream(Slot startChannel, const BufferStream& stream); // not a command, just unroll into a loop of setInputBuffer
 
     void setIndexBuffer(Type type, const BufferPointer& buffer, Offset offset);
 
@@ -87,6 +88,9 @@ public:
     void setViewTransform(const Transform& view);
     void setProjectionTransform(const Transform& proj);
 
+    // Shader Stage
+    void setUniformBuffer(uint32 slot, const BufferPointer& buffer, Offset offset, Offset size);
+    void setUniformBuffer(uint32 slot, const BufferView& view); // not a command, just a shortcut from a BufferView
 
     // TODO: As long as we have gl calls explicitely issued from interface
     // code, we need to be able to record and batch these calls. THe long 
@@ -117,6 +121,7 @@ public:
     void _glUseProgram(GLuint program);
     void _glUniform1f(GLint location, GLfloat v0);
     void _glUniform2f(GLint location, GLfloat v0, GLfloat v1);
+    void _glUniform4fv(GLint location, GLsizei count, const GLfloat* value);
     void _glUniformMatrix4fv(GLint location, GLsizei count, GLboolean transpose, const GLfloat* value);
 
     void _glMatrixMode(GLenum mode); 
@@ -161,6 +166,8 @@ public:
         COMMAND_setViewTransform,
         COMMAND_setProjectionTransform,
 
+        COMMAND_setUniformBuffer,
+
         // TODO: As long as we have gl calls explicitely issued from interface
         // code, we need to be able to record and batch these calls. THe long 
         // term strategy is to get rid of any GL calls in favor of the HIFI GPU API
@@ -187,6 +194,7 @@ public:
         COMMAND_glUseProgram,
         COMMAND_glUniform1f,
         COMMAND_glUniform2f,
+        COMMAND_glUniform4fv,
         COMMAND_glUniformMatrix4fv,
 
         COMMAND_glMatrixMode,
diff --git a/libraries/gpu/src/gpu/Context.h b/libraries/gpu/src/gpu/Context.h
index 3a0fffb4ef..6839f9480a 100644
--- a/libraries/gpu/src/gpu/Context.h
+++ b/libraries/gpu/src/gpu/Context.h
@@ -12,7 +12,6 @@
 #define hifi_gpu_Context_h
 
 #include <assert.h>
-#include "GPUConfig.h"
 
 #include "Resource.h"
 
diff --git a/libraries/gpu/src/gpu/Format.h b/libraries/gpu/src/gpu/Format.h
index d216495b4c..aeeece8a62 100644
--- a/libraries/gpu/src/gpu/Format.h
+++ b/libraries/gpu/src/gpu/Format.h
@@ -12,8 +12,6 @@
 #define hifi_gpu_Format_h
 
 #include <assert.h>
-#include "GPUConfig.h"
-
 
 namespace gpu {
 
@@ -94,7 +92,8 @@ static const int DIMENSION_COUNT[NUM_DIMENSIONS] = {
 // Semantic of an Element
 // Provide information on how to use the element
 enum Semantic {
-    RGB = 0,
+    RAW = 0, // used as RAW memory
+    RGB,
     RGBA,
     XYZ,
     XYZW,
@@ -104,6 +103,8 @@ enum Semantic {
     DIR_XYZ,
     UV,
     R8,
+    INDEX, //used by index buffer of a mesh
+    PART, // used by part buffer of a mesh
 
     NUM_SEMANTICS,
 };
@@ -119,7 +120,7 @@ public:
         _type(type) 
     {}
     Element() :
-        _semantic(R8),
+        _semantic(RAW),
         _dimension(SCALAR),
         _type(INT8)
     {}
diff --git a/libraries/gpu/src/gpu/GLBackend.cpp b/libraries/gpu/src/gpu/GLBackend.cpp
index 90639929ca..f5f998d0d9 100644
--- a/libraries/gpu/src/gpu/GLBackend.cpp
+++ b/libraries/gpu/src/gpu/GLBackend.cpp
@@ -31,6 +31,8 @@ GLBackend::CommandCall GLBackend::_commandCalls[Batch::NUM_COMMANDS] =
     (&::gpu::GLBackend::do_setViewTransform),
     (&::gpu::GLBackend::do_setProjectionTransform),
 
