Peredonov

I've been following the PSIO closely for the past 3 years. It's definitely been a rollercoaster type of development path, and though I felt pretty disappointed especially last year after multiple regressions were introduced, I am now feeling more optimistic than not about it. The 99.9% compatibility they have advertised for all these years has been false advertisement, which is frustrating. That said, it appears to be closer to being true now than ever, though besides a small number of known incompatibilities in specific titles, there remain at least two significant inaccuracy issues that impact all games in general, rather than specific titles. One is that the stereo channels can be randomly flipped on start or re-start of an audio track. This affects CDDA audio, XA audio, and STR audio (used in FMVs). Base SPU audio doesn't seem to be affected though. PSIO has to decode all of such audio because the original process for doing so cannot be accessed from the parallel I/O port. This means that accuracy of such audio will depend entirely on the quality of Cybdyn's code. This issue has been known at least since last year, and there are no updates showing any progress, though the team is reportedly investigating it still. The other general issue was introduced in one of the firmware updates last year, which causes some games to show frame stuttering in certain conditions (more so than on a real and good-condition CD-ROM). There are also no updates about progress in fixing this issue at this point. Response to individual title bugs has been very good, however, and almost all seem to be fixed for the next unreleased update (which hopefully will not introduce regressions). One of the devs even brought up the possibility of a patch for Vib Ribbon to run on PSIO. So that 99.9% compatibility claim will probably be true eventually, but the degree to which "compatibility" and "accuracy" will overlap is still uncertain.

You wouldn't want to market it with the term "clone console" or "hardware clone" because it doesn't sound sexy at all. Something like "FPGA recreation" sounds better. Speaking of sounding better, one of the most difficult things to get right with old fashioned hardware cloning as well as FPGA "cloning" is sound reproduction. This is mainly because sound from old systems does not only depend on the chip generating the audio signals, but also resistor networks, amps, DACs, analogue distortions, filters, etc. Going back to the 1-CHIP, we know that it sounds identical to the original console with discrete S-SMP chip (again, as far as I know). Now take the case of the Famicom/NES, which has had more "clone" attempts than any other system, probably, focusing only on audio. This can help us explore degrees of separation in the original-clone spectrum. The NES uses the same chips as the original Famicom, but obviously with a different mobo and some modifications like AV output and lack of pins for cartridge expansion audio. Late Famicom consoles and early Famicom Twins sound different from the early Famicom due to different resistor arrays, especially in regard to expansion audio balance (in the case of the early TF, a manufacturing mistake). AV Famicom and Toaster use different chip revisions for the CPU, and the sound has a distinct signature. Then perhaps there is the original non-mini Analogue NT, which used original Famicom chips, with audio very close to an original Famicom, though not identical (for starters, with less analogue buzz) and user configurable for channel separation. There have been Chinese clones that used harvested original chips also. Then there are the Dendy clones, which sound obviously different, as well as currently available Chinese clones that use Dendy chips. So you have clone consoles with clone chips. There are also hybrid Chinese consoles that use Dendy clone CPUs but an emulated PPU using FPGA. And then you arrive at something like the NT Mini which uses the FPGA for all components, and audio output is adjusted at the software level.

Just want to add to the current discussion that it's a bit disingenuous to equate a "hardware clone" such as an FPGA-based console or platform to a 1-CHIP SNES/SFC as an official "clone" console. As far as I know, there aren't any full incompatibility issues with 1-CHIPs, in the sense that there's no official titles that won't run on it. Instead, there are only a handful of minor inaccuracies. A helicopter or airplane's shadow that doesn't display correctly in some action game I can't recall, some flickering small line in a Street Fighter game... what else is there? In any case, the crucial difference would be not only the type of implementation of such a "clone", but the fact that it is officially vetted by the original manufacturer.

If you install this on one of the consoles first, play through the entire library in a few years and then move it to the other console and play that library, you really got your money's worth ? Seriously, if compatibility is at least as good as it is on Phoebe/Rhea for Saturn and USB-GDROM (which is slightly higher than GDEMU), then this device is definitely worth considering given what it brings uniquely to the table (better file format support and better UI). One question I have is whether these devices would preserve the ability to use the original disc player interface of each system. Those are nice to keep from a purist perspective, and they also allow to play music CDs, or play back game discs as music CDs to listen to "hidden" messages on some of the audio tracks (usually funny messages by the cast telling you not to use the disc on music CD players).

The PCE/TG16 core on MiSTer is already pretty solid, and getting closer to virtually complete accuracy. They even decapped and scanned the CPU recently for it. And with all of that, it doesn't seem trivial to add CD-ROM support at this time. It's only a matter of time though, and if Analogue were interested in a fully featured PCE FPGA clone with CD library support they would definitely take a look at the open source code at least. But as others have said, Analogue will think twice about such a product given that the popularity in the West of the PCE (especially the CD library which is mostly obscure and in Japanese) is limited beyond a couple of luminaire games like Rondo of Blood. And the Polymega is just a good software emulation machine with a great UI and PnP capabilities. It's not really in the same category as any FPGA-based gaming device.

Hi everyone, It looks like this is still a great place to ask for advice about the Nt Mini. As I have one on the way and contemplate whether it will become my main NES/Famicom machine, I am wondering if a couple of issues regarding audio that early reviews pointed out were resolved later or have workarounds. First, regarding (low pass) filtering on audio from carts/RAM adapter. I have an RAM Adapter+FDSStick, which I would prefer to use instead of the internal emulated FDS sound, which is apparently very good but not identical to the real thing. I was always confused when it was reported that the Nt Mini lacks the resources to add a low pass filter for FDS games, but I'm not sure if that referred to FDS games loaded from SD card, from a RAM Adapter, or both. In any case, I gather that internal audio and external audio from the RAM adapter--and perhaps also from carts with expansion audio--will sound different as one will have filtering while the other does not. Is this correct? And if so, has there been an update or workaround to address it? I strongly prefer to rely on expansion audio from real hardware rather than the emulated chips, for 100% sound accuracy. Second, I recall that expansion audio chips need to be enabled every time a cart is used, and that this setting cannot be saved permanently but is reset on each boot. Is this still the case? Now that the Nt Mini is being re-released I'm wondering if kevtris will revisit the NES core. It sounded like there were a couple of sound issues fixed on the latest official firmware releases (https://support.analogue.co/hc/en-us/articles/115001947108-Nt-mini-Firmware-Update-v2-3), but details were not specified and those may just refer to changes also available in the last jailbreak firmware (2.3 official vs. 2.0 jailbreak, not sure if the numbering correlates). Thanks for any info from anyone in the know!