We present a study of the trade-off between depth and resolution using a large number of U -band imaging observations in the GOODS-North field ( Giavalisco et al .
2004 ) from the Large Binocular Camera ( LBC ) on the Large Binocular Telescope ( LBT ) .
Having acquired over 30 hours of data ( 315 images with 5-6 mins exposures ) , we generated multiple image mosaics , starting with the best atmospheric seeing images ( FWHM \lesssim 0 \farcs 8 ) , which constitute \sim 10 % of the total data set .
For subsequent mosaics , we added in data with larger seeing values until the final , deepest mosaic included all images with FWHM \lesssim 1 \farcs 8 ( \sim 94 % of the total data set ) .
From the mosaics , we made object catalogs to compare the optimal-resolution , yet shallower image to the lower-resolution but deeper image .
We show that the number counts for both images are \sim 90 % complete to U _ { AB } \lesssim 26 .
Fainter than U _ { AB } \sim 27 , the object counts from the optimal-resolution image start to drop-off dramatically ( 90 % between U _ { AB } = 27 and 28 mag ) , while the deepest image with better surface-brightness sensitivity ( \mu ^ { AB } _ { U } \lesssim 32 mag arcsec ^ { -2 } ) show a more gradual drop ( 10 % between U _ { AB } \simeq 27 and 28 mag ) .

For the brightest galaxies within the GOODS-N field , structure and clumpy features within the galaxies are more prominent in the optimal-resolution image compared to the deeper mosaics .
To further investigate how the seeing conditions affect the mosaics , we combined the images by weighting based on the image FWHM .
In these weighted stacks , a larger number of small faint galaxies are detected .
We conclude that for studies of brighter galaxies and features within them , the optimal-resolution image should be used .
However , to fully explore and understand the faintest objects , the deeper imaging with lower resolution are also required .
Finally , we find — for 220 brighter galaxies with U _ { AB } \lesssim 24 mag — only marginal differences in total flux between the optimal-resolution and lower-resolution light-profiles to \mu ^ { AB } _ { U } \lesssim 32 mag arcsec ^ { -2 } .
In only 10 % of the cases are the total-flux differences larger than 0.5 mag .
This helps constrain how much flux can be missed from galaxy outskirts , which is important for studies of the Extragalactic Background Light .