Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
Nice, but with that method the shot from the guy in the next room (that we need to skip the second barrier) is much, much slower, so that's definitely not an improvement over all. I currently don't see any way to get both the quick elevator and the quick bullet. If you can improve the second transition after the second barrier though that would be a definite improvement.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
http://tasvideos.org/userfiles/info/44367370341976366
Ok that silly error is fixed, please replace the movie file with this one.
@Challenger: thanks for the temp! sorry I messed it up now but at least it's only the last 2 minutes!
@Aglar: yes both transition areas after the barriers are different (and slower) then the oriignal. I was completely unable to recreate the strats nElifox used there, I think due to different positioning and scrolling at those points. If someone else can somehow improve them by all means do.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
Woah cool, thank you for all the testing Sonia!
This will greatly help me in sorting out FDS issues, obviously it isn't quite up to speed with the rest of NESHawk.
Also I don't know anything about multi-disk compatibility, I didn't even know such games existed.
EDIT: fixed grand prix II and time twist, those were just IRQs firing a couple cycles too early. I don't know why exactly IRQs should be delayed by a couple of CPU cycles, but this is the case for these games to display correctly.
Other games that only show black screens all seem to have the same problem, in the same form. They scan through the disk but the disk reaches the end before the routine they are in finishes. They are IRQ based, so once the disk runs out there is no more IRQs happening so it just kind of loops forever. Not sure what the solution here is yet but once it's figured out all those games should work.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
I fixed Silviana. The FDS core was not setting $4032 bit 1 (disk ready status) correctly.
I am pretty sure my fix is correct, but if anyone interested in FDS games can try some out and make sure there are no regressions I would appreciate it. So far I don't see any though.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
I do agree that this run is trivial in the sense that it does not stand out from unassisted play, but I'm glad for the opportunity to have more clarification on what aspects go into deciding that.
In partiuclar, does performing glitches that don't save time (supposing they are very difficult to perform real time) make the run 'stand out' in the sense we are interested in here? This seems to be a large part of the disagreement here, from what I can see.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
This is fascinating.
I don't know anything about CPU architecture or computer/information security, but I found those remarkably simple to understand (or if not understand, at least get the gist of), it's pretty surprising such inherent flaws went un-detected for so long, and also a little scary.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
I have code in there as a stub, it doesn't actually do anything. I didn't research what was needed since I wasn't that interested and wanted to move on to other things.
So, if someone handed me what I needed to do to get those games to look right, I would implement it, but otherwise no.
The Camerica titles can't have 0x701 be 0, otherwise they won't boot (or will boot incorrectly.)
For dancing block, those 2 addresses cannot both be 0xFF, but I forget exactly why.
The game Cybernoid also relies on un-initialized RAM for several things, the most obvious one being whether or not music is playing at the start, as well as what difficulty is initially selected.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
You'll want to use a Lua script for that. You'll find the Lua interface in the menues under Tools -> Lua Console.
If you are a beginner in Lua, there are many sample scripts included in BizHawk to help get you started. You'll find them in the Lua folder.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
It is stored in the movie header, or do you mean something else?
I just added suport for korean BIOS. I'll try to make region choice a little more clear cut as well going forward.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
I bare the blame here for GB not being very TASable in the most recent release. I do believe the current dev build has returned to being very stable though:
https://ci.appveyor.com/project/zeromus/bizhawk-udexo/build/artifacts
If you need help with anything feel free to pm me.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
Thanks to some speedy work by judges and publishers, we can now cross off 3 more runs from here!
This is further then I expected to have gotten by this time, maybe 2018 can see this wrapped up!
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
I fixed a couple bugs in C64, including the exceptions above.
I would encourage anyone interested in C64 to download the dev build and test things out and make sure there are no regressions.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
That's interesting. Based on the comments in those threads, it probably expects un-initialized data to be in a certain form, and BizHawk is probably not doing what it expects. I'll look through and see if this is the case, and set things up properly if needed.
EDIT: the game thinks WRAM initialized with 0 is some kind of valid data. initializing instead to 0xFF does not have any files when the game loads, so this seems to be the solution here. I jus commited a Hash based check to make this the case for this game.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
I don't think the rule (as written) is intended to seperate out games based on triviality. No argument against 'Math Blaster' has been about how trivial the game is or even whether it's a game at all.
I would say the the one word that sums up the current rule and judgement is 'style.' I think the intent of the rule , and certainly the the way it is used in this judgement, would be more clearly stated as:
'Games in the style of educational games or board games are not accepted in the vault.'
Then, at least it's clear that we aren't differentiating games from non-games here, nor trying to define some notion of 'serious game', nor judging the game based strictly on it's content. The style alone is what is at issue. Then you could even use 'math blaster' as an example.
I still wouldn't think this is a good or necessary rule (I think it can be safely scrapped altogether), and 'style' would probably still be contentious, but at least it's clear about what's being judged.
On the notion of triviality, I think that aspect should be re-inforced even more then it is. The current 'Porky's' run for example can be almost exactly reproduced real time with a little practice and only 1/4 chance luck.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
(I replied to this on GitHub but it seemed worth it to put a similar reply here.)
It's actually the background that isn't rendering here, but that's besides the point.
It's not expected, and actually I totally forgot that even happened.
Being able to load GB games in GBC mode in gambatte was a derivative of having BIOS support. Originally Gambatte did not have this capability (I had to load up 1.13.1 to even realize this was the case.)
So, I reverted it so that GB games will not load into GBC mode in Gambatte, they will only load into GB mode.
In the event that a GBCHawk core ever comes into existence, the ability to play GB games in GBC mode will be once again added.