diff --git a/cmake/modules/FindGLEW.cmake b/cmake/modules/FindGLEW.cmake
index e86db3fdac..b1789fb614 100644
--- a/cmake/modules/FindGLEW.cmake
+++ b/cmake/modules/FindGLEW.cmake
@@ -38,5 +38,19 @@ if (WIN32)
   find_package_handle_standard_args(GLEW DEFAULT_MSG GLEW_INCLUDE_DIRS GLEW_LIBRARIES GLEW_DLL_PATH)
   
   add_paths_to_fixup_libs(${GLEW_DLL_PATH})
+elseif (APPLE)
+else ()
+   find_path(GLEW_INCLUDE_DIR GL/glew.h)
+    find_library(GLEW_LIBRARY NAMES GLEW glew32 glew glew32s PATH_SUFFIXES lib64)
+
+    set(GLEW_INCLUDE_DIRS ${GLEW_INCLUDE_DIR})
+    set(GLEW_LIBRARIES ${GLEW_LIBRARY})
+
+    include(FindPackageHandleStandardArgs)
+    find_package_handle_standard_args(GLEW
+                                  REQUIRED_VARS GLEW_INCLUDE_DIR GLEW_LIBRARY)
+
+    mark_as_advanced(GLEW_INCLUDE_DIR GLEW_LIBRARY)
+ 
 endif ()
 
diff --git a/interface/CMakeLists.txt b/interface/CMakeLists.txt
index e2b9fcd396..cf5285a9d3 100644
--- a/interface/CMakeLists.txt
+++ b/interface/CMakeLists.txt
@@ -208,33 +208,13 @@ else (APPLE)
     "${PROJECT_SOURCE_DIR}/resources"
     $<TARGET_FILE_DIR:${TARGET_NAME}>/resources
   )
-    
-  find_package(OpenGL REQUIRED)
-  
-  if (${OPENGL_INCLUDE_DIR})
-    include_directories(SYSTEM "${OPENGL_INCLUDE_DIR}")
-  endif ()
-  
-  target_link_libraries(${TARGET_NAME} "${OPENGL_LIBRARY}")
-  
+
   # link target to external libraries
   if (WIN32)
-    add_dependency_external_projects(glew)
-    find_package(GLEW REQUIRED)
-    target_include_directories(${TARGET_NAME} PRIVATE ${GLEW_INCLUDE_DIRS})
-
-    target_link_libraries(${TARGET_NAME} ${GLEW_LIBRARIES} wsock32.lib opengl32.lib Winmm.lib)
-
-    if (USE_NSIGHT)
-      # try to find the Nsight package and add it to the build if we find it
-      find_package(NSIGHT)
-      if (NSIGHT_FOUND)
-        include_directories(${NSIGHT_INCLUDE_DIRS})
-        add_definitions(-DNSIGHT_FOUND)
-        target_link_libraries(${TARGET_NAME} "${NSIGHT_LIBRARIES}")
-      endif ()
-    endif()
-
+  #  target_link_libraries(${TARGET_NAME} wsock32.lib Winmm.lib)
+    target_link_libraries(${TARGET_NAME} wsock32.lib Winmm.lib)
+  else (WIN32)
+    # Nothing else required on linux apparently
   endif()
 endif (APPLE)
 
diff --git a/interface/src/Application.cpp b/interface/src/Application.cpp
index c9cbfe6066..2bd5aa46e3 100644
--- a/interface/src/Application.cpp
+++ b/interface/src/Application.cpp
@@ -797,8 +797,9 @@ void Application::initializeGL() {
     }
     #endif
 
-    // Where the gpuContext is created and where the TRUE Backend is created and assigned
-    _gpuContext = std::make_shared<gpu::Context>(new gpu::GLBackend());
+    // Where the gpuContext is initialized and where the TRUE Backend is created and assigned
+    gpu::Context::init<gpu::GLBackend>();
+    _gpuContext = std::make_shared<gpu::Context>();
 
     initDisplay();
     qCDebug(interfaceapp, "Initialized Display.");
diff --git a/interface/src/Stars.cpp b/interface/src/Stars.cpp
index c1e65086bf..42b1a3f2e2 100644
--- a/interface/src/Stars.cpp
+++ b/interface/src/Stars.cpp
@@ -14,8 +14,6 @@
 #include <mutex>
 
 #include <QElapsedTimer>
-#include <gpu/GPUConfig.h>
-#include <gpu/Batch.h>
 #include <gpu/Context.h>
 #include <NumericalConstants.h>
 #include <DependencyManager.h>
@@ -208,8 +206,6 @@ void Stars::render(RenderArgs* renderArgs, float alpha) {
     batch._glUniform1f(_timeSlot, secs);
     geometryCache->renderUnitCube(batch);
 
-    glHint(GL_POINT_SMOOTH_HINT, GL_NICEST);
-    
     static const size_t VERTEX_STRIDE = sizeof(StarVertex);
     size_t offset = offsetof(StarVertex, position);
     gpu::BufferView posView(vertexBuffer, offset, vertexBuffer->getSize(), VERTEX_STRIDE, positionElement);
@@ -218,10 +214,11 @@ void Stars::render(RenderArgs* renderArgs, float alpha) {
     
     // Render the stars
     batch.setPipeline(_starsPipeline);
+
     batch.setInputFormat(streamFormat);
     batch.setInputBuffer(VERTICES_SLOT, posView);
     batch.setInputBuffer(COLOR_SLOT, colView);
     batch.draw(gpu::Primitive::POINTS, STARFIELD_NUM_STARS);
-    
+
     renderArgs->_context->render(batch);
 }
diff --git a/libraries/entities-renderer/src/RenderablePolyVoxEntityItem.cpp b/libraries/entities-renderer/src/RenderablePolyVoxEntityItem.cpp
index d3ee312311..681702bb07 100644
--- a/libraries/entities-renderer/src/RenderablePolyVoxEntityItem.cpp
+++ b/libraries/entities-renderer/src/RenderablePolyVoxEntityItem.cpp
@@ -35,7 +35,7 @@
 #include <PolyVoxCore/Material.h>
 
