Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
Has anyone else noticed that the console name for N64 runs is listed as "Mupen64" instead of "Nintendo 64"? It could be intentional, but I wouldn't normally call an emulator a console.
Rikku wrote:
Twelvepack wrote:
Edit: Please fix up the file name if possable.
In which name should I change the name ?
I don't see anything wrong with using "Super Mario 64.m64" either.
comicalflop wrote:
it doesn't take too long to get 40 posts...
JXQ wrote:
Maybe votes should be based on a length of time after registering. It would discourage spam. It would also stop penalizing those who don't post thirty times a day just for the hell of it.
spockybiemmichab wrote:
Actually, I'm pretty sure that the hurdle is 17 posts
I was under the impression that all voting priveleges for new users here are granted manually, not automatically based on post count. But I'm only guessing, like everyone else seems to be.
EDIT: comicalflop is almost at 1000 posts already? I wonder if anyone has more posts per day than that. Sorry this is so off-topic but I found it surprising.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
I didn't know VBA allowed you to open multiple memory viewers to get around that problem. That's easier than scrolling back and forth all the time. But yes, someone (possibly me) should add a more compact memory viewer to VBA.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
Bug: If you change the data type or size (or misaligned checking) while the RAM search has results shown, their addresses all get reset to starting from 00A00000.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
Pu7o wrote:
You could have gotten a faster Hidden Palace if you constantly spin-dashed with Tails in the place where Knuckles was gonna fall.
Could you explain that better? Tails can't damage Knuckles when Knuckles is landing from being hit, and Tails doesn't have control once the floor falls.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
The improvements were very noticeable all around. But why hit the signpost and jump in Hydrocity 1? It didn't appear to change anything. Also, I wonder if part of that Flying Battery 1 route could save time for Sonic...
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
Is this video some sort of protest?
Mukki wrote:
If there was that big a glitch in the game Nitsuja would have found it. End of.
Well I missed at least one timesaver there (4 seconds in Aquatic Ruin 2), and a few in S3&K (at least 20 seconds). But I don't think this one is possible. The boundary checking on the left side of the level isn't so bad that you can hope to bypass it with a regular jump.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
I look away for a week and miss 3 Mario movies? I'll have to watch this when I get a chance, the WIPs I saw of the first few levels were great and it could only get more interesting after those.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
Confirmed that bulletproof re-recording is still broken. In a way it got worse, because the warning is gone and it still desyncs.
EDIT: More specifically, it doesn't work correctly if the current movie length is less than the movie length in the state being loaded. Here are some easy steps to reproduce the desync when recording a movie:
1) Save slot 1
2) Walk right for 5 seconds
3) Save slot 2
4) Load slot 1
5) Walk left for 2 seconds
6) Load slot 2
7) Toggle mode to read-only
8) Load slot 1
The result should be that you see the character walk right for 5 seconds. Instead, you will see the character walk left for 2 seconds, then turn around and walk right for 3 seconds and stop. This was working correctly before...
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
AKA wrote:
Is there a Auto-hold and remove Auto-hold feature
spezzafer: Signed integer display is something that's been fixed in a later version. Sometime I'd like to transfer fixes like that back into this version, almost everything except changes to the emulation/timing. I don't know about mapping the other banks but it doesn't sound hard to do... I can't work on it now, but I brought it up on the snes9x board.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
If you mean the AVI itself: There was a change to how sound is generated in Snes9x 1.5 (the original one with no Windows version) that breaks the sound in AVIs made in the Windows port. I didn't catch this until after 1.502. It'll be fixed in 1.51, until then you could use 1.43. EDIT: So apparently using FMOD sound driver sidesteps the problem, I didn't know that.
If you mean while playing a game it's scratchy: Try changing the buffer length in the sound options, or turning off synchronize sound with CPU.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
upthorn wrote:
While it's true that you must have accidentally activated read-only
It happens even when read-only is off, so he might not have. That's why I thought it seemed intentional, because that warning should only happen in read-only mode, and the warning actually correctly informed me of when I was going to have a desync. But I can see now how both of those could be accidental consequences of some code rearrangement.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
Major problem: Bulletproof rerecording is not working at all. Loading any savestate invalidates all later savestates, and makes them give an "inconsistent state" desync warning even when recording. That warning indicates this change is intentional or at least anticipated, but I can't find any option to re-enable bulletproof rerecording.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
RAM search and watch are eating up CPU time like crazy, because the minimum (real-)time between visible updates has apparently been removed. That was intended to prevent those dialogs from slowing things down too much, and should at least be in there as an option, similar to the "Hog CPU" sound option. It's especially bad if you try using fast-forward while either or both of those windows are open (new input goes ignored for hundreds of frames), but even at 100% speed it's too taxing. The CPU intensive part is specifically the redrawing of the window's contents, not any of the actual real work done by RAM search/watch.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
adelikat wrote:
2) That it is difficult to achieve this frame perfection.
