diff --git a/.gitignore b/.gitignore
index 7fbfa51..11eb870 100755
--- a/.gitignore
+++ b/.gitignore
@@ -1,2 +1,4 @@
 /cmake-build-*/
-tests/cmake-build-*/
\ No newline at end of file
+tests/cmake-build-*/
+.vs/
+out/
\ No newline at end of file
diff --git a/CMakeLists.txt b/CMakeLists.txt
index c862ed4..6532c31 100755
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -117,8 +117,12 @@ if(${BUILD_TESTS})
 
     target_link_libraries(BLT_TESTS BLT)
 
-    target_compile_options(BLT_TESTS PRIVATE -Wall -Werror -Wpedantic -Wno-comment)
-    target_link_options(BLT_TESTS PRIVATE -Wall -Werror -Wpedantic -Wno-comment)
+    if(MSVC)
+
+    else()
+        target_compile_options(BLT_TESTS PRIVATE -Wall -Werror -Wpedantic -Wno-comment)
+        target_link_options(BLT_TESTS PRIVATE -Wall -Werror -Wpedantic -Wno-comment)
+    endif()
 
     if (${ENABLE_ADDRSAN} MATCHES ON)
         target_compile_options(BLT_TESTS PRIVATE -fsanitize=address)
diff --git a/include/blt/compatibility.h b/include/blt/compatibility.h
index 359cd6b..fe5a97b 100644
--- a/include/blt/compatibility.h
+++ b/include/blt/compatibility.h
@@ -30,6 +30,8 @@
     #undef BLT_USE_CPP20
 #endif
 
+#define BLT_CONTAINS_IF(container, value) std::find_if(container.begin(), container.end(), value) != container.end()
+
 #define INCLUDE_FS \
     #if defined(CXX17_FILESYSTEM) || defined (CXX17_FILESYSTEM_LIBFS) \
         #include <filesystem>\
diff --git a/include/blt/math/math.h b/include/blt/math/math.h
index 7c2ff50..9e0b78d 100755
--- a/include/blt/math/math.h
+++ b/include/blt/math/math.h
@@ -10,22 +10,35 @@
 #include <blt/math/vectors.h>
 #include <blt/math/matrix.h>
 
-namespace blt {
+namespace blt
+{
+    
+    static inline constexpr double PI = 3.141592653589793238462643383279502884197;
+    
+    template<typename T>
+    static inline T toRadians(T deg)
+    {
+        constexpr double CONV = PI / 180.0;
+        return deg * CONV;
+    }
     
     /**
      * fast number integer
      */
-    static inline unsigned int f_randi(unsigned int seed) {
+    static inline unsigned int f_randi(unsigned int seed)
+    {
         seed = (seed << 13) ^ seed;
         return ((seed * (seed * seed * 15731 + 789221) + 1376312589) & 0x7fffffff);
     }
 
 #pragma GCC diagnostic push
 #pragma GCC diagnostic ignored "-Wstrict-aliasing"
+    
     /**
      * fast inverse sqrt
      */
-    static inline float fsqrt(float n){
+    static inline float fsqrt(float n)
+    {
         int i;
         float x, y;
         x = n * 0.5f;
@@ -41,7 +54,8 @@ namespace blt {
 #pragma GCC diagnostic pop
     
     
-    static inline constexpr double pow(int b, int p) {
+    static inline constexpr double pow(int b, int p)
+    {
         int collection = 1;
         for (int i = 0; i < p; i++)
             collection *= b;
@@ -55,9 +69,10 @@ namespace blt {
      * @return
      */
     template<int decimal_places>
-    static inline double round_up(double value) {
+    static inline double round_up(double value)
+    {
         constexpr double multiplier = pow(10, decimal_places);
-        return ((int)(value * multiplier) + 1) / multiplier;
+        return ((int) (value * multiplier) + 1) / multiplier;
     }
     
