diff --git a/OpenCL/inc_common.cl b/OpenCL/inc_common.cl
index 3e3fe573c..df4beee06 100644
--- a/OpenCL/inc_common.cl
+++ b/OpenCL/inc_common.cl
@@ -2051,6 +2051,31 @@ void make_utf16le (const u32x in[4], u32x out1[4], u32x out2[4])
   #endif
 }
 
+void make_utf16leN (const u32x in[4], u32x out1[4], u32x out2[4])
+{
+  #ifdef IS_NV
+  out2[3] = __byte_perm (in[3], 0, 0x7170);
+  out2[2] = __byte_perm (in[3], 0, 0x7372);
+  out2[1] = __byte_perm (in[2], 0, 0x7170);
+  out2[0] = __byte_perm (in[2], 0, 0x7372);
+  out1[3] = __byte_perm (in[1], 0, 0x7170);
+  out1[2] = __byte_perm (in[1], 0, 0x7372);
+  out1[1] = __byte_perm (in[0], 0, 0x7170);
+  out1[0] = __byte_perm (in[0], 0, 0x7372);
+  #endif
+
+  #if defined IS_AMD || defined IS_GENERIC
+  out2[3] = ((in[3] << 8) & 0x00FF0000) | ((in[3] >>  0) & 0x000000FF);
+  out2[2] = ((in[3] >> 8) & 0x00FF0000) | ((in[3] >> 16) & 0x000000FF);
+  out2[1] = ((in[2] << 8) & 0x00FF0000) | ((in[2] >>  0) & 0x000000FF);
+  out2[0] = ((in[2] >> 8) & 0x00FF0000) | ((in[2] >> 16) & 0x000000FF);
+  out1[3] = ((in[1] << 8) & 0x00FF0000) | ((in[1] >>  0) & 0x000000FF);
+  out1[2] = ((in[1] >> 8) & 0x00FF0000) | ((in[1] >> 16) & 0x000000FF);
+  out1[1] = ((in[0] << 8) & 0x00FF0000) | ((in[0] >>  0) & 0x000000FF);
+  out1[0] = ((in[0] >> 8) & 0x00FF0000) | ((in[0] >> 16) & 0x000000FF);
+  #endif
+}
+
 void undo_utf16be (const u32x in1[4], const u32x in2[4], u32x out[4])
 {
   #ifdef IS_NV
@@ -32930,31 +32955,6 @@ void make_utf16be_S (const u32 in[4], u32 out1[4], u32 out2[4])
   #endif
 }
 
-void make_utf16beN_S (const u32 in[4], u32 out1[4], u32 out2[4])
-{
-  #ifdef IS_NV
-  out2[3] = __byte_perm_S (in[3], 0, 0x1707);
-  out2[2] = __byte_perm_S (in[3], 0, 0x3727);
-  out2[1] = __byte_perm_S (in[2], 0, 0x1707);
-  out2[0] = __byte_perm_S (in[2], 0, 0x3727);
-  out1[3] = __byte_perm_S (in[1], 0, 0x1707);
-  out1[2] = __byte_perm_S (in[1], 0, 0x3727);
-  out1[1] = __byte_perm_S (in[0], 0, 0x1707);
-  out1[0] = __byte_perm_S (in[0], 0, 0x3727);
-  #endif
-
-  #if defined IS_AMD || defined IS_GENERIC
-  out2[3] = ((in[3] << 16) & 0xFF000000) | ((in[3] << 8) & 0x0000FF00);
-  out2[2] = ((in[3] >>  0) & 0xFF000000) | ((in[3] >> 8) & 0x0000FF00);
-  out2[1] = ((in[2] << 16) & 0xFF000000) | ((in[2] << 8) & 0x0000FF00);
-  out2[0] = ((in[2] >>  0) & 0xFF000000) | ((in[2] >> 8) & 0x0000FF00);
-  out1[3] = ((in[1] << 16) & 0xFF000000) | ((in[1] << 8) & 0x0000FF00);
-  out1[2] = ((in[1] >>  0) & 0xFF000000) | ((in[1] >> 8) & 0x0000FF00);
-  out1[1] = ((in[0] << 16) & 0xFF000000) | ((in[0] << 8) & 0x0000FF00);
-  out1[0] = ((in[0] >>  0) & 0xFF000000) | ((in[0] >> 8) & 0x0000FF00);
-  #endif
-}
-
 void make_utf16le_S (const u32 in[4], u32 out1[4], u32 out2[4])
 {
   #ifdef IS_NV
diff --git a/OpenCL/inc_hash_sha1.cl b/OpenCL/inc_hash_sha1.cl
index 47fe4691d..4cf6c280f 100644
