diff --git a/OpenCL/m01000_a3-pure.cl b/OpenCL/m01000_a3-pure.cl
index cc5635b7f..4aea55ae1 100644
--- a/OpenCL/m01000_a3-pure.cl
+++ b/OpenCL/m01000_a3-pure.cl
@@ -52,6 +52,18 @@ KERNEL_FQ void m01000_mxx (KERN_ATTR_VECTOR ())
 
     w[0] = w0;
 
+    #if VECT_SIZE == 1
+
+    md4_ctx_t ctx;
+
+    md4_init (&ctx);
+
+    md4_update_utf16le (&ctx, w, pw_len);
+
+    md4_final (&ctx);
+
+    #else
+
     md4_ctx_vector_t ctx;
 
     md4_init_vector (&ctx);
@@ -60,6 +72,8 @@ KERNEL_FQ void m01000_mxx (KERN_ATTR_VECTOR ())
 
     md4_final_vector (&ctx);
 
+    #endif
+
     const u32x r0 = ctx.h[DGST_R0];
     const u32x r1 = ctx.h[DGST_R1];
     const u32x r2 = ctx.h[DGST_R2];
@@ -119,6 +133,18 @@ KERNEL_FQ void m01000_sxx (KERN_ATTR_VECTOR ())
 
     w[0] = w0;
 
+    #if VECT_SIZE == 1
+
+    md4_ctx_t ctx;
+
+    md4_init (&ctx);
+
+    md4_update_utf16le (&ctx, w, pw_len);
+
+    md4_final (&ctx);
+
+    #else
+
     md4_ctx_vector_t ctx;
 
     md4_init_vector (&ctx);
@@ -127,6 +153,8 @@ KERNEL_FQ void m01000_sxx (KERN_ATTR_VECTOR ())
 
     md4_final_vector (&ctx);
 
+    #endif
+
     const u32x r0 = ctx.h[DGST_R0];
     const u32x r1 = ctx.h[DGST_R1];
     const u32x r2 = ctx.h[DGST_R2];
diff --git a/OpenCL/m01100_a3-pure.cl b/OpenCL/m01100_a3-pure.cl
index 643bd0cef..afa425501 100644
--- a/OpenCL/m01100_a3-pure.cl
+++ b/OpenCL/m01100_a3-pure.cl
@@ -61,6 +61,33 @@ KERNEL_FQ void m01100_mxx (KERN_ATTR_VECTOR ())
 
     w[0] = w0;
 
+    #if VECT_SIZE == 1
+
+    md4_ctx_t ctx0;
+
+    md4_init (&ctx0);
+
+    md4_update_utf16le (&ctx0, w, pw_len);
+
+    md4_final (&ctx0);
+
+    md4_ctx_t ctx;
+
+    md4_init (&ctx);
+
+    ctx.w0[0] = ctx0.h[0];
+    ctx.w0[1] = ctx0.h[1];
+    ctx.w0[2] = ctx0.h[2];
+    ctx.w0[3] = ctx0.h[3];
+
+    ctx.len = 16;
+
+    md4_update_utf16le (&ctx, s, salt_len);
+
+    md4_final (&ctx);
+
+    #else
+
     md4_ctx_vector_t ctx0;
 
     md4_init_vector (&ctx0);
@@ -84,6 +111,8 @@ KERNEL_FQ void m01100_mxx (KERN_ATTR_VECTOR ())
 
     md4_final_vector (&ctx);
 
+    #endif
+
     const u32x r0 = ctx.h[DGST_R0];
     const u32x r1 = ctx.h[DGST_R1];
     const u32x r2 = ctx.h[DGST_R2];
@@ -152,6 +181,33 @@ KERNEL_FQ void m01100_sxx (KERN_ATTR_VECTOR ())
 
     w[0] = w0;
 
