How E.S.P. works is you just get a onetime code with the synth, and it can only be redeemed once ever.

Nuked used a non commercial license specifically to get Roland to look the other way, and it seems to have worked. But a non commercial license will probably limit the usefulness of this, because it's clearly intended to be used in music production, which is commercial use.

And legally, this is indeed more problematic than the use case for nuked-sc55 (playback old midis).

But they haven't stopped mame from trying to emulate their hardware yet. which gives hope that the bring your own rom will cause no trouble.

If Yamaha is informed of this before public distribution, the hope is that they adapt it to the state that it will playback all the midis properly, and then sell it at a fair price. I do think as long as this is not publically distributed before Yamaha sees it, an arrangement might get worked out. if attention is gotten by means of public dsitribution, that is likely to go much more wrong if they care.

zaphod77 wrote on 2026-06-03, 14:17:

In particular, the reverb, chorus, filter, and envelope stuff *must* be correct, because these parameters modify the original sa […]
Show full quote

In particular, the reverb, chorus, filter, and envelope stuff *must* be correct, because these parameters modify the original samples to create the default patches.

Can we do better inside a daw? sure. but to playback existing midi data, the already existing stuff needs to be implemented correctly. and that's what i want. I'd like to see this project playback the entire XG ff7 soundtrack correctly.

I would like to see this project to end up being dual licensed, with yamaha being given a separate license that lets them (and them only) do whatever they want with it (like sell us an s-yxg2000), while the free download will not come with roms, and have a basic license that requires that you own the synth that you dumped the ROMs for. Yamaha's ESP only comes as a free license with the actual hardware, so this new project should work the same. if you own real hardware, you can use it, while if you don't you can't. I think that as long as it works that way, Yamaha won't have an issue. I am curious if enough can be dumped from the actual synth using sysex instead of a rom dump, though i'd guess you can't extract the samples this way.

Then Yamaha could maybe make a new s-yxg50, s-yxg80, s-yxg100, s-yxg125, and s-yxg2000. with the prices going up as the numbers do. S-yxg2000 i could see having a similar price to Sound Canvas VA, with s-yxg50 being anywhere from 20$ or so to free. Most people who just want to play midi files would use the remade s-yxg50, while serious xg music makers would want the 2000.

It's a common misconception that XG FX are used as part of a programs 'sound' internally. There's no MEG element involved. All programs are dry and processed with AEG and FEG1 to 4 only depending on element count, no FX DSP is used by XG internally, it's purely user FX as send and variation.

I've reverse engineered all the FX DSP from syxg50, so Chorus and Reverb FX are faithful, just sound better due to not limited by hardware constraints of the 90s

The things is, this project is for music production, so the focus is on the sound, not the FX, but 90s dsp is well documented, it won't be difficult to emulate. Although Yamaha were quite clever in using the same DSP for many uses, it's not like there's Reverb DSP, Chorus DSP, Flanger DSP or Delay DSP etc, it's all basically the same thing sharing common DSP.

Bu, all sysex and midi is built in and all FX do route to real FX DSP that does something so it will play a midi file or stream and respond similarly. I've used all existing parameters so the FX are constrained to the limits of the hardware. It all matches in that regard.

All the licencing stuff is only going to be known after I send them the final synth. It is already sounding so much better than the hardware and takes many aspects of it beyond what was possible in hardware, so we will see. If not, I'll have quite a nice synth nobody on the planet has to make music with which has value in itself.

that's odd, because other people i've chatted with do state that the filter, and the envelope generators have default settings.

And these default settings are applied to the parts. the adjustments done by the mid controllers are relative to these defaults set for the part!