--- a/OpenCL/inc_hash_sha1.cl
+++ b/OpenCL/inc_hash_sha1.cl
@@ -508,6 +508,124 @@ void sha1_update_utf16le_swap (sha1_ctx_t *ctx, const u32 *w, const int len)
   sha1_update_64 (ctx, w0, w1, w2, w3, (len - pos1) * 2);
 }
 
+void sha1_update_utf16be (sha1_ctx_t *ctx, const u32 *w, const int len)
+{
+  u32 w0[4];
+  u32 w1[4];
+  u32 w2[4];
+  u32 w3[4];
+
+  int pos1;
+  int pos4;
+
+  for (pos1 = 0, pos4 = 0; pos1 < len - 32; pos1 += 32, pos4 += 8)
+  {
+    w0[0] = w[pos4 + 0];
+    w0[1] = w[pos4 + 1];
+    w0[2] = w[pos4 + 2];
+    w0[3] = w[pos4 + 3];
+    w1[0] = w[pos4 + 4];
+    w1[1] = w[pos4 + 5];
+    w1[2] = w[pos4 + 6];
+    w1[3] = w[pos4 + 7];
+
+    make_utf16be_S (w1, w2, w3);
+    make_utf16be_S (w0, w0, w1);
+
+    sha1_update_64 (ctx, w0, w1, w2, w3, 32 * 2);
+  }
+
+  w0[0] = w[pos4 + 0];
+  w0[1] = w[pos4 + 1];
+  w0[2] = w[pos4 + 2];
+  w0[3] = w[pos4 + 3];
+  w1[0] = w[pos4 + 4];
+  w1[1] = w[pos4 + 5];
+  w1[2] = w[pos4 + 6];
+  w1[3] = w[pos4 + 7];
+
+  make_utf16be_S (w1, w2, w3);
+  make_utf16be_S (w0, w0, w1);
+
+  sha1_update_64 (ctx, w0, w1, w2, w3, (len - pos1) * 2);
+}
+
+void sha1_update_utf16be_swap (sha1_ctx_t *ctx, const u32 *w, const int len)
+{
+  u32 w0[4];
+  u32 w1[4];
+  u32 w2[4];
+  u32 w3[4];
+
+  int pos1;
+  int pos4;
+
+  for (pos1 = 0, pos4 = 0; pos1 < len - 32; pos1 += 32, pos4 += 8)
+  {
+    w0[0] = w[pos4 + 0];
+    w0[1] = w[pos4 + 1];
+    w0[2] = w[pos4 + 2];
+    w0[3] = w[pos4 + 3];
+    w1[0] = w[pos4 + 4];
+    w1[1] = w[pos4 + 5];
+    w1[2] = w[pos4 + 6];
+    w1[3] = w[pos4 + 7];
+
+    make_utf16be_S (w1, w2, w3);
+    make_utf16be_S (w0, w0, w1);
+
+    w0[0] = swap32_S (w0[0]);
+    w0[1] = swap32_S (w0[1]);
+    w0[2] = swap32_S (w0[2]);
+    w0[3] = swap32_S (w0[3]);
+    w1[0] = swap32_S (w1[0]);
+    w1[1] = swap32_S (w1[1]);
+    w1[2] = swap32_S (w1[2]);
+    w1[3] = swap32_S (w1[3]);
+    w2[0] = swap32_S (w2[0]);
+    w2[1] = swap32_S (w2[1]);
+    w2[2] = swap32_S (w2[2]);
+    w2[3] = swap32_S (w2[3]);
+    w3[0] = swap32_S (w3[0]);
+    w3[1] = swap32_S (w3[1]);
+    w3[2] = swap32_S (w3[2]);
+    w3[3] = swap32_S (w3[3]);
+
+    sha1_update_64 (ctx, w0, w1, w2, w3, 32 * 2);
+  }
+
+  w0[0] = w[pos4 + 0];
+  w0[1] = w[pos4 + 1];
+  w0[2] = w[pos4 + 2];
+  w0[3] = w[pos4 + 3];
+  w1[0] = w[pos4 + 4];
+  w1[1] = w[pos4 + 5];
+  w1[2] = w[pos4 + 6];
+  w1[3] = w[pos4 + 7];
+
+  make_utf16be_S (w1, w2, w3);
+  make_utf16be_S (w0, w0, w1);
+
+  w0[0] = swap32_S (w0[0]);
+  w0[1] = swap32_S (w0[1]);
+  w0[2] = swap32_S (w0[2]);
+  w0[3] = swap32_S (w0[3]);
+  w1[0] = swap32_S (w1[0]);
+  w1[1] = swap32_S (w1[1]);
+  w1[2] = swap32_S (w1[2]);
+  w1[3] = swap32_S (w1[3]);
+  w2[0] = swap32_S (w2[0]);
+  w2[1] = swap32_S (w2[1]);
+  w2[2] = swap32_S (w2[2]);
+  w2[3] = swap32_S (w2[3]);
+  w3[0] = swap32_S (w3[0]);
+  w3[1] = swap32_S (w3[1]);
