static double x264_ssim( double ssim )
{
- return -10.0 * log10( 1 - ssim );
+ double inv_ssim = 1 - ssim;
+ if( inv_ssim <= 0.0000000001 ) /* Max 100dB */
+ return 100;
+
+ return -10.0 * log10( inv_ssim );
}
static void x264_frame_dump( x264_t *h )
FILE *f = fopen( h->param.psz_dump_yuv, "r+b" );
if( !f )
return;
+
/* Write the frame in display order */
- fseek( f, (uint64_t)h->fdec->i_frame * h->param.i_height * h->param.i_width * 3/2 * sizeof(pixel), SEEK_SET );
- for( int y = 0; y < h->param.i_height; y++ )
- fwrite( &h->fdec->plane[0][y*h->fdec->i_stride[0]], sizeof(pixel), h->param.i_width, f );
- int cw = h->param.i_width>>1;
- int ch = h->param.i_height>>1;
- pixel *planeu = x264_malloc( (cw*ch*2+32)*sizeof(pixel) );
- pixel *planev = planeu + cw*ch + 16;
- h->mc.plane_copy_deinterleave( planeu, cw, planev, cw, h->fdec->plane[1], h->fdec->i_stride[1], cw, ch );
- fwrite( planeu, 1, cw*ch*sizeof(pixel), f );
- fwrite( planev, 1, cw*ch*sizeof(pixel), f );
- x264_free( planeu );
+ int frame_size = FRAME_SIZE( h->param.i_height * h->param.i_width * sizeof(pixel) );
+ fseek( f, (uint64_t)h->fdec->i_frame * frame_size, SEEK_SET );
+ for( int p = 0; p < (CHROMA444 ? 3 : 1); p++ )
+ for( int y = 0; y < h->param.i_height; y++ )
+ fwrite( &h->fdec->plane[p][y*h->fdec->i_stride[p]], sizeof(pixel), h->param.i_width, f );
+ if( !CHROMA444 )
+ {
+ int cw = h->param.i_width>>1;
+ int ch = h->param.i_height>>CHROMA_V_SHIFT;
+ pixel *planeu = x264_malloc( (cw*ch*2+32)*sizeof(pixel) );
+ pixel *planev = planeu + cw*ch + 16;
+ h->mc.plane_copy_deinterleave( planeu, cw, planev, cw, h->fdec->plane[1], h->fdec->i_stride[1], cw, ch );
+ fwrite( planeu, 1, cw*ch*sizeof(pixel), f );
+ fwrite( planev, 1, cw*ch*sizeof(pixel), f );
+ x264_free( planeu );
+ }
fclose( f );
}
-
/* Fill "default" values */
static void x264_slice_header_init( x264_t *h, x264_slice_header_t *sh,
x264_sps_t *sps, x264_pps_t *pps,
if( sh->pps->b_pic_order && !sh->b_field_pic )
bs_write_se( s, sh->i_delta_poc_bottom );
}
- else if( sh->sps->i_poc_type == 1 && !sh->sps->b_delta_pic_order_always_zero )
- {
- bs_write_se( s, sh->i_delta_poc[0] );
- if( sh->pps->b_pic_order && !sh->b_field_pic )
- bs_write_se( s, sh->i_delta_poc[1] );
- }
if( sh->pps->b_redundant_pic_cnt )
bs_write_ue( s, sh->i_redundant_pic_cnt );
static int x264_bitstream_check_buffer( x264_t *h )
{
uint8_t *bs_bak = h->out.p_bitstream;
- if( (h->param.b_cabac && (h->cabac.p_end - h->cabac.p < 2500)) ||
- (h->out.bs.p_end - h->out.bs.p < 2500) )
+ int max_mb_size = 2500 << SLICE_MBAFF;
+ if( (h->param.b_cabac && (h->cabac.p_end - h->cabac.p < max_mb_size)) ||
+ (h->out.bs.p_end - h->out.bs.p < max_mb_size) )
{
h->out.i_bitstream += 100000;
CHECKED_MALLOC( h->out.p_bitstream, h->out.i_bitstream );
{
x264_log( h, X264_LOG_ERROR, "your cpu does not support SSE1, but x264 was compiled with asm support\n");
#else
- if( b_open && !(x264_cpu_detect() & X264_CPU_MMXEXT) )
+ if( b_open && !(x264_cpu_detect() & X264_CPU_MMX2) )
{
x264_log( h, X264_LOG_ERROR, "your cpu does not support MMXEXT, but x264 was compiled with asm support\n");
#endif
return -1;
}
#endif
+
+#if HAVE_INTERLACED
+ h->param.b_interlaced = !!PARAM_INTERLACED;
+#else
+ if( h->param.b_interlaced )
+ {
+ x264_log( h, X264_LOG_ERROR, "not compiled with interlaced support\n" );
+ return -1;
+ }
+#endif
+
if( h->param.i_width <= 0 || h->param.i_height <= 0 )
{
x264_log( h, X264_LOG_ERROR, "invalid width x height (%dx%d)\n",
return -1;
}
- if( h->param.i_width % 2 || h->param.i_height % 2 )
+ int i_csp = h->param.i_csp & X264_CSP_MASK;
+#if X264_CHROMA_FORMAT
+ if( CHROMA_FORMAT != CHROMA_420 && i_csp >= X264_CSP_I420 && i_csp <= X264_CSP_NV12 )
{
- x264_log( h, X264_LOG_ERROR, "width or height not divisible by 2 (%dx%d)\n",
- h->param.i_width, h->param.i_height );
+ x264_log( h, X264_LOG_ERROR, "not compiled with 4:2:0 support\n" );
return -1;
}
- int i_csp = h->param.i_csp & X264_CSP_MASK;
+ else if( CHROMA_FORMAT != CHROMA_422 && i_csp >= X264_CSP_I422 && i_csp <= X264_CSP_NV16 )
+ {
+ x264_log( h, X264_LOG_ERROR, "not compiled with 4:2:2 support\n" );
+ return -1;
+ }
+ else if( CHROMA_FORMAT != CHROMA_444 && i_csp >= X264_CSP_I444 && i_csp <= X264_CSP_RGB )