THE reverb and chorus are separate. you are correct there. but the following parameters are relative.

nn 15 1 00 - 7F VIBRATO RATE -64 - +63 40
nn 16 1 00 - 7F VIBRATO DEPTH -64 - +63 40 (drum part ignores)
nn 17 1 00 - 7F VIBRATO DELAY -64 - +63 40 (drum part ignores)
nn 18 1 00 - 7F FILTER CUTOFF FREQUENCY -64 - +63 40
nn 19 1 00 - 7F FILTER RESONANCE -64 - +63 40
nn 1A 1 00 - 7F EG ATTACK TIME -64 - +63 40
nn 1B 1 00 - 7F EG DECAY TIME -64 - +63 40
nn 1C 1 00 - 7F EG RELEASE TIME -64 - +63

these are all relative. they are changes from what is in the patch.

say you grab that silly xg by dls file. it has the samples. but they are recored with the filter and envelopes alreacy applied. so if you try to playback midi with them that actually messed with them, things sound completely wrong.

can mu2k vsti record the midis?

the vsti work won't take longer, isn't it?

hockinsk wrote on 2026-06-05, 08:27:

It's a common misconception that XG FX are used as part of a programs 'sound' internally. There's no MEG element involved. All p […]
Show full quote

zaphod77 wrote on 2026-06-03, 14:17:

In particular, the reverb, chorus, filter, and envelope stuff *must* be correct, because these parameters modify the original sa […]
Show full quote

In particular, the reverb, chorus, filter, and envelope stuff *must* be correct, because these parameters modify the original samples to create the default patches.

Can we do better inside a daw? sure. but to playback existing midi data, the already existing stuff needs to be implemented correctly. and that's what i want. I'd like to see this project playback the entire XG ff7 soundtrack correctly.

I would like to see this project to end up being dual licensed, with yamaha being given a separate license that lets them (and them only) do whatever they want with it (like sell us an s-yxg2000), while the free download will not come with roms, and have a basic license that requires that you own the synth that you dumped the ROMs for. Yamaha's ESP only comes as a free license with the actual hardware, so this new project should work the same. if you own real hardware, you can use it, while if you don't you can't. I think that as long as it works that way, Yamaha won't have an issue. I am curious if enough can be dumped from the actual synth using sysex instead of a rom dump, though i'd guess you can't extract the samples this way.

Then Yamaha could maybe make a new s-yxg50, s-yxg80, s-yxg100, s-yxg125, and s-yxg2000. with the prices going up as the numbers do. S-yxg2000 i could see having a similar price to Sound Canvas VA, with s-yxg50 being anywhere from 20$ or so to free. Most people who just want to play midi files would use the remade s-yxg50, while serious xg music makers would want the 2000.

It's a common misconception that XG FX are used as part of a programs 'sound' internally. There's no MEG element involved. All programs are dry and processed with AEG and FEG1 to 4 only depending on element count, no FX DSP is used by XG internally, it's purely user FX as send and variation.

I've reverse engineered all the FX DSP from syxg50, so Chorus and Reverb FX are faithful, just sound better due to not limited by hardware constraints of the 90s

The things is, this project is for music production, so the focus is on the sound, not the FX, but 90s dsp is well documented, it won't be difficult to emulate. Although Yamaha were quite clever in using the same DSP for many uses, it's not like there's Reverb DSP, Chorus DSP, Flanger DSP or Delay DSP etc, it's all basically the same thing sharing common DSP.

Bu, all sysex and midi is built in and all FX do route to real FX DSP that does something so it will play a midi file or stream and respond similarly. I've used all existing parameters so the FX are constrained to the limits of the hardware. It all matches in that regard.

All the licencing stuff is only going to be known after I send them the final synth. It is already sounding so much better than the hardware and takes many aspects of it beyond what was possible in hardware, so we will see. If not, I'll have quite a nice synth nobody on the planet has to make music with which has value in itself.

PLEASE release something before you send to Yamaha because otherwise your work would be lost. We do not want hoarders to exist, period. I do not even see how on Earth it is a good idea to let Yamaha in on any of this BEFORE helping us. Oh and please don't leave your videos unlisted, and PLEASE give people who do NOT want to reveal their real names by making Facebook accounts the MU2000 English Manual you got translated. You're going about ALL of this the exact wrong way. Also PLEASE make the code FOSS.

For context I'm trying to resurrect Silicon SoundFonts (Section 11 of the SF2 spec) in a planned ROMpler (when I can actually afford to buy stuff as pricey as custom chips from China) that makes HQ chip sounds, but SC-8850/VA-76 GS, albeit a subset, and NOT the SC-55 subset (keep in mind that in terms of SCs I ONLY appreciate CM-500 or SC-88 and above, up to VA-76), and ITS sample set is libre and sort of like the PLG150-DX Factory patch set, albeit generated quite differently than it or a proper DX7. I even have a BIOS font image made, as well as the audio ROM. As such, I'm not naive when it comes to ROMplers, even though I've currently only achieved the firmware creation yet. I also own a Roland SC-D70, Korg DS-8, Roland D-10, and Casio WK-210, all of which are decent MIDI devices. I'm currently hunting around for a good deal on an MU2000EX to see how much GS it has.