+  w3[2] = swap32_S (w3[2]);
+  w3[3] = swap32_S (w3[3]);
+
+  sha1_update_64 (ctx, w0, w1, w2, w3, (len - pos1) * 2);
+}
+
 void sha1_update_global (sha1_ctx_t *ctx, const __global u32 *w, const int len)
 {
   u32 w0[4];
@@ -764,6 +882,124 @@ void sha1_update_global_utf16le_swap (sha1_ctx_t *ctx, const __global u32 *w, co
   sha1_update_64 (ctx, w0, w1, w2, w3, (len - pos1) * 2);
 }
 
+void sha1_update_global_utf16be (sha1_ctx_t *ctx, const __global u32 *w, const int len)
+{
+  u32 w0[4];
+  u32 w1[4];
+  u32 w2[4];
+  u32 w3[4];
+
+  int pos1;
+  int pos4;
+
+  for (pos1 = 0, pos4 = 0; pos1 < len - 32; pos1 += 32, pos4 += 8)
+  {
+    w0[0] = w[pos4 + 0];
+    w0[1] = w[pos4 + 1];
+    w0[2] = w[pos4 + 2];
+    w0[3] = w[pos4 + 3];
+    w1[0] = w[pos4 + 4];
+    w1[1] = w[pos4 + 5];
+    w1[2] = w[pos4 + 6];
+    w1[3] = w[pos4 + 7];
+
+    make_utf16be_S (w1, w2, w3);
+    make_utf16be_S (w0, w0, w1);
+
+    sha1_update_64 (ctx, w0, w1, w2, w3, 32 * 2);
+  }
+
+  w0[0] = w[pos4 + 0];
+  w0[1] = w[pos4 + 1];
+  w0[2] = w[pos4 + 2];
+  w0[3] = w[pos4 + 3];
+  w1[0] = w[pos4 + 4];
+  w1[1] = w[pos4 + 5];
+  w1[2] = w[pos4 + 6];
+  w1[3] = w[pos4 + 7];
+
+  make_utf16be_S (w1, w2, w3);
+  make_utf16be_S (w0, w0, w1);
+
+  sha1_update_64 (ctx, w0, w1, w2, w3, (len - pos1) * 2);
+}
+
+void sha1_update_global_utf16be_swap (sha1_ctx_t *ctx, const __global u32 *w, const int len)
+{
+  u32 w0[4];
+  u32 w1[4];
+  u32 w2[4];
+  u32 w3[4];
+
+  int pos1;
+  int pos4;
+
+  for (pos1 = 0, pos4 = 0; pos1 < len - 32; pos1 += 32, pos4 += 8)
+  {
+    w0[0] = w[pos4 + 0];
+    w0[1] = w[pos4 + 1];
+    w0[2] = w[pos4 + 2];
+    w0[3] = w[pos4 + 3];
+    w1[0] = w[pos4 + 4];
+    w1[1] = w[pos4 + 5];
+    w1[2] = w[pos4 + 6];
+    w1[3] = w[pos4 + 7];
+
+    make_utf16be_S (w1, w2, w3);
+    make_utf16be_S (w0, w0, w1);
+
+    w0[0] = swap32_S (w0[0]);
+    w0[1] = swap32_S (w0[1]);
+    w0[2] = swap32_S (w0[2]);
+    w0[3] = swap32_S (w0[3]);
+    w1[0] = swap32_S (w1[0]);
+    w1[1] = swap32_S (w1[1]);
+    w1[2] = swap32_S (w1[2]);
+    w1[3] = swap32_S (w1[3]);
+    w2[0] = swap32_S (w2[0]);
+    w2[1] = swap32_S (w2[1]);
+    w2[2] = swap32_S (w2[2]);
+    w2[3] = swap32_S (w2[3]);
+    w3[0] = swap32_S (w3[0]);
+    w3[1] = swap32_S (w3[1]);
+    w3[2] = swap32_S (w3[2]);
+    w3[3] = swap32_S (w3[3]);
+
+    sha1_update_64 (ctx, w0, w1, w2, w3, 32 * 2);
+  }
+
+  w0[0] = w[pos4 + 0];
+  w0[1] = w[pos4 + 1];
+  w0[2] = w[pos4 + 2];
+  w0[3] = w[pos4 + 3];
+  w1[0] = w[pos4 + 4];
+  w1[1] = w[pos4 + 5];
+  w1[2] = w[pos4 + 6];
+  w1[3] = w[pos4 + 7];
+
+  make_utf16be_S (w1, w2, w3);
+  make_utf16be_S (w0, w0, w1);
+
+  w0[0] = swap32_S (w0[0]);
+  w0[1] = swap32_S (w0[1]);
+  w0[2] = swap32_S (w0[2]);
+  w0[3] = swap32_S (w0[3]);
+  w1[0] = swap32_S (w1[0]);
+  w1[1] = swap32_S (w1[1]);
+  w1[2] = swap32_S (w1[2]);
