*
* Authors: Laurent Aimar <fenrir@via.ecp.fr>
* Loren Merritt <lorenm@u.washington.edu>
+ * Fiona Glaser <fiona@x264.com>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
#define bs_write_ue bs_write_ue_big
-static void x264_encoder_frame_end( x264_t *h, x264_t *thread_current,
- x264_nal_t **pp_nal, int *pi_nal,
- x264_picture_t *pic_out );
+static int x264_encoder_frame_end( x264_t *h, x264_t *thread_current,
+ x264_nal_t **pp_nal, int *pi_nal,
+ x264_picture_t *pic_out );
/****************************************************************************
*
/* If effective qp <= 15, deblocking would have no effect anyway */
if( param->b_deblocking_filter
&& ( h->mb.b_variable_qp
- || 15 < i_qp + 2 * X264_MAX(param->i_deblocking_filter_alphac0, param->i_deblocking_filter_beta) ) )
+ || 15 < i_qp + 2 * X264_MIN(param->i_deblocking_filter_alphac0, param->i_deblocking_filter_beta) ) )
{
sh->i_disable_deblocking_filter_idc = 0;
}
{
bs_write_ue( s, sh->ref_pic_list_order[0][i].idc );
bs_write_ue( s, sh->ref_pic_list_order[0][i].arg );
-
+
}
bs_write_ue( s, 3 );
}
/* If we are within a reasonable distance of the end of the memory allocated for the bitstream, */
/* reallocate, adding an arbitrary amount of space (100 kilobytes). */
-static void x264_bitstream_check_buffer( x264_t *h )
+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 ) )
{
- uint8_t *bs_bak = h->out.p_bitstream;
intptr_t delta;
int i;
h->out.i_bitstream += 100000;
- h->out.p_bitstream = x264_realloc( h->out.p_bitstream, h->out.i_bitstream );
+ CHECKED_MALLOC( h->out.p_bitstream, h->out.i_bitstream );
+ h->mc.memcpy_aligned( h->out.p_bitstream, bs_bak, (h->out.i_bitstream - 100000) & ~15 );
delta = h->out.p_bitstream - bs_bak;
h->out.bs.p_start += delta;
for( i = 0; i <= h->out.i_nal; i++ )
h->out.nal[i].p_payload += delta;
+ x264_free( bs_bak );
}
+ return 0;
+fail:
+ x264_free( bs_bak );
+ return -1;
}
/****************************************************************************
#ifndef HAVE_PTHREAD
x264_log( h, X264_LOG_WARNING, "not compiled with pthread support!\n");
h->param.i_threads = 1;
-#else
- if( h->param.i_scenecut_threshold >= 0 )
- h->param.b_pre_scenecut = 1;
#endif
}
}
}
+ /* Detect default ffmpeg settings and terminate with an error. */
+ {
+ int score = 0;
+ score += h->param.analyse.i_me_range == 0;
+ score += h->param.rc.i_qp_step == 3;
+ score += h->param.i_keyint_max == 12;
+ score += h->param.rc.i_qp_min == 2;
+ score += h->param.rc.i_qp_max == 31;
+ score += h->param.rc.f_qcompress == 0.5;
+ score += fabs(h->param.rc.f_ip_factor - 1.25) < 0.01;
+ score += fabs(h->param.rc.f_pb_factor - 1.25) < 0.01;
+ score += h->param.analyse.inter == 0 && h->param.analyse.i_subpel_refine == 8;
+ if( score >= 5 )
+ {
+ x264_log( h, X264_LOG_ERROR, "broken ffmpeg default settings detected\n" );
+ x264_log( h, X264_LOG_ERROR, "use an encoding preset (vpre)\n" );
+ return -1;
+ }
+ }
+
if( h->param.rc.i_rc_method < 0 || h->param.rc.i_rc_method > 2 )
{
x264_log( h, X264_LOG_ERROR, "no ratecontrol method specified\n" );
h->param.rc.i_rc_method = X264_RC_CQP;
h->param.rc.f_ip_factor = 1;
h->param.rc.f_pb_factor = 1;
- h->param.analyse.b_transform_8x8 = 0;
h->param.analyse.b_psnr = 0;
h->param.analyse.b_ssim = 0;
h->param.analyse.i_chroma_qp_offset = 0;
h->param.analyse.i_trellis = 0;
h->param.analyse.b_fast_pskip = 0;
h->param.analyse.i_noise_reduction = 0;
- h->param.analyse.i_subpel_refine = x264_clip3( h->param.analyse.i_subpel_refine, 1, 6 );
+ h->param.analyse.f_psy_rd = 0;
+ h->param.i_bframe = 0;
+ /* 8x8dct is not useful at all in CAVLC lossless */
+ if( !h->param.b_cabac )
+ h->param.analyse.b_transform_8x8 = 0;
}
if( h->param.rc.i_rc_method == X264_RC_CQP )
{
h->param.rc.i_qp_min = x264_clip3( (int)(X264_MIN3( qp_p, qp_i, qp_b )), 0, 51 );
h->param.rc.i_qp_max = x264_clip3( (int)(X264_MAX3( qp_p, qp_i, qp_b ) + .999), 0, 51 );
h->param.rc.i_aq_mode = 0;
+ h->param.rc.b_mb_tree = 0;
}
+ h->param.rc.i_qp_max = x264_clip3( h->param.rc.i_qp_max, 0, 51 );
+ h->param.rc.i_qp_min = x264_clip3( h->param.rc.i_qp_min, 0, h->param.rc.i_qp_max );
if( ( h->param.i_width % 16 || h->param.i_height % 16 )
&& h->param.i_height != 1080 && !h->mb.b_lossless )
// There's nothing special about 1080 in that the warning still applies to it,
// but chances are the user can't help it if his content is already 1080p,
// so there's no point in warning in that case.