+    #if VECT_SIZE == 1
+
+    md4_ctx_t ctx0;
+
+    md4_init (&ctx0);
+
+    md4_update_utf16le (&ctx0, w, pw_len);
+
+    md4_final (&ctx0);
+
+    md4_ctx_t ctx;
+
+    md4_init (&ctx);
+
+    ctx.w0[0] = ctx0.h[0];
+    ctx.w0[1] = ctx0.h[1];
+    ctx.w0[2] = ctx0.h[2];
+    ctx.w0[3] = ctx0.h[3];
+
+    ctx.len = 16;
+
+    md4_update_utf16le (&ctx, s, salt_len);
+
+    md4_final (&ctx);
+
+    #else
+
     md4_ctx_vector_t ctx0;
 
     md4_init_vector (&ctx0);
@@ -175,6 +231,8 @@ KERNEL_FQ void m01100_sxx (KERN_ATTR_VECTOR ())
 
     md4_final_vector (&ctx);
 
+    #endif
+
     const u32x r0 = ctx.h[DGST_R0];
     const u32x r1 = ctx.h[DGST_R1];
     const u32x r2 = ctx.h[DGST_R2];
diff --git a/OpenCL/m05500_a3-pure.cl b/OpenCL/m05500_a3-pure.cl
index 3ef7460da..ab19a80bc 100644
--- a/OpenCL/m05500_a3-pure.cl
+++ b/OpenCL/m05500_a3-pure.cl
@@ -586,6 +586,18 @@ KERNEL_FQ void m05500_mxx (KERN_ATTR_VECTOR ())
 
     w[0] = w0;
 
+    #if VECT_SIZE == 1
+
+    md4_ctx_t ctx;
+
+    md4_init (&ctx);
+
+    md4_update_utf16le (&ctx, w, pw_len);
+
+    md4_final (&ctx);
+
+    #else
+
     md4_ctx_vector_t ctx;
 
     md4_init_vector (&ctx);
@@ -594,6 +606,8 @@ KERNEL_FQ void m05500_mxx (KERN_ATTR_VECTOR ())
 
     md4_final_vector (&ctx);
 
+    #endif
+
     const u32x a = ctx.h[0];
     const u32x b = ctx.h[1];
     const u32x c = ctx.h[2];
@@ -742,6 +756,18 @@ KERNEL_FQ void m05500_sxx (KERN_ATTR_VECTOR ())
 
     w[0] = w0;
 
+    #if VECT_SIZE == 1
+
+    md4_ctx_t ctx;
+
+    md4_init (&ctx);
+
+    md4_update_utf16le (&ctx, w, pw_len);
+
+    md4_final (&ctx);
+
+    #else
+
     md4_ctx_vector_t ctx;
 
     md4_init_vector (&ctx);
@@ -750,6 +776,8 @@ KERNEL_FQ void m05500_sxx (KERN_ATTR_VECTOR ())
 
     md4_final_vector (&ctx);
 
+    #endif
+
     const u32x a = ctx.h[0];
     const u32x b = ctx.h[1];
     const u32x c = ctx.h[2];
diff --git a/docs/changes.txt b/docs/changes.txt
index 8f67fa0fc..e88502478 100644
--- a/docs/changes.txt
+++ b/docs/changes.txt
@@ -36,6 +36,7 @@
 - Fixed keys extraction in luks2hashcat - now extracts all active keys
 - Prevent Hashcat from hanging by checking during startup that the output file is a named pipe
 - Fixed debug mode 5 by adding the missing colon between original-word and finding-rule
+- Skip generated rule that was the result of chaining rule operation and caused this generated rule to exceed the maximum number of function calls
 
 ##
 ## Technical
diff --git a/docs/limits.txt b/docs/limits.txt
index e5fece93e..256e2df0a 100644
--- a/docs/limits.txt
+++ b/docs/limits.txt
@@ -35,6 +35,10 @@ because we basically add a zero byte every second byte within the kernel convers
 
 Since hashcat v6.2.1 there is true UTF-16 support for pure kernels. UTF-16 is fully supported.
 