+  w1[3] = swap32_S (w1[3]);
+  w2[0] = swap32_S (w2[0]);
+  w2[1] = swap32_S (w2[1]);
+  w2[2] = swap32_S (w2[2]);
+  w2[3] = swap32_S (w2[3]);
+  w3[0] = swap32_S (w3[0]);
+  w3[1] = swap32_S (w3[1]);
+  w3[2] = swap32_S (w3[2]);
+  w3[3] = swap32_S (w3[3]);
+
+  sha1_update_64 (ctx, w0, w1, w2, w3, (len - pos1) * 2);
+}
+
 void sha1_final (sha1_ctx_t *ctx)
 {
   const int pos = ctx->len & 63;
@@ -1699,6 +1935,48 @@ void sha1_update_vector_utf16le_swap (sha1_ctx_vector_t *ctx, const u32x *w, con
   sha1_update_vector_64 (ctx, w0, w1, w2, w3, (len - pos1) * 2);
 }
 
+void sha1_update_vector_utf16leN (sha1_ctx_vector_t *ctx, const u32x *w, const int len)
+{
+  u32x w0[4];
+  u32x w1[4];
+  u32x w2[4];
+  u32x w3[4];
+
+  int pos1;
+  int pos4;
+
+  for (pos1 = 0, pos4 = 0; pos1 < len - 32; pos1 += 32, pos4 += 8)
+  {
+    w0[0] = w[pos4 + 0];
+    w0[1] = w[pos4 + 1];
+    w0[2] = w[pos4 + 2];
+    w0[3] = w[pos4 + 3];
+    w1[0] = w[pos4 + 4];
+    w1[1] = w[pos4 + 5];
+    w1[2] = w[pos4 + 6];
+    w1[3] = w[pos4 + 7];
+
+    make_utf16leN (w1, w2, w3);
+    make_utf16leN (w0, w0, w1);
+
+    sha1_update_vector_64 (ctx, w0, w1, w2, w3, 32 * 2);
+  }
+
+  w0[0] = w[pos4 + 0];
+  w0[1] = w[pos4 + 1];
+  w0[2] = w[pos4 + 2];
+  w0[3] = w[pos4 + 3];
+  w1[0] = w[pos4 + 4];
+  w1[1] = w[pos4 + 5];
+  w1[2] = w[pos4 + 6];
+  w1[3] = w[pos4 + 7];
+
+  make_utf16leN (w1, w2, w3);
+  make_utf16leN (w0, w0, w1);
+
+  sha1_update_vector_64 (ctx, w0, w1, w2, w3, (len - pos1) * 2);
+}
+
 void sha1_update_vector_utf16beN (sha1_ctx_vector_t *ctx, const u32x *w, const int len)
 {
   u32x w0[4];
diff --git a/OpenCL/m15500_a0.cl b/OpenCL/m15500_a0.cl
new file mode 100644
index 000000000..c59255b70
--- /dev/null
+++ b/OpenCL/m15500_a0.cl
@@ -0,0 +1,168 @@
+/**
+ * Author......: See docs/credits.txt
+ * License.....: MIT
+ */
+
+//#define NEW_SIMD_CODE
+
+#include "inc_vendor.cl"
+#include "inc_hash_constants.h"
+#include "inc_hash_functions.cl"
+#include "inc_types.cl"
+#include "inc_common.cl"
+#include "inc_rp.h"
+#include "inc_rp.cl"
+#include "inc_scalar.cl"
+#include "inc_hash_sha1.cl"
+
+__kernel void m15500_mxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+{
+  /**
+   * modifier
+   */
+
+  const u32 lid = get_local_id (0);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * base
+   */
+
+  const u32 pw_len = pws[gid].pw_len;
+
+  const u32 pw_lenv = ceil ((float) pw_len / 4);
+
+  u32 w[64] = { 0 };
+
+  for (int idx = 0; idx < pw_lenv; idx++)
+  {
+    w[idx] = pws[gid].i[idx];
+  }
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32 s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = swap32_S (salt_bufs[salt_pos].salt_buf[idx]);
+  }
+
+  /**
+   * loop
+   */
+
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos++)
+  {
+    // todo: add rules engine
+
+    sha1_ctx_t ctx;
+
