- --
Viewing Issue Advanced Details
ID | Category [?] | Severity [?] | Reproducibility | Date Submitted | Last Update |
---|---|---|---|---|---|
06133 | Documentation | Typo | Always | Feb 2, 2016, 22:49 | 11 days ago |
Tester | ketburai | View Status | Public | Platform | MAME (Official Binary) |
Assigned To | Resolution | Open | OS | ||
Status [?] | Acknowledged | Driver | |||
Version | 0.170 | Fixed in Version | Build | ||
Fixed in Git Commit | Github Pull Request # | ||||
Summary | 06133: gunlock, rayforcej, rayforce, dariusgx, dariuse, futaribl, lightbr, dungeonm, dungeonmu, lightbrj, rachero: The year problem | ||||
Description |
Here's one that's made itself even more apparent recently. For the longest time, this has been referred to as a 1993 game, even though the game itself says it was completed in 1994 and all Japanese sources say 1994 (because English sources frequently use MAME as a source). At some point, the program dates for some Taito games were added to the display name, making this all the more apparent. Darius Gaiden Extra Version is similar. The program date has 1995, yet it's listed as a 1994 release. Also, I'm not really sure we need the "Official Hack" badge, it's a bit of an oxymoron. It's no more of a "hack" than any other version of any other game. Darius Extra Version is also similar. Extra Version was a revised edition of the game based on player feedback, and is generally known as a 1987 release. It simply couldn't have been released in 1986, as the original game was released so late into 1986 (likely late December) that pretty much all print material relating to Darius was written in 1987. Finally, we have Mushihimesama Futari Black Label Another Ver., another case of clear program date. At worst, it might have been released in 2010 somehow. edit: Light Bringer. I wonder just how many of these there are. |
||||
Steps To Reproduce | |||||
Additional Information | |||||
Github Commit | |||||
Flags | |||||
Regression Version | |||||
Affected Sets / Systems | gunlock, rayforcej, rayforce, dariusgx, dariuse, futaribl, lightbr, dungeonm, dungeonmu, lightbrj, rachero | ||||
Attached Files
|
|||||
Relationships
There are no relationship linked to this issue. |
Notes
10
No.12382
NekoEd Senior Tester
Feb 3, 2016, 20:48
|
If you can reference any additional proof, it will help your case. |
---|---|
No.12387
ketburai Tester
Feb 5, 2016, 18:56
|
It's kinda hard for a game that was completed in 1994 to be released in 1993. Added Futari BL Another too, that's a similar issue. |
No.12388
NekoEd Senior Tester
Feb 5, 2016, 19:06
|
This one seems very hotly debated. Let's get some developer opinions on this one if we can. As it is, MAME policy is to use the date presented on the title screen as the canonical date. |
No.12389
ketburai Tester
Feb 5, 2016, 19:14
edited on: Feb 5, 2016, 19:30 |
Where's the argument? This is people seriously trying to tell me that what the developer actually did does not matter in the slightest, and that some tiny legal graphic is all-important. This is people seriously trying to tell me that they are okay with outright lying to everyone that uses this program. The copyright years never make sense. They make many assumptions about the future of the game's development, and they have nothing to do with the game's actual release. Japan is pretty good about actually tracking when arcade games get released, and a lot of later games have clear program dates that we can crossreference with sources for arcade releases. The copyright year literally means nothing when what is clearly a 2009 version of the game is going to be referred to as a 2007 game forever. Noone who uses MAME cares about "copyright years". They see that year in MAME match up with that year on the title screen and that game becomes whatever that year is for them. Pretty much every English source for arcade games completely relies on MAME for everything about the game, so this kind of thing persists for YEARS unless you can manage to convince people that MAME is inaccurate. This is not a "debate". This has gone from a typo to a serious problem that needs to be fixed as soon as possible. |
No.12401
Fujix Administrator
Feb 10, 2016, 08:37
|
**DEJA VU** I feel I ve seen the same argument more than a decade ago. |
No.12496
AJR Developer
Mar 30, 2016, 14:58
|
There are already a few cases where MAME has broken with the "title screen as the canonical date" rule. fantzn2x and shinfz have the date field set to 2008 in spite of being copyrighted to 1987 and 2006 respectively. |
No.12498
Tafoid Administrator
Mar 30, 2016, 17:59
|
In those cases, there are reasons in the driver comments regarding the GAME entries.segas16b.cpp: If there is a difference there is usually a reason why. |
No.12510
ketburai Tester
Apr 3, 2016, 21:53
edited on: Apr 15, 2016, 18:14 |
Those look like very similar cases to what we have here. All you did was recognize that the program date is well beyond the copyright date, the exact same as I'm doing. edit: Added Light Bringer. |
No.22491
ketburai Tester
14 days ago
edited on: 14 days ago |
Yeah, just came across 04411 (https://mametesters.org/view.php?id=4411). The title screen is NOT a "never-fail type of documentation". When a title screen is referring a game by a completely different year than the code and all contemporary sources provide, it's clearly failing. This then leads to databases, and everyone using those databases, to simply have incorrect information until someone can be convinced to change it. And this all starts with MAME. MAME is simply wrong about this issue, and it's a BIG issue that desperately needs to be fixed. It's not at all clear why the exceptions listed above got special treatment, as the arguments given are exactly the same as given in 04411 and here in my report. There are simply a LOT more exceptions out there, that's all. Yes, there are going to be incredibly obscure games where finding any information on them whatsoever will be difficult, and the title screen may be all we have. But for well-known games with lots of documentation, there's no sense in relying on information that is provably wrong. It would be much better to research this more than to deny it. On that note, it would be nice to be able to expose internal dates in a searchable way. The internal date is not necessarily when a game might release, and there are certainly times when the date is "frozen" (especially for regional releases), but it too is more reliable than what really amounts to a copyright date. As long as it's clearly marked as "Internal Date" or some other qualifier, this info would be useful or at least interesting, but I guess this is a topic for another time. That aside, I added Racing Hero, because it's another example of a game with a clearly incorrect year. Every source I've found points toward the game releasing in 1990. The game was originally revealed at the 1990 AOU Show (February), advertising for the game's release appears in at least April magazines, and sales data started to appear in May. It seems pretty likely that the game released in March or April 1990, and Famitsu's Sega Arcade History explicitly claims March. The year is what's important, though it'd be nice if months were opened up eventually. |
No.22502
cuavas Administrator
11 days ago
|
You aren’t going to get anywhere by trying to claim, “it's a BIG issue that desperately needs to be fixed” – that’s clearly hyperbole. Metadata in MAME is informative, and best effort. It should not be considered authoritative. Historically, displayed copyright dates were used most of the time because it’s something fairly easy to confirm and cross-reference, and it gives a decent estimate of when the game was completed. The bigger problem is people reading more into MAME metadata than they should. The years are not guaranteed to be accurate documentation of the years games were released. Historically it was just the displayed copyright year. |