+    (&::gpu::GLBackend::do_setUniformBuffer),
+
     (&::gpu::GLBackend::do_glEnable),
     (&::gpu::GLBackend::do_glDisable),
 
@@ -54,6 +56,7 @@ GLBackend::CommandCall GLBackend::_commandCalls[Batch::NUM_COMMANDS] =
     (&::gpu::GLBackend::do_glUseProgram),
     (&::gpu::GLBackend::do_glUniform1f),
     (&::gpu::GLBackend::do_glUniform2f),
+    (&::gpu::GLBackend::do_glUniform4fv),
     (&::gpu::GLBackend::do_glUniformMatrix4fv),
 
     (&::gpu::GLBackend::do_glMatrixMode),
@@ -483,6 +486,25 @@ void GLBackend::updateTransform() {
     }
 }
 
+void GLBackend::do_setUniformBuffer(Batch& batch, uint32 paramOffset) {
+    GLuint slot = batch._params[paramOffset + 3]._uint;
+    BufferPointer uniformBuffer = batch._buffers.get(batch._params[paramOffset + 2]._uint);
+    GLintptr rangeStart = batch._params[paramOffset + 1]._uint;
+    GLsizeiptr rangeSize = batch._params[paramOffset + 0]._uint;
+#if defined(Q_OS_MAC)
+    GLfloat* data = (GLfloat*) (uniformBuffer->getData() + rangeStart);
+    glUniform4fv(slot, rangeSize / sizeof(GLfloat[4]), data);
+#else
+    GLuint bo = getBufferID(*uniformBuffer);
+    glBindBufferRange(GL_UNIFORM_BUFFER, slot, bo, rangeStart, rangeSize);
+
+   // glUniformBufferEXT(_shader._program, slot, bo);
+
+    //glBindBufferBase(GL_UNIFORM_BUFFER, slot, bo);
+#endif
+    CHECK_GL_ERROR();
+}
+
 // TODO: As long as we have gl calls explicitely issued from interface
 // code, we need to be able to record and batch these calls. THe long 
 // term strategy is to get rid of any GL calls in favor of the HIFI GPU API
@@ -672,7 +694,10 @@ void Batch::_glUseProgram(GLuint program) {
     DO_IT_NOW(_glUseProgram, 1);
 }
 void GLBackend::do_glUseProgram(Batch& batch, uint32 paramOffset) {
-    glUseProgram(batch._params[paramOffset]._uint);
+    
+    _shader._program = batch._params[paramOffset]._uint;
+    glUseProgram(_shader._program);
+
     CHECK_GL_ERROR();
 }
 
@@ -708,6 +733,25 @@ void GLBackend::do_glUniform2f(Batch& batch, uint32 paramOffset) {
     CHECK_GL_ERROR();
 }
 
+void Batch::_glUniform4fv(GLint location, GLsizei count, const GLfloat* value) {
+    ADD_COMMAND_GL(glUniform4fv);
+
+    const int VEC4_SIZE = 4 * sizeof(float);
+    _params.push_back(cacheData(count * VEC4_SIZE, value));
+    _params.push_back(count);
+    _params.push_back(location);
+
+    DO_IT_NOW(_glUniform4fv, 3);
+}
+void GLBackend::do_glUniform4fv(Batch& batch, uint32 paramOffset) {
+    glUniform4fv(
+        batch._params[paramOffset + 2]._int,
+        batch._params[paramOffset + 1]._uint,
+        (const GLfloat*)batch.editData(batch._params[paramOffset + 0]._uint));
+
+    CHECK_GL_ERROR();
+}
+
 void Batch::_glUniformMatrix4fv(GLint location, GLsizei count, GLboolean transpose, const GLfloat* value) {
     ADD_COMMAND_GL(glUniformMatrix4fv);
 
diff --git a/libraries/gpu/src/gpu/GLBackend.h b/libraries/gpu/src/gpu/GLBackend.h
index 5a40e9ca36..be7a0a594b 100644
--- a/libraries/gpu/src/gpu/GLBackend.h
+++ b/libraries/gpu/src/gpu/GLBackend.h
@@ -122,6 +122,19 @@ protected:
             _lastMode(GL_TEXTURE) {}
     } _transform;
 