+    sha1_init (&ctx);
+
+    sha1_update_utf16be_swap (&ctx, w, pw_len);
+
+    sha1_update (&ctx, s, salt_len);
+
+    sha1_final (&ctx);
+
+    ctx.h[0] &= 0xff000000;
+    ctx.h[1] &= 0x0000ffff;
+    ctx.h[2] &= 0xffffffff;
+    ctx.h[3] &= 0xffffffff;
+    ctx.h[4] &= 0xffffffff;
+
+    const u32 r0 = ctx.h[DGST_R0];
+    const u32 r1 = ctx.h[DGST_R1];
+    const u32 r2 = ctx.h[DGST_R2];
+    const u32 r3 = ctx.h[DGST_R3];
+
+    COMPARE_M_SCALAR (r0, r1, r2, r3);
+  }
+}
+
+__kernel void m15500_sxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+{
+  /**
+   * modifier
+   */
+
+  const u32 lid = get_local_id (0);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * digest
+   */
+
+  const u32 search[4] =
+  {
+    digests_buf[digests_offset].digest_buf[DGST_R0],
+    digests_buf[digests_offset].digest_buf[DGST_R1],
+    digests_buf[digests_offset].digest_buf[DGST_R2],
+    digests_buf[digests_offset].digest_buf[DGST_R3]
+  };
+
+  /**
+   * base
+   */
+
+  const u32 pw_len = pws[gid].pw_len;
+
+  const u32 pw_lenv = ceil ((float) pw_len / 4);
+
+  u32 w[64] = { 0 };
+
+  for (int idx = 0; idx < pw_lenv; idx++)
+  {
+    w[idx] = pws[gid].i[idx];
+  }
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32 s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = swap32_S (salt_bufs[salt_pos].salt_buf[idx]);
+  }
+
+  /**
+   * loop
+   */
+
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos++)
+  {
+    // todo: add rules engine
+
+    sha1_ctx_t ctx;
+
+    sha1_init (&ctx);
+
+    sha1_update_utf16be_swap (&ctx, w, pw_len);
+
+    sha1_update (&ctx, s, salt_len);
+
+    sha1_final (&ctx);
+
+    ctx.h[0] &= 0xff000000;
+    ctx.h[1] &= 0x0000ffff;
+    ctx.h[2] &= 0xffffffff;
+    ctx.h[3] &= 0xffffffff;
+    ctx.h[4] &= 0xffffffff;
+
+    const u32 r0 = ctx.h[DGST_R0];
+    const u32 r1 = ctx.h[DGST_R1];
+    const u32 r2 = ctx.h[DGST_R2];
+    const u32 r3 = ctx.h[DGST_R3];
+
+    COMPARE_S_SCALAR (r0, r1, r2, r3);
+  }
+}
diff --git a/OpenCL/m15500_a1.cl b/OpenCL/m15500_a1.cl
new file mode 100644
index 000000000..a811e89cb
--- /dev/null
+++ b/OpenCL/m15500_a1.cl
@@ -0,0 +1,148 @@
+/**
+ * Author......: See docs/credits.txt
+ * License.....: MIT
+ */
+
+//#define NEW_SIMD_CODE
+
+#include "inc_vendor.cl"
+#include "inc_hash_constants.h"
+#include "inc_hash_functions.cl"
+#include "inc_types.cl"
+#include "inc_common.cl"
+#include "inc_scalar.cl"
+#include "inc_hash_sha1.cl"
+
+__kernel void m15500_mxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+{
+  /**
+   * modifier
+   */
+
+  const u32 lid = get_local_id (0);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * base
+   */
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32 s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = swap32_S (salt_bufs[salt_pos].salt_buf[idx]);
+  }
+
+  sha1_ctx_t ctx0;
+
+  sha1_init (&ctx0);