- x264_log( h, X264_LOG_WARNING,
+ x264_log( h, X264_LOG_WARNING,
"width or height not divisible by 16 (%dx%d), compression will suffer.\n",
h->param.i_width, h->param.i_height );
}
h->param.i_frame_reference = x264_clip3( h->param.i_frame_reference, 1, 16 );
if( h->param.i_keyint_max <= 0 )
h->param.i_keyint_max = 1;
+ if( h->param.i_scenecut_threshold < 0 )
+ h->param.i_scenecut_threshold = 0;
h->param.i_keyint_min = x264_clip3( h->param.i_keyint_min, 1, h->param.i_keyint_max/2+1 );
-
+ if( !h->param.analyse.i_subpel_refine && h->param.analyse.i_direct_mv_pred > X264_DIRECT_PRED_SPATIAL )
+ {
+ x264_log( h, X264_LOG_WARNING, "subme=0 + direct=temporal is not supported\n" );
+ h->param.analyse.i_direct_mv_pred = X264_DIRECT_PRED_SPATIAL;
+ }
h->param.i_bframe = x264_clip3( h->param.i_bframe, 0, X264_BFRAME_MAX );
h->param.i_bframe_bias = x264_clip3( h->param.i_bframe_bias, -90, 100 );
h->param.b_bframe_pyramid = h->param.b_bframe_pyramid && h->param.i_bframe > 1;
- h->param.b_bframe_adaptive = h->param.b_bframe_adaptive && h->param.i_bframe > 0;
+ if( !h->param.i_bframe )
+ h->param.i_bframe_adaptive = X264_B_ADAPT_NONE;
h->param.analyse.b_weighted_bipred = h->param.analyse.b_weighted_bipred && h->param.i_bframe > 0;
+ h->param.rc.i_lookahead = x264_clip3( h->param.rc.i_lookahead, 0, X264_LOOKAHEAD_MAX );
+ h->param.rc.i_lookahead = X264_MIN( h->param.rc.i_lookahead, h->param.i_keyint_max );
+ if( h->param.rc.b_stat_read )
+ h->param.rc.i_lookahead = 0;
+ else if( !h->param.rc.i_lookahead )
+ h->param.rc.b_mb_tree = 0;
+ if( h->param.rc.f_qcompress == 1 )
+ h->param.rc.b_mb_tree = 0;
+
h->mb.b_direct_auto_write = h->param.analyse.i_direct_mv_pred == X264_DIRECT_PRED_AUTO
&& h->param.i_bframe
&& ( h->param.rc.b_stat_write || !h->param.rc.b_stat_read );
- if( h->param.i_scenecut_threshold < 0 )
- h->param.b_pre_scenecut = 0;
h->param.i_deblocking_filter_alphac0 = x264_clip3( h->param.i_deblocking_filter_alphac0, -6, 6 );
h->param.i_deblocking_filter_beta = x264_clip3( h->param.i_deblocking_filter_beta, -6, 6 );
if( h->param.analyse.i_me_method == X264_ME_TESA &&
(h->mb.b_lossless || h->param.analyse.i_subpel_refine <= 1) )
h->param.analyse.i_me_method = X264_ME_ESA;
- h->param.analyse.i_subpel_refine = x264_clip3( h->param.analyse.i_subpel_refine, 1, 7 );
- h->param.analyse.b_bframe_rdo = h->param.analyse.b_bframe_rdo && h->param.analyse.i_subpel_refine >= 6;
+ h->param.analyse.i_subpel_refine = x264_clip3( h->param.analyse.i_subpel_refine, 0, 10 );
h->param.analyse.b_mixed_references = h->param.analyse.b_mixed_references && h->param.i_frame_reference > 1;
h->param.analyse.inter &= X264_ANALYSE_PSUB16x16|X264_ANALYSE_PSUB8x8|X264_ANALYSE_BSUB16x16|
X264_ANALYSE_I4x4|X264_ANALYSE_I8x8;
if( !h->param.b_cabac )
h->param.analyse.i_trellis = 0;
h->param.analyse.i_trellis = x264_clip3( h->param.analyse.i_trellis, 0, 2 );
+ if( !h->param.analyse.b_psy )
+ {
+ h->param.analyse.f_psy_rd = 0;
+ h->param.analyse.f_psy_trellis = 0;
+ }
+ if( !h->param.analyse.i_trellis )
+ h->param.analyse.f_psy_trellis = 0;
+ h->param.analyse.f_psy_rd = x264_clip3f( h->param.analyse.f_psy_rd, 0, 10 );
+ h->param.analyse.f_psy_trellis = x264_clip3f( h->param.analyse.f_psy_trellis, 0, 10 );
+ if( h->param.analyse.i_subpel_refine < 6 )
+ h->param.analyse.f_psy_rd = 0;
+ h->mb.i_psy_rd = FIX8( h->param.analyse.f_psy_rd );
+ /* 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 */
+ /* This can be triggered repeatedly on multiple calls to parameter_validate, but since encoding
+ * uses the pps chroma qp offset not the param chroma qp offset, this is not a problem. */
+ if( h->mb.i_psy_rd )
+ h->param.analyse.i_chroma_qp_offset -= h->param.analyse.f_psy_rd < 0.25 ? 1 : 2;
+ h->mb.i_psy_trellis = FIX8( h->param.analyse.f_psy_trellis / 4 );
+ /* Psy trellis has a similar effect. */
+ if( h->mb.i_psy_trellis )
+ h->param.analyse.i_chroma_qp_offset -= h->param.analyse.f_psy_trellis < 0.25 ? 1 : 2;
+ else
+ h->mb.i_psy_trellis = 0;
+ h->param.analyse.i_chroma_qp_offset = x264_clip3(h->param.analyse.i_chroma_qp_offset, -12, 12);
h->param.rc.i_aq_mode = x264_clip3( h->param.rc.i_aq_mode, 0, 2 );
- if( h->param.rc.f_aq_strength <= 0 )
+ h->param.rc.f_aq_strength = x264_clip3f( h->param.rc.f_aq_strength, 0, 3 );
+ if( h->param.rc.f_aq_strength == 0 )
h->param.rc.i_aq_mode = 0;
- /* VAQ effectively replaces qcomp, so qcomp is raised towards 1 to compensate. */
- if( h->param.rc.i_aq_mode == X264_AQ_GLOBAL )
