Someone really ought to dump Hucard ROMS properly

[SamIAm]

It's come to my attention after a lot of time using my Turbo Everdrive that there are probably a lot of PCE and TG-16 Hucard games that have never been properly dumped.

I'm seeing a number of games glitching, freezing, or not even starting. While it's possible that I've got a hardware problem, I don't think that's it. Too many other games are running correctly when I leave them for overnight tests, including large ones with attract-modes that use data from all over the game. I know that that's not conclusive, but at least it suggests that there are no access problems, which is the only other possible cause of these malfunctions.

Even if the problem is on my end, though, it looks like there has never been much in the way of a unified effort to dump hucards accurately. I'm guessing one issue is that there doesn't seem to be any standard checksum data included within each game like there is for so many other systems' games. That's why the GoodTools people couldn't really give [!] tags to any hucard roms.

Instead, there are only untagged ROMs that are supposed to be good (which really means they just couldn't see anything wrong with the data), along with lots of [b1], [b2] tagged ROMs (b=bad dump) and the dubious [h1], [h2] tagged ROMs (h = hacked). [a1] etc also basically means it was hacked.

The untagged R-Type (U), for example, doesn't work. I have to use the [h1] ROM, and that probably means that the untagged ROM is actually bad and someone hacked it into working. That may work well enough to play, but it's no way to preserve a hucard game. Preservation is my primary concern.

I used to own an SNES copier and spent a lot of hours messing around with dumping and loading games. I was surprised at how even my late-generation copier didn't always dump games like it was supposed to, and was inconsistent. The SNES ROM community back then (remember cherryroms?) put a lot of effort into making sure that each and every game got a proper, accurate dump. I believe their work really paid off, and that gamers everywhere are all the richer for it.

I hope another community can eventually do the same for the PCE.

With no checksum in the games, it seems to me like the only way to establish an accurate dump would be to dump a game repeatedly with multiple dumpers and cross-check them very carefully. Even then, if each dumper is improperly dumping things the same way, we may still not get what we want. It would take real know-how and diligence to do this right.

I can't offer too much other than to raise awareness and hopefully start a conversation. If someone else with an everdrive wants to try some games and see how they work, then by all means, please do so. Making a list of games that look like they weren't dumped right is a good first step.

For everyone, what do you make of this situation?

[MotherGunner]

I wish there was a list that would explain what was hacked for each and so forth.

[SamIAm]

I wish there was a list that would explain what was hacked for each and so forth. 

Me too. It's actually quite possible that some of the [h] tagged games were bad dumps that were modified into being accurate somehow.

Here's a complete list of the tags, by the way:
http://en.wikipedia.org/wiki/GoodTools#Good_codes

[CrackTiger]

I haven't used my Turbo Everdrive for more than a few minutes  but I never had any problems with any roms on my tototek card, except for that bad Bloody Wolf sound rip rom.

[SamIAm]

The next time it's handy, could you check the GoodTools verified untagged R-Type (U)? If that one starts for anyone else, I've got a hardware issue.

[MotherGunner]

Yea I saw the complete list of tags, but I was referring to what content was modified per each game?  I was hoping the hacked rom for Salamander contained the konami code but it didnt.

=(

[Bernie]

Have you tried the no-intro set?

[SamIAm]

Yea I saw the complete list of tags, but I was referring to what content was modified per each game?  I was hoping the hacked rom for Salamander contained the konami code but it didnt.

=(

I think more often than not that most of the [h] modified games were altered simply so that they would work, sometimes with specific copiers or emulators.

The typical thing to do for distributing a hack with something added to it like that has been to use IPS patches, AFAIK.

Have you tried the no-intro set?

Yep. The ones I looked at had the same checksums as the untagged GoodTools versions.

[Keith Courage]

Have you tried removing the game headers? I have a gameofyou 8mb flash card and have yet to find a game that won't work.

[SamIAm]

Most of them work totally fine. It's just a couple that haven't. PC Genjin 2 untagged crashed once I got far in the game. R-Type (U) untagged didn't load. Aldynes on the SGX crashes if you let it sit in attract mode long enough. There's one more I'm forgetting...

EDIT: R-Type did have a header, but it failed to load even after I stripped it.

Further testing shows that other games will work with or without headers, so the everdrive must know how to recognize them.

[SamIAm]

Solved R-Type. The ROM was interleaved! Thank goodness for ucon64. For now, that one looks OK.

EDIT: .....aaaand, no, it's still screwed up. There is massive corruption in stage 2.