+There is an exception for mask attacks combined with compute devices that require vector data-types
+to achieve full performance. For example a modern CPU. For these devices, the recommendation is to use a
+combination of mask processor piped to hashcat. For modern GPUs it works as expected.
+
 ##
 ## The use of --keep-guessing eventually skips reporting duplicate passwords
 ##
diff --git a/src/backend.c b/src/backend.c
index ca229025c..bc8c74fcf 100644
--- a/src/backend.c
+++ b/src/backend.c
@@ -6233,7 +6233,7 @@ int backend_ctx_devices_init (hashcat_ctx_t *hashcat_ctx, const int comptime)
           if (user_options->backend_info == 0)
           {
             event_log_warning (hashcat_ctx, "The device #%d has been disabled as it most likely also exists as an OpenCL device, but it is not possible to automatically map it.", device_id + 1);
-            event_log_warning (hashcat_ctx, "You can use -d #%d to use Metal API instead of OpenCL API. In some rare cases this is more stable.", device_id + 1);
+            event_log_warning (hashcat_ctx, "You can use -d %d to use Metal API instead of OpenCL API. In some rare cases this is more stable.", device_id + 1);
             event_log_warning (hashcat_ctx, NULL);
           }
 
diff --git a/src/hashcat.c b/src/hashcat.c
index 57d91b0d9..dd00dc638 100644
--- a/src/hashcat.c
+++ b/src/hashcat.c
@@ -364,12 +364,12 @@ static int inner2_loop (hashcat_ctx_t *hashcat_ctx)
   if (status_ctx->devices_status == STATUS_EXHAUSTED)
   {
     // the options speed-only and progress-only cause hashcat to abort quickly.
-    // therefore, they will end up (if no other error occured) as STATUS_EXHAUSTED.
+    // therefore, they will end up (if no other error occurred) as STATUS_EXHAUSTED.
     // however, that can create confusion in hashcats RC, because exhausted translates to RC = 1.
-    // but then having RC = 1 does not match our expection if we use for speed-only and progress-only.
+    // but then having RC = 1 does not match our exception if we use for speed-only and progress-only.
     // to get hashcat to return RC = 0 we have to set it to CRACKED or BYPASS
     // note: other options like --show, --left, --benchmark, --keyspace, --backend-info, etc.
-    // not not reach this section of the code, they've returned already with rc 0.
+    // do not reach this section of the code, they've returned already with rc 0.
 
     if ((user_options->speed_only == true) || (user_options->progress_only == true))
     {
@@ -774,7 +774,7 @@ static int outer_loop (hashcat_ctx_t *hashcat_ctx)
   if (combinator_ctx_init (hashcat_ctx) == -1) return -1;
 
   /**
-   * charsets : keep them together for more easy maintainnce
+   * charsets : keep them together for more easy maintenance
    */
 
   if (mask_ctx_init (hashcat_ctx) == -1) return -1;
@@ -787,7 +787,7 @@ static int outer_loop (hashcat_ctx_t *hashcat_ctx)
   {
     if ((mask_ctx->masks_cnt > 1) || (straight_ctx->dicts_cnt > 1))
     {
-      event_log_error (hashcat_ctx, "Use of --skip/--limit is not supported with --increment or mask files.");
+      event_log_error (hashcat_ctx, "Use of --skip/--limit is not supported with --increment, mask files, or --stdout.");
 
       return -1;
     }
@@ -807,6 +807,17 @@ static int outer_loop (hashcat_ctx_t *hashcat_ctx)
     }
   }
 