- h->param.rc.f_qcompress = x264_clip3f(h->param.rc.f_qcompress + h->param.rc.f_aq_strength / 0.7, 0, 1);
+ /* MB-tree requires AQ to be on, even if the strength is zero. */
+ if( !h->param.rc.i_aq_mode && h->param.rc.b_mb_tree )
+ {
+ h->param.rc.i_aq_mode = 1;
+ h->param.rc.f_aq_strength = 0;
+ }
+ if( h->param.rc.b_mb_tree && h->param.b_bframe_pyramid )
+ {
+ x264_log( h, X264_LOG_WARNING, "b-pyramid + mb-tree is not supported\n" );
+ h->param.b_bframe_pyramid = 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) )
+ h->param.analyse.i_subpel_refine = 9;
{
const x264_level_t *l = x264_levels;
- while( l->level_idc != 0 && l->level_idc != h->param.i_level_idc )
- l++;
- if( l->level_idc == 0 )
+ if( h->param.i_level_idc < 0 )
{
- x264_log( h, X264_LOG_ERROR, "invalid level_idc: %d\n", h->param.i_level_idc );
- return -1;
+ 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++ );
+ h->param.rc.i_vbv_max_bitrate = maxrate_bak;
+ }
+ else
+ {
+ while( l->level_idc && l->level_idc != h->param.i_level_idc )
+ l++;
+ if( l->level_idc == 0 )
+ {
+ x264_log( h, X264_LOG_ERROR, "invalid level_idc: %d\n", h->param.i_level_idc );
+ return -1;
+ }
}
if( h->param.analyse.i_mv_range <= 0 )
- h->param.analyse.i_mv_range = l->mv_range;
+ h->param.analyse.i_mv_range = l->mv_range >> h->param.b_interlaced;
else
- h->param.analyse.i_mv_range = x264_clip3(h->param.analyse.i_mv_range, 32, 512);
- if( h->param.analyse.i_direct_8x8_inference < 0 )
- h->param.analyse.i_direct_8x8_inference = l->direct8x8;
+ h->param.analyse.i_mv_range = x264_clip3(h->param.analyse.i_mv_range, 32, 512 >> h->param.b_interlaced);
}
if( h->param.i_threads > 1 )
BOOLIFY( b_deblocking_filter );
BOOLIFY( b_interlaced );
BOOLIFY( analyse.b_transform_8x8 );
- BOOLIFY( analyse.i_direct_8x8_inference );
- BOOLIFY( analyse.b_bidir_me );
BOOLIFY( analyse.b_chroma_me );
BOOLIFY( analyse.b_fast_pskip );
BOOLIFY( rc.b_stat_write );
BOOLIFY( rc.b_stat_read );
+ BOOLIFY( rc.b_mb_tree );
#undef BOOLIFY
return 0;
static void mbcmp_init( x264_t *h )
{
int satd = !h->mb.b_lossless && h->param.analyse.i_subpel_refine > 1;
- memcpy( h->pixf.mbcmp, satd ? h->pixf.satd : h->pixf.sad, sizeof(h->pixf.mbcmp) );
+ 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_4x4 = satd ? h->pixf.intra_satd_x3_4x4 : h->pixf.intra_sad_x3_4x4;
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) );
****************************************************************************/
x264_t *x264_encoder_open ( x264_param_t *param )
{
- x264_t *h = x264_malloc( sizeof( x264_t ) );
+ x264_t *h;
char buf[1000], *p;
int i;
- memset( h, 0, sizeof( x264_t ) );
+ CHECKED_MALLOCZERO( h, sizeof(x264_t) );
/* Create a copy of param */
- memcpy( &h->param, param, sizeof( x264_param_t ) );
+ memcpy( &h->param, param, sizeof(x264_param_t) );
if( x264_validate_parameters( h ) < 0 )
- {
- x264_free( h );
- return NULL;
- }
+ goto fail;
if( h->param.psz_cqm_file )
if( x264_cqm_parse_file( h, h->param.psz_cqm_file ) < 0 )
- {
- x264_free( h );
- return NULL;
- }
+ goto fail;
if( h->param.rc.psz_stat_out )
h->param.rc.psz_stat_out = strdup( h->param.rc.psz_stat_out );
if( h->param.rc.psz_stat_in )
h->param.rc.psz_stat_in = strdup( h->param.rc.psz_stat_in );
- if( h->param.rc.psz_rc_eq )
- h->param.rc.psz_rc_eq = strdup( h->param.rc.psz_rc_eq );
/* VUI */
if( h->param.vui.i_sar_width > 0 && h->param.vui.i_sar_height > 0 )
h->pps = &h->pps_array[0];
x264_pps_init( h->pps, h->param.i_sps_id, &h->param, h->sps);
- x264_validate_levels( h );
+ x264_validate_levels( h, 1 );
if( x264_cqm_init( h ) < 0 )
- {
- x264_free( h );
- return NULL;
- }
-
+ goto fail;
+
h->mb.i_mb_count = h->sps->i_mb_width * h->sps->i_mb_height;
/* Init frames. */
- h->frames.i_delay = h->param.i_bframe + h->param.i_threads - 1;
+ if( h->param.i_bframe_adaptive == X264_B_ADAPT_TRELLIS )
+ h->frames.i_delay = X264_MAX(h->param.i_bframe,3)*4;
+ else
+ h->frames.i_delay = h->param.i_bframe;
+ if( h->param.rc.b_mb_tree )
+ h->frames.i_delay = X264_MAX( h->frames.i_delay, h->param.rc.i_lookahead );
+ h->frames.i_delay += h->param.i_threads - 1;
+ h->frames.i_delay = X264_MIN( h->frames.i_delay, X264_LOOKAHEAD_MAX );
+
h->frames.i_max_ref0 = h->param.i_frame_reference;
h->frames.i_max_ref1 = h->sps->vui.i_num_reorder_frames;
h->frames.i_max_dpb = h->sps->vui.i_max_dec_frame_buffering;
h->frames.b_have_lowres = !h->param.rc.b_stat_read
&& ( h->param.rc.i_rc_method == X264_RC_ABR
|| h->param.rc.i_rc_method == X264_RC_CRF
- || h->param.b_bframe_adaptive