+    // Shader Stage
+    void do_setUniformBuffer(Batch& batch, uint32 paramOffset);
+ 
+    void updateShader();
+    struct ShaderStageState {
+
+        GLuint _program;
+
+        ShaderStageState() :
+            _program(0) {}
+    } _shader;
+
+ 
     // TODO: As long as we have gl calls explicitely issued from interface
     // code, we need to be able to record and batch these calls. THe long 
     // term strategy is to get rid of any GL calls in favor of the HIFI GPU API
@@ -148,6 +161,7 @@ protected:
     void do_glUseProgram(Batch& batch, uint32 paramOffset);
     void do_glUniform1f(Batch& batch, uint32 paramOffset);
     void do_glUniform2f(Batch& batch, uint32 paramOffset);
+    void do_glUniform4fv(Batch& batch, uint32 paramOffset);
     void do_glUniformMatrix4fv(Batch& batch, uint32 paramOffset);
 
     void do_glMatrixMode(Batch& batch, uint32 paramOffset);
diff --git a/libraries/gpu/src/gpu/Resource.cpp b/libraries/gpu/src/gpu/Resource.cpp
index 3a6e47a7ba..085d97184b 100644
--- a/libraries/gpu/src/gpu/Resource.cpp
+++ b/libraries/gpu/src/gpu/Resource.cpp
@@ -67,6 +67,26 @@ Resource::Sysmem::Sysmem(Size size, const Byte* bytes) :
     }
 }
 
+Resource::Sysmem::Sysmem(const Sysmem& sysmem) :
+    _stamp(0),
+    _size(0),
+    _data(NULL)
+{
+    if (sysmem.getSize() > 0) {
+        _size = allocateMemory(&_data, sysmem.getSize());
+        if (_size >= sysmem.getSize()) {
+            if (sysmem.readData()) {
+                memcpy(_data, sysmem.readData(), sysmem.getSize());
+            }
+        }
+    }
+}
+
+Resource::Sysmem& Resource::Sysmem::operator=(const Sysmem& sysmem) {
+    setData(sysmem.getSize(), sysmem.readData());
+    return (*this);
+}
+
 Resource::Sysmem::~Sysmem() {
     deallocateMemory( _data, _size );
     _data = NULL;
@@ -152,9 +172,25 @@ Resource::Size Resource::Sysmem::append(Size size, const Byte* bytes) {
 
 Buffer::Buffer() :
     Resource(),
-    _sysmem(NULL),
+    _sysmem(new Sysmem()),
     _gpuObject(NULL) {
-    _sysmem = new Sysmem();
+}
+
+Buffer::Buffer(Size size, const Byte* bytes) :
+    Resource(),
+    _sysmem(new Sysmem(size, bytes)),
+    _gpuObject(NULL) {
+}
+
+Buffer::Buffer(const Buffer& buf) :
+    Resource(),
+    _sysmem(new Sysmem(buf.getSysmem())),
+    _gpuObject(NULL) {
+}
+
+Buffer& Buffer::operator=(const Buffer& buf) {
+    (*_sysmem) = buf.getSysmem();
+    return (*this);
 }
 
 Buffer::~Buffer() {
diff --git a/libraries/gpu/src/gpu/Resource.h b/libraries/gpu/src/gpu/Resource.h
index 6247efe675..ec528c230b 100644
--- a/libraries/gpu/src/gpu/Resource.h
+++ b/libraries/gpu/src/gpu/Resource.h
@@ -12,13 +12,15 @@
 #define hifi_gpu_Resource_h
 