 #include "model/Geometry.h"
-#include "gpu/GLBackend.h"
+#include "gpu/Context.h"
 #include "EntityTreeRenderer.h"
 #include "RenderablePolyVoxEntityItem.h"
 
diff --git a/libraries/entities-renderer/src/RenderableWebEntityItem.cpp b/libraries/entities-renderer/src/RenderableWebEntityItem.cpp
index ee1fdfae45..5a08abbb22 100644
--- a/libraries/entities-renderer/src/RenderableWebEntityItem.cpp
+++ b/libraries/entities-renderer/src/RenderableWebEntityItem.cpp
@@ -8,7 +8,6 @@
 
 #include "RenderableWebEntityItem.h"
 
-#include <gpu/GPUConfig.h>
 #include <QMouseEvent>
 #include <QQuickItem>
 #include <QQuickWindow>
@@ -24,7 +23,7 @@
 #include <GLMHelpers.h>
 #include <PathUtils.h>
 #include <TextureCache.h>
-#include <gpu/GLBackend.h>
+#include <gpu/Context.h>
 
 #include "EntityTreeRenderer.h"
 
diff --git a/libraries/entities/src/ParticleEffectEntityItem.cpp b/libraries/entities/src/ParticleEffectEntityItem.cpp
index 4dfc9dd436..dc5bbb85ed 100644
--- a/libraries/entities/src/ParticleEffectEntityItem.cpp
+++ b/libraries/entities/src/ParticleEffectEntityItem.cpp
@@ -155,9 +155,9 @@ void ParticleEffectEntityItem::computeAndUpdateDimensions() {
     float yMin = std::min(yApex, yEnd);
 
     // times 2 because dimensions are diameters not radii.
-    glm::vec3 dims(2.0f * std::max(fabs(xMin), fabs(xMax)),
-                   2.0f * std::max(fabs(yMin), fabs(yMax)),
-                   2.0f * std::max(fabs(zMin), fabs(zMax)));
+    glm::vec3 dims(2.0f * std::max(fabsf(xMin), fabsf(xMax)),
+                   2.0f * std::max(fabsf(yMin), fabsf(yMax)),
+                   2.0f * std::max(fabsf(zMin), fabsf(zMax)));
 
     EntityItem::setDimensions(dims);
 }
diff --git a/libraries/gpu/CMakeLists.txt b/libraries/gpu/CMakeLists.txt
index 30949b83e1..89939535b0 100644
--- a/libraries/gpu/CMakeLists.txt
+++ b/libraries/gpu/CMakeLists.txt
@@ -21,23 +21,26 @@ elseif (WIN32)
 
   if (USE_NSIGHT)
     # try to find the Nsight package and add it to the build if we find it
+    # note that this will also enable NSIGHT profilers in all the projects linking gpu
     find_package(NSIGHT)
     if (NSIGHT_FOUND)
-      include_directories(${NSIGHT_INCLUDE_DIRS})
-      add_definitions(-DNSIGHT_FOUND)
+      target_include_directories(${TARGET_NAME} PUBLIC ${NSIGHT_INCLUDE_DIRS})
+      target_compile_definitions(${TARGET_NAME} PUBLIC NSIGHT_FOUND)
       target_link_libraries(${TARGET_NAME} "${NSIGHT_LIBRARIES}")
     endif ()
   endif()
 elseif (ANDROID)
   target_link_libraries(${TARGET_NAME} "-lGLESv3" "-lEGL")
 else ()
+  find_package(GLEW REQUIRED)
+  target_include_directories(${TARGET_NAME} PUBLIC ${GLEW_INCLUDE_DIRS})
+
   find_package(OpenGL REQUIRED)
   
   if (${OPENGL_INCLUDE_DIR})
     include_directories(SYSTEM "${OPENGL_INCLUDE_DIR}")
   endif ()
-  
-  target_link_libraries(${TARGET_NAME} "${OPENGL_LIBRARY}")
-  
-  target_include_directories(${TARGET_NAME} PUBLIC ${OPENGL_INCLUDE_DIR})
+
+  target_link_libraries(${TARGET_NAME} "${GLEW_LIBRARIES}" "${OPENGL_LIBRARY}")
+
 endif (APPLE)
diff --git a/libraries/gpu/src/gpu/Batch.cpp b/libraries/gpu/src/gpu/Batch.cpp
index 8815c1bae1..3ecbc3b2f3 100644
--- a/libraries/gpu/src/gpu/Batch.cpp
+++ b/libraries/gpu/src/gpu/Batch.cpp
@@ -8,13 +8,9 @@
 //  Distributed under the Apache License, Version 2.0.
 //  See the accompanying file LICENSE or http://www.apache.org/licenses/LICENSE-2.0.html
 //
+#include <string.h>
+
 #include "Batch.h"
-#include "GPUConfig.h"
-
-#include <QDebug>
-
-#include <GLMHelpers.h>
-
 
 #if defined(NSIGHT_FOUND)
 #include "nvToolsExt.h"
@@ -292,15 +288,3 @@ void Batch::resetStages() {
     ADD_COMMAND(resetStages);
 }
 
-void push_back(Batch::Params& params, const vec3& v) {
-    params.push_back(v.x);
-    params.push_back(v.y);
-    params.push_back(v.z);
-}
-
-void push_back(Batch::Params& params, const vec4& v) {
-    params.push_back(v.x);
-    params.push_back(v.y);
-    params.push_back(v.z);
-    params.push_back(v.a);
-}
diff --git a/libraries/gpu/src/gpu/Config.slh b/libraries/gpu/src/gpu/Config.slh
index 76be161822..f24b54e5d5 100644
--- a/libraries/gpu/src/gpu/Config.slh
+++ b/libraries/gpu/src/gpu/Config.slh
@@ -21,10 +21,9 @@
     <@def VERSION_HEADER #version 120
 #extension GL_EXT_gpu_shader4 : enable@>
 <@else@>
-    <@def GPU_FEATURE_PROFILE GPU_LEGACY@>
-    <@def GPU_TRANSFORM_PROFILE GPU_LEGACY@>
-    <@def VERSION_HEADER #version 120
-#extension GL_EXT_gpu_shader4 : enable@>
+    <@def GPU_FEATURE_PROFILE GPU_CORE@>
+    <@def GPU_TRANSFORM_PROFILE GPU_CORE@>
+    <@def VERSION_HEADER #version 430 compatibility@>
 <@endif@>
 
