Comments 123

Re: SNES Consoles Appear To Be Getting Faster As They Age

avcrypt

@Sketcz first off, hi, I'm an electrical engineer with around two decades of professional experience in systems design. The ceramic resonator will be one of the last things to completely fail. They lose about half a percent of accuracy every decade, and from the majority of responses, it's been even more favorable. (For instance, 7Hz is 0.0000284% off reference)

If you're a TAS runner, that's a problem, and honestly replacing it with a high quality crystal oscillator should help. For normal people, the electrolyte in your capacitors will dry up long before the ceramic resonator becomes too inaccurate to maintain stability.

Re: This $75 Handheld Could Be The Best Way To Emulate Nintendo DS In 2025

avcrypt

@GravyThief it's a little goofy and small, but I dig it for tate mode games. If I really wanted something to do tate mode on the go, I think this would work well enough. While I could get the original cabs over this, it's much cheaper just to buy the ROM chips from them and dump them myself, and I can't pack a DoDonPachi cabinet in my luggage. (Though I can try)

Re: Company Behind The X68000 Z Range Wants To Know If Global Players Will Buy Them

avcrypt

It's probably good to remind people that this isn't your normal plug and play retro system. This is a heavy duty recreation of the original X68000 system. You can plug in a SCSI CD-ROM drive and a Roland SoundCanvas and have largely the same experience as the early 90s. You also need to have some familiarity with the OS used on these. It's a serious bit of kit.

Do I want one? Yes. Can I afford one? Unsure. Can I get my money's worth? I doubt that. I think a lot of international customers will fit in that box with me.

Re: "Might Be Time To Go Back To A Corporate Job" - Trump's Tariffs Come Into Effect

avcrypt

Had to inform clients of the impending rate increases from their partners once our current buffer in the chain runs dry. I'm sitting in about a hundred emails of people going through the five stages of grief.

A lot of these companies produce goods you need indirectly, so not something you'd buy, but something needed for other products and services you do buy. Most of these I wouldn't consider luxurious at all, addressing an earlier comment, but all of them keep B2B going.

It was kind of like a shockwave where I'm informing the next ring of the process about the hit. And like a shockwave, everybody felt it.

Re: Interview: "We’ve Certainly Made Mistakes" - Limited Run's Boss On Winning Back The Trust Of The Community

avcrypt

I've still got a couple bones with this:

First, was LRG not worried or bothered by shipping D on burnt discs? It should be obvious where LRG exists in the market as a niche, almost boutique supplier of retro games. That quality doesn't rise above cheap repros.

Second, why didn't LRG have the carts of Rugrats and PioPow inspected by an engineer? It's very common to inspect every aspect of a device you receive from a supplier to make sure it meets the expectations of them and their audience. The fact it shipped, just like the D debacle, is an LRG problem. The buck stopped there.

What I'm reading is that LRG lacks the right people to identify quality issues before they reach consumers. Somebody should have pressed the metaphorical E-stop before any of these issues ended up with a shipping label.

Re: Anbernic Pulls Controversial Firmware Update That Allowed ROM Downloads From The Internet

avcrypt

@RetroGames a case this brings to mind is Arista Records LLC v. Lime Group LLC here in the US circa 2010. This is the case that killed limewire. Even though limewire didn't store or ever hold copyrighted material themselves, their platform "induced copyright infringement" for its users as a core part of its product. Even though not every download they offered was bad, the court found that it was overwhelmingly used and considered a tool for piracy.

That case is still referenced decently often, and I imagine it'd be one of the more relevant. Not that this matters, since Anbernic took it down, but just mentally golfing here

Re: Anbernic's New Firmware Has Opened A Can Of Worms That Could Damage The Handheld Emulation Market

avcrypt

@RetroGames the "store" (in this context) needs to be authorized by the copyright holder to distribute their work since they effectively need to generate copies to distribute.

The core of all this boils down to the question "can somebody else distribute something I made without my permission, even if I release it for free?" That answer is generally no. If they wrote on, say, their itchio page "feel free to share this thing with whomever you want" then there's no explicit permission necessary.