+  /**
+   * prevent the user from using -m/--hash-type together with --stdout
+   */
+
+  if (user_options->hash_mode_chgd == true && user_options->stdout_flag == true)
+  {
+    event_log_error (hashcat_ctx, "Use of -m/--hash-type is not supported with --stdout.");
+
+    return -1;
+  }
+
   /**
    * status progress init; needs hashes that's why we have to do it here and separate from status_ctx_init
    */
@@ -1818,7 +1829,7 @@ int hashcat_session_execute (hashcat_ctx_t *hashcat_ctx)
   }
   else if (rc_final == -1)
   {
-    // setup the new negative status code, usefull in test.sh
+    // set up the new negative status code, useful in test.sh
     // -2 is marked as used in status_codes.txt
     if (backend_ctx->runtime_skip_warning  == true)               rc_final = -3;
     if (backend_ctx->memory_hit_warning    == true)               rc_final = -4;
diff --git a/src/rp.c b/src/rp.c
index 3f360fbe7..67ebd9322 100644
--- a/src/rp.c
+++ b/src/rp.c
@@ -905,11 +905,13 @@ int kernel_rules_load (hashcat_ctx_t *hashcat_ctx, kernel_rule_t **out_buf, u32
     return -1;
   }
 
+  u32 invalid_cnt = 0;
+
   for (u32 i = 0; i < kernel_rules_cnt; i++)
   {
     u32 out_pos = 0;
 
-    kernel_rule_t *out = &kernel_rules_buf[i];
+    kernel_rule_t *out = &kernel_rules_buf[i - invalid_cnt];
 
     for (u32 j = 0; j < user_options->rp_files_cnt; j++)
     {
@@ -922,7 +924,9 @@ int kernel_rules_load (hashcat_ctx_t *hashcat_ctx, kernel_rule_t **out_buf, u32
       {
         if (out_pos == RULES_MAX - 1)
         {
-          // event_log_warning (hashcat_ctx, "Truncated chaining of rules %d and %d - maximum functions per rule exceeded.", i, in_off);
+          event_log_warning (hashcat_ctx, "Maximum functions per rule exceeded during chaining of rules, skipping...");
+
+          invalid_cnt++;
 
           break;
         }
@@ -934,6 +938,8 @@ int kernel_rules_load (hashcat_ctx_t *hashcat_ctx, kernel_rule_t **out_buf, u32
 
   hcfree (repeats);
 
+  kernel_rules_cnt -= invalid_cnt;
+
   hcfree (all_kernel_rules_cnt);
   hcfree (all_kernel_rules_buf);
 
diff --git a/tunings/Alias.hctune b/tunings/Alias.hctune
index 79df26641..abd3300ab 100644
--- a/tunings/Alias.hctune
+++ b/tunings/Alias.hctune
@@ -211,9 +211,9 @@ GeForce_MX110                                   ALIAS_nv_sm50_or_higher
 GeForce_MX130                                   ALIAS_nv_sm50_or_higher
 GeForce_MX150                                   ALIAS_nv_sm50_or_higher
 
+GeForce_GTX_1630                                ALIAS_nv_sm50_or_higher
 GeForce_GTX_1650                                ALIAS_nv_sm50_or_higher
 GeForce_GTX_1650_SUPER                          ALIAS_nv_sm50_or_higher
-GeForce_GTX_1650_Ti                             ALIAS_nv_sm50_or_higher
 GeForce_GTX_1660                                ALIAS_nv_sm50_or_higher
 GeForce_GTX_1660_SUPER                          ALIAS_nv_sm50_or_higher
 GeForce_GTX_1660_Ti                             ALIAS_nv_sm50_or_higher
@@ -227,7 +227,6 @@ GeForce_RTX_2080_SUPER                          ALIAS_nv_sm50_or_higher
 GeForce_RTX_2080_Ti                             ALIAS_nv_sm50_or_higher
 
 GeForce_RTX_3050                                ALIAS_nv_sm50_or_higher
-GeForce_RTX_3050_Ti                             ALIAS_nv_sm50_or_higher
 GeForce_RTX_3060                                ALIAS_nv_sm50_or_higher
 GeForce_RTX_3060_Ti                             ALIAS_nv_sm50_or_higher
 GeForce_RTX_3070                                ALIAS_nv_sm50_or_higher