     /*inline std::ostream& operator<<(std::ostream& out, const mat4x4& v) {
@@ -66,7 +81,7 @@ namespace blt {
  << "         {" << v.m20() << ", " << v.m21() << ", " << v.m22() << ", " << v.m23() << "} \n"\
  << "         {" << v.m30() << ", " << v.m31() << ", " << v.m32() << ", " << v.m33() << "} \n";
     }*/
-
+    
 }
 
 #endif //BLT_MATH_H
diff --git a/include/blt/math/vectors.h b/include/blt/math/vectors.h
index 9646086..15232f1 100755
--- a/include/blt/math/vectors.h
+++ b/include/blt/math/vectors.h
@@ -267,6 +267,11 @@ namespace blt {
         return true;
     }
     
+    template<typename T, uint32_t size>
+    inline constexpr bool operator!=(const vec<T, size>& left, const vec<T, size>& right) {
+        return !(left == right);
+    }
+    
     template<typename T, uint32_t size>
     inline constexpr bool operator&&(const vec<T, size>& left, const vec<T, size>& right) {
         for (uint32_t i = 0; i < size; i++)
diff --git a/include/blt/std/memory_util.h b/include/blt/std/memory_util.h
index b7e2e8b..7802e77 100644
--- a/include/blt/std/memory_util.h
+++ b/include/blt/std/memory_util.h
@@ -23,6 +23,7 @@
 #include <array>
 #include <cstring>
 #include <algorithm>
+#include <cstdint>
 
 #if defined(__clang__) || defined(__llvm__) || defined(__GNUC__) || defined(__GNUG__)
     
@@ -32,9 +33,9 @@
     #define SWAP32(val) bswap_32(val)
     #define SWAP64(val) bswap_64(val)
     #if __cplusplus >= 202002L
-    
+        
         #include <bit>
-    
+
         #define ENDIAN_LOOKUP(little_endian) (std::endian::native == std::endian::little && !little_endian) || \
                                              (std::endian::native == std::endian::big && little_endian)
     #else
@@ -109,6 +110,15 @@ namespace blt::mem
             return prev_size * 2;
         return prev_size + default_allocation_block;
     }
+    
+    template<typename R, typename T>
+    inline static R type_cast(T type)
+    {
+        static_assert(sizeof(T) == sizeof(R));
+        R r;
+        std::memcpy(&r, &type, sizeof(type));
+        return r;
+    }
 }
 