+ {
+ x264_log( h, X264_LOG_ERROR, "not compiled with 4:4:4 support\n" );
+ return -1;
+ }
+#endif
if( i_csp <= X264_CSP_NONE || i_csp >= X264_CSP_MAX )
{
- x264_log( h, X264_LOG_ERROR, "invalid CSP (only I420/YV12/NV12 supported)\n" );
+ x264_log( h, X264_LOG_ERROR, "invalid CSP (only I420/YV12/NV12/I422/YV16/NV16/I444/YV24/BGR/BGRA/RGB supported)\n" );
return -1;
}
-#if HAVE_INTERLACED
- h->param.b_interlaced = !!PARAM_INTERLACED;
-#else
- if( h->param.b_interlaced )
+ if( i_csp < X264_CSP_I444 && h->param.i_width % 2 )
{
- x264_log( h, X264_LOG_ERROR, "not compiled with interlaced support\n" );
+ x264_log( h, X264_LOG_ERROR, "width not divisible by 2 (%dx%d)\n",
+ h->param.i_width, h->param.i_height );
+ return -1;
+ }
+
+ if( i_csp < X264_CSP_I422 && PARAM_INTERLACED && h->param.i_height % 4 )
+ {
+ x264_log( h, X264_LOG_ERROR, "height not divisible by 4 (%dx%d)\n",
+ h->param.i_width, h->param.i_height );
+ return -1;
+ }
+
+ if( (i_csp < X264_CSP_I422 || PARAM_INTERLACED) && h->param.i_height % 2 )
+ {
+ x264_log( h, X264_LOG_ERROR, "height not divisible by 2 (%dx%d)\n",
+ h->param.i_width, h->param.i_height );
return -1;
}
-#endif
if( (h->param.crop_rect.i_left + h->param.crop_rect.i_right ) >= h->param.i_width ||
(h->param.crop_rect.i_top + h->param.crop_rect.i_bottom) >= h->param.i_height )
if( h->param.i_threads == X264_THREADS_AUTO )
h->param.i_threads = x264_cpu_num_processors() * (h->param.b_sliced_threads?2:3)/2;
- h->param.i_threads = x264_clip3( h->param.i_threads, 1, X264_THREAD_MAX );
if( h->param.i_threads > 1 )
{
#if !HAVE_THREAD
h->param.i_threads = X264_MIN( h->param.i_threads, max_threads );
}
}
- else
+ h->param.i_threads = x264_clip3( h->param.i_threads, 1, X264_THREAD_MAX );
+ if( h->param.i_threads == 1 )
h->param.b_sliced_threads = 0;
h->i_thread_frames = h->param.b_sliced_threads ? 1 : h->param.i_threads;
if( h->i_thread_frames > 1 )
h->param.rc.f_rf_constant = x264_clip3f( h->param.rc.f_rf_constant, -QP_BD_OFFSET, 51 );
h->param.rc.f_rf_constant_max = x264_clip3f( h->param.rc.f_rf_constant_max, -QP_BD_OFFSET, 51 );
h->param.rc.i_qp_constant = x264_clip3( h->param.rc.i_qp_constant, 0, QP_MAX );
- h->param.analyse.i_subpel_refine = x264_clip3( h->param.analyse.i_subpel_refine, 0, 10 );
+ h->param.analyse.i_subpel_refine = x264_clip3( h->param.analyse.i_subpel_refine, 0, 11 );
h->param.rc.f_ip_factor = X264_MAX( h->param.rc.f_ip_factor, 0.01f );
h->param.rc.f_pb_factor = X264_MAX( h->param.rc.f_pb_factor, 0.01f );
if( h->param.rc.i_rc_method == X264_RC_CRF )
h->param.analyse.inter &= ~X264_ANALYSE_I8x8;
h->param.analyse.intra &= ~X264_ANALYSE_I8x8;
}
- h->param.analyse.i_chroma_qp_offset = x264_clip3(h->param.analyse.i_chroma_qp_offset, -12, 12);
h->param.analyse.i_trellis = x264_clip3( h->param.analyse.i_trellis, 0, 2 );
h->param.rc.i_aq_mode = x264_clip3( h->param.rc.i_aq_mode, 0, 2 );
h->param.rc.f_aq_strength = x264_clip3f( h->param.rc.f_aq_strength, 0, 3 );
h->param.analyse.f_psy_trellis = x264_clip3f( h->param.analyse.f_psy_trellis, 0, 10 );
h->mb.i_psy_rd = h->param.analyse.i_subpel_refine >= 6 ? FIX8( h->param.analyse.f_psy_rd ) : 0;
h->mb.i_psy_trellis = h->param.analyse.i_trellis ? FIX8( h->param.analyse.f_psy_trellis / 4 ) : 0;
+ h->param.analyse.i_chroma_qp_offset = x264_clip3(h->param.analyse.i_chroma_qp_offset, -32, 32);
+ /* In 4:4:4 mode, chroma gets twice as much resolution, so we can halve its quality. */
+ if( b_open && i_csp >= X264_CSP_I444 && i_csp < X264_CSP_BGR && h->param.analyse.b_psy )
+ h->param.analyse.i_chroma_qp_offset += 6;
/* Psy RDO increases overall quantizers to improve the quality of luma--this indirectly hurts chroma quality */
/* so we lower the chroma QP offset to compensate */
if( b_open && h->mb.i_psy_rd )
h->param.rc.f_aq_strength = 0;
}
h->param.analyse.i_noise_reduction = x264_clip3( h->param.analyse.i_noise_reduction, 0, 1<<16 );
- if( h->param.analyse.i_subpel_refine == 10 && (h->param.analyse.i_trellis != 2 || !h->param.rc.i_aq_mode) )
+ if( h->param.analyse.i_subpel_refine >= 10 && (h->param.analyse.i_trellis != 2 || !h->param.rc.i_aq_mode) )
h->param.analyse.i_subpel_refine = 9;
{
int maxrate_bak = h->param.rc.i_vbv_max_bitrate;
