diff --git a/src/emu/emumem.cpp b/src/emu/emumem.cpp
index ca24883e2f393..11f39dcb89a55 100644
--- a/src/emu/emumem.cpp
+++ b/src/emu/emumem.cpp
@@ -53,6 +53,19 @@ void handler_entry::dump_map(std::vector<memory_entry> &map) const
 	fatalerror("dump_map called on non-dispatching class\n");
 }
 
+bool handler_entry::is_handler_in_map(std::vector<memory_entry> &map, offs_t begin, offs_t end, handler_entry *handler) const
+{
+	auto it = std::find_if(map.begin(), map.end(), [handler,begin,end](const memory_entry& e) {
+		return (e.entry == handler) && (e.start == begin) && (e.end == end);
+	} );
+
+	if(it == map.end()) {
+		return false;
+	}
+
+	return true;
+}
+
 void handler_entry::select_a(int slot)
 {
 	fatalerror("select_a called on non-view\n");
diff --git a/src/emu/emumem.h b/src/emu/emumem.h
index 39470a45632d6..c18f58a55e522 100644
--- a/src/emu/emumem.h
+++ b/src/emu/emumem.h
@@ -608,6 +608,7 @@ class handler_entry
 	inline u32 f_get_pt() const { return (m_flags >> F_PT_BITS) & 15; }
 
 	virtual void dump_map(std::vector<memory_entry> &map) const;
+	bool is_handler_in_map(std::vector<memory_entry> &map, offs_t begin, offs_t end, handler_entry *handler) const;
 
 	virtual std::string name() const = 0;
 	virtual void enumerate_references(handler_entry::reflist &refs) const;
diff --git a/src/emu/emumem_hedr.ipp b/src/emu/emumem_hedr.ipp
index 5ac845b0fef58..9444c0c07821e 100644
--- a/src/emu/emumem_hedr.ipp
+++ b/src/emu/emumem_hedr.ipp
@@ -107,41 +107,64 @@ template<int HighBits, int Width, int AddrShift> offs_t handler_entry_read_dispa
 template<int HighBits, int Width, int AddrShift> void handler_entry_read_dispatch<HighBits, Width, AddrShift>::dump_map(std::vector<memory_entry> &map) const
 {
 	if(m_view) {
-		offs_t base_cur = map.empty() ? m_view->m_addrstart & HIGHMASK : map.back().end + 1;
 		for(u32 i = 0; i != m_dispatch_array.size(); i++) {
-			u32 j = map.size();
-			offs_t cur = base_cur;
-			offs_t end = m_global_range.end + 1;
-
+			u32 map_start_index = map.size();
+			offs_t j = 0;
+			offs_t k = j + 1;
+			handler_entry *handle = m_dispatch_array[i][j];
 			do {
-				offs_t entry = (cur >> LowBits) & BITMASK;
-				if(m_dispatch_array[i][entry]->is_dispatch() || m_dispatch_array[i][entry]->is_view())
-					m_dispatch_array[i][entry]->dump_map(map);
-				else
-					map.emplace_back(memory_entry{ m_ranges_array[i][entry].start, m_ranges_array[i][entry].end, m_dispatch_array[i][entry] });
-				cur = map.back().end + 1;
-			} while(cur != end);
+				while((handle == m_dispatch_array[i][k]) && (k < BITMASK)) {
+					k++;
+				}
+
+				k--;
+
+				if(m_dispatch_array[i][j]->is_dispatch() || m_dispatch_array[i][j]->is_view()) {
+					m_dispatch_array[i][j]->dump_map(map);
+				} else {
+					if(!handler_entry::is_handler_in_map(map, m_ranges_array[i][j].start, m_ranges_array[i][j].end, m_dispatch_array[i][j])) {
+						map.emplace_back(memory_entry{ m_ranges_array[i][j].start, m_ranges_array[i][k].end, m_dispatch_array[i][j]});
+					}
+				}
+
+				j = k + 1;
+				k = j + 1;
+				handle = m_dispatch_array[i][j];
+			} while (j < BITMASK);
+
 			if(i == 0) {
-				for(u32 k = j; k != map.size(); k++)
+				for(u32 k = map_start_index; k != map.size(); k++)
 					map[k].context.emplace(map[k].context.begin(), memory_entry_context{ m_view, true, 0 });
 			} else {
 				int slot = m_view->id_to_slot(int(i)-1);
-				for(u32 k = j; k != map.size(); k++)
+				for(u32 k = map_start_index; k != map.size(); k++)
 					map[k].context.emplace(map[k].context.begin(), memory_entry_context{ m_view, false, slot });
 			}
 		}
 	} else {
-		offs_t cur = map.empty() ? 0 : map.back().end + 1;
-		offs_t base = cur & UPMASK;
-		offs_t end = m_global_range.end + 1;
+		offs_t j = 0;
+		offs_t k = j + 1;
+		handler_entry *the_handler;
 		do {
-			offs_t entry = (cur >> LowBits) & BITMASK;
-			if(m_a_dispatch[entry]->is_dispatch() || m_a_dispatch[entry]->is_view())
-				m_a_dispatch[entry]->dump_map(map);
-			else
-				map.emplace_back(memory_entry{ m_a_ranges[entry].start, m_a_ranges[entry].end, m_a_dispatch[entry] });
-			cur = map.back().end + 1;
-		} while(cur != end && !((cur ^ base) & UPMASK));
+			the_handler = m_a_dispatch[j];
+			while((the_handler == m_a_dispatch[k]) && (k < COUNT)) {
+				k++;
+			}
+
+			k--;
+
+			for(unsigned int z = j; z <= k; z++) {
+				if(!handler_entry::is_handler_in_map(map, m_a_ranges[z].start, m_a_ranges[z].end, m_a_dispatch[z])) {
+					if(m_a_dispatch[z]->is_dispatch() || m_a_dispatch[z]->is_view()) {
+						m_a_dispatch[z]->dump_map(map);
+					}
+					else
+						map.emplace_back(memory_entry{m_a_ranges[z].start,m_a_ranges[z].end,m_a_dispatch[z]});
+				}
+			}
+			j = k + 1;
+			k = j + 1;
+		} while(j < COUNT);
 	}
 }
 