 <@endif@>
diff --git a/libraries/gpu/src/gpu/Context.cpp b/libraries/gpu/src/gpu/Context.cpp
index 6730be33bb..239c460c77 100644
--- a/libraries/gpu/src/gpu/Context.cpp
+++ b/libraries/gpu/src/gpu/Context.cpp
@@ -10,13 +10,16 @@
 //
 #include "Context.h"
 
-// this include should disappear! as soon as the gpu::Context is in place
-#include "GLBackend.h"
-
 using namespace gpu;
 
-Context::Context(Backend* backend) :
-    _backend(backend) {
+Context::CreateBackend Context::_createBackendCallback = nullptr;
+Context::MakeProgram Context::_makeProgramCallback = nullptr;
+std::once_flag Context::_initialized;
+
+Context::Context() {
+    if (_createBackendCallback) {
+        _backend.reset(_createBackendCallback());
+    }
 }
 
 Context::Context(const Context& context) {
@@ -26,8 +29,8 @@ Context::~Context() {
 }
 
 bool Context::makeProgram(Shader& shader, const Shader::BindingSet& bindings) {
-    if (shader.isProgram()) {
-        return GLBackend::makeProgram(shader, bindings);
+    if (shader.isProgram() && _makeProgramCallback) {
+        return _makeProgramCallback(shader, bindings);
     }
     return false;
 }
diff --git a/libraries/gpu/src/gpu/Context.h b/libraries/gpu/src/gpu/Context.h
index ab7a1d1c11..7158bd1a6d 100644
--- a/libraries/gpu/src/gpu/Context.h
+++ b/libraries/gpu/src/gpu/Context.h
@@ -12,6 +12,7 @@
 #define hifi_gpu_Context_h
 
 #include <assert.h>
+#include <mutex>
 
 #include "Batch.h"
 
@@ -26,13 +27,12 @@ namespace gpu {
 
 class Backend {
 public:
-
     virtual~ Backend() {};
+
     virtual void render(Batch& batch) = 0;
     virtual void syncCache() = 0;
     virtual void downloadFramebuffer(const FramebufferPointer& srcFramebuffer, const Vec4i& region, QImage& destImage) = 0;
 
-
     class TransformObject {
     public:
         Mat4 _model;
@@ -118,7 +118,21 @@ protected:
 
 class Context {
 public:
-    Context(Backend* backend);
+    typedef Backend* (*CreateBackend)();
+    typedef bool (*MakeProgram)(Shader& shader, const Shader::BindingSet& bindings);
+
+
+    // This one call must happen before any context is created or used (Shader::MakeProgram) in order to setup the Backend and any singleton data needed
+    template <class T>
+    static void init() {
+        std::call_once(_initialized, [] {
+            _createBackendCallback = T::createBackend;
+            _makeProgramCallback = T::makeProgram;
+            T::init();
+        });
+    }
+
+    Context();
     ~Context();
 
     void render(Batch& batch);
@@ -132,13 +146,17 @@ public:
 protected:
     Context(const Context& context);
 
+    std::unique_ptr<Backend> _backend;
+
     // This function can only be called by "static Shader::makeProgram()"
     // makeProgramShader(...) make a program shader ready to be used in a Batch.
     // It compiles the sub shaders, link them and defines the Slots and their bindings.
     // If the shader passed is not a program, nothing happens. 
-    static bool makeProgram(Shader& shader, const Shader::BindingSet& bindings = Shader::BindingSet());
+    static bool makeProgram(Shader& shader, const Shader::BindingSet& bindings);
 
-    std::unique_ptr<Backend> _backend;
+    static CreateBackend _createBackendCallback;
+    static MakeProgram _makeProgramCallback;
+    static std::once_flag _initialized;
 
     friend class Shader;
 };
diff --git a/libraries/gpu/src/gpu/GLBackend.cpp b/libraries/gpu/src/gpu/GLBackend.cpp
index 8db192b827..c74a94e03e 100644
--- a/libraries/gpu/src/gpu/GLBackend.cpp
+++ b/libraries/gpu/src/gpu/GLBackend.cpp
@@ -8,9 +8,9 @@
 //  Distributed under the Apache License, Version 2.0.
 //  See the accompanying file LICENSE or http://www.apache.org/licenses/LICENSE-2.0.html
 //
-#include <mutex>
-#include "GPULogging.h"
 #include "GLBackendShared.h"
+
+#include <mutex>
 #include <glm/gtc/type_ptr.hpp>
 
 using namespace gpu;
@@ -63,12 +63,7 @@ GLBackend::CommandCall GLBackend::_commandCalls[Batch::NUM_COMMANDS] =
     (&::gpu::GLBackend::do_glLineWidth),
 };
 
-GLBackend::GLBackend() :
-    _input(),
-    _transform(),
-    _pipeline(),
-    _output()
-{
+void GLBackend::init() {
     static std::once_flag once;
     std::call_once(once, [] {
         qCDebug(gpulogging) << "GL Version: " << QString((const char*) glGetString(GL_VERSION));
@@ -94,6 +89,13 @@ GLBackend::GLBackend() :
 #endif
 
 #if defined(Q_OS_LINUX)
+        GLenum err = glewInit();
+        if (GLEW_OK != err) {
+            /* Problem: glewInit failed, something is seriously wrong. */
+            qCDebug(gpulogging, "Error: %s\n", glewGetErrorString(err));
+        }
+        qCDebug(gpulogging, "Status: Using GLEW %s\n", glewGetString(GLEW_VERSION));
+
         // TODO: Write the correct  code for Linux...
         /* if (wglewGetExtension("WGL_EXT_swap_control")) {
             int swapInterval = wglGetSwapIntervalEXT();
@@ -101,7 +103,18 @@ GLBackend::GLBackend() :
         }*/
 #endif
     });
+}
 
+Backend* GLBackend::createBackend() {
+    return new GLBackend();
+}
+
+GLBackend::GLBackend() :
+    _input(),
+    _transform(),
+    _pipeline(),
+    _output()
+{
     initInput();
     initTransform();
 }
diff --git a/libraries/gpu/src/gpu/GLBackend.h b/libraries/gpu/src/gpu/GLBackend.h
index 215e01689b..c0bae76d78 100644
--- a/libraries/gpu/src/gpu/GLBackend.h
+++ b/libraries/gpu/src/gpu/GLBackend.h
@@ -18,16 +18,21 @@
 #include "GPUConfig.h"
 
