Computer graphics has been driven by the desire to generate real-time imagery subject to constraints imposed by the human visual system. The future of computer graphics, when off-the-shelf systems have full multimedia capability and when standard computing engines render imagery faster than real-time, remains to be seen. A dedicated pipeline for graphics will be redundant for all but the most demanding applications; imagery available today only on expensive systems will be supported by standard components. Deficiencies in spatial resolution for current head-mounted displays are one obstacle. However, with predictions of 4K X 4K and even 8K X 8K, it seems that most applications will not be limited by the spatial resolution of the screen once another one or two factors of two are achieved in the number of pixels per inch on a screen. With respect to chromatic resolution (number of bits per pixel used to represent color), research is within one or two factors of two from the ultimate chromatic resolution imagined as being necessary for the human visual system. Temporal resolution comes in two forms: refresh rate and update rate. The future of refresh rate is likely to be tied to consumer video, so the one or two factors of two may be a while in coming. Update rate is mostly a question of memory, transmission bandwidth, and computing power, and will increase almost automatically with the general advance of computing technology. With many of the hardware problems close to being solved and the promise of widespread multimedia applications likely to bring costs down, will there be any difference between computer graphics and multimedia? It is proposed that the difference will be significant, and computer graphics will be relegated to the back seat. The multimedia pipeline of the future will involve computing of the images ahead of time (maybe 1/240 of a second before needed), compression using high-speed circuitry, moving the images to secondary memory, fetching them all back when needed (1/240 of a second later), decompression, then pasting them onto the screen. Networking provides access to moving imagery, and largely eliminates the distinction between real-time and pre-computer imagery. So where will this leave computer graphics? It will still be there, but mostly as a producer of imagery on the network and as a tool for augmenting imagery obtained from other sources. (Contains 17 references.) (MAS)