 #endif //BLT_MEMORY_UTIL_H
diff --git a/include/blt/std/thread.h b/include/blt/std/thread.h
new file mode 100644
index 0000000..73bd727
--- /dev/null
+++ b/include/blt/std/thread.h
@@ -0,0 +1,132 @@
+/*
+ *  <Short Description>
+ *  Copyright (C) 2023  Brett Terpstra
+ *
+ *  This program is free software: you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation, either version 3 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program.  If not, see <https://www.gnu.org/licenses/>.
+ */
+
+#ifndef BLT_THREAD_H
+#define BLT_THREAD_H
+
+#include <thread>
+#include <functional>
+#include <string>
+#include <vector>
+#include <queue>
+#include <variant>
+#include <atomic>
+#include <mutex>
+#include <chrono>
+#include <optional>
+#include <blt/std/logging.h>
+
+namespace blt
+{
+    /**
+     * @tparam queue should we use a queue or execute the same function over and over?
+     */
+    template<bool queue = false>
+    class thread_pool
+    {
+        private:
+            typedef std::function<void()> thread_function;
+            volatile std::atomic_bool should_stop = false;
+            volatile std::atomic_uint64_t stopped = 0;
+            std::uint64_t number_of_threads = 0;
+            std::vector<std::thread*> threads;
+            std::variant<std::queue<thread_function>, thread_function> func_queue;
+            std::mutex queue_mutex;
+        public:
+            explicit thread_pool(std::uint64_t number_of_threads = 8, std::optional<thread_function> default_function = {})
+            {
+                if (default_function.has_value())
+                    func_queue = default_function.value();
+                this->number_of_threads = number_of_threads;
+                for (std::uint64_t i = 0; i < number_of_threads; i++)
+                {
+                    threads.push_back(new std::thread([this]() {
+                        while (!should_stop)
+                        {
+                            if constexpr (queue)
+                            {
+                                // should be safe right?
+                                std::unique_lock lock(queue_mutex);
+                                lock.lock();
+                                auto& func_q = std::get<std::queue<thread_function>>(func_queue);
+                                if (func_q.empty())
+                                {
+                                    lock.unlock();
+                                    std::this_thread::sleep_for(std::chrono::milliseconds(16));
+                                    continue;
+                                }
+                                auto func = func_q.front();
+                                func_q.pop();
+                                lock.unlock();
+                                func();
+                            } else
+                            {
+                                {
+                                    std::scoped_lock lock(queue_mutex);
+                                    if (std::holds_alternative<std::queue<thread_function>>(func_queue))
+                                    {
+                                        std::this_thread::sleep_for(std::chrono::milliseconds(16));
+                                        BLT_WARN("Running non queue variant with a queue inside!");
+                                        break;
+                                    }
+                                }
+                                auto& func = std::get<thread_function>(func_queue);
+                                func();
+                            }
+                        }
+                        stopped++;
+                    }));
+                }
+            }
+            
+            inline void execute(const thread_function& func)
+            {
+                std::scoped_lock lock(queue_mutex);
+                if constexpr (queue)
+                {
+                    auto& v = std::get<std::queue<thread_function>>(func_queue);
+                    v.push(func);
+                } else
+                {
+                    func_queue = func;
+                }
+            }
+            
+            [[nodiscard]] inline bool complete() const {
+                return stopped == number_of_threads;
+            }
+            
+            inline void stop()
+            {
+                should_stop = true;
+            }
+            
+            ~thread_pool()
+            {
+                should_stop = true;
+                for (auto* t : threads)
+                {
+                    if (t->joinable())
+                        t->join();
+                    delete t;
+                }
+            }
+    };
+}
+
+#endif //BLT_THREAD_H
diff --git a/include/blt/std/time.h b/include/blt/std/time.h
index 8a43163..daec649 100755
--- a/include/blt/std/time.h
+++ b/include/blt/std/time.h
@@ -73,7 +73,7 @@ namespace blt::system
         clock_getres(CLOCK_PROCESS_CPUTIME_ID, &res);
         return res.tv_sec * static_cast<int64_t>(1e9) + res.tv_nsec;
 #else
-        return CLOCKS_PER_SECOND;
+        return 1;
 #endif
     }
     
diff --git a/include/blt/std/utility.h b/include/blt/std/utility.h
index bbf8679..b007bb0 100644
--- a/include/blt/std/utility.h
+++ b/include/blt/std/utility.h
@@ -20,11 +20,11 @@
 #define BLT_UTILITY_H
 
 #include <string>
+#include <blt/compatibility.h>
 
 #if defined(__GNUC__)
     
     #include <cxxabi.h>
-    #include <blt/compatibility.h>
 
 namespace blt
 {
@@ -238,7 +238,7 @@ namespace blt
 #endif
     
     template<typename T>
-    void BLT_ATTRIB_NO_INLINE black_box(const T& val)
+    BLT_ATTRIB_NO_INLINE void black_box(const T& val)
     {
         static volatile void* hell;
         hell = (void*) &val;
@@ -246,7 +246,7 @@ namespace blt
     }
     
     template<typename T>
-    const T& BLT_ATTRIB_NO_INLINE black_box_ret(const T& val)
+    BLT_ATTRIB_NO_INLINE const T& black_box_ret(const T& val)
     {
         static volatile void* hell;
         hell = (void*) &val;
diff --git a/libraries/parallel-hashmap b/libraries/parallel-hashmap
index 401552d..946ebad 160000
--- a/libraries/parallel-hashmap
+++ b/libraries/parallel-hashmap
@@ -1 +1 @@
-Subproject commit 401552da80b0971f818e648621260720ad40934e
+Subproject commit 946ebad67a21212d11a0dd4deb7cdedc297d47bc