 #include <assert.h>
-#include "GPUConfig.h"
 
 #include "Format.h"
 
 #include <vector>
 
 #include <QSharedPointer>
+#ifdef _DEBUG
+#include <QDebug>
+#endif
 
 namespace gpu {
 
@@ -29,7 +31,7 @@ typedef int  Stamp;
 class Resource {
 public:
     typedef unsigned char Byte;
-    typedef unsigned int  Size;
+    typedef unsigned int Size;
 
     static const Size NOT_ALLOCATED = -1;
 
@@ -47,6 +49,8 @@ protected:
 
         Sysmem();
         Sysmem(Size size, const Byte* bytes);
+        Sysmem(const Sysmem& sysmem); // deep copy of the sysmem buffer
+        Sysmem& operator=(const Sysmem& sysmem); // deep copy of the sysmem buffer
         ~Sysmem();
 
         Size getSize() const { return _size; }
@@ -90,9 +94,6 @@ protected:
         static void deallocateMemory(Byte* memDeallocated, Size size);
 
     private:
-        Sysmem(const Sysmem& sysmem) {}
-        Sysmem &operator=(const Sysmem& other) {return *this;}
-
         Stamp _stamp;
         Size  _size;
         Byte* _data;
@@ -104,12 +105,15 @@ class Buffer : public Resource {
 public:
 
     Buffer();
-    Buffer(const Buffer& buf);
+    Buffer(Size size, const Byte* bytes);
+    Buffer(const Buffer& buf); // deep copy of the sysmem buffer
+    Buffer& operator=(const Buffer& buf); // deep copy of the sysmem buffer
     ~Buffer();
 
     // The size in bytes of data stored in the buffer
     Size getSize() const { return getSysmem().getSize(); }
     const Byte* getData() const { return getSysmem().readData(); }
+    Byte* editData() { return editSysmem().editData(); }
 
     // Resize the buffer
     // Keep previous data [0 to min(pSize, mSize)]
@@ -130,7 +134,7 @@ public:
 
     // Access the sysmem object.
     const Sysmem& getSysmem() const { assert(_sysmem); return (*_sysmem); }
-
+    Sysmem& editSysmem() { assert(_sysmem); return (*_sysmem); }
 
 protected:
 
@@ -138,8 +142,6 @@ protected:
 
     mutable GPUObject* _gpuObject;
 
-    Sysmem& editSysmem() { assert(_sysmem); return (*_sysmem); }
-
     // This shouldn't be used by anything else than the Backend class with the proper casting.
     void setGPUObject(GPUObject* gpuObject) const { _gpuObject = gpuObject; }
     GPUObject* getGPUObject() const { return _gpuObject; }
@@ -149,6 +151,281 @@ protected:
 