 #include "Context.h"
-#include "Batch.h"
-
 
 namespace gpu {
 
 class GLBackend : public Backend {
-public:
+
+    // Context Backend static interface required
+    friend class Context;
+    static void init();
+    static Backend* createBackend();
+    static bool makeProgram(Shader& shader, const Shader::BindingSet& bindings);
 
     explicit GLBackend(bool syncCache);
     GLBackend();
+public:
+
     virtual ~GLBackend();
 
     virtual void render(Batch& batch);
@@ -49,7 +54,6 @@ public:
 
     static void checkGLStackStable(std::function<void()> f);
 
-    static bool makeProgram(Shader& shader, const Shader::BindingSet& bindings = Shader::BindingSet());
     
 
     class GLBuffer : public GPUObject {
@@ -91,9 +95,9 @@ public:
 
 #if (GPU_TRANSFORM_PROFILE == GPU_CORE)
 #else
-        GLuint _transformObject_model = -1;
-        GLuint _transformCamera_viewInverse = -1;
-        GLuint _transformCamera_viewport = -1;
+        GLint _transformObject_model = -1;
+        GLint _transformCamera_viewInverse = -1;
+        GLint _transformCamera_viewport = -1;
 #endif
 
         GLShader();
diff --git a/libraries/gpu/src/gpu/GLBackendQuery.cpp b/libraries/gpu/src/gpu/GLBackendQuery.cpp
index 2bdf7c86ad..5772a09943 100644
--- a/libraries/gpu/src/gpu/GLBackendQuery.cpp
+++ b/libraries/gpu/src/gpu/GLBackendQuery.cpp
@@ -8,7 +8,6 @@
 //  Distributed under the Apache License, Version 2.0.
 //  See the accompanying file LICENSE or http://www.apache.org/licenses/LICENSE-2.0.html
 //
-#include "GPULogging.h"
 #include "GLBackendShared.h"
 
 
diff --git a/libraries/gpu/src/gpu/GLBackendShader.cpp b/libraries/gpu/src/gpu/GLBackendShader.cpp
index fd5bed2e5e..9c0cf7628d 100755
--- a/libraries/gpu/src/gpu/GLBackendShader.cpp
+++ b/libraries/gpu/src/gpu/GLBackendShader.cpp
@@ -8,7 +8,6 @@
 //  Distributed under the Apache License, Version 2.0.
 //  See the accompanying file LICENSE or http://www.apache.org/licenses/LICENSE-2.0.html
 //
-#include "GPULogging.h"
 #include "GLBackendShared.h"
 #include "Format.h"
 
@@ -542,7 +541,12 @@ ElementResource getFormatFromGLUniform(GLenum gltype) {
 };
 
 
-int makeUniformSlots(GLuint glprogram, const Shader::BindingSet& slotBindings, Shader::SlotSet& uniforms, Shader::SlotSet& textures, Shader::SlotSet& samplers) {
+int makeUniformSlots(GLuint glprogram, const Shader::BindingSet& slotBindings,
+    Shader::SlotSet& uniforms, Shader::SlotSet& textures, Shader::SlotSet& samplers
+#if (GPU_FEATURE_PROFILE == GPU_LEGACY)
+    , Shader::SlotSet& fakeBuffers
+#endif
+) {
     GLint uniformsCount = 0;
 
 #if (GPU_FEATURE_PROFILE == GPU_LEGACY)
@@ -583,6 +587,15 @@ int makeUniformSlots(GLuint glprogram, const Shader::BindingSet& slotBindings, S
             }
 
             if (elementResource._resource == Resource::BUFFER) {
+#if (GPU_FEATURE_PROFILE == GPU_LEGACY)
+                // if in legacy profile, we fake the uniform buffer with an array
+                // this is where we detect it assuming it's explicitely assinged a binding
+                auto requestedBinding = slotBindings.find(std::string(sname));
+                if (requestedBinding != slotBindings.end()) {
+                    // found one buffer!
+                    fakeBuffers.insert(Shader::Slot(sname, location, elementResource._element, elementResource._resource));
+                }
+#endif
                 uniforms.insert(Shader::Slot(sname, location, elementResource._element, elementResource._resource));
             } else {
                 // For texture/Sampler, the location is the actual binding value
@@ -640,14 +653,13 @@ int makeUniformBlockSlots(GLuint glprogram, const Shader::BindingSet& slotBindin
         GLchar name[NAME_LENGTH];
         GLint length = 0;
         GLint size = 0;
-        GLenum type = 0;
         GLint binding = -1;
 
         glGetActiveUniformBlockiv(glprogram, i, GL_UNIFORM_BLOCK_NAME_LENGTH, &length);
         glGetActiveUniformBlockName(glprogram, i, NAME_LENGTH, &length, name);
         glGetActiveUniformBlockiv(glprogram, i, GL_UNIFORM_BLOCK_BINDING, &binding);
         glGetActiveUniformBlockiv(glprogram, i, GL_UNIFORM_BLOCK_DATA_SIZE, &size);
-        
+
         GLuint blockIndex = glGetUniformBlockIndex(glprogram, name);
 
         // CHeck if there is a requested binding for this block
@@ -738,8 +750,12 @@ bool GLBackend::makeProgram(Shader& shader, const Shader::BindingSet& slotBindin
         Shader::SlotSet uniforms;
         Shader::SlotSet textures;
         Shader::SlotSet samplers;
+#if (GPU_FEATURE_PROFILE == GPU_CORE)
         makeUniformSlots(object->_program, slotBindings, uniforms, textures, samplers);
-
+#else
+        makeUniformSlots(object->_program, slotBindings, uniforms, textures, samplers, buffers);
+#endif
+        
         Shader::SlotSet inputs;
         makeInputSlots(object->_program, slotBindings, inputs);
 
diff --git a/libraries/gpu/src/gpu/GLBackendShared.h b/libraries/gpu/src/gpu/GLBackendShared.h
index 27f58fcbe3..888fd1164d 100644
--- a/libraries/gpu/src/gpu/GLBackendShared.h
+++ b/libraries/gpu/src/gpu/GLBackendShared.h
@@ -11,11 +11,10 @@
 #ifndef hifi_gpu_GLBackend_Shared_h
 #define hifi_gpu_GLBackend_Shared_h
 
