DDS
DDS raster imageUsed for pixel-based artwork, scans, and photographs.
- Family
- Images
- MIME
- image/vnd.ms-dds
Convert DDS to PSD online. See the real pixel decoding and image re-encoding behavior, preservation limits, and a format-specific acceptance test.
Directdraw surface texture data that can include block compression, mipmaps, cubemap faces, and alpha is decoded into a working pixel canvas first, then written back out under PSD's own color and compression model. Dimensions can stay exactly the same even as encoding, metadata, and fine detail change underneath. DDS loading inventories texture format, block compression, dimensions, mip levels, array slices, cubemap faces, alpha, and channel conventions before selecting one surface. PSD output aims to retain a Photoshop-style composite and editable layers, masks, channels, effects, text, and color data instead of only one flattened bitmap.
Canvas decoded from DirectDraw Surface texture data that can include block compression, mipmaps, cubemap faces, and alpha. Image written as a Photoshop document that can contain layers, masks, channels, type, and adjustment data.
Used for pixel-based artwork, scans, and photographs.
Designed for pixel-based artwork, scans, and photographs.
PSD wins this route whenever its specific mix of compatibility, alpha, animation, and file size fits the destination better than DDS does.
A flat image export may keep only the top mip level or one face and can mishandle normal-map channels.
Most non-PSD outputs flatten layers, while unsupported smart objects or effects may render differently.
a Photoshop document that can contain layers, masks, channels, type, and adjustment data is what gets written, labeled as image/vnd.adobe.photoshop. Whether that actually opens depends on the receiving application, not on the file extension.
Inspect alpha, channel order, every required cubemap face, mip level, and texture orientation in the target engine.
Open the result in the intended editor and inspect layer count, masks, typography, blend modes, and canvas size.