ALSO tcaud and zaphod thinking humoring Yamaha is wise isn't exactly something wise to speak of. Sorry for the triple post, but God, this thread makes me so peeved, it's not even funny. The sheer amount of people barking up the wrong tree is getting to me.

the mu2000ex is a licensed GS implementation.

but i can pretty much guarantee you it's the same implementation their TG300 has in B mode.

All the change to the EX did was allow them to actually show the GS logo.

And yes, it includes the mt-32 voices. exact same instrument names as the mu-50.

https://archive.org/details/manualsbase-id-224151/mode/2up

The s-yxg50 is the same. they just paid roland to be able to actually use the GS logo.

This set is 100% the sc-88 soundmap, but done with yamaha samples.

every mu series synth suports the sc-88 instruments in the GS mode.

stgiga wrote on 2026-06-11, 03:54:

For context I'm trying to resurrect Silicon SoundFonts (Section 11 of the SF2 spec) in a planned ROMpler (when I can actually afford to buy stuff as pricey as custom chips from China) that makes HQ chip sounds, but SC-8850/VA-76 GS, albeit a subset, and NOT the SC-55 subset (keep in mind that in terms of SCs I ONLY appreciate CM-500 or SC-88 and above, up to VA-76), and ITS sample set is libre and sort of like the PLG150-DX Factory patch set, albeit generated quite differently than it or a proper DX7. I even have a BIOS font image made, as well as the audio ROM. As such, I'm not naive when it comes to ROMplers, even though I've currently only achieved the firmware creation yet. I also own a Roland SC-D70, Korg DS-8, Roland D-10, and Casio WK-210, all of which are decent MIDI devices. I'm currently hunting around for a good deal on an MU2000EX to see how much GS it has.

Exhibit-A why the Japanese don't fuck around with their IP...

tcaud wrote on 2026-06-11, 23:19:

stgiga wrote on 2026-06-11, 03:54:

For context I'm trying to resurrect Silicon SoundFonts (Section 11 of the SF2 spec) in a planned ROMpler (when I can actually afford to buy stuff as pricey as custom chips from China) that makes HQ chip sounds, but SC-8850/VA-76 GS, albeit a subset, and NOT the SC-55 subset (keep in mind that in terms of SCs I ONLY appreciate CM-500 or SC-88 and above, up to VA-76), and ITS sample set is libre and sort of like the PLG150-DX Factory patch set, albeit generated quite differently than it or a proper DX7. I even have a BIOS font image made, as well as the audio ROM. As such, I'm not naive when it comes to ROMplers, even though I've currently only achieved the firmware creation yet. I also own a Roland SC-D70, Korg DS-8, Roland D-10, and Casio WK-210, all of which are decent MIDI devices. I'm currently hunting around for a good deal on an MU2000EX to see how much GS it has.

Exhibit-A why the Japanese don't fuck around with their IP...

What do you mean by that? I didn't say this chip would use Roland or Yamaha samples.

zaphod77 wrote on 2026-06-05, 21:17:

that's odd, because other people i've chatted with do state that the filter, and the envelope generators have default settings. […]
Show full quote

that's odd, because other people i've chatted with do state that the filter, and the envelope generators have default settings.

And these default settings are applied to the parts. the adjustments done by the mid controllers are relative to these defaults set for the part!

THE reverb and chorus are separate. you are correct there. but the following parameters are relative.

nn 15 1 00 - 7F VIBRATO RATE -64 - +63 40
nn 16 1 00 - 7F VIBRATO DEPTH -64 - +63 40 (drum part ignores)
nn 17 1 00 - 7F VIBRATO DELAY -64 - +63 40 (drum part ignores)
nn 18 1 00 - 7F FILTER CUTOFF FREQUENCY -64 - +63 40
nn 19 1 00 - 7F FILTER RESONANCE -64 - +63 40
nn 1A 1 00 - 7F EG ATTACK TIME -64 - +63 40
nn 1B 1 00 - 7F EG DECAY TIME -64 - +63 40
nn 1C 1 00 - 7F EG RELEASE TIME -64 - +63

these are all relative. they are changes from what is in the patch.

say you grab that silly xg by dls file. it has the samples. but they are recored with the filter and envelopes alreacy applied. so if you try to playback midi with them that actually messed with them, things sound completely wrong.

