orange808 wrote:WinUAE exists. It's just not getting hyped on YouTube.
So, your argument about more development (on a given hardware platform) eventually matching achievements of other projects (on the same platform) and reproducing real hardware behaviors can be applied directly to software emulation as well.
It can, but as I said hardware simulation is not the same as emulation. It has many advantages over emulation. Like being able to use the actual code and requiring incredibly less resources to be hardware accurate. Also once you make an accurate simulation of of a chip as an fpga any other project that uses that chip will be able to integrate that code and be more accurate as a result.
Byuu is honestly just being biased towards emulators because he has put so much time into them and his feelings are hurt. I've read his anti fpga posts before and they are mostly him crying about stuff and saying things that simply are not correct.
orange808 wrote:WinUAE exists. It's just not getting hyped on YouTube.
So, your argument about more development (on a given hardware platform) eventually matching achievements of other projects (on the same platform) and reproducing real hardware behaviors can be applied directly to software emulation as well.
It can,
Yep.
Wolf_ wrote:
but
Shouldn't have added a "but."
Wolf_ wrote:
as I said hardware simulation is not the same as emulation. It has many advantages over emulation.
You mean potential future advantages when better FPGA hardware is available at affordable prices, right?
Wolf_ wrote:
Like being able to use the actual code and requiring incredibly less resources to be hardware accurate. Also once you make an accurate simulation of of a chip as an fpga any other project that uses that chip will be able to integrate that code and be more accurate as a result.
That's complete jibberish.
All emulators run "actual code". All emulators use potentially reusable bits. That's not restricted to FPGA.
Wolf_ wrote:
Byuu is honestly just being biased towards emulators because he has put so much time into them and his feelings are hurt. I've read his anti fpga posts before and they are mostly him crying about stuff and saying things that simply are not correct.
On the contrary it's all correct. You read his work, but you didn't understand it.
orange808 wrote:WinUAE exists. It's just not getting hyped on YouTube.
So, your argument about more development (on a given hardware platform) eventually matching achievements of other projects (on the same platform) and reproducing real hardware behaviors can be applied directly to software emulation as well.
It can,
Yep.
Wolf_ wrote:
but
Shouldn't have added a "but."
Wolf_ wrote:
as I said hardware simulation is not the same as emulation. It has many advantages over emulation.
You mean potential future advantages when better FPGA hardware is available at affordable prices, right?
Wolf_ wrote:
Like being able to use the actual code and requiring incredibly less resources to be hardware accurate. Also once you make an accurate simulation of of a chip as an fpga any other project that uses that chip will be able to integrate that code and be more accurate as a result.
That's complete jibberish.
All emulators run "actual code". All emulators use potentially reusable bits. That's not restricted to FPGA.
Wolf_ wrote:
Byuu is honestly just being biased towards emulators because he has put so much time into them and his feelings are hurt. I've read his anti fpga posts before and they are mostly him crying about stuff and saying things that simply are not correct.
On the contrary it's all correct. You read his work, but you didn't understand it.
Just because you don't understand it doesn't mean it's gibberish. Just running the accuracy version of Byuus snes emulator takes an incredibly beefy computer unlike the Super NT and MiSTer. It's very obvious that fpga accuracy requires less resources to achieve accurate hardware simulation. No one is arguing against that. And as I said once you accurately map out a chip in fpga language (like the fairly recent 6800k mapping) anything that uses that chip can implement that code and be more accurate as a result. Please let me know if anything else I've said has confused you and I'll be more than happy to continue explaining it to you.
Yep. The ability to produce good **emulation** at relatively low prices was never a point of contention. I didn't say FPGA's couldn't do acceptable emulation for less money. I didn't say FPGAs couldn't do it with less power.
However, you're back on the accuracy thing again and that's pure bullshit.
Your insistence on focusing solely on SuperNT versus Higan is also bullshit. Byuu has famously struggled to implement WASAPI and he basically thumbed his nose at Rejhon and beam racing, so Byuu is choosing to make a slow emulator.
Byuu's choice has nothing to do with the potential of software emulation.
orange808 wrote:Yep. The ability to produce good **emulation** at relatively low prices was never a point of contention. I didn't say FPGA's couldn't do acceptable emulation for less money. I didn't say FPGAs couldn't do it with less power.
However, you're back on the accuracy thing again and that's pure bullshit.