- || h->param.b_pre_scenecut );
+ || h->param.i_bframe_adaptive
+ || h->param.i_scenecut_threshold
+ || h->param.rc.b_mb_tree );
h->frames.b_have_lowres |= (h->param.rc.b_stat_read && h->param.rc.i_vbv_buffer_size > 0);
+ h->frames.b_have_sub8x8_esa = !!(h->param.analyse.inter & X264_ANALYSE_PSUB8x8);
h->frames.i_last_idr = - h->param.i_keyint_max;
h->frames.i_input = 0;
h->i_ref0 = 0;
h->i_ref1 = 0;
+ h->chroma_qp_table = i_chroma_qp_table + 12 + h->pps->i_chroma_qp_index_offset;
+
x264_rdo_init( );
/* 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_8x8_init( h->param.cpu, h->predict_8x8 );
+ 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_init_vlc_tables();
x264_pixel_init( h->param.cpu, &h->pixf );
x264_dct_init( h->param.cpu, &h->dctf );
x264_zigzag_init( h->param.cpu, &h->zigzagf, h->param.b_interlaced );
if( !strcmp(x264_cpu_names[i].name, "SSE3")
&& (param->cpu & X264_CPU_SSSE3 || !(param->cpu & X264_CPU_CACHELINE_64)) )
continue;
+ if( !strcmp(x264_cpu_names[i].name, "SSE4.1")
+ && (param->cpu & X264_CPU_SSE42) )
+ continue;
if( (param->cpu & x264_cpu_names[i].flags) == x264_cpu_names[i].flags
&& (!i || x264_cpu_names[i].flags != x264_cpu_names[i-1].flags) )
p += sprintf( p, " %s", x264_cpu_names[i].name );
h->thread[0] = h;
h->i_thread_num = 0;
for( i = 1; i < h->param.i_threads; i++ )
- h->thread[i] = x264_malloc( sizeof(x264_t) );
+ CHECKED_MALLOC( h->thread[i], sizeof(x264_t) );
for( i = 0; i < h->param.i_threads; i++ )
{
if( i > 0 )
*h->thread[i] = *h;
h->thread[i]->fdec = x264_frame_pop_unused( h );
- h->thread[i]->out.p_bitstream = x264_malloc( h->out.i_bitstream );
+ if( !h->thread[i]->fdec )
+ goto fail;
+ CHECKED_MALLOC( h->thread[i]->out.p_bitstream, h->out.i_bitstream );
if( x264_macroblock_cache_init( h->thread[i] ) < 0 )
- return NULL;
+ goto fail;
}
if( x264_ratecontrol_new( h ) < 0 )
- return NULL;
+ goto fail;
+
+ if( x264_lowres_context_alloc( h ) )
+ goto fail;
if( h->param.psz_dump_yuv )
{
else
{
x264_log( h, X264_LOG_ERROR, "can't write to fdec.yuv\n" );
- x264_free( h );
- return NULL;
+ goto fail;
}
}
+ x264_log( h, X264_LOG_INFO, "profile %s, level %d.%d\n",
+ h->sps->i_profile_idc == PROFILE_BASELINE ? "Baseline" :
+ h->sps->i_profile_idc == PROFILE_MAIN ? "Main" :
+ h->sps->i_profile_idc == PROFILE_HIGH ? "High" :
+ "High 4:4:4 Predictive", h->sps->i_level_idc/10, h->sps->i_level_idc%10 );
+
return h;
+fail:
+ x264_free( h );
+ return NULL;
}
/****************************************************************************
#define COPY(var) h->param.var = param->var
COPY( i_frame_reference ); // but never uses more refs than initially specified
COPY( i_bframe_bias );
- if( h->param.i_scenecut_threshold >= 0 && param->i_scenecut_threshold >= 0 )
+ if( h->param.i_scenecut_threshold )
COPY( i_scenecut_threshold ); // can't turn it on or off, only vary the threshold
COPY( b_deblocking_filter );
COPY( i_deblocking_filter_alphac0 );
COPY( analyse.intra );
COPY( analyse.inter );
COPY( analyse.i_direct_mv_pred );
- COPY( analyse.i_me_method );
- COPY( analyse.i_me_range );
+ /* Scratch buffer prevents me_range from being increased for esa/tesa */
+ if( h->param.analyse.i_me_method < X264_ME_ESA || param->analyse.i_me_range < h->param.analyse.i_me_range )
+ COPY( analyse.i_me_range );
COPY( analyse.i_noise_reduction );
- COPY( analyse.i_subpel_refine );
+ /* We can't switch out of subme=0 during encoding. */
+ if( h->param.analyse.i_subpel_refine )
+ COPY( analyse.i_subpel_refine );
COPY( analyse.i_trellis );
- COPY( analyse.b_bidir_me );
- COPY( analyse.b_bframe_rdo );
COPY( analyse.b_chroma_me );
COPY( analyse.b_dct_decimate );
COPY( analyse.b_fast_pskip );
COPY( analyse.b_mixed_references );
+ COPY( analyse.f_psy_rd );
+ COPY( analyse.f_psy_trellis );
// can only twiddle these if they were enabled to begin with:
+ if( h->param.analyse.i_me_method >= X264_ME_ESA || param->analyse.i_me_method < X264_ME_ESA )
+ COPY( analyse.i_me_method );
+ if( h->param.analyse.i_me_method >= X264_ME_ESA && !h->frames.b_have_sub8x8_esa )
+ h->param.analyse.inter &= ~X264_ANALYSE_PSUB8x8;
if( h->pps->b_transform_8x8_mode )
COPY( analyse.b_transform_8x8 );
if( h->frames.i_max_ref1 > 1 )
{
/* identify ourself */
x264_nal_start( h, NAL_SEI, NAL_PRIORITY_DISPOSABLE );
- x264_sei_version_write( h, &h->out.bs );
+ if( x264_sei_version_write( h, &h->out.bs ) )
+ return -1;
x264_nal_end( h );
/* generate sequence parameters */
if( b_hpel )
{
x264_frame_expand_border( h, h->fdec, min_y, b_end );
