diff --git a/src/cpu.cpp b/src/cpu.cpp
index 81a28db..d2e8307 100644
--- a/src/cpu.cpp
+++ b/src/cpu.cpp
@@ -67,26 +67,26 @@ void CPU::update()
 		print("running opcode: {:#x}\n", opcode);
 		switch (opcode) {
 
-		case 0x20: jr8(); break;
-		case 0x01: ld16(); break;
+		case 0x01: ldi16(); break;
 		case 0x02: ldr8(); break;
 		case 0x06: ldi8(); break;
-		case 0x08: ld16(); break;
+		case 0x08: lda16(); break;
 		case 0x0a: ldr8(); break;
 		case 0x0d: dec8(); break;
 		case 0x0e: ldi8(); break;
-		case 0x11: ld16(); break;
+		case 0x11: ldi16(); break;
 		case 0x12: ldr8(); break;
 		case 0x16: ldi8(); break;
 		case 0x1a: ldr8(); break;
 		case 0x1e: ldi8(); break;
-		case 0x21: ld16(); break;
+		case 0x20: jr8(); break;
+		case 0x21: ldi16(); break;
 		case 0x22: ldr8(); break;
 		case 0x26: ldi8(); break;
 		case 0x2a: ldr8(); break;
 		case 0x2e: ldi8(); break;
 		case 0x2f: misc(); break;
-		case 0x31: ld16(); break;
+		case 0x31: ldi16(); break;
 		case 0x32: ldr8(); break;
 		case 0x36: ldi8(); break;
 		case 0x3a: ldr8(); break;
@@ -98,8 +98,8 @@ void CPU::update()
 		case 0xcd: call(); break;
 		case 0xe0: ldffi8(); break;
 		case 0xf0: ldffi8(); break;
-		case 0xf8: ld16(); break;
-		case 0xf9: ld16(); break;
+		case 0xf8: ldr16(); break;
+		case 0xf9: ldr16(); break;
 
 		default:
 			print("opcode {:#x} not implemented\n", opcode);
@@ -175,6 +175,67 @@ void CPU::xor8()
 	}
 }
 