So, no matter how technically perfect the movie is, it will never technically be perfect? A technical rating less than 10 tells me "at least one person thinks this movie could be faster."
I don't think something like that is a problem, however. The rating system seems to be working quite well, with very few unfathomable average ratings for anything. Honestly, my biggest problem with the ratings is that there is no way to submit a "null" rating for one's own movies to get them out of the non-rated movies section.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
I tested in Mario 3 with FCEU 0.98.15 and 0.98.16 and that method worked to find 0x0090 as the most precise X position value. My guess is that it's really finding the values that increased and you are thinking it found ones that didn't change because it sets original equal to current after a search.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
You should enter initials at the end of the movie so that it reaches the "THE END" screen. It's a minor change, so you don't need to cancel this (PM it Bisqwit or whatever people do to update a movie file).
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
You can test mouse/superscope recording in this Snes9x; it's not an official version and probably can't be accepted for publication (due to incompatibility and emulation inaccuracy issues), but the next official version will record mouse input similarly, barring any major changes before its release. The only mouse-related change already made after that version is to allow you to use the keys or joypad buttons assigned to the SNES controller (like A and B) to activate clicks and triggers, or autofire/autohold those buttons, instead of having to click on your actual mouse each time.
The game has to be running to see where the mouse will go, so the way to get instant and precise movement is to put the mouse where you want it to go, then load a savestate and frame advance, and the mouse will teleport there. Moving it near the edge of the screen can be tricky because any input that accumulates to put the mouse outside of the screen will be entirely rejected until you move it back inside and try again, or at least that's what Mario Paint does.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
The hack would be a lot slower than anything you could do with the mouse. The mouse can move anywhere between -127 and 127 pixels in x and/or y per frame with perfect pixel precision. All SNES games have that 2 frame delay before being able to respond to any input.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
I don't know if I would have watched it all without fast-forwarding through the long and numerous transitions, but most of the solutions were incredibly impressive and interesting to watch unfold.
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
Writing most of it from scratch is probably a necessity anyway when you're going between such different hardware while both ends need to be written in assembly and emulation is impossible. You would think they would still have the source for one of the Sonic Advance games, but my guess is that they couldn't use it because it would be technically inadequate to handle the (ironically) more demanding Sonic 1.
As for the physics, they are so bafflingly different that the only explanation I can think of is that they are placeholder physics that they never got a chance to replace with the real ones. They couldn't have lost the source to practically every 2D Sonic game in existence and also failed to notice that Sonic handled differently.
upthorn wrote:
I'm guessing that Sonic: Genesis is just what happens when you have the current Sonic Team rewrite the Sonic engine from scratch.
And/or when someone greatly underestimates how much time it would take them to do it and pushing back the release isn't possible and they can't spare people from other more important projects to work on this one. I heard that it also had to do with Yuji Naka leaving, but I don't know how much that really had to do with it.
In any case, there are better Sonic 1 hacks for TASing than this, albeit none of them for GBA [yet; I wouldn't be surprised if someone is able to do better with no budget], so...
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
Apparently you can simply choose Restart at any checkpoint to reset the clock to 0, which makes it trivial to get an in-game completion time of 3 seconds on Green Hill 1. A TAS would ignore that, but I don't think this version even deserves one.
VANDAL wrote:
What were they thinking?
That's not such a mystery, but a better question is "What happened?"
Emulator Coder, Experienced Forum User, Published Author, Skilled player
(1300)
Joined: 12/21/2004
Posts: 2687
Zurreco wrote:
Oddly, though, Ogre's solution works. When I had desynchs before, deleting the save file would remedy the problem. Sure, the problem is somewhere else, but removing the save file was a simpler solution.
But are you sure it wouldn't have worked if you had tried again in the same way without actually deleting the save file? If it's simply unreliable then you could do just about anything and try again and get the impression that what you did fixed the desync.
Well, if it does somehow help, then I would think upthorn's suggestion above would also work, or replacing it with a 0-byte read-only file. Unless that would make the emulator give periodic error messages when saving fails...
spezzafer: Signed integer display is something that's been fixed in a later version. Sometime I'd like to transfer fixes like that back into this version, almost everything except changes to the emulation/timing. I don't know about mapping the other banks but it doesn't sound hard to do... I can't work on it now, but I brought it up on the snes9x board.