if( h->param.rc.i_rc_method == X264_RC_ABR && h->param.rc.i_vbv_buffer_size <= 0 )
h->param.rc.i_vbv_max_bitrate = h->param.rc.i_bitrate * 2;
- h->sps = h->sps_array;
x264_sps_init( h->sps, h->param.i_sps_id, &h->param );
do h->param.i_level_idc = l->level_idc;
while( l[1].level_idc && x264_validate_levels( h, 0 ) && l++ );
memcpy( h->pixf.mbcmp, satd ? h->pixf.satd : h->pixf.sad_aligned, sizeof(h->pixf.mbcmp) );
memcpy( h->pixf.mbcmp_unaligned, satd ? h->pixf.satd : h->pixf.sad, sizeof(h->pixf.mbcmp_unaligned) );
h->pixf.intra_mbcmp_x3_16x16 = satd ? h->pixf.intra_satd_x3_16x16 : h->pixf.intra_sad_x3_16x16;
- h->pixf.intra_mbcmp_x3_8x8c = satd ? h->pixf.intra_satd_x3_8x8c : h->pixf.intra_sad_x3_8x8c;
+ h->pixf.intra_mbcmp_x3_8x16c = satd ? h->pixf.intra_satd_x3_8x16c : h->pixf.intra_sad_x3_8x16c;
+ h->pixf.intra_mbcmp_x3_8x8c = satd ? h->pixf.intra_satd_x3_8x8c : h->pixf.intra_sad_x3_8x8c;
h->pixf.intra_mbcmp_x3_8x8 = satd ? h->pixf.intra_sa8d_x3_8x8 : h->pixf.intra_sad_x3_8x8;
h->pixf.intra_mbcmp_x3_4x4 = satd ? h->pixf.intra_satd_x3_4x4 : h->pixf.intra_sad_x3_4x4;
+ h->pixf.intra_mbcmp_x9_4x4 = h->param.b_cpu_independent || h->mb.b_lossless ? NULL
+ : satd ? h->pixf.intra_satd_x9_4x4 : h->pixf.intra_sad_x9_4x4;
+ h->pixf.intra_mbcmp_x9_8x8 = h->param.b_cpu_independent || h->mb.b_lossless ? NULL
+ : satd ? h->pixf.intra_sa8d_x9_8x8 : h->pixf.intra_sad_x9_8x8;
satd &= h->param.analyse.i_me_method == X264_ME_TESA;
memcpy( h->pixf.fpelcmp, satd ? h->pixf.satd : h->pixf.sad, sizeof(h->pixf.fpelcmp) );
memcpy( h->pixf.fpelcmp_x3, satd ? h->pixf.satd_x3 : h->pixf.sad_x3, sizeof(h->pixf.fpelcmp_x3) );
memcpy( h->pixf.fpelcmp_x4, satd ? h->pixf.satd_x4 : h->pixf.sad_x4, sizeof(h->pixf.fpelcmp_x4) );
}
+static void chroma_dsp_init( x264_t *h )
+{
+ memcpy( h->luma2chroma_pixel, x264_luma2chroma_pixel[CHROMA_FORMAT], sizeof(h->luma2chroma_pixel) );
+
+ switch( CHROMA_FORMAT )
+ {
+ case CHROMA_420:
+ memcpy( h->predict_chroma, h->predict_8x8c, sizeof(h->predict_chroma) );
+ h->mc.prefetch_fenc = h->mc.prefetch_fenc_420;
+ h->loopf.deblock_chroma[0] = h->loopf.deblock_h_chroma_420;
+ h->loopf.deblock_chroma_intra[0] = h->loopf.deblock_h_chroma_420_intra;
+ h->loopf.deblock_chroma_mbaff = h->loopf.deblock_chroma_420_mbaff;
+ h->loopf.deblock_chroma_intra_mbaff = h->loopf.deblock_chroma_420_intra_mbaff;
+ h->pixf.intra_mbcmp_x3_chroma = h->pixf.intra_mbcmp_x3_8x8c;
+ h->quantf.coeff_last[DCT_CHROMA_DC] = h->quantf.coeff_last4;
+ h->quantf.coeff_level_run[DCT_CHROMA_DC] = h->quantf.coeff_level_run4;
+ break;
+ case CHROMA_422:
+ memcpy( h->predict_chroma, h->predict_8x16c, sizeof(h->predict_chroma) );
+ h->mc.prefetch_fenc = h->mc.prefetch_fenc_422;
+ h->loopf.deblock_chroma[0] = h->loopf.deblock_h_chroma_422;
+ h->loopf.deblock_chroma_intra[0] = h->loopf.deblock_h_chroma_422_intra;
+ h->loopf.deblock_chroma_mbaff = h->loopf.deblock_chroma_422_mbaff;
+ h->loopf.deblock_chroma_intra_mbaff = h->loopf.deblock_chroma_422_intra_mbaff;
+ h->pixf.intra_mbcmp_x3_chroma = h->pixf.intra_mbcmp_x3_8x16c;
+ h->quantf.coeff_last[DCT_CHROMA_DC] = h->quantf.coeff_last8;
+ h->quantf.coeff_level_run[DCT_CHROMA_DC] = h->quantf.coeff_level_run8;
+ break;
+ case CHROMA_444:
+ h->mc.prefetch_fenc = h->mc.prefetch_fenc_422; /* FIXME: doesn't cover V plane */
+ h->loopf.deblock_chroma_mbaff = h->loopf.deblock_luma_mbaff;
+ h->loopf.deblock_chroma_intra_mbaff = h->loopf.deblock_luma_intra_mbaff;
+ break;
+ }
+}
+
static void x264_set_aspect_ratio( x264_t *h, x264_param_t *param, int initial )
{
/* VUI */
goto fail;
}
- h->sps = &h->sps_array[0];
x264_sps_init( h->sps, h->param.i_sps_id, &h->param );
-
- h->pps = &h->pps_array[0];
x264_pps_init( h->pps, h->param.i_sps_id, &h->param, h->sps );
x264_set_aspect_ratio( h, &h->param, 1 );
h->mb.i_mb_width = h->sps->i_mb_width;
h->mb.i_mb_height = h->sps->i_mb_height;
h->mb.i_mb_count = h->mb.i_mb_width * h->mb.i_mb_height;
+
+ h->mb.chroma_h_shift = CHROMA_FORMAT == CHROMA_420 || CHROMA_FORMAT == CHROMA_422;
+ h->mb.chroma_v_shift = CHROMA_FORMAT == CHROMA_420;
+
/* Adaptive MBAFF and subme 0 are not supported as we require halving motion
* vectors during prediction, resulting in hpel mvs.