+
+  sha1_update_global_utf16be_swap (&ctx0, pws[gid].i, pws[gid].pw_len);
+
+  /**
+   * loop
+   */
+
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos++)
+  {
+    sha1_ctx_t ctx = ctx0;
+
+    sha1_update_global_utf16be_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len);
+
+    sha1_update (&ctx, s, salt_len);
+
+    sha1_final (&ctx);
+
+    ctx.h[0] &= 0xff000000;
+    ctx.h[1] &= 0x0000ffff;
+    ctx.h[2] &= 0xffffffff;
+    ctx.h[3] &= 0xffffffff;
+    ctx.h[4] &= 0xffffffff;
+
+    const u32 r0 = ctx.h[DGST_R0];
+    const u32 r1 = ctx.h[DGST_R1];
+    const u32 r2 = ctx.h[DGST_R2];
+    const u32 r3 = ctx.h[DGST_R3];
+
+    COMPARE_M_SCALAR (r0, r1, r2, r3);
+  }
+}
+
+__kernel void m15500_sxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+{
+  /**
+   * modifier
+   */
+
+  const u32 lid = get_local_id (0);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * digest
+   */
+
+  const u32 search[4] =
+  {
+    digests_buf[digests_offset].digest_buf[DGST_R0],
+    digests_buf[digests_offset].digest_buf[DGST_R1],
+    digests_buf[digests_offset].digest_buf[DGST_R2],
+    digests_buf[digests_offset].digest_buf[DGST_R3]
+  };
+
+  /**
+   * base
+   */
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32 s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = swap32_S (salt_bufs[salt_pos].salt_buf[idx]);
+  }
+
+  sha1_ctx_t ctx0;
+
+  sha1_init (&ctx0);
+
+  sha1_update_global_utf16be_swap (&ctx0, pws[gid].i, pws[gid].pw_len);
+
+  /**
+   * loop
+   */
+
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos++)
+  {
+    sha1_ctx_t ctx = ctx0;
+
+    sha1_update_global_utf16be_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len);
+
+    sha1_update (&ctx, s, salt_len);
+
+    sha1_final (&ctx);
+
+    ctx.h[0] &= 0xff000000;
+    ctx.h[1] &= 0x0000ffff;
+    ctx.h[2] &= 0xffffffff;
+    ctx.h[3] &= 0xffffffff;
+    ctx.h[4] &= 0xffffffff;
+
+    const u32 r0 = ctx.h[DGST_R0];
+    const u32 r1 = ctx.h[DGST_R1];
+    const u32 r2 = ctx.h[DGST_R2];
+    const u32 r3 = ctx.h[DGST_R3];
+
+    COMPARE_S_SCALAR (r0, r1, r2, r3);
+  }
+}
diff --git a/OpenCL/m15500_a3.cl b/OpenCL/m15500_a3.cl
new file mode 100644
index 000000000..66605f171
--- /dev/null
+++ b/OpenCL/m15500_a3.cl
@@ -0,0 +1,178 @@
+/**
+ * Author......: See docs/credits.txt
+ * License.....: MIT
+ */
+
+#define NEW_SIMD_CODE
+
+#include "inc_vendor.cl"
+#include "inc_hash_constants.h"
+#include "inc_hash_functions.cl"
+#include "inc_types.cl"
+#include "inc_common.cl"
+#include "inc_simd.cl"
+#include "inc_hash_sha1.cl"
+
+__kernel void m15500_mxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __constant const u32x *words_buf_r, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+{
+  /**
+   * modifier
+   */
+
+  const u32 lid = get_local_id (0);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * base