Not sure what the reply relates to, I assume MEG? Yes, you're correct AEG and FEG you list above are bespoke per program/voice. MEG isn't, MEG is purely a user only DSP applied. all programs have a default hall1 reverb, but that's not per voice, it's simply a global value for the send amount across everything.

For those interested this is the dsp flow i'm using in MUXG2K and follows the hardware faithfully

Per-sample, per-voice:

Streaming — read sample from wave ROM (or pre-baked PCM blob). DPCM decoded if format=3. Resampled via 32-tap Kaiser sinc (FUTURE) at the pitch step m_pitch + LFO pitch mod.

Filter stage 1 + stage 2 — two Chamberlin SVF / direct-form IIR cells in series. Cutoff/resonance driven by the filter envelope timeline (captured from chip on note-on, replayed per-sample). FUTURE mode adds 1 ms coefficient glide.

IIR1 — extra one-pole for final tone shaping (used by ~all voices, often as a soft HF damper).

AEG — 4-stage envelope ATTACK → DECAY1 → DECAY2 → RELEASE produces a 14-bit attenuation level. FUTURE mode adds 2 ms LPF on the level output (dezippers slow sweeps).

LFO amp mod added on top of AEG level → volume_apply exponentiates this into a multiplier applied to the filtered sample.

Mixer — apply_simple runs the chip's vol/route encoding for pan (L/R), or float future_pan_L/R if a KSP voice. Then per-part CC#7 / CC#10 / CC#11 extra-att stages (FUTURE: clean exp gain + per-sample smoothed; FAITHFUL: chip's att encoder with cliffs).

Voice sum — all active voices summed into master L/R + parallel FX accumulators (CC#91/93/94 scale per-voice into reverb/chorus/variation send buses).

Float normalise — int → float via kS17Norm × kChipLevelNorm (≈ −12 dB chip-natural amplitude).

Future Warmth — gentle 2× oversampled soft saturation on the dry master only.

FX wet — Variation → Chorus → Reverb (chip-faithful MEG when state files loaded, modelled DSP fallback). Returns + pans mixed back to L/R via smoothed gains.

Master — XG master volume × attenuator × +6 dB plugin makeup, all through a single per-sample-smoothed gain. Hard clamp ±1.0.

silvally03@ wrote on 2026-06-06, 12:46:

can mu2k vsti record the midis?

Record the midis? It accepts all the XG midi and deals with it the same as the hardware does if that's what you mean. There's a midi-thru port so you can use it as an editor of other XG devices if that's what you mean?

stgiga wrote on 2026-06-11, 03:48:

PLEASE release something before you send to Yamaha because otherwise your work would be lost. We do not want hoarders to exist, period. I do not even see how on Earth it is a good idea to let Yamaha in on any of this BEFORE helping us. Oh and please don't leave your videos unlisted, and PLEASE give people who do NOT want to reveal their real names by making Facebook accounts the MU2000 English Manual you got translated. You're going about ALL of this the exact wrong way. Also PLEASE make the code FOSS.

It started out as a personal project because I wanted something easier to use than the SYXG50 VSTi, and without relying on cracked software like most of the available options do. I'm not willing to take risks with someone else's intellectual property, whether that's through a FOSS release or putting it out underground and dealing with the stress that comes with that. If this ever goes anywhere, I want Yamaha involved.

At the end of the day, it's just a synth, and the original hardware already exists and MAME too if someone really needs the full XG sound set in hardware or software format. This isn't about hoarding anything. I'm the one who's spent a huge amount of time over the last year figuring this out, so I get to decide what happens with it.

It's simply not in my nature to take someone else's property, reverse engineer it, and then release it for everyone else to benefit from while potentially damaging Yamaha's interests. The project only exists because Yamaha created the hardware in the first place but never really explored the software potential of XG and the MU series.