-#include "GLBackend.h"
-
 #include <QDebug>
 
-#include "Batch.h"
+#include "GPULogging.h"
+#include "GLBackend.h"
 
 static const GLenum _primitiveToGLmode[gpu::NUM_PRIMITIVES] = {
     GL_POINTS,
diff --git a/libraries/gpu/src/gpu/GLBackendState.cpp b/libraries/gpu/src/gpu/GLBackendState.cpp
index bd683e0136..22c61b2365 100644
--- a/libraries/gpu/src/gpu/GLBackendState.cpp
+++ b/libraries/gpu/src/gpu/GLBackendState.cpp
@@ -484,6 +484,7 @@ void GLBackend::syncPipelineStateCache() {
     glEnable(GL_TEXTURE_CUBE_MAP_SEAMLESS);
 
     // Point size is always on
+    glHint(GL_POINT_SMOOTH_HINT, GL_NICEST);
     glEnable(GL_PROGRAM_POINT_SIZE_EXT);
     glEnable(GL_VERTEX_PROGRAM_POINT_SIZE);
 
diff --git a/libraries/gpu/src/gpu/GPUConfig.h b/libraries/gpu/src/gpu/GPUConfig.h
index 9cc9002d1e..3a4dbecefb 100644
--- a/libraries/gpu/src/gpu/GPUConfig.h
+++ b/libraries/gpu/src/gpu/GPUConfig.h
@@ -37,8 +37,7 @@
 
 #else
 
-#include <GL/gl.h>
-#include <GL/glext.h>
+#include <GL/glew.h>
 
 #define GPU_FEATURE_PROFILE GPU_CORE
 #define GPU_TRANSFORM_PROFILE GPU_CORE
diff --git a/libraries/model/src/model/Material.h b/libraries/model/src/model/Material.h
index 392fd918a1..a1a17d29e9 100755
--- a/libraries/model/src/model/Material.h
+++ b/libraries/model/src/model/Material.h
@@ -13,13 +13,13 @@
 
 #include <bitset>
 #include <map>
+#include <qurl.h>
 
 #include <glm/glm.hpp>
 
 #include "gpu/Resource.h"
 #include "gpu/Texture.h"
 
-#include <qurl.h>
 
 namespace model {
 
diff --git a/libraries/render-utils/src/AmbientOcclusionEffect.cpp b/libraries/render-utils/src/AmbientOcclusionEffect.cpp
index 720f765ae4..5f0afd37d1 100644
--- a/libraries/render-utils/src/AmbientOcclusionEffect.cpp
+++ b/libraries/render-utils/src/AmbientOcclusionEffect.cpp
@@ -9,15 +9,12 @@
 //  See the accompanying file LICENSE or http://www.apache.org/licenses/LICENSE-2.0.html
 //
 
-// include this before QOpenGLFramebufferObject, which includes an earlier version of OpenGL
-#include <gpu/GPUConfig.h>
-
-#include <gpu/GLBackend.h>
 
 #include <glm/gtc/random.hpp>
 
 #include <PathUtils.h>
 #include <SharedUtil.h>
+#include <gpu/Context.h>
 
 #include "gpu/StandardShaderLib.h"
 #include "AmbientOcclusionEffect.h"
@@ -176,7 +173,6 @@ void AmbientOcclusion::run(const render::SceneContextPointer& sceneContext, cons
     assert(renderContext->args);
     assert(renderContext->args->_viewFrustum);
     RenderArgs* args = renderContext->args;
-    auto& scene = sceneContext->_scene;
 
     gpu::Batch batch;
 
diff --git a/libraries/render-utils/src/DeferredLightingEffect.cpp b/libraries/render-utils/src/DeferredLightingEffect.cpp
index c14bbfcb1d..3fc0a4c46a 100644
--- a/libraries/render-utils/src/DeferredLightingEffect.cpp
+++ b/libraries/render-utils/src/DeferredLightingEffect.cpp
@@ -12,12 +12,12 @@
 #include "DeferredLightingEffect.h"
 
 #include <GLMHelpers.h>
-#include <gpu/GPUConfig.h>
+#include <PathUtils.h>
+#include <ViewFrustum.h>
+
 #include <gpu/Batch.h>
 #include <gpu/Context.h>
 #include <gpu/StandardShaderLib.h>
-#include <PathUtils.h>
-#include <ViewFrustum.h>
 
 #include "AbstractViewStateInterface.h"
 #include "GeometryCache.h"
@@ -291,7 +291,7 @@ void DeferredLightingEffect::render(RenderArgs* args) {
                 locations = &_directionalAmbientSphereLightCascadedShadowMapLocations;
             }
             batch.setPipeline(program);
-            batch._glUniform3fv(locations->shadowDistances, 1, (const GLfloat*) &_viewState->getShadowDistances());
+            batch._glUniform3fv(locations->shadowDistances, 1, (const float*) &_viewState->getShadowDistances());
         
         } else {
             if (useSkyboxCubemap) {
@@ -325,7 +325,7 @@ void DeferredLightingEffect::render(RenderArgs* args) {
                 sh = (*_skybox->getCubemap()->getIrradiance());
             }
             for (int i =0; i <gpu::SphericalHarmonics::NUM_COEFFICIENTS; i++) {
-               batch._glUniform4fv(locations->ambientSphere + i, 1, (const GLfloat*) (&sh) + i * 4);
+               batch._glUniform4fv(locations->ambientSphere + i, 1, (const float*) (&sh) + i * 4);
             }
         }
     
@@ -340,7 +340,7 @@ void DeferredLightingEffect::render(RenderArgs* args) {
         if (_atmosphere && (locations->atmosphereBufferUnit >= 0)) {
             batch.setUniformBuffer(locations->atmosphereBufferUnit, _atmosphere->getDataBuffer());
         }
-        batch._glUniformMatrix4fv(locations->invViewMat, 1, false, reinterpret_cast< const GLfloat* >(&invViewMat));
+        batch._glUniformMatrix4fv(locations->invViewMat, 1, false, reinterpret_cast< const float* >(&invViewMat));
     }
 
     float left, right, bottom, top, nearVal, farVal;
@@ -419,9 +419,9 @@ void DeferredLightingEffect::render(RenderArgs* args) {
         batch._glUniform2f(_pointLightLocations.depthTexCoordOffset, depthTexCoordOffsetS, depthTexCoordOffsetT);
         batch._glUniform2f(_pointLightLocations.depthTexCoordScale, depthTexCoordScaleS, depthTexCoordScaleT);
         