+   */
+
+  const u32 pw_len = pws[gid].pw_len;
+
+  const u32 pw_lenv = ceil ((float) pw_len / 4);
+
+  u32x w[64] = { 0 };
+
+  for (int idx = 0; idx < pw_lenv; idx++)
+  {
+    w[idx] = pws[gid].i[idx];
+  }
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32x s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = swap32_S (salt_bufs[salt_pos].salt_buf[idx]);
+  }
+
+  /**
+   * loop
+   */
+
+  u32x w0l = w[0];
+
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
+  {
+    const u32x w0r = words_buf_r[il_pos / VECT_SIZE];
+
+    const u32x w0 = w0l | w0r;
+
+    w[0] = w0;
+
+    sha1_ctx_vector_t ctx;
+
+    sha1_init_vector (&ctx);
+
+    sha1_update_vector_utf16leN (&ctx, w, pw_len);
+
+    sha1_update_vector (&ctx, s, salt_len);
+
+    sha1_final_vector (&ctx);
+
+    ctx.h[0] &= 0xff000000;
+    ctx.h[1] &= 0x0000ffff;
+    ctx.h[2] &= 0xffffffff;
+    ctx.h[3] &= 0xffffffff;
+    ctx.h[4] &= 0xffffffff;
+
+    const u32x r0 = ctx.h[DGST_R0];
+    const u32x r1 = ctx.h[DGST_R1];
+    const u32x r2 = ctx.h[DGST_R2];
+    const u32x r3 = ctx.h[DGST_R3];
+
+    COMPARE_M_SIMD (r0, r1, r2, r3);
+  }
+}
+
+__kernel void m15500_sxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __constant const u32x *words_buf_r, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+{
+  /**
+   * modifier
+   */
+
+  const u32 lid = get_local_id (0);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * digest
+   */
+
+  const u32 search[4] =
+  {
+    digests_buf[digests_offset].digest_buf[DGST_R0],
+    digests_buf[digests_offset].digest_buf[DGST_R1],
+    digests_buf[digests_offset].digest_buf[DGST_R2],
+    digests_buf[digests_offset].digest_buf[DGST_R3]
+  };
+
+  /**
+   * base
+   */
+
+  const u32 pw_len = pws[gid].pw_len;
+
+  const u32 pw_lenv = ceil ((float) pw_len / 4);
+
+  u32x w[64] = { 0 };
+
+  for (int idx = 0; idx < pw_lenv; idx++)
+  {
+    w[idx] = pws[gid].i[idx];
+  }
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32x s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = swap32_S (salt_bufs[salt_pos].salt_buf[idx]);
+  }
+
+  /**
+   * loop
+   */
+
+  u32x w0l = w[0];
+
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
+  {
+    const u32x w0r = words_buf_r[il_pos / VECT_SIZE];
+
+    const u32x w0 = w0l | w0r;
+
+    w[0] = w0;
+
+    sha1_ctx_vector_t ctx;
+
+    sha1_init_vector (&ctx);
+
+    sha1_update_vector_utf16leN (&ctx, w, pw_len);
+
+    sha1_update_vector (&ctx, s, salt_len);
+
+    sha1_final_vector (&ctx);
+
+    ctx.h[0] &= 0xff000000;
+    ctx.h[1] &= 0x0000ffff;
+    ctx.h[2] &= 0xffffffff;
+    ctx.h[3] &= 0xffffffff;
+    ctx.h[4] &= 0xffffffff;
+
+    const u32x r0 = ctx.h[DGST_R0];
+    const u32x r1 = ctx.h[DGST_R1];
+    const u32x r2 = ctx.h[DGST_R2];
+    const u32x r3 = ctx.h[DGST_R3];
+
+    COMPARE_S_SIMD (r0, r1, r2, r3);
+  }
+}