diff --git a/docs/Development.md b/docs/Development.md
index e00f88bddc..33c488d4ed 100644
--- a/docs/Development.md
+++ b/docs/Development.md
@@ -57,6 +57,13 @@ Some examples:
 Before adding a new setting, consider:
 - Does the piece of code that the setting pertains to, make a significant difference if it's on/off?
 - Can it be auto-detected?
+- Does the option behave differently on some platforms?
+- Is it a feature specific to a given platform?"
+
+While not a silver bullet, settings are mainly suitable for:
+- Core: CPU, memory, input, scheduler, audio, or other base emulator logic.
+- UI: Anything involving menus, dialogs shows/removal, bind settings.
+- Graphics: Rendering backends (OpenGL, Vulkan) Visual fixes that only work on some games and may break others.
 
 # IDE setup
 
diff --git a/src/qt_common/shared_translation.cpp b/src/qt_common/shared_translation.cpp
index 55a493d4fc..7d0a3cc170 100644
--- a/src/qt_common/shared_translation.cpp
+++ b/src/qt_common/shared_translation.cpp
@@ -113,40 +113,35 @@ std::unique_ptr<TranslationMap> InitializeTranslations(QObject* parent)
 
     // Cpu Unsafe
     INSERT(Settings, cpuopt_unsafe_host_mmu, tr("Enable Host MMU Emulation (fastmem)"),
-           tr("This optimization speeds up memory accesses by the guest program.\nEnabling it causes guest memory reads/writes to be done directly into memory and make use of Host's MMU.\nDisabling this forces all memory accesses to use Software MMU Emulation."));
+           tr("Enabling causes guest memory operations to be done directly into host memory with an on-demand paging mechanism.\nDisabling this forces all memory accesses to use a slower translation dispatch."));
     INSERT(
         Settings,
         cpuopt_unsafe_unfuse_fma,
-        tr("Unfuse FMA (improve performance on CPUs without FMA)"),
-        tr("This option improves speed by reducing accuracy of fused-multiply-add instructions on "
-           "CPUs without native FMA support."));
+        tr("Unfuse FMA"),
+        tr("Prevents materialisation of precise fused-multiply-add instructions."));
     INSERT(
         Settings,
         cpuopt_unsafe_reduce_fp_error,
         tr("Faster FRSQRTE and FRECPE"),
-        tr("This option improves the speed of some approximate floating-point functions by using "
-           "less accurate native approximations."));
+        tr("Uses (innacurate but fast) native instructions as opposed to checked ones."));
     INSERT(Settings,
            cpuopt_unsafe_ignore_standard_fpcr,
            tr("Faster ASIMD instructions (32 bits only)"),
-           tr("This option improves the speed of 32 bits ASIMD floating-point functions by running "
-              "with incorrect rounding modes."));
+           tr("Removes checks for FPCR rounding mode on floating-point instructions."));
     INSERT(Settings,
            cpuopt_unsafe_inaccurate_nan,
            tr("Inaccurate NaN handling"),
-           tr("This option improves speed by removing NaN checking.\nPlease note this also reduces "
-              "accuracy of certain floating-point instructions."));
+           tr("Removes checks for NaNs on all floating-point instructions."));
     INSERT(Settings,
            cpuopt_unsafe_fastmem_check,
            tr("Disable address space checks"),
-           tr("This option improves speed by eliminating a safety check before every memory "
-              "read/write in guest.\nDisabling it may allow a game to read/write the emulator's "
-              "memory."));
+           tr("Removes the safety check before every memory operation in the guest.\n"
+            "Disabling will allow the guest to read/write all memory."));
     INSERT(
         Settings,
         cpuopt_unsafe_ignore_global_monitor,
         tr("Ignore global monitor"),
-        tr("This option improves speed by relying only on the semantics of cmpxchg to ensure "
+        tr("This option improves speed by relying on the semantics of CMPXCHG to ensure "
            "safety of exclusive access instructions.\nPlease note this may result in deadlocks and "
            "other race conditions."));
 
@@ -168,7 +163,7 @@ std::unique_ptr<TranslationMap> InitializeTranslations(QObject* parent)
            resolution_setup,
            tr("Resolution:"),
            tr("Forces the guest to render at a different resolution.\n"
-              "Higher resolutions require more VRAM and bandwidth.\n"
+              "Higher resolutions require more VRAM.\n"
               "Options lower than 1X can cause artifacts."));
     INSERT(Settings, scaling_filter, tr("Window Adapting Filter:"), QString());
     INSERT(Settings,
@@ -183,10 +178,8 @@ std::unique_ptr<TranslationMap> InitializeTranslations(QObject* parent)
     INSERT(Settings,
            fullscreen_mode,
            tr("Fullscreen Mode:"),
-           tr("The method used to render the window in fullscreen.\nBorderless offers the best "
-              "compatibility with the on-screen keyboard that some games request for "
-              "input.\nExclusive "
-              "fullscreen may offer better performance and better Freesync/Gsync support."));
+           tr("Borderless offers the best compatibility with the on-screen keyboard.\n"
+              "Exclusive fullscreen offers better Freesync/Gsync support."));
     INSERT(Settings,
            aspect_ratio,
            tr("Aspect Ratio:"),
@@ -196,14 +189,11 @@ std::unique_ptr<TranslationMap> InitializeTranslations(QObject* parent)
     INSERT(Settings,
            use_disk_shader_cache,
            tr("Use persistent pipeline cache"),
-           tr("Allows saving shaders to storage for faster loading on following game "
-              "boots.\nDisabling it is only intended for debugging."));
+           tr("Allows saving shaders for faster boot times - but will infinitely load if the cache is corrupt."));
     INSERT(Settings,
            optimize_spirv_output,
            tr("Optimize SPIRV shader"),
-           tr("Runs an additional optimization pass over generated SPIRV shaders.\n"
-              "Will increase time required for shader compilation.\nMay slightly improve "
-              "performance.\nThis feature is experimental."));
+           tr("Runs an optimisation pass on all compiled SPIRV shaders.\n"));
     INSERT(
         Settings,
         use_asynchronous_gpu_emulation,