What's new
I've been following development of the MISTer for a while. I agree that it is interesting, and when FPGA cores are well-developed then they are superior to software emulation. But it seems like the MISTer cores aren't quite there yet - many of the cores have problems and inaccuracies, and they're not at the polished level of the Kevtris / Analog FPGA solutions. The current upscaling algorithms for HDMI output add at least two frames of lag as well.

I think it'll continue to improve, but I'm sticking with my original hardware and mature, well-developed PC emulators for the time being. I'll probably jump on the MISTer train in another year or so.

I'd definitely take a MISTer over a Raspberry Pi though!
 
It's interesting, and I see the potential, but it's hardly the second coming of Christ that everyone seems to be making it out to be.

I'm calling it now, in a year all the guys who go on about how amazing their Pi is and why are you even spending money on games when the Pi exists are going to be saying the same thing with three dozen broken MISTer cores instead of choppy SNES and CPS2 emulation...

If it improves in time, I'll end up picking one up, if even just for 486 core. But it's so much in its infancy that I don't understand why anyone is excited about it (unless of course you're interested in writing the cores yourself.)
 
I'm ready too for the MiSTer project and yes, i will have to spare 3 IO boards and 3 SDRam boards.
 

Attachments

  • DE10-Nano-MiSTer.JPG
    DE10-Nano-MiSTer.JPG
    751.5 KB · Views: 127
The MiSTer for console cores, give it a few years for accuracy as people learn how to code for the consoles.. the Mist has been around for a long time for non consoles PC type emulation and those cores were ported to the MiSTer and that is what I am looking to play on the MiSTer, more specifically Amiga stuff. The MiSTer will introduce me to his Amiga and we will become Amigos:)..I'll let myself out :D
 
I am an atheist otherwise this would have been the first FPGA prophet. I am thinking mainly from latency point. PI with Retrotink is perfect except for the latency.

I did get my board yesterday but you need a OTG cable to run anything input/output since the dev board doesn't have a full size USB port. The OTG hub I got today does not work.

This would be more expensive compared to a PI setup (when you add SDRAM board and LL Cooljoy), but I really want to see if we can get lower latency than PI.

Anyone who just plays Adventure/Strategy/RPG/Puzzle or other games which relies less on reflex doesn't need to get this to be honest.
 
Last edited:
I have one and I love it!

15khz PC Engine SuprGfx gaming :thumbsup:
The Amiga I never had :thumbup:
Finally play 128kb Spectrum :D
Possibly soon AES/MVS core =O
Slowly arcade games are added - in few years I expect bigger arcade games to be there too :)

(Yes you can do SW emulation, but experience with HW is better)
 
Seriously that people are comparing hardware emulation with software?
 
Seriously that people are comparing hardware emulation with software?
I would call it Hardware Simulation instead of Emulation :)

And why would that comparison be a thing for you? Something running on an FPGA can be just as inaccurate as SW Emulation.
 
This look interesting. Is there a list of available or wip arcade cores?
 
Seriously that people are comparing hardware emulation with software?
Based on the last time I looked at MiSTer, technically speaking almost all of the cores implemented are no better than emulation simply because they are not cycle accurate. That is the same problem with software based emulate, it is not cycle accurate.

It takes a serious investment of time and skill to create cycle accurate cpu cores. As mentioned above about kevtris and his work on 6502, that took him 3 years to create. He taught me a lot about verilog programming and design.
 
And why would that comparison be a thing for you? Something running on an FPGA can be just as inaccurate as SW Emulation.
^ This. When I first started looking into the MISTer, I thought that FPGA was the Magic Solution where everything would be accurate to the original hardware with no lag. Unfortunately that's not the case - many of the cores have graphical, audio and timing issues. For example, the Sega Genesis / Mega Drive core is looking decent today, but it was a mess just 3 months ago. The Sharp X68000 core currently doesn't run anything at a playable speed.

Recommended reading: the author of the high-accuracy SNES emulators BSNES and Higan, wrote an excellent article talking about why FPGA isn't the perfect solution that it appears to be:
https://byuu.org/articles/fpgas-arent-magic/
 
Seriously that people are comparing hardware emulation with software?
I would call it Hardware Simulation instead of Emulation :)
And why would that comparison be a thing for you? Something running on an FPGA can be just as inaccurate as SW Emulation.
When was it said by me that there would be no problems? The fact is that it can not be compared because the logic of implementation is at opposite levels, and there can not be this kind of equivalence.
 
Recommended reading: the author of the high-accuracy SNES emulators BSNES and Higan, wrote an excellent article talking about why FPGA isn't the perfect solution that it appears to be:
https://byuu.org/articles/fpgas-arent-magic/
Very nice read indeed.


When was it said by me that there would be no problems? The fact is that it can not be compared because the logic of implementation is at opposite levels, and there can not be this kind of equivalence.
They are different, yes. Yet, you can still compare them - both are implementations by humans. The article that @ShootTheCore linked in his above post the comparison is reflected pretty good.
 
They are different, yes. Yet, you can still compare them - both are implementations by humans. The article that @ShootTheCore linked in his above post the comparison is reflected pretty good.
I respect Byuu very much, but his article is very shallow about implementation, only being clear on the issue about input lag. There is more rage than computational engineering in your words.
 
They are different, yes. Yet, you can still compare them - both are implementations by humans. The article that @ShootTheCore linked in his above post the comparison is reflected pretty good.
I respect Byuu very much, but his article is very shallow about implementation, only being clear on the issue about input lag. There is more rage than computational engineering in your words.
Where on Earth did you find rage in saying they're different but can still be compared?
 
Back
Top