diff --git a/renderdoc/driver/d3d11/d3d11_shaderdebug.cpp b/renderdoc/driver/d3d11/d3d11_shaderdebug.cpp
index d80981a84..dbd63ed84 100644
--- a/renderdoc/driver/d3d11/d3d11_shaderdebug.cpp
+++ b/renderdoc/driver/d3d11/d3d11_shaderdebug.cpp
@@ -41,8 +41,6 @@
 
 #include "data/hlsl/hlsl_cbuffers.h"
 
-RDOC_EXTERN_CONFIG(bool, D3D_Hack_EnableGroups);
-
 class D3D11DebugAPIWrapper : public DXBCDebug::DebugAPIWrapper
 {
 public:
@@ -2318,9 +2316,8 @@ ShaderDebugTrace *D3D11Replay::DebugThread(uint32_t eventId,
   uint32_t activeIndex = 0;
   if(dxbc->GetThreadScope() == DXBC::ThreadScope::Workgroup)
   {
-    if(D3D_Hack_EnableGroups())
-      activeIndex = threadid[0] + threadid[1] * refl.dispatchThreadsDimension[0] +
-                    threadid[2] * refl.dispatchThreadsDimension[0] * refl.dispatchThreadsDimension[1];
+    activeIndex = threadid[0] + threadid[1] * refl.dispatchThreadsDimension[0] +
+                  threadid[2] * refl.dispatchThreadsDimension[0] * refl.dispatchThreadsDimension[1];
   }
 
   DXBCDebug::InterpretDebugger *interpreter = new DXBCDebug::InterpretDebugger;
diff --git a/renderdoc/driver/d3d12/d3d12_shaderdebug.cpp b/renderdoc/driver/d3d12/d3d12_shaderdebug.cpp
index a4ee73952..f2ebb436a 100644
--- a/renderdoc/driver/d3d12/d3d12_shaderdebug.cpp
+++ b/renderdoc/driver/d3d12/d3d12_shaderdebug.cpp
@@ -40,8 +40,6 @@
 
 #include "data/hlsl/hlsl_cbuffers.h"
 
-RDOC_EXTERN_CONFIG(bool, D3D_Hack_EnableGroups);
-
 using namespace DXBCBytecode;
 
 static bool IsShaderParameterVisible(DXBC::ShaderType shaderType,
@@ -1922,7 +1920,7 @@ ShaderDebugTrace *D3D12Replay::DebugVertex(uint32_t eventId, uint32_t vertid, ui
   bytebuf staticData[D3D12_IA_VERTEX_INPUT_RESOURCE_SLOT_COUNT];
 
   // if we're fetching from the GPU anyway, don't grab any buffer data
-  if(D3D_Hack_EnableGroups() && (dxbc->GetThreadScope() & DXBC::ThreadScope::Subgroup))
+  if(dxbc->GetThreadScope() & DXBC::ThreadScope::Subgroup)
     vertexbuffers.clear();
 
   for(auto it = vertexbuffers.begin(); it != vertexbuffers.end(); ++it)
@@ -2211,7 +2209,7 @@ ShaderDebugTrace *D3D12Replay::DebugVertex(uint32_t eventId, uint32_t vertid, ui
     ret->constantBlocks = global.constantBlocks;
     ret->inputs = state.inputs;
   }
-  else if(D3D_Hack_EnableGroups() && (dxbc->GetThreadScope() & DXBC::ThreadScope::Subgroup))
+  else if(dxbc->GetThreadScope() & DXBC::ThreadScope::Subgroup)
   {
     DXDebug::InputFetcherConfig cfg;
     DXDebug::InputFetcher fetcher;
@@ -2830,7 +2828,7 @@ ShaderDebugTrace *D3D12Replay::DebugPixel(uint32_t eventId, uint32_t x, uint32_t
   cfg.maxWaveSize = 4;
   cfg.outputSampleCount = RDCMAX(1U, pipeDesc.SampleDesc.Count);
 
-  if(D3D_Hack_EnableGroups() && (dxbc->GetThreadScope() & DXBC::ThreadScope::Subgroup))
+  if(dxbc->GetThreadScope() & DXBC::ThreadScope::Subgroup)
     cfg.maxWaveSize = m_pDevice->GetOpts1().WaveLaneCountMax;
 
   DXDebug::CreateInputFetcher(dxbc, prevDxbc, cfg, fetcher);
@@ -3358,10 +3356,8 @@ ShaderDebugTrace *D3D12Replay::DebugThread(uint32_t eventId,
     uint32_t activeIndex = 0;
     if(dxbc->GetThreadScope() & DXBC::ThreadScope::Workgroup)
     {
-      if(D3D_Hack_EnableGroups())
-        activeIndex =
-            threadid[0] + threadid[1] * refl.dispatchThreadsDimension[0] +
-            threadid[2] * refl.dispatchThreadsDimension[0] * refl.dispatchThreadsDimension[1];
+      activeIndex = threadid[0] + threadid[1] * refl.dispatchThreadsDimension[0] +
+                    threadid[2] * refl.dispatchThreadsDimension[0] * refl.dispatchThreadsDimension[1];
     }
 