-        batch._glUniformMatrix4fv(_pointLightLocations.invViewMat, 1, false, reinterpret_cast< const GLfloat* >(&invViewMat));
+        batch._glUniformMatrix4fv(_pointLightLocations.invViewMat, 1, false, reinterpret_cast< const float* >(&invViewMat));
 
-        batch._glUniformMatrix4fv(_pointLightLocations.texcoordMat, 1, false, reinterpret_cast< const GLfloat* >(&texcoordMat));
+        batch._glUniformMatrix4fv(_pointLightLocations.texcoordMat, 1, false, reinterpret_cast< const float* >(&texcoordMat));
 
         for (auto lightID : _pointLights) {
             auto& light = _allocatedLights[lightID];
@@ -467,9 +467,9 @@ void DeferredLightingEffect::render(RenderArgs* args) {
         batch._glUniform2f(_spotLightLocations.depthTexCoordOffset, depthTexCoordOffsetS, depthTexCoordOffsetT);
         batch._glUniform2f(_spotLightLocations.depthTexCoordScale, depthTexCoordScaleS, depthTexCoordScaleT);
         
-        batch._glUniformMatrix4fv(_spotLightLocations.invViewMat, 1, false, reinterpret_cast< const GLfloat* >(&invViewMat));
+        batch._glUniformMatrix4fv(_spotLightLocations.invViewMat, 1, false, reinterpret_cast< const float* >(&invViewMat));
 
-        batch._glUniformMatrix4fv(_spotLightLocations.texcoordMat, 1, false, reinterpret_cast< const GLfloat* >(&texcoordMat));
+        batch._glUniformMatrix4fv(_spotLightLocations.texcoordMat, 1, false, reinterpret_cast< const float* >(&texcoordMat));
 
         for (auto lightID : _spotLights) {
             auto light = _allocatedLights[lightID];
@@ -489,7 +489,7 @@ void DeferredLightingEffect::render(RenderArgs* args) {
             if ((eyeHalfPlaneDistance > -nearRadius) &&
                 (glm::distance(eyePoint, glm::vec3(light->getPosition())) < expandedRadius + nearRadius)) {
                 coneParam.w = 0.0f;
-                batch._glUniform4fv(_spotLightLocations.coneParam, 1, reinterpret_cast< const GLfloat* >(&coneParam));
+                batch._glUniform4fv(_spotLightLocations.coneParam, 1, reinterpret_cast< const float* >(&coneParam));
 
                 Transform model;
                 model.setTranslation(glm::vec3(0.0f, 0.0f, -1.0f));
@@ -509,7 +509,7 @@ void DeferredLightingEffect::render(RenderArgs* args) {
                 batch.setViewTransform(viewMat);
             } else {
                 coneParam.w = 1.0f;
-                batch._glUniform4fv(_spotLightLocations.coneParam, 1, reinterpret_cast< const GLfloat* >(&coneParam));
+                batch._glUniform4fv(_spotLightLocations.coneParam, 1, reinterpret_cast< const float* >(&coneParam));
 
                 Transform model;
                 model.setTranslation(light->getPosition());
@@ -595,9 +595,9 @@ void DeferredLightingEffect::loadLightProgram(const char* vertSource, const char
     slotBindings.insert(gpu::Shader::Binding(std::string("depthMap"), 3));
     slotBindings.insert(gpu::Shader::Binding(std::string("shadowMap"), 4));
     slotBindings.insert(gpu::Shader::Binding(std::string("skyboxMap"), 5));
-    const GLint LIGHT_GPU_SLOT = 3;
+    const int LIGHT_GPU_SLOT = 3;
     slotBindings.insert(gpu::Shader::Binding(std::string("lightBuffer"), LIGHT_GPU_SLOT));
-    const GLint ATMOSPHERE_GPU_SLOT = 4;
+    const int ATMOSPHERE_GPU_SLOT = 4;
     slotBindings.insert(gpu::Shader::Binding(std::string("atmosphereBufferUnit"), ATMOSPHERE_GPU_SLOT));
 
     gpu::Shader::makeProgram(*program, slotBindings);
@@ -614,13 +614,8 @@ void DeferredLightingEffect::loadLightProgram(const char* vertSource, const char
     locations.texcoordMat = program->getUniforms().findLocation("texcoordMat");
     locations.coneParam = program->getUniforms().findLocation("coneParam");
 
-#if (GPU_FEATURE_PROFILE == GPU_CORE)
     locations.lightBufferUnit = program->getBuffers().findLocation("lightBuffer");
     locations.atmosphereBufferUnit = program->getBuffers().findLocation("atmosphereBufferUnit");
-#else
-    locations.lightBufferUnit = program->getUniforms().findLocation("lightBuffer");
-    locations.atmosphereBufferUnit = program->getUniforms().findLocation("atmosphereBufferUnit");
-#endif
 
     auto state = std::make_shared<gpu::State>();
     if (lightVolume) {
@@ -677,10 +672,10 @@ model::MeshPointer DeferredLightingEffect::getSpotLightMesh() {
         int ringFloatOffset = slices * 3;
 
 
-        GLfloat* vertexData = new GLfloat[verticesSize];
-        GLfloat* vertexRing0 = vertexData;
-        GLfloat* vertexRing1 = vertexRing0 + ringFloatOffset;
-        GLfloat* vertexRing2 = vertexRing1 + ringFloatOffset;
+        float* vertexData = new float[verticesSize];
+        float* vertexRing0 = vertexData;
+        float* vertexRing1 = vertexRing0 + ringFloatOffset;
+        float* vertexRing2 = vertexRing1 + ringFloatOffset;
         
         for (int i = 0; i < slices; i++) {
             float theta = TWO_PI * i / slices;
@@ -746,7 +741,7 @@ model::MeshPointer DeferredLightingEffect::getSpotLightMesh() {
             *(index++) = capVertex;
         }
 