I also like to think that when you've spent over a year working on something and approach a company openly and transparently, the answer isn't always an automatic closed door. That's not how life works, in my opinion. Sure, the odds are probably against Yamaha agreeing to license the ROM, and if they say no, that's completely fine. It's their property, and I respect that.

@hockinsk That's a shame that this project is now officially completely dead in the water. It would have been of benefit to all, including the Japanese themselves.

These Japanese corporations have never been known for being friendly to foreigners, no matter how nicely you approach them. To make matters worse, Yamaha in particular does have an "automatic closed door" policy toward just this sort of thing, but do as you wish because, as you say, this is your work to do with as you like.

I would have loved to have seen this approached as an open source project, so the inner workings of XG would not continue to be hidden through obscurity. It's very doubtful that anyone else will ever be so inspired to put this amount of effort into such a project again after all of these years have passed since XG was still relevant. The generation of people that actually care anything about this will soon die off and that will be the end of this era of MIDI being properly preserved through emulation. Very sad indeed.

hockinsk wrote on Today, 11:18:

It started out as a personal project because I wanted something easier to use than the SYXG50 VSTi, and without relying on crack […]
Show full quote

stgiga wrote on 2026-06-11, 03:48:

PLEASE release something before you send to Yamaha because otherwise your work would be lost. We do not want hoarders to exist, period. I do not even see how on Earth it is a good idea to let Yamaha in on any of this BEFORE helping us. Oh and please don't leave your videos unlisted, and PLEASE give people who do NOT want to reveal their real names by making Facebook accounts the MU2000 English Manual you got translated. You're going about ALL of this the exact wrong way. Also PLEASE make the code FOSS.

It started out as a personal project because I wanted something easier to use than the SYXG50 VSTi, and without relying on cracked software like most of the available options do. I'm not willing to take risks with someone else's intellectual property, whether that's through a FOSS release or putting it out underground and dealing with the stress that comes with that. If this ever goes anywhere, I want Yamaha involved.

At the end of the day, it's just a synth, and the original hardware already exists and MAME too if someone really needs the full XG sound set in hardware or software format. This isn't about hoarding anything. I'm the one who's spent a huge amount of time over the last year figuring this out, so I get to decide what happens with it.

It's simply not in my nature to take someone else's property, reverse engineer it, and then release it for everyone else to benefit from while potentially damaging Yamaha's interests. The project only exists because Yamaha created the hardware in the first place but never really explored the software potential of XG and the MU series.

I also like to think that when you've spent over a year working on something and approach a company openly and transparently, the answer isn't always an automatic closed door. That's not how life works, in my opinion. Sure, the odds are probably against Yamaha agreeing to license the ROM, and if they say no, that's completely fine. It's their property, and I respect that.