Re: Anbernic's New Firmware Has Opened A Can Of Worms That Could Damage The Handheld Emulation Market

avcrypt

@RetroGames my first response still applies. When a developer uploads to itchio, they grant you the right to download from itch (Microsoft, etc. aren't part of that relationship), and itch the right to distribute though their own methods. These are detailed in the terms of service.

For all other forms of distribution, you'd need to get permission from the author unless they've explicitly said it's okay to redistribute.

As to the part about it being okay to play on certain devices, copyright holders generally cannot mandate how you interact with their work. (They can make it harder to do so, but generally can't stop you.) That's why Analogue or the RetroN can play Pokemon/Mario carts and Nintendo has no recourse to stop it. Copyright only controls how information is shared and what rights each party has, but once somebody has the copyrighted material, they can generally (giant asterisk here) do whatever they want with it privately. (For clarity, talking about the end user here, not Anbernic. Anbernic would be like a store here. They'd deliver the copyrighted material under agreement, but whatever you do with it from there is up to you.)

Re: Here Are The Specs For The Retroid Pocket Flip 2

avcrypt

I really liked the Pocket Flip 1 with some minor gripes about the stick placement, and this addresses those almost to the T. Also, I understand where people come from on it looking uncomfortable, but it's honestly not bad for long sessions. Retroid has traditionally had a good feel for ergonomics.

Re: Retro-Bit Apologises For Using Fan-Translations Without Permission

avcrypt

@Quick_Man "There's no legal way to claim a fan translation, really, considering it's an illegal modification of a game that is also most likely downloaded illegally."

There's two statements here I wanted to address: that it's an illegal modification and that it follows that the game is downloaded illegally.

The second point I think addresses another issue entirely that I don't think it's material to what's being discussed here as fan translations are rarely shared as full games. Rather they're patches to be applied.

On the first point, this is a huge legal grey area. Is it distributing copyrighted work? Yes and no. Yes in the sense that it is distributing translations of copyrighted dialog as a derivative work. No in the sense that it's only usable by presumed owners of the original work under non-commercial terms for the purposes of accessibility, and so they stand a chance at sliding just under the bar of fair use... potentially. As all fair use cases go, it's not really fair use until a judge says so.

Important here is that it's not an illegal modification per se. That and illegal modifications aren't really a thing in copyright. Whistler tips and catalytic converter deletes are illegal modifications. Translations are derivative works. It's the distribution that's potentially illegal.

I want to recognize a case in Sweden in 2017 that is referenced a lot about this issue where they did focus heavily on the pirate aspect. I live in the US where we factor in that sort of situation differently and have different factors for fair use.

Re: Retro-Bit Accused Of Plagiarising Existing Fan-Translations

avcrypt

@RetroGames As a commercial entity, I wouldn't be fine with just a credit. I'd want compensation, or I'd want affected portions with my work redone without infringing on my own because I may not want to be associated. It's free for fans to grab, but not for companies to profit on. Retrobit, if the allegations are true (and they seem to be), are flying dangerously close to the sun on this in terms of copyright infringement.

Re: "This Cartridge Is A Tiny Time Bomb" - Limited Run Accused Of Selling Carts Which Can Damage Your NES

avcrypt

@Zeebor15 News now is they're blaming their supplier which is wild. That either means their supplier drop ships directly to the customer without LRG inspection so they couldn't inspect the board, or more likely that LRG didn't inspect the boards before approving them.

Needless to say, LRG is placing blame on others (again) for problems that should never happen which isn't surprising but very tiring. It's their duty to perform final inspection on the product. This is inexcusable.

Re: "This Cartridge Is A Tiny Time Bomb" - Limited Run Accused Of Selling Carts Which Can Damage Your NES

avcrypt

@Hastor I did look it up and it is indeed abusing the clamping diodes. So that isn't fantastic.

"what would keep such chips from undervolting the signal back to the NES, which isn't good for it either?"
This isn't an actual concern. If the NES was harmed by 3.3v signaling, I'd blame Nintendo for building a truly awful product. The truth is that a 3.3v signal is a valid level-HI in TTL logic. Sending 5v signals to a 3.3v device will wear it down, but sending 3.3v signals to a 5v device is fine, with the only risk being the possibility, however unlikely, that an out of spec TTL chip floated on 3.3v. I'd be surprised if that were the case by 1985 TTL.