-        _spotLightMesh->setIndexBuffer(gpu::BufferView(new gpu::Buffer(sizeof(GLushort) * indices, (gpu::Byte*) indexData), gpu::Element::INDEX_UINT16));
+        _spotLightMesh->setIndexBuffer(gpu::BufferView(new gpu::Buffer(sizeof(unsigned short) * indices, (gpu::Byte*) indexData), gpu::Element::INDEX_UINT16));
         delete[] indexData;
 
         model::Mesh::Part part(0, indices, 0, model::Mesh::TRIANGLES);
diff --git a/libraries/render-utils/src/FramebufferCache.cpp b/libraries/render-utils/src/FramebufferCache.cpp
index 601d99108d..d6ebd001d2 100644
--- a/libraries/render-utils/src/FramebufferCache.cpp
+++ b/libraries/render-utils/src/FramebufferCache.cpp
@@ -18,7 +18,6 @@
 #include <QMap>
 #include <QQueue>
 #include <gpu/Batch.h>
-#include <gpu/GPUConfig.h>
 #include "RenderUtilsLogging.h"
 
 static QQueue<gpu::FramebufferPointer> _cachedFramebuffers;
diff --git a/libraries/render-utils/src/GeometryCache.cpp b/libraries/render-utils/src/GeometryCache.cpp
index 26b9a933c4..def5f38db4 100644
--- a/libraries/render-utils/src/GeometryCache.cpp
+++ b/libraries/render-utils/src/GeometryCache.cpp
@@ -9,22 +9,22 @@
 //  See the accompanying file LICENSE or http://www.apache.org/licenses/LICENSE-2.0.html
 //
 
-// include this before QOpenGLBuffer, which includes an earlier version of OpenGL
+#include "GeometryCache.h"
+
 #include <cmath>
 
 #include <QNetworkReply>
 #include <QRunnable>
 #include <QThreadPool>
 
-#include <gpu/Batch.h>
-#include <gpu/GLBackend.h>
-
 #include <FSTReader.h>
 #include <NumericalConstants.h>
 
+#include <gpu/Batch.h>
+#include <gpu/GLBackend.h>
+
 #include "TextureCache.h"
 #include "RenderUtilsLogging.h"
-#include "GeometryCache.h"
 
 #include "standardTransformPNTC_vert.h"
 #include "standardDrawTexture_frag.h"
diff --git a/libraries/render-utils/src/Model.cpp b/libraries/render-utils/src/Model.cpp
index 67eb85edfc..1e4f3f7190 100644
--- a/libraries/render-utils/src/Model.cpp
+++ b/libraries/render-utils/src/Model.cpp
@@ -18,20 +18,18 @@
 
 #include <CapsuleShape.h>
 #include <GeometryUtil.h>
-#include <gpu/Batch.h>
-#include <gpu/GLBackend.h>
 #include <PathUtils.h>
 #include <PerfStat.h>
-#include "PhysicsEntity.h"
 #include <ShapeCollider.h>
 #include <SphereShape.h>
 #include <ViewFrustum.h>
+#include <render/Scene.h>
+#include <gpu/Batch.h>
 
 #include "AbstractViewStateInterface.h"
 #include "AnimationHandle.h"
 #include "DeferredLightingEffect.h"
 #include "Model.h"
-#include "RenderUtilsLogging.h"
 
 #include "model_vert.h"
 #include "model_shadow_vert.h"
@@ -96,7 +94,7 @@ Model::~Model() {
 }
 
 Model::RenderPipelineLib Model::_renderPipelineLib;
-const GLint MATERIAL_GPU_SLOT = 3;
+const int MATERIAL_GPU_SLOT = 3;
 
 void Model::RenderPipelineLib::addRenderPipeline(Model::RenderKey key,
                                                  gpu::ShaderPointer& vertexShader,
@@ -189,13 +187,9 @@ void Model::RenderPipelineLib::initLocations(gpu::ShaderPointer& program, Model:
     locations.specularTextureUnit = program->getTextures().findLocation("specularMap");
     locations.emissiveTextureUnit = program->getTextures().findLocation("emissiveMap");
 
-#if (GPU_FEATURE_PROFILE == GPU_CORE)
     locations.materialBufferUnit = program->getBuffers().findLocation("materialBuffer");
     locations.lightBufferUnit = program->getBuffers().findLocation("lightBuffer");
-#else
-    locations.materialBufferUnit = program->getUniforms().findLocation("materialBuffer");
-    locations.lightBufferUnit = program->getUniforms().findLocation("lightBuffer");
-#endif
+
     locations.clusterMatrices = program->getUniforms().findLocation("clusterMatrices");
 
     locations.clusterIndices = program->getInputs().findLocation("clusterIndices");;
diff --git a/libraries/render-utils/src/RenderDeferredTask.cpp b/libraries/render-utils/src/RenderDeferredTask.cpp
index f4b6d9b0a1..07a8b63e6f 100755
--- a/libraries/render-utils/src/RenderDeferredTask.cpp
+++ b/libraries/render-utils/src/RenderDeferredTask.cpp
@@ -11,12 +11,10 @@
 //
 #include "RenderDeferredTask.h"
 
-#include <gpu/GPUConfig.h>
-#include <gpu/Batch.h>
-#include <gpu/Context.h>
 #include <PerfStat.h>
 #include <RenderArgs.h>
 #include <ViewFrustum.h>
+#include <gpu/Context.h>
 
 #include "FramebufferCache.h"
 #include "DeferredLightingEffect.h"
diff --git a/libraries/render-utils/src/TextureCache.cpp b/libraries/render-utils/src/TextureCache.cpp
index 1a6ea97b64..d6a9bf5b36 100644
--- a/libraries/render-utils/src/TextureCache.cpp
+++ b/libraries/render-utils/src/TextureCache.cpp
@@ -16,16 +16,15 @@
 #include <glm/glm.hpp>
 #include <glm/gtc/random.hpp>
 
-#include <gpu/Batch.h>
-#include <gpu/GLBackend.h>
-#include <gpu/GPUConfig.h>
-
 #include <QNetworkReply>
 #include <QPainter>
 #include <QRunnable>
 #include <QThreadPool>
 #include <qimagereader.h>
 
+#include <gpu/Batch.h>
+
+
 
 #include "RenderUtilsLogging.h"
 
diff --git a/libraries/render/src/render/DrawStatus.cpp b/libraries/render/src/render/DrawStatus.cpp
index f84d212112..cf3616a83a 100644
--- a/libraries/render/src/render/DrawStatus.cpp
+++ b/libraries/render/src/render/DrawStatus.cpp
@@ -15,15 +15,11 @@
 #include <assert.h>
 
 #include <PerfStat.h>
-
-#include <gpu/Batch.h>
-#include <gpu/Context.h>
-#include <gpu/GPUConfig.h>
-#include <gpu/GPULogging.h>
-
 #include <ViewFrustum.h>
 #include <RenderArgs.h>
 