Your insistence on focusing solely on SuperNT versus Higan is also bullshit. Byuu has famously struggled to implement WASAPI and he basically thumbed his nose at Rejhon and beam racing, so Byuu is choosing to make a slow emulator.
Byuu's choice has nothing to do with the potential of software emulation.
Actually you're the only one focused soley on the Super NT. I mentioned both the MiSTer and the Super NT. In fact the MiSTer is a far more amazing fpga snes because it can also handle the effect chips as well and will be getting more relatively soon. Well deserved hype.
And of course needing a thousand dollar machine to emulate a snes accurately is relevant when a >$200 setup can do the same. Not to mention the ability to use real hardware and ect. Last I checked Byuus emulator doesn't offer support for running carts on it (or backing them up) like the Super NT does.
Now if anything I've said confuses you feel free to just ask about it. You don't have to call it gibberish because it is above your head.
Wow, this escalated quickly. WHO HERE LIKES VIDYA GAMES, SAY YEAAAAAAAAH!!!
Few of us are technical enough to have informed opinions on the accuracy of any given emulator/fpga solution. Here's something to think about: FPGA can reproduce results that are extremely close to original hardware, at a very reasonable price, without the overhead of a full desktop OS and all of the pros and cons that come with one.
Maybe it's just me, but I want akin to an appliance when I'm playing games. This, of course doesn't mention the classic computing platforms that are also hugely beneficial as original hardware begins to go by the wayside.
At the end of the day, there isn't a single perfect solution for everyone. Software emulation can do amazing things, despite its hardware being vastly different from the source console/computer. FPGA simulation can do incredible things for retro gaming and computing, despite its low-powered nature.
Dont have any history with Ash just noticed a few of his projects are unfinished.
Theres no documented info on BlisSTer yet.
Board design is being worked on and for now we are just connecting a blissbox or gamer pro to mister via SPI.
Sorlig formalized the SPI code recently too.
Want more info or keen to help out testing or coding feel free to join the Blissbox discord.
Do you have an invitation to the Discord server? I'd love to check it out.
Also, I've been looking at the Bliss Box for years. Do many people own them? I'm not sure what I'd charge for the LL Cool Joy, but it'll definitely be the most expensive IO board yet, as it has 47 more components than the IO board (118 vs 71) and they're densely populated. Considering the BlissBox is roughly $150 with the base set of adapters, and the LL Cool Joy would need MC Cthulu RJ45 adapters for many controllers, it'll be interesting to see how everything settles.
I'm filling out a BOM for the current revision of the LL Cool Joy in hopes of soldering and testing for Ash.
thrasherx wrote:Do you have an invitation to the Discord server? I'd love to check it out.
Also, I've been looking at the Bliss Box for years. Do many people own them? I'm not sure what I'd charge for the LL Cool Joy, but it'll definitely be the most expensive IO board yet, as it has 47 more components than the IO board (118 vs 71) and they're densely populated. Considering the BlissBox is roughly $150 with the base set of adapters, and the LL Cool Joy would need MC Cthulu RJ45 adapters for many controllers, it'll be interesting to see how everything settles.
I'm filling out a BOM for the current revision of the LL Cool Joy in hopes of soldering and testing for Ash.
I have the BlissBox from the original failed kickstarter and it's sitting in a box. I have all the cables, I even bought newly released cables, but haven't bothered to try it. I don't really know why, I suppose it's just easier to use a universal USB controller than to switch in and out?
I'm looking into trying out one of these, especially for MSX/2 and X68000 games. How are those two cores doing at the moment? Has anyone heard anything about the possibilities of an NEC PC98 core?
And does any keyboard plugged through USB work on the Mister? I only have a Mac keyboard, would be nice if it works.
fernan1234 wrote:I'm looking into trying out one of these, especially for MSX/2 and X68000 games. How are those two cores doing at the moment? Has anyone heard anything about the possibilities of an NEC PC98 core?
And does any keyboard plugged through USB work on the Mister? I only have a Mac keyboard, would be nice if it works.
I believe that MSX/2 is working. Search Youtube for examples. I don't know anythign about a PC98 core.
X68000 isn't close. Supposedly there's a Japanese developer that has a decent FPGA core but it would need to be ported and there's a language barrier.
All USB keyboards should work providing: you have a USB OTG adapter, and you don't have LED backlighting. The LED lights can draw a substantial amount of current and would merit a separate power supply connected to a dedicated MiSTer USB Hub.