* The chosen solution is to make MBAFF non-adaptive in this case. */
/* init CPU functions */
x264_predict_16x16_init( h->param.cpu, h->predict_16x16 );
x264_predict_8x8c_init( h->param.cpu, h->predict_8x8c );
+ x264_predict_8x16c_init( h->param.cpu, h->predict_8x16c );
x264_predict_8x8_init( h->param.cpu, h->predict_8x8, &h->predict_8x8_filter );
x264_predict_4x4_init( h->param.cpu, h->predict_4x4 );
- if( h->param.b_cabac )
- x264_cabac_init();
- else
- x264_cavlc_init();
x264_pixel_init( h->param.cpu, &h->pixf );
x264_dct_init( h->param.cpu, &h->dctf );
x264_zigzag_init( h->param.cpu, &h->zigzagf_progressive, &h->zigzagf_interlaced );
x264_quant_init( h, h->param.cpu, &h->quantf );
x264_deblock_init( h->param.cpu, &h->loopf, PARAM_INTERLACED );
x264_bitstream_init( h->param.cpu, &h->bsf );
- x264_dct_init_weights();
+ if( h->param.b_cabac )
+ x264_cabac_init( h );
+ else
+ x264_cavlc_init( h );
mbcmp_init( h );
+ chroma_dsp_init( h );
p = buf + sprintf( buf, "using cpu capabilities:" );
for( int i = 0; x264_cpu_names[i].flags; i++ )
else
h->thread[i]->fdec = h->thread[0]->fdec;
- h->thread[i]->sps = &h->thread[i]->sps_array[0];
- h->thread[i]->pps = &h->thread[i]->pps_array[0];
-
CHECKED_MALLOC( h->thread[i]->out.p_bitstream, h->out.i_bitstream );
/* Start each thread with room for init_nal_count NAL units; it'll realloc later if needed. */
CHECKED_MALLOC( h->thread[i]->out.nal, init_nal_count*sizeof(x264_nal_t) );
h->sps->i_profile_idc == PROFILE_MAIN ? "Main" :
h->sps->i_profile_idc == PROFILE_HIGH ? "High" :
h->sps->i_profile_idc == PROFILE_HIGH10 ? (h->sps->b_constraint_set3 == 1 ? "High 10 Intra" : "High 10") :
- "High 4:4:4 Predictive";
+ h->sps->i_profile_idc == PROFILE_HIGH422 ? (h->sps->b_constraint_set3 == 1 ? "High 4:2:2 Intra" : "High 4:2:2") :
+ h->sps->b_constraint_set3 == 1 ? "High 4:4:4 Intra" : "High 4:4:4 Predictive";
char level[4];
snprintf( level, sizeof(level), "%d.%d", h->sps->i_level_idc/10, h->sps->i_level_idc%10 );
- if( h->sps->i_level_idc == 9 || ( h->sps->i_level_idc == 11 && h->sps->b_constraint_set3 ) )
+ if( h->sps->i_level_idc == 9 || ( h->sps->i_level_idc == 11 && h->sps->b_constraint_set3 &&
+ (h->sps->i_profile_idc >= PROFILE_BASELINE && h->sps->i_profile_idc <= PROFILE_EXTENDED) ) )
strcpy( level, "1b" );
if( h->sps->i_profile_idc < PROFILE_HIGH10 )
}
else
{
- x264_log( h, X264_LOG_INFO, "profile %s, level %s, bit depth %d\n",
- profile, level, BIT_DEPTH );
+ static const char * const subsampling[4] = { "4:0:0", "4:2:0", "4:2:2", "4:4:4" };
+ x264_log( h, X264_LOG_INFO, "profile %s, level %s, %s %d-bit\n",
+ profile, level, subsampling[CHROMA_FORMAT], BIT_DEPTH );
}
return h;
static int x264_nal_end( x264_t *h )
{
x264_nal_t *nal = &h->out.nal[h->out.i_nal];
- nal->i_payload = &h->out.p_bitstream[bs_pos( &h->out.bs ) / 8] - nal->p_payload;
+ uint8_t *end = &h->out.p_bitstream[bs_pos( &h->out.bs ) / 8];
+ nal->i_payload = end - nal->p_payload;
+ /* nal_escape_mmx reads past the end of the input.
+ * While undefined padding wouldn't actually affect the output, it makes valgrind unhappy. */
+ memset( end, 0xff, 32 );
if( h->param.nalu_process )
h->param.nalu_process( h, nal );
h->out.i_nal++;
/* generate picture parameters */
x264_nal_start( h, NAL_PPS, NAL_PRIORITY_HIGHEST );
- x264_pps_write( &h->out.bs, h->pps );
+ x264_pps_write( &h->out.bs, h->sps, h->pps );
if( x264_nal_end( h ) )
return -1;
/* Duplication is a hack to compensate for crappy rounding in motion compensation.
* With high bit depth, it's not worth doing, so turn it off except in the case of
* unweighted dupes. */
- if( BIT_DEPTH > 8 && w != weight_none )
+ if( BIT_DEPTH > 8 && w != x264_weight_none )
return -1;
newframe = x264_frame_pop_blank_unused( h );
{
/* for now no analysis and set all weights to nothing */
for( int i_ref = 0; i_ref < h->i_ref[0]; i_ref++ )
- h->fenc->weighted[i_ref] = h->fref[0][i_ref]->filtered[0];
+ h->fenc->weighted[i_ref] = h->fref[0][i_ref]->filtered[0][0];
// FIXME: This only supports weighting of one reference frame
// and duplicates of that frame.