I get personal projects and all, but at the very least you could contribute to MAME. Having said that, for the past 11 years I've been an SF2 maker precisely due to how non-crossplatform VSTs are. I just wish MAME was mature enough. I AM however looking for an MU2000EX on Yahoo Japan Auctions that isn't listed as damaged, and isn't ludicrously upmarked. Really what I want to know is what stuff like "Nashville" sounds like. I must underscore that I sort of already did quite a few things over the years that by all means should have angered the synth companies and nobody said anything. Also one of my fans is a Roland employee. So I agree, it's not hopeless. Plus Yamaha even made DIY boards for their YMF825 chip. So I think Yamaha complaining is unlikely, even though they limited the reach of the ESS ESFM but that chip was distinct enough (it faked feedback FM via recursion of 29 hops) and did things that not even post-OPL3 FM chips did, AND in the Ultrasound Extreme served to outpace the OPL4. Now to be completely fair: over the past decade and then some, I've done stuff that you'd expect would anger the companies but apparently does not. I dabbled with XG1 years ago. My HiDef and Tyroland SF2s as well as ancient prototypes of the same idea used Roland and Yamaha ROMpler samples, and it's been years and nobody said nothing even though they are popular. If we're talking *libre* SF2s, I recently came into the interesting fact that my JummBox SoundFont, a CC-BY-SA4 (due to inherited licenses from components, and I acknowledge and support that CC-BY-SA4 is compatible with GPL) SF2, has a lot in common with the PLG150-DX board's factory patches, which are GM and XG. Mine favors a Roland approach to going beyond GM, and in fact some of the patch stuff doesn't fit XG. Now for the record, the melodic patches are all browser-based FM, and I'll set it straight that according to Shaktool, the BeepBox (and by extension JummBox) GM patches are supposed to be based on SGM 2.01 and Timbres of Heaven, though Shan barely sees a resemblance. Shaktool also said that the patches are LPF plus OPN-type (4OP sine operators FM with no Detune2 like on IBM PC Music Feature/FB01 and the synths using OPZ and OPZ2, and with more intricate frequency control like on Channel 3 of OPN chips) FM, except that the result somehow sounds beyond ESFM. It even goes toe-to-toe with the DX, and in some cases, like the SNES-esque-but-still-FM strings. Now, Yamaha DID make later MA chips have filters. Now, by the time Shaktool wrote BeepBox many FM patents expired. But it's still technically competing with Yamaha even though they don't make many FM chips these days because I'm using Silicon SoundFonts to burn it to a proper FM chip. Now the drums are interesting in that they are a fusion of the OPL3 Win9x FatMan patches (the dominant set, does not come from Yamaha) with the OPL2 ones for some lower notes, as well as Piconica (an app that simulates NES+N163) for the Low Beep sound (a square, drum note 22), as well as an SF2 of the YM2612, used for notes above the OPL3 range, and Open Triangle (drum note 81) of that was replaced with the ESFM's default patches' version from a Japanese SF2 of the ES1968 laptop version in an IBM machine, and to cover the SC-88x (including SC-D70, VA-7(6) and VA-5) drum range, the OPLL's Bass Drum is used for the SC-8850 kicks, and above the YM2612+ESFM samples is where two more OPLL rhythm sound samples are mapped to the SC-8850's snare range. Also the SF2 is a subset of SC-8850/8820/SC-D70/VA-7(6)/VA-5 GS, not SC-55 GS as you would assume from there being 212 presets. What matters here is A: real chips were partially involved, B: ESFM was used, C: the OPLL rhythm patches are hardcoded into all OPL-type FM chips, including the ESFM in OPL3 mode. So any one of these factors could have angered Yamaha, but if anything I'd be flattered by any person who puts this particular SF2 to use, because it's WAY less prone to licensing issues than most other SF2s and it has some unique traits that are not common amongst SF2s.

So like, most of my SF2s are technically a bit sketchy, but I made out all right. I honestly cannot stomach the idea of killing one's work. I'm also against the whole delayed release concept, and as much as I'm prone to ignoring terms and merging stuff as I please (and not just SF2s, my VTuber+MMD model, code (though my most polished code ironically turned out libre by sheer accident, as did my 3D art, and the JummBox bank too), art, and song (it is SoundTrap loop city), among many other things, are significantly derivative, transformative, and indiscriminate in terms of components) and run Windows more often than I run Linux, I AM a *strong* proponent of open-source software to Stallman levels (not that I agree with his non-GNU stuff), and this affects how I feel about the whole thing even if it gets released, if it isn't open-source I'd still be unhappy until MAME fully works. Heck, I've even made SF2s targeting the Pokemon Emerald MIDI hoard from 2020. I've done MANY wild things from mixing stuff together at will. The key to doing this if you're forbidden to is to be SO transformative the original input is unrecognizable. And of course you COULD go for a Pokemon Prism esque strategy for release. Now I have to admit, my ethics are a bit... unstructured to say the least. I'm just a person who when I get a vision I get a bit too dedicated to realizing it for my own good. I'm just too resourceful for my own good. Never mind that I'm quite unconventional in that I really do not care too much what people do with my content, and I provide ALL project files free of charge so people can make forks. Heck, I'm even fine with being trained, jinriki'd (UTAU term for a voicebank made of stuff like video clips) or in tasteful deepfakes and all that. As such, I suppose my standards on sharing aren't exactly the norm, and perhaps I'm likely holding you to an expectation or such, but like, I don't exactly feel like an "I have this and you can't have this" is unhealthy AND comes across as arrogant regardless of your intentions. Personally I don't think the Tegra X1 bug that lead to Switch hax should have ever been disclosed to those whose response to it lead to Mariko revisions of the Switch. Yeah, I mostly care about the common user having no arbitrary barriers to stuff that a barrier for just comes off as tacky. It's why even if I legally could, I won't do paywalls on my content. The reason I am resourceful is that I never had an allowance as a kid and even now am flat broke, so I had to get creative to achieve my goals. And let me tell you, site ads and paywalls to this day are annoying AF. So to not anger my current and past self, I morally cannot justify ever paywalling what I do. I think of the common person with no ability to pay for something intangible. And because I care about the common person, I want the people who can't pay 200 USD + for an MU2000EX. Now, let me be clear: Tyroland and HiDef have the honor of emulating SC-88+ so that VSC/SCVA is not needed, nor is a physical device, and both do things a regular device cannot do. In short, I *have* made things that DO compete with established devices for the low price of nothing, and it got me popular. Meanwhile I expect the blowback of not releasing stuff to not ingratiate you with quite a lot of people in the MIDI community given the rarity and elusiveness of XG3 and we don't even know what a LOT of it is supposed to be. I had to guess on the SFX when making Tyroland have XG SFX. So like, PLEASE don't die on the hill of "I made cool stuff you don't get", regardless of the answer, and next time just don't even bother asking. That said I DO try to be anonymous online so that's how I handle my online life being wild AF, but IRL I'm a studious compsci student. Ultimately, PLEASE think twice about bait-and-switching the community or being the paramilitary arm of a giant corporation in a region that does not like foreigners. I don't think that action would go over well, regardless of the reasoning. You'll likely get pestered and badgered for years if you do so, by all the people wondering "Where is my XG3 VST?!" Like legitimately the optics aren't good if you do that. So don't.

