attempt to convert SSE2 Fast Corner score code to ARM Neon

I was trying to port some SSE2 code (fast corner detector score computation) using ARM Neon instruction. The code is quite simple at first sight but the results are different for some reason. The thing is that sometimes the difference may be quite significant and sometimes differs by 2 or 3 values. It would be great if someone can explain why it can happen.

Here is the code

Original SSE2:

__m128i q0 = _mm_set1_epi16(-1000), q1 = _mm_set1_epi16(1000);
for( k = 0; k < 16; k += 8 )
{
    __m128i v0 = _mm_loadu_si128((__m128i*)(d+k+1));
    __m128i v1 = _mm_loadu_si128((__m128i*)(d+k+2));
    __m128i a = _mm_min_epi16(v0, v1);
    __m128i b = _mm_max_epi16(v0, v1);
    v0 = _mm_loadu_si128((__m128i*)(d+k+3));
    a = _mm_min_epi16(a, v0);
    b = _mm_max_epi16(b, v0);
    v0 = _mm_loadu_si128((__m128i*)(d+k+4));
    a = _mm_min_epi16(a, v0);
    b = _mm_max_epi16(b, v0);
    v0 = _mm_loadu_si128((__m128i*)(d+k+5));
    a = _mm_min_epi16(a, v0);
    b = _mm_max_epi16(b, v0);
    v0 = _mm_loadu_si128((__m128i*)(d+k+6));
    a = _mm_min_epi16(a, v0);
    b = _mm_max_epi16(b, v0);
    v0 = _mm_loadu_si128((__m128i*)(d+k+7));
    a = _mm_min_epi16(a, v0);
    b = _mm_max_epi16(b, v0);
    v0 = _mm_loadu_si128((__m128i*)(d+k+8));
    a = _mm_min_epi16(a, v0);
    b = _mm_max_epi16(b, v0);
    v0 = _mm_loadu_si128((__m128i*)(d+k));
    q0 = _mm_max_epi16(q0, _mm_min_epi16(a, v0));
    q1 = _mm_min_epi16(q1, _mm_max_epi16(b, v0));
    v0 = _mm_loadu_si128((__m128i*)(d+k+9));
    q0 = _mm_max_epi16(q0, _mm_min_epi16(a, v0));
    q1 = _mm_min_epi16(q1, _mm_max_epi16(b, v0));
}
q0 = _mm_max_epi16(q0, _mm_sub_epi16(_mm_setzero_si128(), q1));
q0 = _mm_max_epi16(q0, _mm_unpackhi_epi64(q0, q0));
q0 = _mm_max_epi16(q0, _mm_srli_si128(q0, 4));
q0 = _mm_max_epi16(q0, _mm_srli_si128(q0, 2));
threshold = (short)_mm_cvtsi128_si32(q0) - 1;

HERE IS WORKING ARM NEON CODE:

int16x8_t q0 = vdupq_n_s16(-1000), q1 = vdupq_n_s16(1000);
int16x8_t zero = vdupq_n_s16(0);
for( k = 0; k < 16; k += 8 )
{
    int16x8_t v0 = vld1q_s16((const int16_t*)(d+k+1));
    int16x8_t v1 = vld1q_s16((const int16_t*)(d+k+2));
    int16x8_t a = vminq_s16(v0, v1);
    int16x8_t b = vmaxq_s16(v0, v1);
    v0 = vld1q_s16((const int16_t*)(d+k+3));
    a = vminq_s16(a, v0);
    b = vmaxq_s16(b, v0);
    v0 = vld1q_s16((const int16_t*)(d+k+4));
    a = vminq_s16(a, v0);
    b = vmaxq_s16(b, v0);
    v0 = vld1q_s16((const int16_t*)(d+k+5));
    a = vminq_s16(a, v0);
    b = vmaxq_s16(b, v0);
    v0 = vld1q_s16((const int16_t*)(d+k+6));
    a = vminq_s16(a, v0);
    b = vmaxq_s16(b, v0);
    v0 = vld1q_s16((const int16_t*)(d+k+7));
    a = vminq_s16(a, v0);
    b = vmaxq_s16(b, v0);
    v0 = vld1q_s16((const int16_t*)(d+k+8));
    a = vminq_s16(a, v0);
    b = vmaxq_s16(b, v0);
    v0 = vld1q_s16((const int16_t*)(d+k));
    q0 = vmaxq_s16(q0, vminq_s16(a, v0));
    q1 = vminq_s16(q1, vmaxq_s16(b, v0));
    v0 = vld1q_s16((const int16_t*)(d+k+9));
    q0 = vmaxq_s16(q0, vminq_s16(a, v0));
    q1 = vminq_s16(q1, vmaxq_s16(b, v0));
}
q0 = vmaxq_s16(q0, vsubq_s16(zero, q1));
// first mistake it produce wrong result
//q0 = vmaxq_s16(q0, vzipq_s16(q0, q0).val[1]);
// may be someone knows faster/better way?
int16x4_t a_hi = vget_high_s16(q0);
q1 = vcombine_s16(a_hi, a_hi);
q0 = vmaxq_s16(q0, q1);

// this is _mm_srli_si128(q0, 4)
q1 = vextq_s16(q0, zero, 2);

q0 = vmaxq_s16(q0, q1);

// this is _mm_srli_si128(q0, 2)
q1 = vextq_s16(q0, zero, 1);
q0 = vmaxq_s16(q0, q1);

// read the result
int16_t __attribute__ ((aligned (16))) x[8];
vst1q_s16(x, q0);

threshold = x[0] - 1;

This code seems to be correct, so if anyone will find it useful... fast corners detector is the fastest at the moment. And with such small optimizations it may run even faster on mobiles.