shmups.system11.org

Here is the article
https://db-electronics.ca/2017/07/05/th ... mment-2809

What guys do you think ?

I think it's bad form and repairs should be done at cost if possible. If not that, giving owners of affected carts a significant discount on properly designed replacements. If detailed sale records aren't kept, mailing them in should suffice.


No future carts should be out of spec if making it right isn't going to happen .

Krikzz's response can be seen here:

http://krikzz.com/forum/index.php?topic ... 5#msg51195

Krikzz's response comes across as really arrogant...

"There haven't been problems" doesn't mean the design isn't incredibly poor.

Since he's switched to level translation on the newer Everdrives, it's clear he realized the problem, yet he STILL won't admit he was wrong...

Hard to say what damage is being done but it is theoretically possible.

It is much the same as certain scart cables not properly attenuating Csync to 75 ohms or just passing it through. All the rage these days all of a sudden lol.

People saying it works on their setup for x years is literally what Rene prefaced with his "I've smoked for 2 years and don't have cancer, therefor smoking doesn't cause cancer" quote.Saying we haven't seen a case yet kinda comes off as pompous instead of acknowledging yeah it can potentially.

What developers of homebrew should take from this is that it can be a "potential" failure of a system and that future designs should have proper voltage transitions on all I/O.

Right or wrong, there isn't really anything Krikzz can do about it at this point: as a business, he couldn't afford to recall all existing Everdrives, and if all future products will have the issue fixed, there's not much more that he could do.

Admitting fault has legal ramifications that he may not be able to afford.

That said, he really ought to address the issue on all his existing products so that all future manufacturing runs of the older products are also corrected.

His cancer analogy is pretty poor, there's no manufacturer safety margin on smoking cigarettes like there are for electrical specs. I do agree that this isn't good and krikzz should fix it going forward.
If it works on his existing carts (take megadrive for example) and we're not hearing about mass failure, then he got lucky and the protection diodes can likely handle the current.

Pretty sure the EEVblog guy ran chips out of spec for his electrical meter, but I think he also contacted the manufacturer to confirm it was ok.

I agree, It would be ludicrous for him to refund people for what they have already bought.

also I took the analogy that it wasn't cigarettes but more it works on my setup guys everything is fine and kosher aka confirmational bias. Many people smoke cigarettes and don't get cancer, but it certainly increases the chances of it.

If the risk of damage is due to excess current through the input diodes, would an ok/acceptable fix (since we can't send them back to be fixed) be to replace the resistor arrays with higher value parts?

I think rene talked about this on retro gaming roundtable and the answer is yes can modify these carts to have all I/O converted to 3.3v, but it would look like shit.

I'm just suggesting to replace the input resistors with higher value ones. It would look identical if soldered properly.

I'm guessing the podcast reference is talking about modifying the carts to include proper voltage level translation chips, which would indeed look like shit.

Ludicrous maybe but also something you accept as a business when you knowingly release a product you may need to recall or offer repairs on.

This certainly shouldn't be ignored or brushed off just because he's not a big corporation and consoles haven't started dropping like flies. (I'm not accusing anyone of this, only stating what I think.)

As in the mod would look like shit? If it's inside the cartridge (i.e. just mod the board) would anyone really care?

The really scary shit is the 161-in-1 and other related Neo-Geo carts.

If I recall correctly there is a similar voltage problem with the Columbus Circle Famicom releases (Kira Kira Star Knight, and the two 8 bit music power carts). I would not be surprised if this is common on a lot of newer homebrew carts too.

IIRC, during a review that GameTech did about a clone toploader, the included multicart specifically said it would not function on real hardware. What do you wanna bet they're feeding the voltage directly to the chips on there too?

Yup, I heard about the two above, and it's really common on a lot of china bootlegs, too. I never realised this was an issue with Everdrives, and this is the first time I've ever heard of it. Is it only now that people discovered it?

It's common knowledge in the NES homebrew community that using 3v components in cartridges is bad, so this is really poor form. You can probably use it for years and years with no issue, but knowing this I would definitely never buy one.

I wouldn't care if it had wires all over it for an extra piece of mind. Out of interest, what would potentially fail on the console end, I take it wouldn't always be the same part or some cheap fuse?

A lot of different parts could be at risk.

I'm honestly not worried. But if I was plugging my everdrive into LaserDisc Pac or Wondermega, it might make me thing twice.

So glad I haven't bought any Everdrives yet.

When is the GB[C] ED and N64 ED getting savestates?

FYI. NeoSD wrote the following following to my ask about the Neo Geo multicart (given neo geo is 5V and the flash they use is 3.3v):

NeoSD uses voltage translators to convert all the cart 5V signals into 3.3V for internal use and back. Also those translators have a fairly high ESD protection rating (they exceed the JEDEC standard rating), so they also protect the internal circutry from electrostatic discharge from the cartridge connector.

Not all of them have the design issue though...

So for me the two everdrives incriminated are : the turbo everdrive and the mega everdrive V2 (X7), using them since 2013 nothing to report for now...well damages could be there or not...maybe my consoles will keep up until another thing would affect or this would be caused by everdrives ?

Who could tells

We need bigger tests, on very large scales to determinate if these products could affect consoles immediatly, or in the near future

For now Krikkz seems to have included this fact on his newers versions (well the versions who will comes...) just hope the multi neo & cie would not gives headhache to their owners

I think the DarkSoft one is safe...

I think we can rule that out, based on the fact that there are no reports of consoles that got damaged by an Everdrive. There may be problems on the long run, or maybe it's all fine.

It's good to point out flaws like this, but personally I'm not too worried about my Super Everdrive.

Some consoles might have died, we don't really know. Users would chalk it up to the these machines be 25+ years old and not blame the everdrive.

I had a Sega Genesis model 1 that was working fine until I used a mega ever drive and then it died (still powered on, no picture). Could’ve been anything I guess, but this thread has me thinking twice.

I've been saying all this time the SD2SNES was better than the Super Everdrive.

dude its pure gold it can't get better than that. All other flash carts get wrecked son.

Exactly what I was thinking. Heck, that'd be the first thing that came to my mind.

I don't think I saw it mentioned, but does anyone know if this effects repro carts too?

It would be on a case-by-case basis depending on the repro cart design. In theory, a repro/homebrew cart can be as bad as the worst console-killing multicarts. Take that 8-bit music power cart for example, it fed 5v directly to the 3.3v flash. That's a recipe for a quick death for the console.