//scale full resolution frame
if( h->param.i_threads == 1 )
{
- pixel *src = h->fref[0][j]->filtered[0] - h->fref[0][j]->i_stride[0]*i_padv - PADH;
+ pixel *src = h->fref[0][j]->filtered[0][0] - h->fref[0][j]->i_stride[0]*i_padv - PADH;
pixel *dst = h->fenc->weighted[j] - h->fenc->i_stride[0]*i_padv - PADH;
int stride = h->fenc->i_stride[0];
int width = h->fenc->i_width[0] + PADH*2;
{
SET_WEIGHT( h->fenc->weight[0][0], 1, 1, 0, h->fenc->weight[0][0].i_offset );
}
- x264_weighted_reference_duplicate( h, 0, weight_none );
+ x264_weighted_reference_duplicate( h, 0, x264_weight_none );
if( h->fenc->weight[0][0].i_offset > -128 )
{
w[0] = h->fenc->weight[0][0];
* but the actual image data is equivalent. For now, maintain this
* consistency by copying deblocked pixels between planes. */
if( PARAM_INTERLACED )
- for( int p = 0; p < 2; p++ )
- for( int i = minpix_y>>p; i < maxpix_y>>p; i++ )
+ for( int p = 0; p < h->fdec->i_plane; p++ )
+ for( int i = minpix_y>>(CHROMA_V_SHIFT && p); i < maxpix_y>>(CHROMA_V_SHIFT && p); i++ )
memcpy( h->fdec->plane_fld[p] + i*h->fdec->i_stride[p],
h->fdec->plane[p] + i*h->fdec->i_stride[p],
h->mb.i_mb_width*16*sizeof(pixel) );
}
if( SLICE_MBAFF )
- for( int i = 0; i < 2; i++ )
+ for( int i = 0; i < 3; i++ )
{
XCHG( pixel *, h->intra_border_backup[0][i], h->intra_border_backup[3][i] );
XCHG( pixel *, h->intra_border_backup[1][i], h->intra_border_backup[4][i] );
maxpix_y = X264_MIN( maxpix_y, h->param.i_height );
if( h->param.analyse.b_psnr )
{
- uint64_t ssd_y = x264_pixel_ssd_wxh( &h->pixf,
- h->fdec->plane[0] + minpix_y * h->fdec->i_stride[0], h->fdec->i_stride[0],
- h->fenc->plane[0] + minpix_y * h->fenc->i_stride[0], h->fenc->i_stride[0],
- h->param.i_width, maxpix_y-minpix_y );
- uint64_t ssd_u, ssd_v;
- x264_pixel_ssd_nv12( &h->pixf,
- h->fdec->plane[1] + (minpix_y>>1) * h->fdec->i_stride[1], h->fdec->i_stride[1],
- h->fenc->plane[1] + (minpix_y>>1) * h->fenc->i_stride[1], h->fenc->i_stride[1],
- h->param.i_width>>1, (maxpix_y-minpix_y)>>1, &ssd_u, &ssd_v );
- h->stat.frame.i_ssd[0] += ssd_y;
- h->stat.frame.i_ssd[1] += ssd_u;
- h->stat.frame.i_ssd[2] += ssd_v;
+ for( int p = 0; p < (CHROMA444 ? 3 : 1); p++ )
+ h->stat.frame.i_ssd[p] += x264_pixel_ssd_wxh( &h->pixf,
+ h->fdec->plane[p] + minpix_y * h->fdec->i_stride[p], h->fdec->i_stride[p],
+ h->fenc->plane[p] + minpix_y * h->fenc->i_stride[p], h->fenc->i_stride[p],
+ h->param.i_width, maxpix_y-minpix_y );
+ if( !CHROMA444 )
+ {
+ uint64_t ssd_u, ssd_v;
+ int v_shift = CHROMA_V_SHIFT;
+ x264_pixel_ssd_nv12( &h->pixf,
+ h->fdec->plane[1] + (minpix_y>>v_shift) * h->fdec->i_stride[1], h->fdec->i_stride[1],
+ h->fenc->plane[1] + (minpix_y>>v_shift) * h->fenc->i_stride[1], h->fenc->i_stride[1],
+ h->param.i_width>>1, (maxpix_y-minpix_y)>>v_shift, &ssd_u, &ssd_v );
+ h->stat.frame.i_ssd[1] += ssd_u;
+ h->stat.frame.i_ssd[2] += ssd_v;
+ }
}
if( h->param.analyse.b_ssim )
{
+ int ssim_cnt;
x264_emms();
/* offset by 2 pixels to avoid alignment of ssim blocks with dct blocks,
* and overlap by 4 */
x264_pixel_ssim_wxh( &h->pixf,
h->fdec->plane[0] + 2+minpix_y*h->fdec->i_stride[0], h->fdec->i_stride[0],
h->fenc->plane[0] + 2+minpix_y*h->fenc->i_stride[0], h->fenc->i_stride[0],
- h->param.i_width-2, maxpix_y-minpix_y, h->scratch_buffer );
+ h->param.i_width-2, maxpix_y-minpix_y, h->scratch_buffer, &ssim_cnt );
+ h->stat.frame.i_ssim_cnt += ssim_cnt;
}
}
}
h->fdec->i_delta_poc[0] = h->sh.i_delta_poc_bottom == -1;
h->fdec->i_delta_poc[1] = h->sh.i_delta_poc_bottom == 1;
}
- else if( h->sps->i_poc_type == 1 )
- {
- /* FIXME TODO FIXME */
- }
else
{
/* Nothing to do ? */
bs_align_1( &h->out.bs );
/* init cabac */
- x264_cabac_context_init( &h->cabac, h->sh.i_type, x264_clip3( h->sh.i_qp-QP_BD_OFFSET, 0, 51 ), h->sh.i_cabac_init_idc );
+ x264_cabac_context_init( h, &h->cabac, h->sh.i_type, x264_clip3( h->sh.i_qp-QP_BD_OFFSET, 0, 51 ), h->sh.i_cabac_init_idc );
x264_cabac_encode_init ( &h->cabac, h->out.bs.p, h->out.bs.p_end );
last_emu_check = h->cabac.p;
}
i_mb_y = h->sh.i_first_mb / h->mb.i_mb_width;
i_mb_x = h->sh.i_first_mb % h->mb.i_mb_width;
i_skip = 0;
- int mb_size[2];
while( 1 )
{
mb_xy = i_mb_x + i_mb_y * h->mb.i_mb_width;
int mb_spos = bs_pos(&h->out.bs) + x264_cabac_pos(&h->cabac);
- if( x264_bitstream_check_buffer( h ) )
- return -1;
-
