Posts for Kirkq


1 2
7 8 9 31 32
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
If you mean the trigger for leveling up after a battle, I was unable to find it quickly. I'd assume it has a value that increments much like the elements related to the percentage on the pause screen. It may also have a subcounter related to how much experience the enemies give. In order to find the step counter (float) watch the agility counter. Right after it increments, step counter should be close to 0. Use the other comparison operations in order to find it. You want to compare it to previous value. So you do a new data search from scratch, then as long as you don't move, you can keep selecting "equal to previous" in a subsearch. Walk just a bit and stop. The value is now "greater than previous" (Previous is the value last time you did a subsearch for clarification.) Once the agility counter rolls over it should be "less than previous" You have enough to hack the stats for testing. The first thing I would look for is minimum conditions to be able to beat bosses, especially those that have attacks that never miss. Be sure to post your findings here when you get around to it. ^_^
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Maybe it's operating system difference. I'm using Windows 7. MP Level Counter (Byte) - increases by one every time a spell hits an enemy successfully, it takes a lot to roll over and get to 16 MP. It is related to the percentage shown on the stat screen as well. HP Level Counter (Byte) - starts at 0 and increases by 1 each time you get hit by some object. Once it increments after 9 you get to 51 HP. It is related to the percentage shown on the stat screen as well. Defense Level Counter (Byte) - same as HP level counter except once it increments after 19 you get to 5 defense. Agility Level Counter (Byte) - similar to previous except for increments of agility. I think the only trigger is running. It is related to the percentage shown on the stat screen as well. Step Counter (Float) - Since this is a floating point values it increases by various decimals. It increments from 0 to 2000 as you walk a fair distance. Once it hits 2000, it rolls back over to 0 and increments Agility Level Counter by 1.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
I think it is unfortunate that the motivation behind arcade games may tend to influence people to put in more money, but I don't think you can make a solid rule outright banning putting in more money for an entire system. The motivation causes the game creators to put in exploits that otherwise wouldn't exist, as shown in the submission. I'll present a hypothetical scenario though. Say a game gives you 100 points per coin or something small, and there is an event triggered by score (maybe which level you are on or something.) Then jamming in 1 coin per frame is the fastest way to beat the game, which is no good. I think it is best to consider it on a case by case basis. In the case of continuing in general, we have said it is okay for NES games. We've also seemed to agree that it is okay to replenish ammo. It seems questionable to do and get a bonus gun. What if the character started with a random weapon? Would it be okay to die to switch weapons? What if the default gun was better in some parts than a better gun? Is it okay to die and alter your gun to default even though it is better in that case? I think it is for the most part okay as long as it doesn't work to the effect that 100 coins are jammed in per frame. I think it should be considered on a case by case basis. I think the case presented the author may have been correct even though it detracted from entertainment.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
The author has shown consistent improvement on this run. It seems to be reasonably well optimized. (with a bit of unoptimized randomness in any beginner's TAS) This run contains plenty of well-planned damage boosting and drop manipulation at the end. The game isn't great, but it's an alright watch. The bosses are quite strange. Destroying an onion ring with a baseball bat is typically interesting. One critique. Is it necessary to get hit by the final boss? It looks bad if it isn't saving time. The submission can be updated if necessary so it isn't a big deal to fix it if it can be better. Also, I don't really think this submission "uses warps" in the sense that other games do. I'm thinking it would be better not labeled as such. (I also agree with Apathos's post below that a little more could have been done with the music. You did move/fly pretty close to enemies where applicable though, which was good.) I'm a bit unclear how this could look "like normal gameplay." I could understand if it wasn't found greatly entertaining, but I've seen plenty of normal gameplay in my day. (I watched the encode not the movie file.) Yes vote.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Finding a byte in Mupen 64: We're going to assume we know the behavior. We will use the current HP value to find since it is probably easiest here. 1: Download MHS here: http://memoryhacking.com/download.php 2: Open MHS. Go to File->Open Process. Choose Mupen 3: Go to Search->Data Type Search 4: Choose 'Short', exact value, and 50 (assuming you start with 50 current HP). (The data type is short for HP, the rest seem to be bytes.) 5: Click search (It should say something about 2,000-5,000 values remaining.) 6: Go get hit by something 7: Now go to Window->Found Addresses. This should open the 'Found Addresses' window if it isn't already open. 8: Click the second icon in this window or tab. It looks like two magnifying glasses. It should open a window called "Sub Search" 9: Choose exact value, type in your new current HP for the value (48 or whatever) This should narrow it down to less than 50 or so values. 10: Repeat steps 6 through 9 as necessary until you determine which one is your HP. Once you have your current HP address, (I'm assuming it is different from mine), the relative location of the other addresses should be the same. For example, just add two to get the MP Level and add four to get the current MP. The other values may be bytes or shorts though.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
