"Those" refers to the "DPM" to which you refer, and I'm just asking which of your favorite MCU's has an internal dual ported memory to which the "loader" would have access so your MCU could do the rest of the work and keep up with the pixel rate?

Once you find you have to add complex hardware, it's not so likely you'll be able to justify an MCU for the subordinate tasks. It's easier, particularly with video display circuits, to do that in programmable logic.

BTW, a RAMDAC, as is used on nearly every SVGA board, has much of that logic built in, so you can load the translation buffer, then synchronously feed pixel data to the thing in whatever its word-width is, and it will generate the R, G, and B, in the color and shade that you've specified. It does, of course, have to proceed at one word per pixel, since each pixel requires a color value specification. The video display and timing logic provides all that, though.