- if( back_up_bitstream && (!SLICE_MBAFF || (i_mb_y&1) == 0) )
+ if( !(i_mb_y & SLICE_MBAFF) )
{
- mv_bits_bak = h->stat.frame.i_mv_bits;
- tex_bits_bak = h->stat.frame.i_tex_bits;
- /* We don't need the contexts because flushing the CABAC encoder has no context
- * dependency and macroblocks are only re-encoded in the case where a slice is
- * ended (and thus the content of all contexts are thrown away). */
- if( h->param.b_cabac )
- {
- memcpy( &cabac_bak, &h->cabac, offsetof(x264_cabac_t, f8_bits_encoded) );
- /* x264's CABAC writer modifies the previous byte during carry, so it has to be
- * backed up. */
- cabac_prevbyte_bak = h->cabac.p[-1];
- }
- else
+ if( x264_bitstream_check_buffer( h ) )
+ return -1;
+
+ if( back_up_bitstream )
{
- bs_bak = h->out.bs;
- i_skip_bak = i_skip;
+ mv_bits_bak = h->stat.frame.i_mv_bits;
+ tex_bits_bak = h->stat.frame.i_tex_bits;
+ /* We don't need the contexts because flushing the CABAC encoder has no context
+ * dependency and macroblocks are only re-encoded in the case where a slice is
+ * ended (and thus the content of all contexts are thrown away). */
+ if( h->param.b_cabac )
+ {
+ memcpy( &cabac_bak, &h->cabac, offsetof(x264_cabac_t, f8_bits_encoded) );
+ /* x264's CABAC writer modifies the previous byte during carry, so it has to be
+ * backed up. */
+ cabac_prevbyte_bak = h->cabac.p[-1];
+ }
+ else
+ {
+ bs_bak = h->out.bs;
+ i_skip_bak = i_skip;
+ }
}
}
if( !(i_mb_y&1) )
{
/* FIXME: VSAD is fast but fairly poor at choosing the best interlace type. */
- int stride = h->fenc->i_stride[0];
- pixel *fenc = h->fenc->plane[0] + 16 * (i_mb_x + i_mb_y * stride);
- h->mb.b_interlaced = x264_field_vsad( h, fenc, stride );
+ h->mb.b_interlaced = x264_field_vsad( h, i_mb_x, i_mb_y );
memcpy( &h->zigzagf, MB_INTERLACED ? &h->zigzagf_interlaced : &h->zigzagf_progressive, sizeof(h->zigzagf) );
+ if( !MB_INTERLACED && (i_mb_y+2) == h->mb.i_mb_height )
+ x264_expand_border_mbpair( h, i_mb_x, i_mb_y );
}
}
h->mb.field[mb_xy] = MB_INTERLACED;
}
int total_bits = bs_pos(&h->out.bs) + x264_cabac_pos(&h->cabac);
- mb_size[i_mb_y&1] = total_bits - mb_spos;
+ int mb_size = total_bits - mb_spos;
if( slice_max_size )
{
h->stat.frame.i_mb_count[h->mb.i_type]++;
int b_intra = IS_INTRA( h->mb.i_type );
+ int b_skip = IS_SKIP( h->mb.i_type );
if( h->param.i_log_level >= X264_LOG_INFO || h->param.rc.b_stat_write )
{
- if( !b_intra && !IS_SKIP( h->mb.i_type ) && !IS_DIRECT( h->mb.i_type ) )
+ if( !b_intra && !b_skip && !IS_DIRECT( h->mb.i_type ) )
{
if( h->mb.i_partition != D_8x8 )
h->stat.frame.i_mb_partition[h->mb.i_partition] += 4;
{
if( h->mb.i_cbp_luma | h->mb.i_cbp_chroma )
{
- int cbpsum = (h->mb.i_cbp_luma&1) + ((h->mb.i_cbp_luma>>1)&1)
- + ((h->mb.i_cbp_luma>>2)&1) + (h->mb.i_cbp_luma>>3);
- h->stat.frame.i_mb_cbp[!b_intra + 0] += cbpsum;
- h->stat.frame.i_mb_cbp[!b_intra + 2] += !!h->mb.i_cbp_chroma;
- h->stat.frame.i_mb_cbp[!b_intra + 4] += h->mb.i_cbp_chroma >> 1;
+ if( CHROMA444 )
+ {
+ for( int i = 0; i < 4; i++ )
+ if( h->mb.i_cbp_luma & (1 << i) )
+ for( int p = 0; p < 3; p++ )
+ {
+ int s8 = i*4+p*16;
+ int nnz8x8 = M16( &h->mb.cache.non_zero_count[x264_scan8[s8]+0] )
+ | M16( &h->mb.cache.non_zero_count[x264_scan8[s8]+8] );
+ h->stat.frame.i_mb_cbp[!b_intra + p*2] += !!nnz8x8;
+ }
+ }
+ else
+ {
+ int cbpsum = (h->mb.i_cbp_luma&1) + ((h->mb.i_cbp_luma>>1)&1)
+ + ((h->mb.i_cbp_luma>>2)&1) + (h->mb.i_cbp_luma>>3);
+ h->stat.frame.i_mb_cbp[!b_intra + 0] += cbpsum;
+ h->stat.frame.i_mb_cbp[!b_intra + 2] += !!h->mb.i_cbp_chroma;
+ h->stat.frame.i_mb_cbp[!b_intra + 4] += h->mb.i_cbp_chroma >> 1;
+ }
}
if( h->mb.i_cbp_luma && !b_intra )
{
else //if( h->mb.i_type == I_4x4 )
for( int i = 0; i < 16; i++ )
h->stat.frame.i_mb_pred_mode[2][h->mb.cache.intra4x4_pred_mode[x264_scan8[i]]]++;
- h->stat.frame.i_mb_pred_mode[3][x264_mb_pred_mode8x8c_fix[h->mb.i_chroma_pred_mode]]++;
+ h->stat.frame.i_mb_pred_mode[3][x264_mb_chroma_pred_mode_fix[h->mb.i_chroma_pred_mode]]++;
}
+ h->stat.frame.i_mb_field[b_intra?0:b_skip?2:1] += MB_INTERLACED;
}
/* calculate deblock strength values (actual deblocking is done per-row along with hpel) */
if( b_deblock )
x264_macroblock_deblock_strength( h );
- if( SLICE_MBAFF )
- {
- /* update ratecontrol per-mbpair in MBAFF */
- if( i_mb_y&1 )
- x264_ratecontrol_mb( h, mb_size[0]+mb_size[1] );
- }
- else
