First lie: "branching and jumping is slow". Let's evaluate "BNE" (branch when not equal) on 6502, and the Z80 equivalent "JP NZ". BNE requires 2 clockcycles if not jumping, or 3 clockcycles when jumping; JP NZ requires 4 clockcycles if not jumping, or 10 clockcycles if jumping. The Spectrum is faster than the C64. Demonstration: - JP NZ: 4 and 10 clockcycles at 3.5MHz, that is 1.14ns and 2.85ns
- BNE: 2 and 3 clockcycles at 982kHz, that is 2.03ns and 3.05ns
- result: branch/jump is slow on the C64; in fact Zx Spectrum shows to be
7% to 78% faster than Commodore 64.
- 6502: RTS requires 6 clockcycles at 982kHz, that is 6.11ns
- Z80: RET requires 10 clockcycles at 3.5MHz, that is 2.857ns
- result: Zx Spectrum requires 53.2% less time than Commodore 64
In all other cases, Spectrum shows the expected figures: 2.2-2.3 means that the Z80A of the Spectrum performs at 155%-165% the Commodore 64 "fast" speed; the almost missing "LDIR" test performs at 225% the C64 "fast memory access" speed... |
In Italian language, ciofeca (read as "choo-fay-kah") means "some coffee substitute, vaguely barley-based". You can imagine what does that mean in the Country of the True Coffee!
Well, the Commodore 64 is a great ciofeca.
Most Commodorian guys tried hard to find out some trick to make it appear at least fast as a Zx Spectrum.
Alas, the Zx Spectrum was always faster than the Commodore 64, even in the fields of "page zero", "fast instructions", "memory access", "graphics, "speed"...!
Click above for a full debunking of one of the most Commodorian-cited "sources". Most of the Standard Commodorian-Fan Speech is based on factual inaccuracies, fake assumptions, unproven assertions, stupid slogans and so on.
The truth is, the Zx Spectrum largely outperforms the Commodore 64.