Talk:Zilog eZ80

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

about zilog ez80[edit]

what is the main use of zilog ez80? —Preceding unsigned comment added by 124.105.41.86 (talk) 15:44, 7 January 2009 (UTC)[reply]

A general purpose microcontroller or CPU has no main use, it's up to you! —Preceding unsigned comment added by 83.255.43.245 (talk) 23:06, 1 March 2009 (UTC)[reply]
But if I recall correctly Zilog suggested that you use it to put web servers in everything like industrial machines, dishwashers, washing machines, toasters et.c. // Liftarn (talk)
Why hasn't anybody made a lightning fast ZX Spectrum clone? :D 84.49.75.226 (talk) 10:12, 10 February 2014 (UTC)[reply]

n Times faste than Z80?[edit]

The article states:

Available at up to 50 MHz (2004), the performance is comparable to a Z80 clocked at 200 MHz if fast memory is used […] or even higher in some applications (a 16-bit addition is 11 times as fast as in the original). <Omission and markup added by discussant.>

Keeping in mind that a Z80 is usually clocked with 2,5, 4 or maybe 8 MHz, the last bold striked clause seems to be irritating (at least 200 Mhz means we are talking about being 25 upto 80 times faster). I guess that there is meant that the eZ80 is 11 times faster at same clock speed as a Z80 regarding 16 bit addition. Please clarify. -- Pemu (talk) 11:42, 3 May 2015 (UTC)[reply]

@Pemu The ADD HL, rr takes 1 cycle on the eZ80, but 3 cycles on the Z80. In this case, it is 3x the speed on this instruction alone. It doesn't make sense either. Imclevor (talk) 12:38, 21 June 2023 (UTC)[reply]
It actually does to me. If the clock speed is 2-3x faster and the addition instruction is 3x faster, then it would be approximately 10x faster. Imclevor (talk) 12:41, 21 June 2023 (UTC)[reply]

24bit?[edit]

@Maury Markowitz: I would argue that the eZ80 is both an 8-bit and a 24-bit processor. The ALU is 24 bit. The registers are 24 bit although they can be used as three 8-bit registers. The instruction set consists of instructions for 8-bit data (legacy Z80) and instructions for 24-bit data. Similarly, the NEC V20 / V30 would be both 16-bit processors and 8-bit processors because they have an 8080 mode. Cheers, Drahtlos (talk) 12:54, 29 October 2021 (UTC)[reply]

This has been hashed over repeatedly for the last 20 years - the ALU size does not define the categorization. The Nova was a 16-bit machine in spite of having a 4-bit ALU, and the Z80 was an 8-bit machine in spite of having 16-bit instructions. This example a further extension of the original Z80 model in every way and the ISA remains very 8-bit oriented. Maury Markowitz (talk) 13:14, 29 October 2021 (UTC)[reply]
24-bit registers are split into three sections, two of which can be directly used. For example, the 24-bit HL register is comprised of three 8-bit sections: H, L, and HLU. HLU is the middle bytewhichbut cannot be used directly. In theory, it is possible to use HLU by bit- shifting, but it is easier just to use 3 8-bit registers for three 8-bit values. Imclevor (talk) 12:13, 21 June 2023 (UTC)[reply]

I have to agree with Drahtlos. All 24 bit instructions work like any 24 bit machine. There is even a 24 bit stack. It pushes 3 bytes so we are back on 8 bits at that point. Th 8 bit bus doesnt make it an 8 bit CPU. I's a mix. Please see https://www.youtube.com/watch?v=71bDpegZJTs Robotics1 (talk) 14:47, 29 October 2021 (UTC)[reply]