- x264_frame_filter( h, h->fdec, min_y, b_end );
- x264_frame_expand_border_filtered( h, h->fdec, min_y, b_end );
+ if( h->param.analyse.i_subpel_refine )
+ {
+ x264_frame_filter( h, h->fdec, min_y, b_end );
+ x264_frame_expand_border_filtered( h, h->fdec, min_y, b_end );
+ }
}
if( h->param.i_threads > 1 && h->fdec->b_kept_as_ref )
x264_pixel_ssim_wxh( &h->pixf,
h->fdec->plane[0] + 2+min_y*h->fdec->i_stride[0], h->fdec->i_stride[0],
h->fenc->plane[0] + 2+min_y*h->fenc->i_stride[0], h->fenc->i_stride[0],
- h->param.i_width-2, max_y-min_y );
+ h->param.i_width-2, max_y-min_y, h->scratch_buffer );
}
}
-static inline void x264_reference_update( x264_t *h )
+static inline int x264_reference_update( x264_t *h )
{
int i;
{
x264_frame_push_unused( h, h->fdec );
h->fdec = x264_frame_pop_unused( h );
+ if( !h->fdec )
+ return -1;
}
- return;
+ return 0;
}
/* move lowres copy of the image to the ref frame */
if( h->frames.reference[h->frames.i_max_dpb] )
x264_frame_push_unused( h, x264_frame_shift( h->frames.reference ) );
h->fdec = x264_frame_pop_unused( h );
+ if( !h->fdec )
+ return -1;
+ return 0;
}
static inline void x264_reference_reset( x264_t *h )
x264_macroblock_slice_init( h );
}
-static void x264_slice_write( x264_t *h )
+static int x264_slice_write( x264_t *h )
{
int i_skip;
int mb_xy, i_mb_x, i_mb_y;
* Slice I: choose I_4x4 or I_16x16 mode
* Slice P: choose between using P mode or intra (4x4 or 16x16)
* */
- x264_macroblock_analyse( h );
+ if( x264_macroblock_analyse( h ) )
+ return -1;
/* encode this macroblock -> be careful it can change the mb type to P_SKIP if needed */
x264_macroblock_encode( h );
- x264_bitstream_check_buffer( h );
+ if( x264_bitstream_check_buffer( h ) )
+ return -1;
if( h->param.b_cabac )
{
h->stat.frame.i_mb_count_ref[i_list][i_ref] ++;
}
}
+ 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);
+ int b_intra = IS_INTRA(h->mb.i_type);
+ h->stat.frame.i_mb_cbp[!b_intra + 0] += cbpsum;
+ h->stat.frame.i_mb_cbp[!b_intra + 2] += h->mb.i_cbp_chroma >= 1;
+ h->stat.frame.i_mb_cbp[!b_intra + 4] += h->mb.i_cbp_chroma == 2;
+ }
if( h->mb.i_cbp_luma && !IS_INTRA(h->mb.i_type) )
{
h->stat.frame.i_mb_count_8x8dct[0] ++;
/* Compute misc bits */
h->stat.frame.i_misc_bits = bs_pos( &h->out.bs )
+ NALU_OVERHEAD * 8
- - h->stat.frame.i_itex_bits
- - h->stat.frame.i_ptex_bits
- - h->stat.frame.i_hdr_bits;
+ - h->stat.frame.i_tex_bits
+ - h->stat.frame.i_mv_bits;
+ return 0;
}
static void x264_thread_sync_context( x264_t *dst, x264_t *src )
// copy everything except the per-thread pointers and the constants.
memcpy( &dst->i_frame, &src->i_frame, offsetof(x264_t, mb.type) - offsetof(x264_t, i_frame) );
- memcpy( &dst->mb.i_type, &src->mb.i_type, offsetof(x264_t, rc) - offsetof(x264_t, mb.i_type) );
dst->stat = src->stat;
}
memcpy( &dst->stat.i_slice_count, &src->stat.i_slice_count, sizeof(dst->stat) - sizeof(dst->stat.frame) );
}
-static int x264_slices_write( x264_t *h )
+static void *x264_slices_write( x264_t *h )
{
int i_frame_size;
+#ifdef HAVE_MMX
+ /* Misalign mask has to be set separately for each thread. */
+ if( h->param.cpu&X264_CPU_SSE_MISALIGN )
+ x264_cpu_mask_misalign_sse();
+#endif
+
#if VISUALIZE
if( h->param.b_visualize )
- x264_visualize_init( h );
+ if( x264_visualize_init( h ) )
+ return (void *)-1;
#endif
- x264_stack_align( x264_slice_write, h );
+ if( x264_stack_align( x264_slice_write, h ) )
+ return (void *)-1;
i_frame_size = h->out.nal[h->out.i_nal-1].i_payload;
#if VISUALIZE
#endif
h->out.i_frame_size = i_frame_size;
- return 0;
+ return (void *)0;
}
/****************************************************************************
else
{
thread_current =
- thread_prev =
thread_oldest = h;
}
// ok to call this before encoding any frames, since the initial values of fdec have b_kept_as_ref=0
- x264_reference_update( h );
+ if( x264_reference_update( h ) )
+ return -1;
h->fdec->i_lines_completed = -1;
/* no data out */
{
/* 1: Copy the picture to a frame and move it to a buffer */
x264_frame_t *fenc = x264_frame_pop_unused( h );
+ if( !fenc )
+ return -1;
if( x264_frame_copy_picture( h, fenc, pic_in ) < 0 )
return -1;
if( h->frames.b_have_lowres )
x264_frame_init_lowres( h, fenc );
+ if( h->param.rc.b_mb_tree && h->param.rc.b_stat_read )
+ {
+ if( x264_macroblock_tree_read( h, fenc ) )
+ return -1;
+ }
+ else if( h->param.rc.i_aq_mode )
+ x264_adaptive_quant_frame( h, fenc );
+
if( h->frames.i_input <= h->frames.i_delay + 1 - h->param.i_threads )
