Journal Alioth's Journal: Pacman coding contest 2
To celebrate 30 years of "Pac Man", at this year's Retromañï½Âa (University of Zaragoza, 8th - 12th November) there's a Pacman programming contest (with prizes!). Information about the contest and rules can be found here:
There's two categories - one for games programmed beforehand (so that people who can't go to Retromañï½Âa can put in an entry) and for those programmed during the event.
I'll myself be off to Retromañï½Âa, not for anything to do with the Pac man coding contest, but to demonstrate and explain the ethernet hardware that I've made for the ZX Spectrum (itself coming close to its 30th anniversary). Indeed that entire week is going to be an orgy of retro geekyness, the weekend leading up to it is R3PLAY in Blackpool (retrogaming event), then I'm off to Zaragoza for Retromañï½Âa, and then on the way back I go via Madrid for a small gathering of retroafficionados in a bar somewhere in Madrid. Three events in less than 10 days.
ethernet hardware that I've made for the ZX (Score:2)
If it wasn't so sad, I'd have laughed at you a couple of days ago. Last night I was looking at the hardware which this week's client has supplied, and realised that the 100-base desktops and laptops in this office communicate with the rest of the building (and thence to the rest of the world) through 2-mbps co-ax cable.
Still, it isn't that much worse than surfing at home.
Do you think that your Speccy will be able to saturate your ethernet link.
Oblig : imag
Re: (Score:2)
It can't even come close to saturating a 100baseTX network :-)
It does operate at the full memory bandwidth of of the Speccy (or at least, of the Z80's LDIR instruction). If I were to add a Z80 DMA, I could probably get the throughput up to something like 14Mbit/sec in uncontended memory, slightly less sustained (assuming the DMA can load a byte from the ethernet hardware and transfer it to RAM in 2 T-states with no wait states, I'm assuming here that the DMA's memory access cycle is longer than half a T-sta