MSX core is pretty decent, no TurboR, laserdisc, MIDI or network support. It's based off the OneChipMSX project so it's pretty comperable to Zenmix minus real hardware support.
x68000 core isn't all that good, best to stick with software emulation on that front still not the same as the real thing.
I recently set up an I/O board and I gotta say, the MiSTer is now the value proposition of the universe. It's still missing some modern conveniences (like a ******' power switch) but there's no beating the form factor, compatibility and lag-free output options.
Although this does have Bliss-Box ports on it, it is being made to offer many features. Like LLAPI switching mode, in out power plug, a power toggle button, linkage port to the main board (used with that bridge board), many LEDS to know what is going on, USB plugs and headers, a remote power header, 5 usb and 2 bliss-box ports, and a strong nice made product by a trusted manufacture.
Galgomite wrote:I recently set up an I/O board and I gotta say, the MiSTer is now the value proposition of the universe. It's still missing some modern conveniences (like a ******' power switch) but there's no beating the form factor, compatibility and lag-free output options.
Even now more affordable as with direct video over HDmi the io board is no longer necessary for analog out. Bettet quality too!
Sent from my SAMSUNG-SM-G891A using Tapatalk
Displays I currently own:
LG 83C1(OLED),LG 77C2(OLED), LG 42C2(OLED),TCL 75R635(MiniLED),Apple Studio Monitor 21(PCCRT),SONY 34XBR960x2(HDCRT)
SONY 32XBR250,Samsung UBJ590(LED),Panasonic P50VT20(Plasma),JVC NZ8
Even now more affordable as with direct video over HDmi the io board is no longer necessary for analog out. Bettet quality too!
you mean output in 15khz over HDMI and then just add generic HDMI to analogue converter?
But is the HDMI output really AS flexible as as the output using the i/o board? I mean in terms like non standard resolutions or refresh rates? I don't have an overview of the currently available arcade cores, but just think of a possible Irem core in the future where we'd have 256p55 output.
Even now more affordable as with direct video over HDmi the io board is no longer necessary for analog out. Bettet quality too!
you mean output in 15khz over HDMI and then just add generic HDMI to analogue converter?
But is the HDMI output really AS flexible as as the output using the i/o board? I mean in terms like non standard resolutions or refresh rates? I don't have an overview of the currently available arcade cores, but just think of a possible Irem core in the future where we'd have 256p55 output.
Yeah using a generic converter. Sorgelig doesn't really go into how flexible the resolution/refresh rate he can send out of the HDMI is but from this comment it sounds like he might be able to send whatever? From the thread: Original analog video without I/O board? Possible!
Core sends original video to HDMI without any scaling, as-is. Then HDMI-VGA converter acts as a DAC for video and audio.
No any frame buffers are used in VGA converter, so it converts online.
Basically you get zero-lag video. Same as you would use VGA from I/O board.
This video is even better because you get 8bits per color while I/O board gives only 6bits per color. Audio output is also better than one used on I/O board.
I skimmed the rest of the thread and there was some posts mentioning interlace being a bit flaky (More so with the SNES core) although I'm not sure if that's an issue with outputting via HDMI or the DACs people are using.
This would be fantastic if possible. On a Raspberry Pi HDMI does allow outputting 15khz, but, in my experience, only at 60Hz. That may just be because the HDMI output sticks to standard CEA modes (and using custom video modes can lead to vsync problems, and even with custom modes refresh rate can only be set to 60 or 50).
because TO MY KNOWLEDGE, there's also a HDMI transmitter chip on these boards and 15khz timings without horizontal "super resolutions" can fall below the minimum output specs for those chips.
That is true. The cores on MiSTer usually use a high clock for the video. For SNES that is 2 times the master clock at 42.95MHz. This is also what it outputs in direct video mode. So it is outputting 2728x262 at 60.09Hz.
It maxes out the horizontal size of 2048 on my Datapath VisionRGB-E1.
And how are those generic HDMI to VGA/component converters handing these resolutions? Is there any side effect to this which is visible on a CRT afterwards?
Fudoh wrote:And how are those generic HDMI to VGA/component converters handing these resolutions? Is there any side effect to this which is visible on a CRT afterwards?
there should be no effect on the CRT, we just had a thread discussing this at length