If you're on windows you need this emulator. Try it, it may be a different version. http://code.google.com/p/mupen64-rr/ If you're on Linux (or for various other reasons) I may have to describe how to find the byte offset.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Are the values showing up correctly to start with? They should match the game stats exactly. If they aren't that is the first problem. The solution would be to either match my emulator or figure out the offset of the emulator you're using from the one I'm using. If step 5 isn't working, I'd guess you might be reading it wrong. What I wrote should be accurate. Also, I'm using MHS 6.1. If you have a newer version, there is a small chance the fields are labeled differently.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
To my knowledge the DW3 item glitch run simply corrupts memory, which has little to do with debug codes. The "Wild Warps" Crystalis run is no longer published, and was obsoleted by a slower run without using the debug code.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
I don't think this can obsolete any of the currently published movies. I'm of the opinion it has to be a new category. I think this submission needs to address two main points. 1: When is a debug code okay? 2: How many categories is too many categories? I think the point at which the debug code was discovered is largely irrelevant. If this had been known about in 2005 vs 2010, I think it should be viewed the same. (This seems to be a debug code; the author has addressed it as a code. Whether or not it is actually a debug code may still be up for debate.) I think utilizing what seems in this case to be a debug code would typically force a new category. The next question is, does the debug code actually provide suitably different and entertaining gameplay? In this case, the answer seems to be mostly yes according to user feedback. In the Earthbound submission, it was decided that accessing a debug code through abnormal means was acceptable in a new category. I think the general site rule against debug codes may be interpreted as "don't use easy to access debug codes to make the gameplay stupid" or something. In Zelda 2, we said that using 2nd controller Up+A to teleport to the beginning instead of system reset is okay. It would be stupid two have two runs separated by such a trivial detail. In Crystalis we changed our minds once and decided that using 1P A+B + 2P A to teleport to the next town was too broken. (I am now of the opinion that both Crystalis categories may be valid, though I wasn't at the time, but I guess it depends how different the run actually becomes. I do think it was stupid at the time to have the debug code run not allow the legitimate run.) I think the function of the debug code and the effect on the run are the largest deciding factors. The site's goal to provide entertainment. Why not feature an amount of material as is reasonable for each game? I think more categories may be better than less. I also am under the impression that maybe the site has moved in a new direction since the 14% run was rejected, but that is another discussion. Should any well defined, well played, suitably different, entertaining run be published on the site? I don't know. Why not? If someone likes it, he can watch it. If someone else thinks using debug codes is stupid, he can watch the other run. I haven't watched the submission yet. This is just open discussion on what I see as the main issues regarding this submission and some relation to previous site decisions.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
I type all my essays with hotkeys and macros. :D
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Someone asked me to do this, so I posted the results here: Relevant Addresses 00967XXX HP Level: 504 (Short maybe?) Curr HP: 506 MP Level: 508 Curr MP: 50A Def Level: 50C Agi Level: 50E Wind Level: 524 Water Level: 525 Earth Level: 526 Fire Level: 527 MP Exp Counter: 528 HP Exp Counter: 52A Def Exp Counter: 52C Agi Exp Counter: 52E Step Counter (Agi): 69C (Float) Values tend to match the in-game values. For example: Fire Level 1 has a value of 1, not 0. How to use this information 1: Download MHS here: http://memoryhacking.com/download.php 2: Open MHS. Go to File->Open Process. Choose Mupen 3: Go to File-> Open Save File. Choose this file I've uploaded: http://www.mediafire.com/?8tswxbucwvseecb 4: You can now observe memory while playing. 5: To overwrite a value for testing, right click on the value and choose to Modify Selected. Under Cur Value, type the value you want it to be set to. Click OK. This value is now written to the target value one time. The emulator will further modify the value as it normally would. In the case of element levels, if you hack to 20, it will increment to 21 when you level it again. Certain stats may have triggers stored elsewhere and could revert to an original value + 1. Just be aware. I confirmed that one magic attack hitting two enemies gives 2 MP increments, and a wind cutter level 3 gives 1 MP increment for each successful hit. It seems to shoot 6 times. Also MP takes forever to level as it is. I would generally recommend getting hit as often as is practical in order to gain HP and defense. Especially if it is only 1 damage.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
3-1 and 3-2 are awesome highlights. Great run! ^_^
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
I feel it would make more sense to list Simon as "Simon Sternis" under the authors, unless Simon particularly prefers to be listed as currently. Enjoyed the run, great job.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