diff --git a/src/emu/emumem_hedw.ipp b/src/emu/emumem_hedw.ipp
index a3d8f78bb07c6..c85a0dc0a86d0 100644
--- a/src/emu/emumem_hedw.ipp
+++ b/src/emu/emumem_hedw.ipp
@@ -108,40 +108,64 @@ template<int HighBits, int Width, int AddrShift> offs_t handler_entry_write_disp
 template<int HighBits, int Width, int AddrShift> void handler_entry_write_dispatch<HighBits, Width, AddrShift>::dump_map(std::vector<memory_entry> &map) const
 {
 	if(m_view) {
-		offs_t base_cur = map.empty() ? m_view->m_addrstart & HIGHMASK : map.back().end + 1;
 		for(u32 i = 0; i != m_dispatch_array.size(); i++) {
-			u32 j = map.size();
-			offs_t cur = base_cur;
-			offs_t end = m_global_range.end + 1;
+			u32 map_start_index = map.size();
+			offs_t j = 0;
+			offs_t k = j + 1;
+			handler_entry *handle = m_dispatch_array[i][j];
 			do {
-				offs_t entry = (cur >> LowBits) & BITMASK;
-				if(m_dispatch_array[i][entry]->is_dispatch() || m_dispatch_array[i][entry]->is_view())
-					m_dispatch_array[i][entry]->dump_map(map);
-				else
-					map.emplace_back(memory_entry{ m_ranges_array[i][entry].start, m_ranges_array[i][entry].end, m_dispatch_array[i][entry] });
-				cur = map.back().end + 1;
-			} while(cur != end);
+				while((handle == m_dispatch_array[i][k]) && (k < BITMASK)) {
+					k++;
+				}
+
+				k--;
+
+				if(m_dispatch_array[i][j]->is_dispatch() || m_dispatch_array[i][j]->is_view()) {
+					m_dispatch_array[i][j]->dump_map(map);
+				} else {
+					if(!handler_entry::is_handler_in_map(map, m_ranges_array[i][j].start, m_ranges_array[i][j].end, m_dispatch_array[i][j])) {
+						map.emplace_back(memory_entry{ m_ranges_array[i][j].start, m_ranges_array[i][k].end, m_dispatch_array[i][j]});
+					}
+				}
+
+				j = k + 1;
+				k = j + 1;
+				handle = m_dispatch_array[i][j];
+			} while (j < BITMASK);
+
 			if(i == 0) {
-				for(u32 k = j; k != map.size(); k++)
+				for(u32 k = map_start_index; k != map.size(); k++)
 					map[k].context.emplace(map[k].context.begin(), memory_entry_context{ m_view, true, 0 });
 			} else {
 				int slot = m_view->id_to_slot(int(i)-1);
-				for(u32 k = j; k != map.size(); k++)
+				for(u32 k = map_start_index; k != map.size(); k++)
 					map[k].context.emplace(map[k].context.begin(), memory_entry_context{ m_view, false, slot });
 			}
 		}
 	} else {
-		offs_t cur = map.empty() ? 0 : map.back().end + 1;
-		offs_t base = cur & UPMASK;
-		offs_t end = m_global_range.end + 1;
+		offs_t j = 0;
+		offs_t k = j + 1;
+		handler_entry *the_handler;
 		do {
-			offs_t entry = (cur >> LowBits) & BITMASK;
-			if(m_a_dispatch[entry]->is_dispatch() || m_a_dispatch[entry]->is_view())
-				m_a_dispatch[entry]->dump_map(map);
-			else
-				map.emplace_back(memory_entry{ m_a_ranges[entry].start, m_a_ranges[entry].end, m_a_dispatch[entry] });
-			cur = map.back().end + 1;
-		} while(cur != end && !((cur ^ base) & UPMASK));
+			the_handler = m_a_dispatch[j];
+			while((the_handler == m_a_dispatch[k]) && (k < COUNT)) {
+				k++;
+			}
+
+			k--;
+
+			for(unsigned int z = j; z <= k; z++) {
+				if(!handler_entry::is_handler_in_map(map, m_a_ranges[z].start, m_a_ranges[z].end, m_a_dispatch[z])) {
+					if(m_a_dispatch[z]->is_dispatch() || m_a_dispatch[z]->is_view()) {
+						m_a_dispatch[z]->dump_map(map);
+					}
+					else
+						map.emplace_back(memory_entry{m_a_ranges[z].start,m_a_ranges[z].end,m_a_dispatch[z]});
+				}
+			}
+			j = k + 1;
+			k = j + 1;
+		} while(j < COUNT);
 	}
 }