DragonSlayer wrote on Today, 17:40:

@hockinsk That's a shame that this project is now officially completely dead in the water. It would have been of benefit to all, including the Japanese themselves.

These Japanese corporations have never been known for being friendly to foreigners, no matter how nicely you approach them. To make matters worse, Yamaha in particular does have an "automatic closed door" policy toward just this sort of thing, but do as you wish because, as you say, this is your work to do with as you like.

I would have loved to have seen this approached as an open source project, so the inner workings of XG would not continue to be hidden through obscurity. It's very doubtful that anyone else will ever be so inspired to put this amount of effort into such a project again after all of these years have passed since XG was still relevant. The generation of people that actually care anything about this will soon die off and that will be the end of this era of MIDI being properly preserved through emulation. Very sad indeed.

See, THIS is why you NEVER hoard stuff, EVER!

YAMAHA HAS AN AUTOMATIC CLOSED DOOR POLICY, so don't even bother asking, just release it anyways!.

People like this also are the type of people who buy prototypes and rare media and REFUSE to dump them, and this even happened to the Pokemon Center New York distribution machines. BREAKING NEWS: "I have it and you can't" only makes you come across badly. Do the right thing and release the VST!

Sorry for being heated, but CHRIST do your statements sound "Na na na you can't have this", and as someone who was bullied as a kid it gives me war flashbacks. Do you really want to come across that way? Argh... this thread is SO ridiculous.

I do apologize if I'm heated, but as someone who has been in the preservation scene for 9 years, and as someone who has used emulators since I was very young, this whole thread REALLY gets to me AND grinds my gears. Reminds me of the author of Devolution who explicitly forced real discs, even on Wii U whose drive hates GameCube games, and was an absolute pedant about it on GBATemp. Don't be the people who defend Berne, Sonny Bono, or anything similar because doing so has EXTREMELY bad optics. I'm trying my best not to make any statements I may regret later, but this whole situation has played out before and the people involved in those past situations were never liked again by even their own friends and workmates. But even if Vogons isn't out for blood, it would be the newest early teen to experiment with MIDI. Nobody likes when people say they have something and refuse to ship it. So seriously, PLEASE reconsider what you are doing. It's just going to anger everyone if I'm being honest. And that of course is not a good thing. Also I already deal with enough problems as it is, and trying to interrupt someone while they are making a mistake was NOT a roll on my Tarot+Ouija d120. So like please do the right thing here, and don't hold this whole thing over our heads, thank you. It's just not advisable. Legitimately. Just don't. /srs

PS: my avatar on certain sites is Nintendo fanart made by me, other times its also derivative but in different ways. I guess I'm a bit too passionate.