+#include <gpu/Context.h>
+
 #include "drawItemBounds_vert.h"
 #include "drawItemBounds_frag.h"
 #include "drawItemStatus_vert.h"
@@ -152,17 +148,17 @@ void DrawStatus::run(const SceneContextPointer& sceneContext, const RenderContex
     const unsigned int VEC3_ADRESS_OFFSET = 3;
 
     for (int i = 0; i < nbItems; i++) {
-        batch._glUniform3fv(_drawItemBoundPosLoc, 1, (const GLfloat*) (itemAABox + i));
-        batch._glUniform3fv(_drawItemBoundDimLoc, 1, ((const GLfloat*) (itemAABox + i)) + VEC3_ADRESS_OFFSET);
+        batch._glUniform3fv(_drawItemBoundPosLoc, 1, (const float*) (itemAABox + i));
+        batch._glUniform3fv(_drawItemBoundDimLoc, 1, ((const float*) (itemAABox + i)) + VEC3_ADRESS_OFFSET);
 
         batch.draw(gpu::LINES, 24, 0);
     }
 
     batch.setPipeline(getDrawItemStatusPipeline());
     for (int i = 0; i < nbItems; i++) {
-        batch._glUniform3fv(_drawItemStatusPosLoc, 1, (const GLfloat*) (itemAABox + i));
-        batch._glUniform3fv(_drawItemStatusDimLoc, 1, ((const GLfloat*) (itemAABox + i)) + VEC3_ADRESS_OFFSET);
-        batch._glUniform4iv(_drawItemStatusValueLoc, 1, (const GLint*) (itemStatus + i));
+        batch._glUniform3fv(_drawItemStatusPosLoc, 1, (const float*) (itemAABox + i));
+        batch._glUniform3fv(_drawItemStatusDimLoc, 1, ((const float*) (itemAABox + i)) + VEC3_ADRESS_OFFSET);
+        batch._glUniform4iv(_drawItemStatusValueLoc, 1, (const int*) (itemStatus + i));
 
         batch.draw(gpu::TRIANGLES, 24, 0);
     }
@@ -171,4 +167,4 @@ void DrawStatus::run(const SceneContextPointer& sceneContext, const RenderContex
     args->_context->syncCache();
     renderContext->args->_context->syncCache();
     args->_context->render((batch));
-}
\ No newline at end of file
+}
diff --git a/libraries/render/src/render/DrawTask.cpp b/libraries/render/src/render/DrawTask.cpp
index 0aef913d50..36ff302952 100755
--- a/libraries/render/src/render/DrawTask.cpp
+++ b/libraries/render/src/render/DrawTask.cpp
@@ -14,12 +14,10 @@
 #include <algorithm>
 #include <assert.h>
 
-#include <gpu/Batch.h>
-#include <gpu/Context.h>
-#include <gpu/GPUConfig.h>
 #include <PerfStat.h>
 #include <RenderArgs.h>
 #include <ViewFrustum.h>
+#include <gpu/Context.h>
 
 using namespace render;
 
diff --git a/tests/render-utils/src/main.cpp b/tests/render-utils/src/main.cpp
index 3b7eb18368..9abd533650 100644
--- a/tests/render-utils/src/main.cpp
+++ b/tests/render-utils/src/main.cpp
@@ -8,30 +8,35 @@
 //  See the accompanying file LICENSE or http://www.apache.org/licenses/LICENSE-2.0.html
 //
 
-#include <gpu/GPUConfig.h>
-
-#include <QWindow>
-#include <QFile>
-#include <QTime>
-#include <QImage>
-#include <QTimer>
-#include <QElapsedTimer>
-#include <QOpenGLContext>
-#include <QOpenGLBuffer>
-#include <QOpenGLShaderProgram>
-#include <QResizeEvent>
-#include <QLoggingCategory>
-#include <QOpenGLTexture>
-#include <QOpenGLVertexArrayObject>
-#include <QApplication>
-#include <QOpenGLDebugLogger>
-
 #include <unordered_map>
 #include <memory>
+
 #include <glm/glm.hpp>
 
-#include <PathUtils.h>
+#include <QApplication>
 #include <QDir>
+#include <QElapsedTimer>
+#include <QFile>
+#include <QImage>
+#include <QLoggingCategory>
+
+#include <gpu/Context.h>
+#include <gpu/GLBackend.h>
+
+#include <QOpenGLBuffer>
+#include <QOpenGLContext>
+#include <QOpenGLDebugLogger>
+#include <QOpenGLShaderProgram>
+#include <QOpenGLTexture>
+#include <QOpenGLVertexArrayObject>
+#include <QResizeEvent>
+#include <QTime>
+#include <QTimer>
+#include <QWindow>
+
+
+
+#include <PathUtils.h>
 
 class RateCounter {
     std::vector<float> times;
@@ -119,6 +124,10 @@ public:
         show();
         makeCurrent();
 
+        gpu::Context::init<gpu::GLBackend>();
+
+
+
         {
             QOpenGLDebugLogger* logger = new QOpenGLDebugLogger(this);
             logger->initialize(); // initializes in the current context, i.e. ctx
@@ -130,23 +139,6 @@ public:
         }
         qDebug() << (const char*)glGetString(GL_VERSION);
 
-#ifdef WIN32
-        glewExperimental = true;
-        GLenum err = glewInit();
-        if (GLEW_OK != err) {
-            /* Problem: glewInit failed, something is seriously wrong. */
-            const GLubyte * errStr = glewGetErrorString(err);
-            qDebug("Error: %s\n", errStr);
-        }
-        qDebug("Status: Using GLEW %s\n", glewGetString(GLEW_VERSION));
-
-        if (wglewGetExtension("WGL_EXT_swap_control")) {
-            int swapInterval = wglGetSwapIntervalEXT();
-            qDebug("V-Sync is %s\n", (swapInterval > 0 ? "ON" : "OFF"));
-        }
-        glGetError();
-#endif
-
         //_textRenderer[0] = TextRenderer::getInstance(SANS_FONT_FAMILY, 12, false);
         //_textRenderer[1] = TextRenderer::getInstance(SERIF_FONT_FAMILY, 12, false,
         //    TextRenderer::SHADOW_EFFECT);