{
/* Nothing yet to encode */
/* 2: Select frame types */
if( h->frames.next[0] == NULL )
{
- x264_encoder_frame_end( thread_oldest, thread_current, pp_nal, pi_nal, pic_out );
+ if( x264_encoder_frame_end( thread_oldest, thread_current, pp_nal, pi_nal, pic_out ) < 0 )
+ return -1;
return 0;
}
- x264_slicetype_decide( h );
+ x264_stack_align( x264_slicetype_decide, h );
/* 3: move some B-frames and 1 non-B to encode queue */
while( IS_X264_TYPE_B( h->frames.next[bframes]->i_type ) )
return 0;
}
-do_encode:
-
if( h->fenc->i_type == X264_TYPE_IDR )
{
h->frames.i_last_idr = h->fenc->i_frame;
{
/* identify ourself */
x264_nal_start( h, NAL_SEI, NAL_PRIORITY_DISPOSABLE );
- x264_sei_version_write( h, &h->out.bs );
+ if( x264_sei_version_write( h, &h->out.bs ) )
+ return -1;
x264_nal_end( h );
}
/* Write frame */
if( h->param.i_threads > 1 )
{
- x264_pthread_create( &h->thread_handle, NULL, (void*)x264_slices_write, h );
+ if( x264_pthread_create( &h->thread_handle, NULL, (void*)x264_slices_write, h ) )
+ return -1;
h->b_thread_active = 1;
}
else
- x264_slices_write( h );
-
- /* restore CPU state (before using float again) */
- x264_emms();
-
- if( h->sh.i_type == SLICE_TYPE_P && !h->param.rc.b_stat_read
- && h->param.i_scenecut_threshold >= 0
- && !h->param.b_pre_scenecut )
- {
- const int *mbs = h->stat.frame.i_mb_count;
- int i_mb_i = mbs[I_16x16] + mbs[I_8x8] + mbs[I_4x4];
- int i_mb_p = mbs[P_L0] + mbs[P_8x8];
- int i_mb_s = mbs[P_SKIP];
- int i_mb = h->sps->i_mb_width * h->sps->i_mb_height;
- int64_t i_inter_cost = h->stat.frame.i_inter_cost;
- int64_t i_intra_cost = h->stat.frame.i_intra_cost;
-
- float f_bias;
- int i_gop_size = h->fenc->i_frame - h->frames.i_last_idr;
- float f_thresh_max = h->param.i_scenecut_threshold / 100.0;
- /* magic numbers pulled out of thin air */
- float f_thresh_min = f_thresh_max * h->param.i_keyint_min
- / ( h->param.i_keyint_max * 4 );
- if( h->param.i_keyint_min == h->param.i_keyint_max )
- f_thresh_min= f_thresh_max;
-
- /* macroblock_analyse() doesn't further analyse skipped mbs,
- * so we have to guess their cost */
- if( h->stat.frame.i_mbs_analysed > 0 )
- i_intra_cost = i_intra_cost * i_mb / h->stat.frame.i_mbs_analysed;
-
- if( i_gop_size < h->param.i_keyint_min / 4 )
- f_bias = f_thresh_min / 4;
- else if( i_gop_size <= h->param.i_keyint_min )
- f_bias = f_thresh_min * i_gop_size / h->param.i_keyint_min;
- else
- {
- f_bias = f_thresh_min
- + ( f_thresh_max - f_thresh_min )
- * ( i_gop_size - h->param.i_keyint_min )
- / ( h->param.i_keyint_max - h->param.i_keyint_min );
- }
- f_bias = X264_MIN( f_bias, 1.0 );
-
- /* Bad P will be reencoded as I */
- if( h->stat.frame.i_mbs_analysed > 0 &&
- i_inter_cost >= (1.0 - f_bias) * i_intra_cost )
- {
- int b;
-
- x264_log( h, X264_LOG_DEBUG, "scene cut at %d Icost:%.0f Pcost:%.0f ratio:%.4f bias:%.4f gop:%d (imb:%d pmb:%d smb:%d)\n",
- h->fenc->i_frame,
- (double)i_intra_cost, (double)i_inter_cost,
- 1. - (double)i_inter_cost / i_intra_cost,
- f_bias, i_gop_size,
- i_mb_i, i_mb_p, i_mb_s );
-
- /* Restore frame num */
- h->i_frame_num--;
-
- for( b = 0; h->frames.current[b] && IS_X264_TYPE_B( h->frames.current[b]->i_type ); b++ );
- if( b > 0 )
- {
- /* If using B-frames, force GOP to be closed.
- * Even if this frame is going to be I and not IDR, forcing a
- * P-frame before the scenecut will probably help compression.
- *
- * We don't yet know exactly which frame is the scene cut, so
- * we can't assign an I-frame. Instead, change the previous
- * B-frame to P, and rearrange coding order. */
-
- if( h->param.b_bframe_adaptive || b > 1 )
- h->fenc->i_type = X264_TYPE_AUTO;
- x264_frame_sort_pts( h->frames.current );
- x264_frame_unshift( h->frames.next, h->fenc );
- h->fenc = h->frames.current[b-1];
- h->frames.current[b-1] = NULL;
- h->fenc->i_type = X264_TYPE_P;
- x264_frame_sort_dts( h->frames.current );
- }
- /* Do IDR if needed */
- else if( i_gop_size >= h->param.i_keyint_min )
- {
- /* Reset */
- h->i_frame_num = 0;
-
- /* Reinit field of fenc */
- h->fenc->i_type = X264_TYPE_IDR;
- h->fenc->i_poc = 0;
-
- /* Put enqueued frames back in the pool */
- while( h->frames.current[0] )
- x264_frame_push( h->frames.next, x264_frame_shift( h->frames.current ) );
- x264_frame_sort_pts( h->frames.next );
- }
- else
- {
- h->fenc->i_type = X264_TYPE_I;
- }
- goto do_encode;
- }
- }
+ if( (intptr_t)x264_slices_write( h ) )
+ return -1;
- x264_encoder_frame_end( thread_oldest, thread_current, pp_nal, pi_nal, pic_out );