 typedef QSharedPointer<Buffer> BufferPointer;
 typedef std::vector< BufferPointer > Buffers;
+
+
+class BufferView {
+public:
+    typedef Resource::Size Size;
+    typedef int Index;
+
+    BufferPointer _buffer;
+    Size _offset;
+    Size _size;
+    Element _element;
+    uint16 _stride;
+
+    BufferView() :
+        _buffer(NULL),
+        _offset(0),
+        _size(0),
+        _element(gpu::SCALAR, gpu::UINT8, gpu::RAW),
+        _stride(1)
+    {};
+
+    BufferView(const Element& element) :
+        _buffer(NULL),
+        _offset(0),
+        _size(0),
+        _element(element),
+        _stride(uint16(element.getSize()))
+    {};
+
+    // create the BufferView and own the Buffer
+    BufferView(Buffer* newBuffer, const Element& element = Element(gpu::SCALAR, gpu::UINT8, gpu::RAW)) :
+        _buffer(newBuffer),
+        _offset(0),
+        _size(newBuffer->getSize()),
+        _element(element),
+        _stride(uint16(element.getSize()))
+    {};
+    BufferView(const BufferPointer& buffer, const Element& element = Element(gpu::SCALAR, gpu::UINT8, gpu::RAW)) :
+        _buffer(buffer),
+        _offset(0),
+        _size(buffer->getSize()),
+        _element(element),
+        _stride(uint16(element.getSize()))
+    {};
+    BufferView(const BufferPointer& buffer, Size offset, Size size, const Element& element = Element(gpu::SCALAR, gpu::UINT8, gpu::RAW)) :
+        _buffer(buffer),
+        _offset(offset),
+        _size(size),
+        _element(element),
+        _stride(uint16(element.getSize()))
+    {};
+    ~BufferView() {}
+    BufferView(const BufferView& view) = default;
+    BufferView& operator=(const BufferView& view) = default;
+
+    Size getNumElements() const { return _size / _element.getSize(); }
+
+    //Template iterator with random access on the buffer sysmem
+    template<typename T>
+    class Iterator : public std::iterator<std::random_access_iterator_tag,
+                                               T,
+                                               Index,
+                                               T*,
+                                               T&>
+    {
+    public:
+
+        Iterator(T* ptr = NULL) { _ptr = ptr; }
+        Iterator(const Iterator<T>& iterator) = default;
+        ~Iterator() {}
+
+        Iterator<T>& operator=(const Iterator<T>& iterator) = default;
+        Iterator<T>& operator=(T* ptr) {
+            _ptr = ptr;
+            return (*this);
+        }
+
+        operator bool() const
+        {
+            if(_ptr)
+                return true;
+            else
+                return false;
+        }
+
+        bool operator==(const Iterator<T>& iterator) const { return (_ptr == iterator.getConstPtr()); }
+        bool operator!=(const Iterator<T>& iterator) const { return (_ptr != iterator.getConstPtr()); }
+
+        Iterator<T>& operator+=(const Index& movement) {
+            _ptr += movement;
+            return (*this);
+        }
+        Iterator<T>& operator-=(const Index& movement) {
+            _ptr -= movement;
+            return (*this);
+        }
+        Iterator<T>& operator++() {
+            ++_ptr;
+            return (*this);
+        }
+        Iterator<T>& operator--() {
+            --_ptr;
+            return (*this);
+        }
+        Iterator<T> operator++(Index) {
+            auto temp(*this);
+            ++_ptr;
+            return temp;
+        }
+        Iterator<T> operator--(Index) {
+            auto temp(*this);
+            --_ptr;
+            return temp;
+        }
+        Iterator<T> operator+(const Index& movement) {
+            auto oldPtr = _ptr;
+            _ptr += movement;
+            auto temp(*this);
+            _ptr = oldPtr;
+            return temp;
+        }
+        Iterator<T> operator-(const Index& movement) {
+            auto oldPtr = _ptr;
+            _ptr -= movement;
+            auto temp(*this);
+            _ptr = oldPtr;
+            return temp;
+        }
+
+        Index operator-(const Iterator<T>& iterator) { return (iterator.getPtr() - this->getPtr())/sizeof(T); }
+
+        T& operator*(){return *_ptr;}
+        const T& operator*()const{return *_ptr;}
+        T* operator->(){return _ptr;}
+
+        T* getPtr()const{return _ptr;}
+        const T* getConstPtr()const{return _ptr;}
+
+    protected:
+
+        T* _ptr;
+    };
+
+    template <typename T> Iterator<T> begin() { return Iterator<T>(&edit<T>(0)); }
+    template <typename T> Iterator<T> end() { return Iterator<T>(&edit<T>(getNum<T>())); }
+    template <typename T> Iterator<const T> cbegin() const { return Iterator<const T>(&get<T>(0)); }
+    template <typename T> Iterator<const T> cend() const { return Iterator<const T>(&get<T>(getNum<T>())); }
+
+    // the number of elements of the specified type fitting in the view size