+void CPU::lda8()
+{
+	uint8_t opcode = pcRead();
+	switch (opcode) {
+	case 0x22: { // LD (HL+),A == LD (HLI),A == LDI (HL),A
+		m_wait_cycles += 8;
+
+		// Put A into memory address in HL
+		uint32_t address = hl();
+		write(address, m_a);
+
+		// Increment HL
+		address = (address + 1) & 0xffff;
+		m_l = address & 0xff;
+		m_h = address >> 8;
+		break;
+	}
+	case 0x2a: { // LD A,(HL+) == LD A,(HLI) == LDI A,(HL)
+		m_wait_cycles += 8;
+
+		// Put value at address in HL into A
+		uint32_t address = hl();
+		m_a = read(address);
+
+		// Increment HL
+		address = (address + 1) & 0xffff;
+		m_l = address & 0xff;
+		m_h = address >> 8;
+		break;
+	}
+	case 0x32: { // LD (HL-),A == LD (HLD),A == LDD (HL),A
+		m_wait_cycles += 8;
+
+		// Put A into memory address in HL
+		uint32_t address = hl();
+		write(address, m_a);
+
+		// Decrement HL
+		address = (address - 1) & 0xffff;
+		m_l = address & 0xff;
+		m_h = address >> 8;
+		break;
+	}
+	case 0x3a: { // LD A,(HL-) == LD A,(HLD) == LDD A,(HL)
+		m_wait_cycles += 8;
+
+		// Put value at address in HL into A
+		uint32_t address = hl();
+		m_a = read(address);
+
+		// Decrement HL
+		address = (address + 1) & 0xffff;
+		m_l = address & 0xff;
+		m_h = address >> 8;
+		break;
+	}
+	default:
+		VERIFY_NOT_REACHED();
+	}
+}
+
 void CPU::ldi8()
 {
 	uint8_t opcode = pcRead();
@@ -231,58 +292,6 @@ void CPU::ldr8()
 		m_wait_cycles += 8;
 		m_a = read(de());
 		break;
-	case 0x22: { // LD (HL+),A == LD (HLI),A == LDI (HL),A
-		m_wait_cycles += 8;
-
-		// Put A into memory address in HL
-		uint32_t address = hl();
-		write(address, m_a);
-
-		// Increment HL
-		address = (address + 1) & 0xffff;
-		m_l = address & 0xff;
-		m_h = address >> 8;
-		break;
-	}
-	case 0x2a: { // LD A,(HL+) == LD A,(HLI) == LDI A,(HL)
-		m_wait_cycles += 8;
-
-		// Put value at address in HL into A
-		uint32_t address = hl();
-		m_a = read(address);
-
-		// Increment HL
-		address = (address + 1) & 0xffff;
-		m_l = address & 0xff;
-		m_h = address >> 8;
-		break;
-	}
-	case 0x32: { // LD (HL-),A == LD (HLD),A == LDD (HL),A
-		m_wait_cycles += 8;
-
-		// Put A into memory address in HL
-		uint32_t address = hl();
-		write(address, m_a);
-
-		// Decrement HL
-		address = (address - 1) & 0xffff;
-		m_l = address & 0xff;
-		m_h = address >> 8;
-		break;
-	}
-	case 0x3a: { // LD A,(HL-) == LD A,(HLD) == LDD A,(HL)
-		m_wait_cycles += 8;
-
-		// Put value at address in HL into A
-		uint32_t address = hl();
-		m_a = read(address);
-
-		// Decrement HL
-		address = (address + 1) & 0xffff;
-		m_l = address & 0xff;
-		m_h = address >> 8;
-		break;
-	}
 	default:
 		VERIFY_NOT_REACHED();
 	}
@@ -309,15 +318,10 @@ void CPU::ldffi8()
 	}
 }
 
-void CPU::ld16()
+void CPU::lda16()
 {
 	uint8_t opcode = pcRead();
 	switch (opcode) {
-	case 0x01: { // LD BC,i16
-		m_wait_cycles += 12;
-		write(bc(), pcRead16());
-		break;
-	}
 	case 0x08: { // LD a16,SP
 		m_wait_cycles += 20;
 
@@ -325,6 +329,20 @@ void CPU::ld16()
 		// TODO
 		break;
 	}
+	default:
+		VERIFY_NOT_REACHED();
+	}
+}
+
+void CPU::ldi16()
+{
+	uint8_t opcode = pcRead();
+	switch (opcode) {
+	case 0x01: { // LD BC,i16
+		m_wait_cycles += 12;
+		write(bc(), pcRead16());
+		break;
+	}
 	case 0x11: // LD DE,i16
 		m_wait_cycles += 12;
 		write(de(), pcRead16());
@@ -338,6 +356,15 @@ void CPU::ld16()
 		m_sp = pcRead16();
 		break;
 	}
+	default:
+		VERIFY_NOT_REACHED();
+	}
+}
+
+void CPU::ldr16()
+{
+	uint8_t opcode = pcRead();
+	switch (opcode) {
 	case 0xf8: { // LD HL,SP + s8 == LDHL SP,s8, flags: 0 0 H C
 		m_wait_cycles += 12;
 
diff --git a/src/cpu.h b/src/cpu.h
index 0ed92cf..43f0fd3 100644
--- a/src/cpu.h
+++ b/src/cpu.h
@@ -45,11 +45,14 @@ public:
 
 	// 8-bit
 	void ldffi8();
+	void lda8();
 	void ldi8();
 	void ldr8();
 
 	// 16-bit
-	void ld16();
+	void lda16();
+	void ldi16();
+	void ldr16();
 
 	// -------------------------------------
 	// Jumps and Subroutines