- return 0;
+ return x264_encoder_frame_end( thread_oldest, thread_current, pp_nal, pi_nal, pic_out );
}
-static void x264_encoder_frame_end( x264_t *h, x264_t *thread_current,
- x264_nal_t **pp_nal, int *pi_nal,
- x264_picture_t *pic_out )
+static int x264_encoder_frame_end( x264_t *h, x264_t *thread_current,
+ x264_nal_t **pp_nal, int *pi_nal,
+ x264_picture_t *pic_out )
{
int i, i_list;
char psz_message[80];
if( h->b_thread_active )
{
- x264_pthread_join( h->thread_handle, NULL );
+ void *ret = NULL;
+ x264_pthread_join( h->thread_handle, &ret );
+ if( (intptr_t)ret )
+ return (intptr_t)ret;
h->b_thread_active = 0;
}
if( !h->out.i_nal )
{
pic_out->i_type = X264_TYPE_AUTO;
- return;
+ return 0;
}
x264_frame_push_unused( thread_current, h->fenc );
pic_out->i_pts = h->fenc->i_pts;
pic_out->img.i_plane = h->fdec->i_plane;
- for(i = 0; i < 4; i++){
+ for(i = 0; i < 3; i++)
+ {
pic_out->img.i_stride[i] = h->fdec->i_stride[i];
pic_out->img.plane[i] = h->fdec->plane[i];
}
/* update rc */
x264_emms();
- x264_ratecontrol_end( h, h->out.i_frame_size * 8 );
+ if( x264_ratecontrol_end( h, h->out.i_frame_size * 8 ) < 0 )
+ return -1;
/* restore CPU state (before using float again) */
x264_emms();
h->stat.i_mb_partition[h->sh.i_type][i] += h->stat.frame.i_mb_partition[i];
for( i = 0; i < 2; i++ )
h->stat.i_mb_count_8x8dct[i] += h->stat.frame.i_mb_count_8x8dct[i];
+ for( i = 0; i < 6; i++ )
+ h->stat.i_mb_cbp[i] += h->stat.frame.i_mb_cbp[i];
if( h->sh.i_type != SLICE_TYPE_I )
for( i_list = 0; i_list < 2; i_list++ )
for( i = 0; i < 32; i++ )
h->stat.i_mb_count_ref[h->sh.i_type][i_list][i] += h->stat.frame.i_mb_count_ref[i_list][i];
+ if( h->sh.i_type == SLICE_TYPE_P )
+ h->stat.i_consecutive_bframes[h->fdec->i_frame - h->fref0[0]->i_frame - 1]++;
if( h->sh.i_type == SLICE_TYPE_B )
{
h->stat.i_direct_frames[ h->sh.b_direct_spatial_mv_pred ] ++;
psz_message[0] = '\0';
if( h->param.analyse.b_psnr )
{
- int64_t sqe[3] = {
+ int64_t ssd[3] = {
h->stat.frame.i_ssd[0],
h->stat.frame.i_ssd[1],
h->stat.frame.i_ssd[2],
};
- h->stat.i_sqe_global[h->sh.i_type] += sqe[0] + sqe[1] + sqe[2];
- h->stat.f_psnr_average[h->sh.i_type] += x264_psnr( sqe[0] + sqe[1] + sqe[2], 3 * h->param.i_width * h->param.i_height / 2 );
- h->stat.f_psnr_mean_y[h->sh.i_type] += x264_psnr( sqe[0], h->param.i_width * h->param.i_height );
- h->stat.f_psnr_mean_u[h->sh.i_type] += x264_psnr( sqe[1], h->param.i_width * h->param.i_height / 4 );
- h->stat.f_psnr_mean_v[h->sh.i_type] += x264_psnr( sqe[2], h->param.i_width * h->param.i_height / 4 );
+ h->stat.i_ssd_global[h->sh.i_type] += ssd[0] + ssd[1] + ssd[2];
+ h->stat.f_psnr_average[h->sh.i_type] += 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] += x264_psnr( ssd[0], h->param.i_width * h->param.i_height );
+ h->stat.f_psnr_mean_u[h->sh.i_type] += x264_psnr( ssd[1], h->param.i_width * h->param.i_height / 4 );
+ h->stat.f_psnr_mean_v[h->sh.i_type] += 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",
- x264_psnr( sqe[0], h->param.i_width * h->param.i_height ),
- x264_psnr( sqe[1], h->param.i_width * h->param.i_height / 4),
- x264_psnr( sqe[2], h->param.i_width * h->param.i_height / 4) );
+ 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) );
}
if( h->param.analyse.b_ssim )
" SSIM Y:%.5f", ssim_y );
}
psz_message[79] = '\0';
-
+
x264_log( h, X264_LOG_DEBUG,
"frame=%4d QP=%.2f NAL=%d Slice:%c Poc:%-3d I:%-4d P:%-4d SKIP:%-4d size=%d bytes%s\n",
h->i_frame,
if( h->param.psz_dump_yuv )
x264_frame_dump( h );
+
+ return 0;
}
static void x264_print_intra( int64_t *i_mb_count, double i_count, int b_print_pcm, char *intra )
{
int64_t i_yuv_size = 3 * h->param.i_width * h->param.i_height / 2;
int64_t i_mb_count_size[2][7] = {{0}};
- char intra[40];
+ char buf[200];
int i, j, i_list, i_type;
int b_print_pcm = h->stat.i_mb_count[SLICE_TYPE_I][I_PCM]
|| h->stat.i_mb_count[SLICE_TYPE_P][I_PCM]
(double)h->stat.i_slice_size[i_slice] / i_count,
h->stat.f_psnr_mean_y[i_slice] / i_count, h->stat.f_psnr_mean_u[i_slice] / i_count, h->stat.f_psnr_mean_v[i_slice] / i_count,
h->stat.f_psnr_average[i_slice] / i_count,
- x264_psnr( h->stat.i_sqe_global[i_slice], i_count * i_yuv_size ) );
+ x264_psnr( h->stat.i_ssd_global[i_slice], i_count * i_yuv_size ) );
}
else
{
}
}
}
+ if( h->param.i_bframe && h->stat.i_slice_count[SLICE_TYPE_P] )
+ {
+ char *p = buf;
+ int den = 0;
+ // weight by number of frames (including the P-frame) that are in a sequence of N B-frames