+    template <typename T> Index getNum() const {
+        return Index(_size / sizeof(T));
+    }
+
+    template <typename T> const T& get() const {
+ #if _DEBUG
+        if (_buffer.isNull()) {
+            qDebug() << "Accessing null gpu::buffer!";
+        }
+        if (sizeof(T) > (_buffer->getSize() - _offset)) {
+            qDebug() << "Accessing buffer in non allocated memory, element size = " << sizeof(T) << " available space in buffer at offset is = " << (_buffer->getSize() - _offset);
+        }
+        if (sizeof(T) > _size) {
+            qDebug() << "Accessing buffer outside the BufferView range, element size = " << sizeof(T) << " when bufferView size = " << _size;
+        }
+ #endif
+        const T* t = (reinterpret_cast<const T*> (_buffer->getData() + _offset));
+        return *(t);
+    }
+
+    template <typename T> T& edit() {
+ #if _DEBUG
+        if (_buffer.isNull()) {
+            qDebug() << "Accessing null gpu::buffer!";
+        }
+        if (sizeof(T) > (_buffer->getSize() - _offset)) {
+            qDebug() << "Accessing buffer in non allocated memory, element size = " << sizeof(T) << " available space in buffer at offset is = " << (_buffer->getSize() - _offset);
+        }
+        if (sizeof(T) > _size) {
+            qDebug() << "Accessing buffer outside the BufferView range, element size = " << sizeof(T) << " when bufferView size = " << _size;
+        }
+ #endif
+        T* t = (reinterpret_cast<T*> (_buffer->editData() + _offset));
+        return *(t);
+    }
+
+    template <typename T> const T& get(const Index index) const {
+        Resource::Size elementOffset = index * sizeof(T) + _offset;
+ #if _DEBUG
+        if (_buffer.isNull()) {
+            qDebug() << "Accessing null gpu::buffer!";
+        }
+        if (sizeof(T) > (_buffer->getSize() - elementOffset)) {
+            qDebug() << "Accessing buffer in non allocated memory, index = " << index << ", element size = " << sizeof(T) << " available space in buffer at offset is = " << (_buffer->getSize() - elementOffset);
+        }
+        if (index > getNum<T>()) {
+            qDebug() << "Accessing buffer outside the BufferView range, index = " << index << " number elements = " << getNum<T>();
+        }
+ #endif
+        return *(reinterpret_cast<const T*> (_buffer->getData() + elementOffset));
+    }
+
+    template <typename T> T& edit(const Index index) const {
+        Resource::Size elementOffset = index * sizeof(T) + _offset;
+ #if _DEBUG
+        if (_buffer.isNull()) {
+            qDebug() << "Accessing null gpu::buffer!";
+        }
+        if (sizeof(T) > (_buffer->getSize() - elementOffset)) {
+            qDebug() << "Accessing buffer in non allocated memory, index = " << index << ", element size = " << sizeof(T) << " available space in buffer at offset is = " << (_buffer->getSize() - elementOffset);
+        }
+        if (index > getNum<T>()) {
+            qDebug() << "Accessing buffer outside the BufferView range, index = " << index << " number elements = " << getNum<T>();
+        }
+ #endif
+        return *(reinterpret_cast<T*> (_buffer->editData() + elementOffset));
+    }
+};
+ 
+
+ // TODO: For now TextureView works with Buffer as a place holder for the Texture.
+ // The overall logic should be about the same except that the Texture will be a real GL Texture under the hood
+class TextureView {
+public:
+    typedef Resource::Size Size;
+    typedef int Index;
+
+    BufferPointer _buffer;
+    Size _offset;
+    Size _size;
+    Element _element;
+    uint16 _stride;
+
+    TextureView() :
+        _buffer(NULL),
+        _offset(0),
+        _size(0),
+        _element(gpu::VEC3, gpu::UINT8, gpu::RGB),
+        _stride(1)
+    {};
+
+    TextureView(const Element& element) :
+        _buffer(NULL),
+        _offset(0),
+        _size(0),
+        _element(element),
+        _stride(uint16(element.getSize()))
+    {};
+
+    // create the BufferView and own the Buffer
+    TextureView(Buffer* newBuffer, const Element& element) :
+        _buffer(newBuffer),
+        _offset(0),
+        _size(newBuffer->getSize()),
+        _element(element),
+        _stride(uint16(element.getSize()))
+    {};
+    TextureView(const BufferPointer& buffer, const Element& element) :
+        _buffer(buffer),
+        _offset(0),
+        _size(buffer->getSize()),
+        _element(element),
+        _stride(uint16(element.getSize()))
+    {};
+    TextureView(const BufferPointer& buffer, Size offset, Size size, const Element& element) :
+        _buffer(buffer),
+        _offset(offset),
+        _size(size),
+        _element(element),
+        _stride(uint16(element.getSize()))
+    {};
+    ~TextureView() {}
+    TextureView(const TextureView& view) = default;
+    TextureView& operator=(const TextureView& view) = default;
+};
+
 };
 