Anyway, Aldynes on SGX consistently crashes in attract mode, though sometimes it needs a while to do so. That's the easiest one to test if anyone has a Supergrafx. PC Genjin 2 crashed as I was entering stage 3, I think.

Googling around, I can see that there's been a lot of talk about improperly dumped games. I wonder what the method of choice to dump games has been so far?

Get a load of this!
http://lh5.ggpht.com/_E4fMqwX-9UE/TTqnBCg2AfI/AAAAAAAADtA/X7JKDCYm43A/PICT5660_thumb%5B1%5D.jpg
http://lh6.ggpht.com/_E4fMqwX-9UE/TTqnCtf90iI/AAAAAAAADtI/cPatOlJ65Aw/PICT5662_thumb%5B4%5D.jpg
http://lh6.ggpht.com/_E4fMqwX-9UE/TTqnD1NY4vI/AAAAAAAADtQ/IjoYaCElqag/PICT5664_thumb%5B2%5D.jpg
http://lh5.ggpht.com/_E4fMqwX-9UE/TTqnFUYNOVI/AAAAAAAADtY/0rIiFoSrcvo/PICT5666_thumb%5B2%5D.jpg
http://lh4.ggpht.com/_E4fMqwX-9UE/TTqnHINJYoI/AAAAAAAADtg/-2rkK5I5xH0/PICT5667_thumb%5B2%5D.jpg

[jperryss]

Have you tried the no-intro set?

Not sure if it was you that mentioned it last time, but this is a great tip. I've yet to find a rom that DOESN'T work.

[CrackTiger]

I've been using the same batch of roms that I downloaded individually during the late 90's. I haven't played most on flashcards, because that's not what I use flashcards for. My R-Type rom works fine on my Turbo Everdrive, but SamIAm's doesn't boot at all. I guess that too many people have been mixing up misc versions over the years that trying to obtain them all now will net a certain percentage of bunk roms.

[SamIAm]

I better stop doing tests for a while or my brain will hurt.

That R-Type ROM I was using was interleaved for some odd reason. I de-interleaved it and it started right up, although it still had graphical errors. I think what I did to de-interleave it isn't exactly the reverse of whatever interleaved it.

I've used that no-intro set, but the results with Aldynes didn't change. It's exactly the same file anyway. Sometimes it happens early, sometimes after a while, but it freezes in the attract mode.

One thing is interesting - Aldynes seems to access the hucard on a constant basis. That or it uses some pins that other games don't. If you lay a finger on the Aldynes hucard (real or everdrive) while it's running, it freezes instantly, but you can jiggle Daimakaimura around to your heart's content and it will be fine most of the time.

Therefore, I might just have a contact problem with the Everdrive. The SGX is more susceptible to dirt, after all. It'd be nice if it were that simple. I'll find out tomorrow.

[ccovell]

All card games access the HuCards on every machine cycle; that's the only way they can possibly run.  So your idea about Aldynes is a bit off.  The game might have been poorly programmed, or have very brittle timing.

Most hacked games are US Turbografx games that simply have their country lockout defeated.  Japanese hacked games are... well, post some examples using DOS FC, as a start, rather than wild speculation.

[SamIAm]

Aha!!

From the Neo Team Forums (of the Neo PCE 128m flashcart):

Note - I've tried all the SGX games, and they all run with one caveat: Aldynes always freezes at the same point in the demo; if you start the game before that point, it plays fine. It doesn't freeze under emulation, so I'm not sure what the issue is on real hardware.

http://www.neoflash.com/forum/index.php/topic,6700.msg50682.html#msg50682

It freezes for me on stage 5, too. Not to brag, but getting that far takes a lot of practice and I wouldn't be surprised if Mr. Chilly Willy hasn't made it that far. Maybe it really does need a redump. Or maybe there's something unique about the hucard.

All card games access the HuCards on every machine cycle; that's the only way they can possibly run.  So your idea about Aldynes is a bit off.  The game might have been poorly programmed, or have very brittle timing.

Most hacked games are US Turbografx games that simply have their country lockout defeated.  Japanese hacked games are... well, post some examples using DOS FC, as a start, rather than wild speculation.

Thanks for the insight. Yeah, I was just speculating. It really is a funny difference, though. I wonder what is going on that Aldynes is so sensitive?

I'm getting it now that a lot of the TG-16 (U) games in the GoodTools set were dumped using equipment with the PCE pin setup, which is why they need the deinterlacing and why the "unhacked" R-Type didn't work. It's definitely best to use the [h1] version of any US ROM when playing from the GoodTools set.