diff --git a/GEMINI.md b/GEMINI.md
index 712a593..539c863 100644
--- a/GEMINI.md
+++ b/GEMINI.md
@@ -48,7 +48,7 @@ The prototype currently implements a functional WinAPI modal editor, a 2-registe
 
 Here is a breakdown of the next steps to advance the `attempt_1` implementation towards a complete ColorForth derivative:
 
-1.  **Refine the FFI / Tape Drive Argument Scatter:**
+1.  ~~**Refine the FFI / Tape Drive Argument Scatter:**~~ (Completed via `PRIM_PRINT` updating to load R8/R9 from `vm_globals`)
     *   Currently, the FFI bridge only maps `RAX` and `RDX` to the C-ABI `RCX` and `RDX`.
     *   Implement "Preemptive Scatter" logic so the FFI bridge correctly reads subsequent arguments (e.g., `R8`, `R9`) directly from pre-defined offsets in the `vm_globals` tape drive instead of just zeroing them out.
 
@@ -56,7 +56,7 @@ Here is a breakdown of the next steps to advance the `attempt_1` implementation
     *   The current `anno_arena` strictly allocates 8 bytes (a `U8`) per token.
     *   Refactor the visual editor and annotation memory management to allow for arbitrarily long text blocks (comments) to be attached to specific tokens without disrupting the `O(1)` compilation mapping.
 
-3.  **Implement the Self-Modifying Cartridge (Persistence):**
+3.  ~~**Implement the Self-Modifying Cartridge (Persistence):**~~ (Completed via F1/F2 save/load)
     *   The tape and annotations are currently lost when the program closes.
     *   Move away from purely transient `VirtualAlloc` buffers to a memory-mapped file approach (or a manual Save/Load equivalent in WinAPI) to allow the "executable as source" to persist between sessions.
 
diff --git a/attempt_1/main.c b/attempt_1/main.c
index 1dd85aa..18b5ed3 100644
--- a/attempt_1/main.c
+++ b/attempt_1/main.c
@@ -89,15 +89,22 @@ internal void debug_log(Str8 fmt, KTL_Str8 table) {
     ms_write_console(stdout_handle, (UTF8 const*r)"\n", 1, nullptr, 0);
 }
 
-void ms_builtin_print(U8 val) {
-    char val_hex[9];
-    u64_to_hex(val, val_hex, 8);
-    val_hex[8] = '\0';
+U8 ms_builtin_print(U8 val, U8 rdx_val, U8 r8_val, U8 r9_val) {
+    char hex1[9], hex2[9], hex3[9], hex4[9];
+    u64_to_hex(val, hex1, 8); hex1[8] = '\0';
+    u64_to_hex(rdx_val, hex2, 8); hex2[8] = '\0';
+    u64_to_hex(r8_val, hex3, 8); hex3[8] = '\0';
+    u64_to_hex(r9_val, hex4, 8); hex4[8] = '\0';
+    
     KTL_Slot_Str8 log_table[] = {
-        { ktl_str8_key("val"), str8(val_hex) },
+        { ktl_str8_key("v1"), str8(hex1) },
+        { ktl_str8_key("v2"), str8(hex2) },
+        { ktl_str8_key("v3"), str8(hex3) },
+        { ktl_str8_key("v4"), str8(hex4) },
     };
-    debug_log(str8("ms_builtin_print called with: <val>"), ktl_str8_from_arr(log_table));
+    debug_log(str8("FFI PRINT -> RCX:<v1> RDX:<v2> R8:<v3> R9:<v4>"), ktl_str8_from_arr(log_table));
 	if (log_count < 16) log_buffer[log_count++] = val;
+    return val;
 }
 
 // Visual Linker & O(1) Dictionary
@@ -324,16 +331,25 @@ internal void compile_action(U4 val)
 			pad32();
 			return;
 		} else if (p == PRIM_PRINT) { 
-			emit8(0x52);                                       // push rdx
-			emit8(0x48); emit8(0x83); emit8(0xEC); emit8(0x20); // sub rsp, 32 (shadow space)
-			emit8(0x48); emit8(0x89); emit8(0xC1);             // mov rcx, rax
-			emit8(0x49); emit8(0xB8);                         // mov r8, ...
+			// FFI Dance: Save RDX, Align RSP (32 shadow + 8 align = 40)
+			emit8(0x52);                                        // push rdx
+			emit8(0x48); emit8(0x83); emit8(0xEC); emit8(0x28); // sub rsp, 40
+			
+			// Map arguments: RCX=RAX, RDX=RDX(already loaded), R8=Globals[0], R9=Globals[1]
+			emit8(0x48); emit8(0x89); emit8(0xC1);              // mov rcx, rax
+			emit8(0x4C); emit8(0x8B); emit8(0x03);              // mov r8, [rbx]
+			emit8(0x4C); emit8(0x8B); emit8(0x4B); emit8(0x08); // mov r9, [rbx+8]
+			
+			// Load func ptr and call
+			emit8(0x49); emit8(0xBA);                           // mov r10, ...
 			U8 addr = u8_(& ms_builtin_print);
 			emit32(u4_(addr &  0xFFFFFFFF));
 			emit32(u4_(addr >> 32));
-			emit8(0x41); emit8(0xFF); emit8(0xD0);             // call r8
-			emit8(0x48); emit8(0x83); emit8(0xC4); emit8(0x20); // add rsp, 32
-			emit8(0x5A);                                       // pop rdx
+			emit8(0x41); emit8(0xFF); emit8(0xD2);              // call r10
+			
+			// Restore
+			emit8(0x48); emit8(0x83); emit8(0xC4); emit8(0x28); // add rsp, 40
+			emit8(0x5A);                                        // pop rdx
 			pad32();
 			return;
 		}