 
diff --git a/libraries/gpu/src/gpu/Stream.cpp b/libraries/gpu/src/gpu/Stream.cpp
index 099cbde24a..73677f715e 100644
--- a/libraries/gpu/src/gpu/Stream.cpp
+++ b/libraries/gpu/src/gpu/Stream.cpp
@@ -18,7 +18,7 @@ void Stream::Format::evaluateCache() {
     _elementTotalSize = 0;
     for(AttributeMap::iterator it = _attributes.begin(); it != _attributes.end(); it++) {
         Attribute& attrib = (*it).second;
-        Channel& channel = _channels[attrib._channel];
+        ChannelInfo& channel = _channels[attrib._channel];
         channel._slots.push_back(attrib._slot);
         channel._stride = std::max(channel._stride, attrib.getSize() + attrib._offset);
         channel._netSize += attrib.getSize();
@@ -41,7 +41,7 @@ BufferStream::BufferStream() :
 BufferStream::~BufferStream() {
 }
 
-void BufferStream::addBuffer(BufferPointer& buffer, Offset offset, Offset stride) {
+void BufferStream::addBuffer(const BufferPointer& buffer, Offset offset, Offset stride) {
     _buffers.push_back(buffer);
     _offsets.push_back(offset);
     _strides.push_back(stride);
diff --git a/libraries/gpu/src/gpu/Stream.h b/libraries/gpu/src/gpu/Stream.h
index 93abfeeca3..b1aaec665f 100644
--- a/libraries/gpu/src/gpu/Stream.h
+++ b/libraries/gpu/src/gpu/Stream.h
@@ -12,7 +12,6 @@
 #define hifi_gpu_Stream_h
 
 #include <assert.h>
-#include "GPUConfig.h"
 
 #include "Resource.h"
 #include "Format.h"
@@ -83,16 +82,16 @@ public:
     public:
         typedef std::map< Slot, Attribute > AttributeMap;
 
-        class Channel {
+        class ChannelInfo {
         public:
             std::vector< Slot > _slots;
             std::vector< Offset > _offsets;
             Offset _stride;
             uint32 _netSize;
 
-            Channel() : _stride(0), _netSize(0) {}
+            ChannelInfo() : _stride(0), _netSize(0) {}
         };
-        typedef std::map< Slot, Channel > ChannelMap;
+        typedef std::map< Slot, ChannelInfo > ChannelMap;
 
         Format() :
             _attributes(),
@@ -104,6 +103,7 @@ public:
 
         uint8  getNumChannels() const { return _channels.size(); }
         const ChannelMap& getChannels() const { return _channels; }
+        const Offset getChannelStride(Slot channel) const { return _channels.at(channel)._stride; }
 
         uint32 getElementTotalSize() const { return _elementTotalSize; }
 
@@ -131,7 +131,7 @@ public:
     BufferStream();
     ~BufferStream();
 
-    void addBuffer(BufferPointer& buffer, Offset offset, Offset stride);
+    void addBuffer(const BufferPointer& buffer, Offset offset, Offset stride);
 
     const Buffers& getBuffers() const { return _buffers; }
     const Offsets& getOffsets() const { return _offsets; }