In consitensy between SAM and pileup format

[wwhlazio]

Hi,
I found a great inconsistency between sam format and pileup format when I analyze the 1000 Genome phase1 data.
Download bam file on target region
samtools view -h -b ftp://ftp.1000genomes.ebi.ac.uk/vol1...e.20110411.bam 11:66191255-66192791 > NA18873_11_66191255_66192791.bam
Index the file
samtools sort download_data/reads_data/NA18873/NA18873_11_66191255_66192791.bam temp/NA18873_11_66191255_66192791.bam.sorted
samtools index temp/NA18873_11_66191255_66192791.bam.sorted.bam
Get the sam file:
samtools view -h -o temp.sam ./temp/NA18873_11_66191255_66192791.bam.sorted.bam
Get the pileup format:
samtools mpileup -A -B -f ./download_data/human_g1k_v37.fasta -d 1500 -Q0 ./temp/NA18873_11_66191255_66192791.bam.sorted.bam > temp.txt

The first sevel lines of sam file:
SRR111961.2999876 147 11 66191180 62 76M = 66191103 -152 AGCAGGTAAGCCACAAGCCAGGGGACTAGGGGGCAGCTGAGGTCGTCATGGAGGAGACACAAATCAGGGAACTGCT ##########?-?.ACFGGCEDFE@H?DFFDDEGCEGAFCFE@@BAG>@EFDEFDF?F?GA@??EBCCC><D=BCS ......
SRR111961.17887438 147 11 66191182 61 76M = 66191139 -118 CAGTAAAGCCACAAGCCAGGGGACTAGGGGGCAGCTGAGGTCGTCATGGAGGAGACACAAATCAGGGAACTGCTCT ########?C=7BDFFDCFFEE?G?CED'DDHDGGAFDEEA?EAG?AEFDEFCF@G?GAA??FBDCD?=D=CD=CS ......
SRR111961.48268320 163 11 66191182 65 76M = 66191195 88 CAGGTAAGCCACAAGCCAGGGGACTAGGGGGCAGCTGAGGTCGTCATGGAGGAGACACAAATCAGGGAACTGCTCT S=BB<;@DCB?C?BFEA@FFFF?DB>GAAAEFAGFCHAGC;F>?D??E@@E?@E?EAEADC>EAGE>ACDCHECES ......
SRR111961.99148939 147 11 66191182 62 75M = 66191143 -113 CAGGTAACCCACAAGCCAGGGGACTAGGGGGCAGCTGAGGTCGTCATGGAGGAGACACAAATCAGGGAACTGCTC ##########<E@AC>:BCCBD=D=1DECDBFBEG@F>DDA=D@D?AEECDEBC>F>=@@>@E@BBC?<D=BC<S ......
SRR111961.29303868 1107 11 66191183 62 76M = 66190976 -282 AGGTAAGCCAAAAGCCAGGGGACTAGGGGGCAGCTGAGGTCGTCATGGAGGAGACACAAATCAGGGAACTGCTCTG ##################E>A?F=CECD@EAAED6EBDCA?6AE?AEFBEFCF?G>GA@?@GBDDFA>F?DE?E=S ......
SRR111961.55807721 83 11 66191183 61 76M = 66190976 -282 AGGAAAGCCAAAAGCCAGGGGACTAGGGGGCAGCTGAGGTCGTCATGGAGGAGACACAAATCAGGGAACTGCTCTG #############FBHCE@EC?F=CEFEFGHCGGADDFFB@FAG@@FGDFFDF@G@GAA@AGCEEFA?F@EF@E>S ......
SRR111961.148279807 1107 11 66191184 62 76M = 66191020 -239 GGAAAGCCACAAGCCAGGGGACTAGGGGGCAGCTGAGGTCGTCATGGAGGAGACACAAATCAGGGAACTGCTCTGG ############D?ECEEFD@H>DFEEEFFDFGAGDEFB@FBH@AFGDFGDG@G@GBA@@GDEEFA?F@EF@F>BS ......
The first several lines of pileup format:
11 66191180 A 1 ^_, #
11 66191181 T 1 g #
11 66191182 C 4 ,^^,^b.^_, ##S#
11 66191183 A 5 ,,.,^^, ##=##
there are two reads start at 66191183 but comparing to 66191182 in the pileup file, #of reads only increase by 1.

I would say pileup missed a lot of reads in sam. Total throughput in sam file:　１２６０９００　(length of reads * #reads) while the throughput in bam files:
602330 (sum of depth on all the pos)

Does anyone have an idea on this problem? Thanks!