The utilization of DNA molecular markers in plant breeding to maximize selection response via marker assisted selection (MAS) and genomic selection (GS) has revolutionized plant breeding. The Nebraska Winter Wheat Breeding Program is interested in developing and utilizing GS. A key factor affecting GS applicability is the choice of marker platform. Genotypying-by-sequencing (GBS) provides large number of SNPs, albeit with high rates of missing data. While SNPs scored using array-based assays are of high quality, but often the cost per sample is substantially higher.  The objectives of the this study were  1) compare genomic prediction accuracy between GBS-derived SNPs and array-based SNPs; 2) compare estimates of genomic kinship and population structure calculated using the two different types of marker data. A diversity panel consisting of 299 winter wheat accessions that is part of ongoing TCAP (Triticeae Coordinated Agricultural Project) multi-year, multi-environments association mapping study was used. The panel was phenotyped in Mead, NE for grain yield, grain volume weight, plant height, and flowering date in 2012 and 2013. The panel was genotyped using SNPs derived from GBS as well as SNPs scored using a 9K SNP assay.  Results from 2012 indicate no difference in prediction accuracy between marker platforms, indicating less costly GBS is appropriate for genomic prediction applications.