+ for( i=0; i<=h->param.i_bframe; i++ )
+ den += (i+1) * h->stat.i_consecutive_bframes[i];
+ for( i=0; i<=h->param.i_bframe; i++ )
+ p += sprintf( p, " %4.1f%%", 100. * (i+1) * h->stat.i_consecutive_bframes[i] / den );
+ x264_log( h, X264_LOG_INFO, "consecutive B-frames:%s\n", buf );
+ }
for( i_type = 0; i_type < 2; i_type++ )
for( i = 0; i < X264_PARTTYPE_MAX; i++ )
{
int64_t *i_mb_count = h->stat.i_mb_count[SLICE_TYPE_I];
double i_count = h->stat.i_slice_count[SLICE_TYPE_I] * h->mb.i_mb_count / 100.0;
- x264_print_intra( i_mb_count, i_count, b_print_pcm, intra );
- x264_log( h, X264_LOG_INFO, "mb I %s\n", intra );
+ x264_print_intra( i_mb_count, i_count, b_print_pcm, buf );
+ x264_log( h, X264_LOG_INFO, "mb I %s\n", buf );
}
if( h->stat.i_slice_count[SLICE_TYPE_P] > 0 )
{
int64_t *i_mb_count = h->stat.i_mb_count[SLICE_TYPE_P];
double i_count = h->stat.i_slice_count[SLICE_TYPE_P] * h->mb.i_mb_count / 100.0;
int64_t *i_mb_size = i_mb_count_size[SLICE_TYPE_P];
- x264_print_intra( i_mb_count, i_count, b_print_pcm, intra );
+ x264_print_intra( i_mb_count, i_count, b_print_pcm, buf );
x264_log( h, X264_LOG_INFO,
"mb P %s P16..4: %4.1f%% %4.1f%% %4.1f%% %4.1f%% %4.1f%% skip:%4.1f%%\n",
- intra,
+ 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),
{
int64_t *i_mb_count = h->stat.i_mb_count[SLICE_TYPE_B];
double i_count = h->stat.i_slice_count[SLICE_TYPE_B] * h->mb.i_mb_count / 100.0;
+ double i_mb_list_count;
int64_t *i_mb_size = i_mb_count_size[SLICE_TYPE_B];
int64_t list_count[3] = {0}; /* 0 == L0, 1 == L1, 2 == BI */
- x264_print_intra( i_mb_count, i_count, b_print_pcm, intra );
+ x264_print_intra( i_mb_count, i_count, b_print_pcm, buf );
for( i = 0; i < X264_PARTTYPE_MAX; i++ )
for( j = 0; j < 2; j++ )
{
- int l0 = x264_mb_type_list0_table[i][j];
- int l1 = x264_mb_type_list1_table[i][j];
+ int l0 = x264_mb_type_list_table[i][0][j];
+ int l1 = x264_mb_type_list_table[i][1][j];
if( l0 || l1 )
list_count[l1+l0*l1] += h->stat.i_mb_count[SLICE_TYPE_B][i] * 2;
}
list_count[1] += h->stat.i_mb_partition[SLICE_TYPE_B][D_L1_8x8];
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;
- const double i_mb_list_count = (list_count[0] + list_count[1] + list_count[2]) / 100.0;
+ 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",
- intra,
+ 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),
if( h->stat.i_slice_count[SLICE_TYPE_I] + h->stat.i_slice_count[SLICE_TYPE_P] + h->stat.i_slice_count[SLICE_TYPE_B] > 0 )
{
+#define SUM3(p) (p[SLICE_TYPE_I] + p[SLICE_TYPE_P] + p[SLICE_TYPE_B])
+#define SUM3b(p,o) (p[SLICE_TYPE_I][o] + p[SLICE_TYPE_P][o] + p[SLICE_TYPE_B][o])
+ int64_t i_i8x8 = SUM3b( h->stat.i_mb_count, I_8x8 );
+ 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);
const int i_count = h->stat.i_slice_count[SLICE_TYPE_I] +
h->stat.i_slice_count[SLICE_TYPE_P] +
h->stat.i_slice_count[SLICE_TYPE_B];
+ int64_t i_mb_count = i_count * h->mb.i_mb_count;
float fps = (float) h->param.i_fps_num / h->param.i_fps_den;
-#define SUM3(p) (p[SLICE_TYPE_I] + p[SLICE_TYPE_P] + p[SLICE_TYPE_B])
-#define SUM3b(p,o) (p[SLICE_TYPE_I][o] + p[SLICE_TYPE_P][o] + p[SLICE_TYPE_B][o])
float f_bitrate = fps * SUM3(h->stat.i_slice_size) / i_count / 125;
if( h->pps->b_transform_8x8_mode )
{
- int64_t i_i8x8 = SUM3b( h->stat.i_mb_count, I_8x8 );
- int64_t i_intra = i_i8x8 + SUM3b( h->stat.i_mb_count, I_4x4 )
- + SUM3b( h->stat.i_mb_count, I_16x16 );
x264_log( h, X264_LOG_INFO, "8x8 transform intra:%.1f%% inter:%.1f%%\n",
100. * i_i8x8 / i_intra,
100. * h->stat.i_mb_count_8x8dct[1] / h->stat.i_mb_count_8x8dct[0] );
h->stat.i_direct_frames[0] * 100. / h->stat.i_slice_count[SLICE_TYPE_B] );
}
+ x264_log( h, X264_LOG_INFO, "coded y,uvDC,uvAC intra:%.1f%% %.1f%% %.1f%% inter:%.1f%% %.1f%% %.1f%%\n",
+ 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 ),
+ 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)) );
+
for( i_list = 0; i_list < 2; i_list++ )
{
int i_slice;
for( i_slice = 0; i_slice < 2; i_slice++ )
{
- char buf[200];
char *p = buf;
int64_t i_den = 0;
int i_max = 0;
SUM3( h->stat.f_psnr_mean_u ) / i_count,
SUM3( h->stat.f_psnr_mean_v ) / i_count,
SUM3( h->stat.f_psnr_average ) / i_count,
- x264_psnr( SUM3( h->stat.i_sqe_global ), i_count * i_yuv_size ),
+ x264_psnr( SUM3( h->stat.i_ssd_global ), i_count * i_yuv_size ),
f_bitrate );
}
else
free( h->param.rc.psz_stat_out );
if( h->param.rc.psz_stat_in )
free( h->param.rc.psz_stat_in );
- if( h->param.rc.psz_rc_eq )
- free( h->param.rc.psz_rc_eq );
x264_cqm_delete( h );