Re: "This Cartridge Is A Tiny Time Bomb" - Limited Run Accused Of Selling Carts Which Can Damage Your NES

avcrypt

@KingMike that's right. 5V ROM chips have been out of production for a little over a decade now, and while we do have 3.3V ROM chips that behave the same way, there's a list of considerations that need to be addressed before plugging them together.

Usually using a transceiver or one-way buffer is enough to get you by, but those are pricey in the grand scheme, so a lot of manufacturers limit the current via resistors low enough that the ESD clamping diodes shunt the excess power away. It's a hack. It's ugly. But it's cheap.

I made a comment above wanting to know the layout because flash manufacturers have caught into this happening and some are now purpose-built to stand up to that abuse for the entire lifetime of the chip without getting excess current draw or heat build-up. LRG may have selected such a part, but it's also LRG, so maybe not.

Re: "This Cartridge Is A Tiny Time Bomb" - Limited Run Accused Of Selling Carts Which Can Damage Your NES

avcrypt

I invite the person who submitted this to show more about what they found. From the description alone, I don't have quite enough information to fully jump on board with this analysis, as there do exist 3.3v flash roms with long life clamping diodes for interfacing with 5v systems, though they should still use resistor packs to limit current.

It's definitely a yellow flag in my mind, but not necessarily egregious. I'm willing to withhold my usual LRG gripes until seeing it.

Re: Creator Of New Open-Source Game Boy Disagrees That FPGA Is Superior To Software Emulation

avcrypt

@Slobbert for the most popular cores, yes. Accuracy falls off on some less popular ones like the X68000, but they're still pretty close. The Saturn core is still polishing edges, so it's not quite to the point of Mednafen Saturn. (I'm sure it will, just not yet.)

And it's not selling FPGA short, just pulling it back to reality. In all honesty Analogue did some damage in their marketing claims that has been hard to reeducate the larger enthusiast audience about. The Mister SNES core authors, for instance, got most of their info from BSNES/Higan, the software emulator renowned for how accurate it is. That's not a secret and they're open about it. I'm not well versed in that core-set, but from what I gathered, they're as accurate as one another.

In regards to the article, I've hardly seen any discussion near devs compared to the more charged discussions I see elsewhere. The author here, an FPGA core dev themself, is just being open and honest about putting things into perspective. I've been chatting with a couple core devs on the side since this article dropped and nobody's felt slighted so far (at least in my little nerd network).

Re: Nintendo DS Emulator DraStic Pulled From Google Play Store

avcrypt

@jbrodack I don't think in this case it was Nintendo directly that caused them to pull it down. However, they've used a weird legal interaction as a poison pill in their new consoles.

In the Wii and upwards, all published games must have their copy protection mechanism disabled in order to play via the console's key. Even if you own the console and the game, in the US at least, the act of extracting or reproducing the key in order to access the game software is considered copy protection circumvention which is illegal under the DMCA. Again, even for the game and system you physically own and don't redistribute. Courts have been warming up to the interpretation that, because circumvention is necessary to load software at all, then all activities (the emulation part) afterward must also be illegal since it's tainted by the initial circumvention.

It's my biggest gripe with the DMCA beyond every other restriction it imposes.

Re: Creator Of New Open-Source Game Boy Disagrees That FPGA Is Superior To Software Emulation

avcrypt

@TenEighty sure, and the statement of experience is true of most things.

"The more experienced developer will develop more accurate and optimized code when working with the lowest level of technology."

I don't think of optimization in that way. The Mister PlayStation emulator is less accurate than Mednafen (more obvious as an LLE) and DuckStation (less obvious as an HLE). The PSX core has to wrangle a lot more than DuckStation does. This causes the PSX core to struggle more because how they approach the problem is so significantly different, requiring a ton of RTL leg work, and because gateware simply can't exploit the benefits of HLE.

It's more about how you approach the emulation, not necessarily how low level you go that impacts how well a given emulation performs. The hotspot of DuckStation is translating PSYQ calls into native system calls. Turns out that's really performant and can be made really accurate (with some tweaks) even though it's very high level.