Editor, Emulator Coder, Experienced Forum User, Site Developer
Joined: 5/11/2011
Posts: 1108
Location: Murka
The description was more or less copied from the previous movie. That being said, I didn't copy blindly; I did read it.
I'm not against changing it, but I want to give a bit of time for more people to provide feedback.
Editor, Emulator Coder, Experienced Forum User, Site Developer
Joined: 5/11/2011
Posts: 1108
Location: Murka
syncs on snes9x-rr 1.51 svn147
File: Super Mario All-Stars + Super Mario World (USA).sfc
CRC-32: f84305b1
MD4: 27f3470732c20cc7a604de9d2d60d8e6
MD5: 7ecaf3e2e021c8a836041c85886f4594
SHA-1: d245e41a2b590f7d63666b0772cbddfb26f254a2
Editor, Emulator Coder, Experienced Forum User, Site Developer
Joined: 5/11/2011
Posts: 1108
Location: Murka
Well, I didn't try anything =p However, I do have a bit of familiarity with the PSXjin code-base and the developers of it. It's not horrid, but it's not perfect, and more importantly, there's no real active development of the emulation core at this point.
Unless it runs correctly in pcsx-rr, the only tas option may be to wait for a new emulator to come along.
Editor, Emulator Coder, Experienced Forum User, Site Developer
Joined: 5/11/2011
Posts: 1108
Location: Murka
nanogyth wrote:
natt wrote:
fceux has an option to enable more than 8 sprites per scanline. guidelines say to leave it off though.
Seems a bit silly when other encodes are rendering at 1440p to remove aliasing.
But I guess the rules are the rules. ;)
/shrug
The Encoder Guidelines are a bit murky on this. Two viewpoints are presented: "As close to original as possible," and "enhanced," and they aren't really reconciled. Of course, there are quite a few problems with that page...
Editor, Emulator Coder, Experienced Forum User, Site Developer
Joined: 5/11/2011
Posts: 1108
Location: Murka
If it is in fact copy protection, it could be reacting poorly simply due to bad emulation in psxjin.
Have you double checked your ISO? Try to match it by checksum to one of the ones on redump.org. If it doesn't match, rip it again.
This should have been run at 50fps. I'm not familiar with hourglass internals, so I don't know whether that's enough of a technical foul for rejection. The game runs at the appropriate speed, just with a 5->6 cadence.
Editor, Emulator Coder, Experienced Forum User, Site Developer
Joined: 5/11/2011
Posts: 1108
Location: Murka
nanogyth wrote:
"fastest death" and "fastest end of input that results in death" have the same meaning for official timing.
But for a viewer unfamiliar with the official definition some clarification would be helpful.
I think part of the problem is that fastest death runs typically aren't accepted, so there's little precedent.
If they were accepted, I personally would think that the fastest time as measured from start to when the death actually occurs would be more interesting than end-of-input, because there are so many games where you'll die eventually with no input as long as you start the first level, so "fastest death" runs in many cases then become a simple menu navigation exercise.
Editor, Emulator Coder, Experienced Forum User, Site Developer
Joined: 5/11/2011
Posts: 1108
Location: Murka
nanogyth wrote:
Famtasia isn't great, there are a number of reason one would want to avoid it when making an encode.
But if it doesn't sync on anything else you might be stuck.
I have a possible work-around, but I don't know what kind of practical limits it runs into.
If the movie syncs on famtasia and the savestate data for each frame could be recorded, could that then be played back with TASEditor?
IE a large collection of single-frame from-savestate movies, using famtasia to recreate its faulty logic and something else to render the A/V.
Edit:
ACK, the famtasia source code is lost, pulling the savestates out of the .exe is probably nigh impossible.
Oh well, plenty of SNES youtubes to encode.
Even with massive massaging, I don't think you'd be able to convert savestates from one emu to another anyways...
Editor, Emulator Coder, Experienced Forum User, Site Developer
Joined: 5/11/2011
Posts: 1108
Location: Murka
Obviously, some sort of encoder blame tagging needs to be available, but that can be accomplished by various metadatas and whatnot.
I'm for this provided the logos created live up to extra-high quality standards.
Editor, Emulator Coder, Experienced Forum User, Site Developer
Joined: 5/11/2011
Posts: 1108
Location: Murka
syncs on desmume 0.9.4 official release
does not sync on desmume 0.9.8 (64bit) official release
sync appears to have FBA levels of variability; dunno if that's usual for desmume...
edit: to clarify the above statement, what i mean is, when played back in desmume 0.9.4, sometimes it syncs, and sometimes it doesn't; a common FBAism.
File: 0121 - Castlevania Dawn of Sorrow (U).nds
CRC-32: 135737f6
MD4: a714bea87bd7445e18d93f9dbc3dae9f
MD5: cc0f25b8783fb83cb4588d1c111bdc18
SHA-1: 47530ff87e608f88105a314fdf36dc385f8dec94