     DXBCDebug::InterpretDebugger *interpreter = new DXBCDebug::InterpretDebugger;
@@ -3455,7 +3451,7 @@ ShaderDebugTrace *D3D12Replay::DebugThread(uint32_t eventId,
     rdcarray<DXILDebug::ThreadProperties> workgroupProperties;
 
     // hard case - with subgroups we want the actual layout so read that from the GPU
-    if(D3D_Hack_EnableGroups() && (dxbc->GetThreadScope() & DXBC::ThreadScope::Subgroup))
+    if(dxbc->GetThreadScope() & DXBC::ThreadScope::Subgroup)
     {
       DXDebug::InputFetcherConfig cfg;
       DXDebug::InputFetcher fetcher;
@@ -3705,7 +3701,7 @@ ShaderDebugTrace *D3D12Replay::DebugThread(uint32_t eventId,
         }
       }
     }
-    else if(D3D_Hack_EnableGroups() && (dxbc->GetThreadScope() & DXBC::ThreadScope::Workgroup))
+    else if(dxbc->GetThreadScope() & DXBC::ThreadScope::Workgroup)
     {
       numThreads = threadDim[0] * threadDim[1] * threadDim[2];
 
diff --git a/renderdoc/driver/shaders/dxbc/dxbc_debug.cpp b/renderdoc/driver/shaders/dxbc/dxbc_debug.cpp
index ec441a564..4c6b0cefb 100644
--- a/renderdoc/driver/shaders/dxbc/dxbc_debug.cpp
+++ b/renderdoc/driver/shaders/dxbc/dxbc_debug.cpp
@@ -33,9 +33,6 @@
 #include "dxbc_bytecode.h"
 #include "dxbc_container.h"
 
-RDOC_DEBUG_CONFIG(bool, D3D_Hack_EnableGroups, false,
-                  "Work in progress allow shaders to be debugged with workgroup requirements.");
-
 using namespace DXBCBytecode;
 using namespace DXDebug;
 
@@ -4884,8 +4881,7 @@ ShaderDebugTrace *InterpretDebugger::BeginDebug(const DXBC::DXBCContainer *dxbcC
   if(dxbc->m_Type == DXBC::ShaderType::Compute &&
      dxbcContainer->GetThreadScope() == DXBC::ThreadScope::Workgroup)
   {
-    if(D3D_Hack_EnableGroups())
-      workgroupSize = numthreads[0] * numthreads[1] * numthreads[2];
+    workgroupSize = numthreads[0] * numthreads[1] * numthreads[2];
   }
 
   for(int i = 0; i < workgroupSize; i++)
diff --git a/renderdoc/driver/shaders/dxil/dxil_reflect.cpp b/renderdoc/driver/shaders/dxil/dxil_reflect.cpp
index e1bce81ee..73a3ae637 100644
--- a/renderdoc/driver/shaders/dxil/dxil_reflect.cpp
+++ b/renderdoc/driver/shaders/dxil/dxil_reflect.cpp
@@ -28,8 +28,6 @@
 #include "dxil_bytecode.h"
 #include "dxil_common.h"
 
-RDOC_EXTERN_CONFIG(bool, D3D_Hack_EnableGroups);
-
 namespace DXIL
 {
 enum class ResourcesTag
@@ -1870,27 +1868,6 @@ rdcstr Program::GetDebugStatus()
                       "Only supported when debugging pixel shaders dx.op call `%s` %s",
                       callFunc->name.c_str(), ToStr(dxOpCode).c_str());
                 continue;
-              case DXOp::WaveIsFirstLane:
-              case DXOp::WaveGetLaneIndex:
-              case DXOp::WaveGetLaneCount:
-              case DXOp::WaveAnyTrue:
-              case DXOp::WaveAllTrue:
-              case DXOp::WaveActiveAllEqual:
-              case DXOp::WaveActiveBallot:
-              case DXOp::WaveReadLaneAt:
-              case DXOp::WaveReadLaneFirst:
-              case DXOp::WaveActiveOp:
-              case DXOp::WaveActiveBit:
-              case DXOp::WavePrefixOp:
-              case DXOp::WavePrefixBitCount:
-              case DXOp::WaveAllBitCount:
-              case DXOp::WaveMatch:
-              case DXOp::WaveMultiPrefixOp:
-              case DXOp::WaveMultiPrefixBitCount:
-                if(!D3D_Hack_EnableGroups())
-                  return StringFormat::Fmt("Unsupported dx.op call `%s` %s", callFunc->name.c_str(),
-                                           ToStr(dxOpCode).c_str());
-                continue;
               case DXOp::TempRegLoad:
               case DXOp::TempRegStore:
               case DXOp::MinPrecXRegLoad: