I would say “yes” mainly because they off-load the main cpu regarding graphics primitives and rendering to the hardware (the display). Here you only send commands to the chip and it does the hard work.
There are no computations to draw a circle, draw a font, and so on. You send orders and the FT8xx executes. Many “orders” are simple integers, so only 4 bytes to transmit for one command. That’s almost nothing.
In such constrained environment, I find rather unfortunate to consume memory and cpu power that could be otherwise used by your application. Even a small STM32F401 can achieve the same level of performance seen on the video (which is using a STM32F427).
I would compare this hardware with onboard graphic chips vs dedicated GPU on PCs, performance wize.
If you want to see more features, you can watch the demo we’ve made a few years ago, now. At 2:00, you can see that it is still running at nearly 60fps with touch and many things on the screen. I don’t think that this kind of demo would be possible with the small CPUs we are using.
Btw, there are other similar alternatives like the Vizic SmartGPU 2. This one is using serial communications instead of SPI.