-# Comparative analysis with Andes Packed ISA proposal
+# Comparative analysis of Andes Packed ISA proposal vs Harmonised RVP (forwards compatible with RV Vector)
+
+## Proposed vector instruction encoding
+
+Register x 2 -> register operations:
+
+| 31 30 29 28 27 26 | 25 | 24 23 22 21 20 | 19 18 17 16 15 | 14 | 13 12 | 11 10 9 8 7 | 6 5 4 3 2 1 0 |
+| ----------------- | -- | -------------- | -------------- | -- | ----- | ----------- | ------------- |
+| func_6 | 0 | rs2 | rs1 | 0 | mm | rd1 | VOP opcode |
+
+Immediate + register -> register operations:
+
+| 31 30 29 | 28 27 26 | 25 | 24 23 22 21 20 | 19 18 17 16 15 | 14 | 13 12 | 11 10 9 8 7 | 6 5 4 3 2 1 0 |
+| -------- | -------- | -- | -------------- | -------------- | -- | ----- | ----------- | ------------- |
+| func_3 | imm[7:5] | 1 | imm[4:0] | rs1 | 0 | mm | rd1 | VOP opcode |
+
+Register x 3 -> register operations:
+
+| 31 30 29 28 27 | 26 25 | 24 23 22 21 20 | 19 18 17 16 15 | 14 | 13 12 | 11 10 9 8 7 | 6 5 4 3 2 1 0 |
+| ----------------------- | -------------- | -------------- | -- | ----- | ----------- | ------------- |
+| rs3 | func_2 | rs2 | rs1 | 1 | mm | rd1 | VOP opcode |
+
+mm values:
+mm = 00 -> use current global saturation or rounding, no mask
+mm = 00 -> force saturation or rounding for this instruction only
+mm = 10 -> use v1 as predicate mask
+mm = 11 -> use ~v1 as predicate mask
## Register file