- x264_ratecontrol_mb( h, mb_size[i_mb_y&1] );
+ x264_ratecontrol_mb( h, mb_size );
if( mb_xy == h->sh.i_last_mb )
break;
for( int j = 0; j < 3; j++ )
h->stat.frame.i_ssd[j] += t->stat.frame.i_ssd[j];
h->stat.frame.f_ssim += t->stat.frame.f_ssim;
+ h->stat.frame.i_ssim_cnt += t->stat.frame.i_ssim_cnt;
}
return 0;
/* generate picture parameters */
x264_nal_start( h, NAL_PPS, NAL_PRIORITY_HIGHEST );
- x264_pps_write( &h->out.bs, h->pps );
+ x264_pps_write( &h->out.bs, h->sps, h->pps );
if( x264_nal_end( h ) )
return -1;
overhead += h->out.nal[h->out.i_nal-1].i_payload + NALU_OVERHEAD;
}
- /* buffering period sei is written in x264_encoder_frame_end */
+ /* when frame threading is used, buffering period sei is written in x264_encoder_frame_end */
+ if( h->i_thread_frames == 1 && h->sps->vui.b_nal_hrd_parameters_present )
+ {
+ x264_hrd_fullness( h );
+ x264_nal_start( h, NAL_SEI, NAL_PRIORITY_DISPOSABLE );
+ x264_sei_buffering_period_write( h, &h->out.bs );
+ if( x264_nal_end( h ) )
+ return -1;
+ overhead += h->out.nal[h->out.i_nal-1].i_payload + NALU_OVERHEAD;
+ }
}
/* write extra sei */
}
x264_emms();
- /* generate sei buffering period and insert it into place */
- if( h->fenc->b_keyframe && h->sps->vui.b_nal_hrd_parameters_present )
+ /* generate buffering period sei and insert it into place */
+ if( h->i_thread_frames > 1 && h->fenc->b_keyframe && h->sps->vui.b_nal_hrd_parameters_present )
{
x264_hrd_fullness( h );
x264_nal_start( h, NAL_SEI, NAL_PRIORITY_DISPOSABLE );
if( pic_out->i_pts < pic_out->i_dts )
x264_log( h, X264_LOG_WARNING, "invalid DTS: PTS is less than DTS\n" );
- pic_out->img.i_csp = X264_CSP_NV12;
+ pic_out->opaque = h->fenc->opaque;
+
+ pic_out->img.i_csp = h->fdec->i_csp;
#if HIGH_BIT_DEPTH
pic_out->img.i_csp |= X264_CSP_HIGH_DEPTH;
#endif
pic_out->img.i_plane = h->fdec->i_plane;
- for( int i = 0; i < 2; i++ )
+ for( int i = 0; i < pic_out->img.i_plane; i++ )
{
pic_out->img.i_stride[i] = h->fdec->i_stride[i] * sizeof(pixel);
pic_out->img.plane[i] = (uint8_t*)h->fdec->plane[i];
for( int i_list = 0; i_list < 2; i_list++ )
for( int i = 0; i < X264_REF_MAX*2; i++ )
h->stat.i_mb_count_ref[h->sh.i_type][i_list][i] += h->stat.frame.i_mb_count_ref[i_list][i];
+ for( int i = 0; i < 3; i++ )
+ h->stat.i_mb_field[i] += h->stat.frame.i_mb_field[i];
if( h->sh.i_type == SLICE_TYPE_P && h->param.analyse.i_weighted_pred >= X264_WEIGHTP_SIMPLE )
{
h->stat.i_wpred[0] += !!h->sh.weight[0][0].weightfn;
h->stat.frame.i_ssd[1],
h->stat.frame.i_ssd[2],
};
+ int luma_size = h->param.i_width * h->param.i_height;
+ int chroma_size = CHROMA_SIZE( luma_size );
+ double psnr_y = x264_psnr( ssd[0], luma_size );
+ double psnr_u = x264_psnr( ssd[1], chroma_size );
+ double psnr_v = x264_psnr( ssd[2], chroma_size );
h->stat.f_ssd_global[h->sh.i_type] += dur * (ssd[0] + ssd[1] + ssd[2]);
- h->stat.f_psnr_average[h->sh.i_type] += dur * x264_psnr( ssd[0] + ssd[1] + ssd[2], 3 * h->param.i_width * h->param.i_height / 2 );
- h->stat.f_psnr_mean_y[h->sh.i_type] += dur * x264_psnr( ssd[0], h->param.i_width * h->param.i_height );
- h->stat.f_psnr_mean_u[h->sh.i_type] += dur * x264_psnr( ssd[1], h->param.i_width * h->param.i_height / 4 );
- h->stat.f_psnr_mean_v[h->sh.i_type] += dur * x264_psnr( ssd[2], h->param.i_width * h->param.i_height / 4 );
+ h->stat.f_psnr_average[h->sh.i_type] += dur * x264_psnr( ssd[0] + ssd[1] + ssd[2], luma_size + chroma_size*2 );
+ h->stat.f_psnr_mean_y[h->sh.i_type] += dur * psnr_y;
+ h->stat.f_psnr_mean_u[h->sh.i_type] += dur * psnr_u;
+ h->stat.f_psnr_mean_v[h->sh.i_type] += dur * psnr_v;
- snprintf( psz_message, 80, " PSNR Y:%5.2f U:%5.2f V:%5.2f",
- x264_psnr( ssd[0], h->param.i_width * h->param.i_height ),
- x264_psnr( ssd[1], h->param.i_width * h->param.i_height / 4),
- x264_psnr( ssd[2], h->param.i_width * h->param.i_height / 4) );
+ snprintf( psz_message, 80, " PSNR Y:%5.2f U:%5.2f V:%5.2f", psnr_y, psnr_u, psnr_v );
}
if( h->param.analyse.b_ssim )
{
double ssim_y = h->stat.frame.f_ssim
- / (((h->param.i_width-6)>>2) * ((h->param.i_height-6)>>2));
+ / h->stat.frame.i_ssim_cnt;
h->stat.f_ssim_mean_y[h->sh.i_type] += ssim_y * dur;
snprintf( psz_message + strlen(psz_message), 80 - strlen(psz_message),
" SSIM Y:%.5f", ssim_y );
****************************************************************************/
void x264_encoder_close ( x264_t *h )
