cheater

Why are you being hostile? Obviously the point was that a cpu can emulate many things just as well as an fpga and better if ns level timing is not necessary. Oh btw, I've been doing electronics for over 20 years, I'm a professional programmer, and I've used fpgas in projects. Maybe next time consider disagreeing /politely/? Sheesh...

Bear in mind in every sich system there are parts done better by an fpga

I disagree - a CPU is just as "reconfigurable" as an FPGA for our purposes. We're talking about having something like an ARM and emulating whatever other CPU on top of it. I bet it'll still be faster than an FPGA could manage with the original CPU implemented in Verilog.

Sure, in the case that the console actually has a separate GPU that can be overclocked too. Yes, you can overclock a CPU implemented in an FPGA. That's not really overclocking, you just run it at a higher speed (which the FPGA still supports, but the original CPU wouldn't have). The idea behind FPGA simulation though is to keep timings as tight as possible, keep delays down at a minimum, etc. This is null and void for situations where CPU and GPU run asynchronously like what you described. So the reason to use an FPGA is void. What I was saying was that at this point you might start using a normal cpu, which will perform much better than an FPGA.

Here's an example where people overclocked the Genesis somewhat... you can only ask yourself what would happen if you could go real real fast. naturally the same stuff goes for consoles newer than the Genesis...

Am I the only person thinking that in consoles newer than the Genesis there shouldn't be a reason to do an fpga impl of the cpu, but instead use a way to run the code as fast as possible, maybe on a traditional cpu? Even on the Genesis, overclocking the cpu just makes games smoother, there are no timing issues. That's because that cpu runs stuff asynchronously to display and audio code which is where timings have to be most accurate.

I have i said homebrews are starting to add internet support and it would be great to have a pinout that you could connect an ethernet port to, but it wouldn't make sense to put it in the BOM for most people. Alternatively an expansion would be cool

Yep I mentioned that in one of my posts as well. Makes a big difference

watch the video from 4:55 to 6:40. He first shows CGA graphics on an LCD and compares them to EGA. EGA looks pretty good and CGA looks crap, it's basically shades of pink. Then he takes out another computer and shows CGA via composite output and it looks just as good as EGA did, with the same colors as EGA had on that other computer. So basically, CGA has to be played via component to be enjoyable. Using it with RGBI (which is a digital 1 bit per channel version of RGB explained in another part of the video) really makes it look crap because it's missing the smoothing that component provided due to limited signal bandwidth and crt smearing. https://youtu.be/niKblgZupOc?t=295

Wow, this is so cool. I wish he had a page that displays what all the settings do with screenshots or short videos.

kevtris: i know you didn't want to do any fancy output processing but this might be somewhat important. the effect blows my mind:

Been watching this: why the hell is it "panel bus driver" and not "Otto"?

This is at a distance if 150 million km, that is, 75 000 000 000 times more than the signal takes from the cathode to your eyes.

I beg to differ. Unless you mean what happens to the signal before it becomes light. That is, the electron beam. The only way for electrons to "bounce around" the inside the CRT like you say (aka secondary electron emission) is if the crt is hugely miscalibrated. At that point you can't see the picture any more, or there's a huge amount of overscan (and the beam bounces off the side) or the dot is extremely bright (unachievable by most TVs anyways, only happens in oscilloscopes). There are no secondary electrons in TVs except for severely broken ones. Here's secondary emission from a too bright dot. The beam is so intensive it knocks out electrons from the fluorescent screen and they end up falling back on the screen again, kind of like when you toss a large rock into a pond, and that splashes water making projectiles that then fall on the surface of the water again. I've never seen a crt monitor or tv that would do that. They simply don't operate at high enough cathode voltages for that effect to happen.

That's a nice idea. You could do the following: for each RGB triad, you could output each subpixel with both the colour of the triad and the colour of the subpixel at the same time. So if you're displaying teal, then display reddish teal, then greenish teal, then blueish teal. That should look CRTish while keeping the brightness levels nice and high, and you could set the mix ratio (e.g. 50% subpixel, 50% triad, or 75%/25%, or 25%/75%, or 12.5%/87.5%, etc.)