{
- int64_t i_yuv_size = 3 * h->param.i_width * h->param.i_height / 2;
+ int64_t i_yuv_size = FRAME_SIZE( h->param.i_width * h->param.i_height );
int64_t i_mb_count_size[2][7] = {{0}};
char buf[200];
int b_print_pcm = h->stat.i_mb_count[SLICE_TYPE_I][I_PCM]
list_count[2] += h->stat.i_mb_partition[SLICE_TYPE_B][D_BI_8x8];
i_mb_count[B_DIRECT] += (h->stat.i_mb_partition[SLICE_TYPE_B][D_DIRECT_8x8]+2)/4;
i_mb_list_count = (list_count[0] + list_count[1] + list_count[2]) / 100.0;
- x264_log( h, X264_LOG_INFO,
- "mb B %s B16..8: %4.1f%% %4.1f%% %4.1f%% direct:%4.1f%% skip:%4.1f%% L0:%4.1f%% L1:%4.1f%% BI:%4.1f%%\n",
- buf,
- i_mb_size[PIXEL_16x16] / (i_count*4),
- (i_mb_size[PIXEL_16x8] + i_mb_size[PIXEL_8x16]) / (i_count*4),
- i_mb_size[PIXEL_8x8] / (i_count*4),
- i_mb_count[B_DIRECT] / i_count,
- i_mb_count[B_SKIP] / i_count,
- list_count[0] / i_mb_list_count,
- list_count[1] / i_mb_list_count,
- list_count[2] / i_mb_list_count );
+ sprintf( buf + strlen(buf), " B16..8: %4.1f%% %4.1f%% %4.1f%% direct:%4.1f%% skip:%4.1f%%",
+ i_mb_size[PIXEL_16x16] / (i_count*4),
+ (i_mb_size[PIXEL_16x8] + i_mb_size[PIXEL_8x16]) / (i_count*4),
+ i_mb_size[PIXEL_8x8] / (i_count*4),
+ i_mb_count[B_DIRECT] / i_count,
+ i_mb_count[B_SKIP] / i_count );
+ if( i_mb_list_count != 0 )
+ sprintf( buf + strlen(buf), " L0:%4.1f%% L1:%4.1f%% BI:%4.1f%%",
+ list_count[0] / i_mb_list_count,
+ list_count[1] / i_mb_list_count,
+ list_count[2] / i_mb_list_count );
+ x264_log( h, X264_LOG_INFO, "mb B %s\n", buf );
}
x264_ratecontrol_summary( h );
int64_t i_intra = i_i8x8 + SUM3b( h->stat.i_mb_count, I_4x4 )
+ SUM3b( h->stat.i_mb_count, I_16x16 );
int64_t i_all_intra = i_intra + SUM3b( h->stat.i_mb_count, I_PCM);
+ int64_t i_skip = SUM3b( h->stat.i_mb_count, P_SKIP )
+ + SUM3b( h->stat.i_mb_count, B_SKIP );
const int i_count = h->stat.i_frame_count[SLICE_TYPE_I] +
h->stat.i_frame_count[SLICE_TYPE_P] +
h->stat.i_frame_count[SLICE_TYPE_B];
+ int64_t i_mb_count = (int64_t)i_count * h->mb.i_mb_count;
+ int64_t i_inter = i_mb_count - i_skip - i_intra;
const double duration = h->stat.f_frame_duration[SLICE_TYPE_I] +
h->stat.f_frame_duration[SLICE_TYPE_P] +
h->stat.f_frame_duration[SLICE_TYPE_B];
- int64_t i_mb_count = (int64_t)i_count * h->mb.i_mb_count;
float f_bitrate = SUM3(h->stat.i_frame_size) / duration / 125;
+ if( PARAM_INTERLACED )
+ {
+ char *fieldstats = buf;
+ fieldstats[0] = 0;
+ if( i_inter )
+ fieldstats += sprintf( fieldstats, " inter:%.1f%%", h->stat.i_mb_field[1] * 100.0 / i_inter );
+ if( i_skip )
+ fieldstats += sprintf( fieldstats, " skip:%.1f%%", h->stat.i_mb_field[2] * 100.0 / i_skip );
+ x264_log( h, X264_LOG_INFO, "field mbs: intra: %.1f%%%s\n",
+ h->stat.i_mb_field[0] * 100.0 / i_intra, buf );
+ }
+
if( h->pps->b_transform_8x8_mode )
{
buf[0] = 0;
}
buf[0] = 0;
+ int csize = CHROMA444 ? 4 : 1;
if( i_mb_count != i_all_intra )
sprintf( buf, " inter: %.1f%% %.1f%% %.1f%%",
h->stat.i_mb_cbp[1] * 100.0 / ((i_mb_count - i_all_intra)*4),
- h->stat.i_mb_cbp[3] * 100.0 / ((i_mb_count - i_all_intra) ),
- h->stat.i_mb_cbp[5] * 100.0 / ((i_mb_count - i_all_intra)) );
- x264_log( h, X264_LOG_INFO, "coded y,uvDC,uvAC intra: %.1f%% %.1f%% %.1f%%%s\n",
+ h->stat.i_mb_cbp[3] * 100.0 / ((i_mb_count - i_all_intra)*csize),
+ h->stat.i_mb_cbp[5] * 100.0 / ((i_mb_count - i_all_intra)*csize) );
+ x264_log( h, X264_LOG_INFO, "coded y,%s,%s intra: %.1f%% %.1f%% %.1f%%%s\n",
+ CHROMA444?"u":"uvDC", CHROMA444?"v":"uvAC",
h->stat.i_mb_cbp[0] * 100.0 / (i_all_intra*4),
- h->stat.i_mb_cbp[2] * 100.0 / (i_all_intra ),
- h->stat.i_mb_cbp[4] * 100.0 / (i_all_intra ), buf );
+ h->stat.i_mb_cbp[2] * 100.0 / (i_all_intra*csize),
+ h->stat.i_mb_cbp[4] * 100.0 / (i_all_intra*csize), buf );
int64_t fixed_pred_modes[4][9] = {{0}};
int64_t sum_pred_modes[4] = {0};
}
for( int i = 0; i <= I_PRED_CHROMA_DC_128; i++ )
{
- fixed_pred_modes[3][x264_mb_pred_mode8x8c_fix[i]] += h->stat.i_mb_pred_mode[3][i];
+ fixed_pred_modes[3][x264_mb_chroma_pred_mode_fix[i]] += h->stat.i_mb_pred_mode[3][i];
sum_pred_modes[3] += h->stat.i_mb_pred_mode[3][i];
}
- if( sum_pred_modes[3] )
+ if( sum_pred_modes[3] && !CHROMA444 )
x264_log( h, X264_LOG_INFO, "i8c dc,h,v,p: %2.0f%% %2.0f%% %2.0f%% %2.0f%%\n",
fixed_pred_modes[3][0] * 100.0 / sum_pred_modes[3],
fixed_pred_modes[3][1] * 100.0 / sum_pred_modes[3],