The emulators only record resets as an input during 1/60th of a second. The reset button on the console is not polled every frame however, it interrupts at whatever cycle or command the system is executing. In this sense, the emulator limits the points you can press reset during. All of the points emulated are theoretically possible, but it is not a complete set. Conceptually, save corruption manipulation should be possible as emulated, but it is doubtful that one could get the cycle accuracy necessary to perform it correctly. See this post for inichi's comments on resetting Chrono Trigger between frames with the debugging version of the emulator: http://tasvideos.org/forum/viewtopic.php?t=386&postdays=0&postorder=asc&start=764 In my opinion it is likely that save corruption is more accurate on emulators than any sort of Megaman 1 NMI race conditions. It is likely that the NMI race conditions can occur during gameplay, but I tend to doubt the emulator's accuracy in the cycle timing. I do think save corruption can take away from certain runs, but I think it is fine to just make more categories due to the fact. Pokemon Yellow is still one of my favorite TASes. Replacing cartridge batteries is pretty common for old games.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
diggidoyo wrote:
-Mega Man Normal -Proto Man Normal -Bass Hard DLC -Mega Man Hard DLC
I am agreeable to this, assuming Megaman Normal is in-game time, and Megaman Hard is real time. Proto Man Normal/Hard I'm not so sure about, but it isn't terribly relevant right now. I was just trying to encourage discussion on the topic and by no means trying to be any sort of final say on this.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
"Aims for in-game time" http://tasvideos.org/Movies-C1012Y.html This may or may not be tagged on relevant movies. Most games don't even really have an explicit in-game timer, but many that do have an in-game run. Also see the Goldeneye and Perfect Dark TASes on the workbench.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Maybe some judges can provide feedback before you get started. If a consensus can be agreed upon before you start, it is likely there will be no problem when you do submit.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
An in-game featured online leaderboard seems categorically different than "what other sites do." We have many runs on this site that aim for in-game time, and the normal mode category might be what is best suited for such. It is another issue whether a given run should aim for in-game time or real-time. If the normal mode run goes for in-game time, and the hard mode run goes for real-time, maybe that is a decent medium? (Do certain items in the shop help certain levels a bit for in-game time, maybe this should be done for the normal mode in-game time run?) Mothrayas: Do you think that both Proto Man normal and Proto Man hard are valid categories? I doubt there is too much difference in the gameplay between them.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
diggidoyo wrote:
Looking at the options available, it seems there are 18 possible categories: Mega Man -Easy -Normal -Hard -Easy w/DLC -Normal w/DLC -Hard w/DLC And the same 6 options for Proto Man and Bass.
Yeah this is a bit of an issue. Since the leaderboard only keeps track of Megaman Normal times, I think it may be allowable to have that as a category. To my understanding DLC has a really minimal impact on the game, so maybe that is something better reserved for youtube demonstrations. IMO there are multiple valid categories. I think these are the best 4 personally: Megaman Normal (In-game timer?) Megaman Hard (Real-time?) Proto Man Hard Bass Hard
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
I'm wondering if we should wait until version updates slow down a bit. If anything regarding the input file or the general game engine is further modified, files won't be backwards compatible. It seems like a bit of an issue to have 36 different versions of Dolphin supported on the site for syncing purposes. Is this going to be a problem?
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
JohnInBerkeley wrote:
to Kirkq) - from http://tasvideos.org/EmulatorResources.html I tried to download NES FCEUX - saw three choices: * FCEUX 2.1.4a win32 Binary * FCEUX 2.1.4a win32 Binary - Windows 2000/98 * FCEUX 2.1.4a src The only one looked any familiar was the second one, the one for windows 98 and 2000 (although son's computer runs Vista) - but when tried to unzip that download, got a virus warning. ?? I'm just lost here. I don't even know what a port is.
You won't really need to deal with computer port issues, as the emulators do not connect to the internet. (EDIT: If you did indeed mean software port, it may be aware that you previously downloaded the wrong version for your operating system.) Your virus software needs to not block anything regarding the emulator. You want the FCEUX 2.1.4a win32 Binary. (Not the windows 2000/98 one) It seems you have extremely overprotective virus software on your computer that you need to bypass to open it. It seems you aren't terribly proficient with computers, so getting started may prove to be difficult. These files are 'binary executable' files, which basically means once you unzip them, they are ready to run. Overprotective virus protection is afraid of unknown executables (.exe files) because it contains executable program code. (An executable could be a malicious virus, but any normal working program is also an executable. A word document or text file cannot be a virus unless you set your computer up to run it as an executable.) The point is that you should always be wary of .exe files you download. All of the emulators linked to on this site are produced and maintained by members of this site and other credible individuals and will never be viruses. If you want to get things working, download the FCEUX 2.1.4a win32 Binary, open the emulator, and try to open a rom. A rom might be named Super Mario Bros (U). The (U) means it is a good version of the US copy. Other versions will still run in the emulator for getting started. Remember, we cannot tell you where to get roms. Really all you need to get started is the emulator and roms. Essentially everything else can be done after that.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
(I was also skeptical of the post, but I chose to respond in the case that it was a legitimate request.) We have had a few motivated 12-13 year old kids in the past who have produced good runs. The way most of us got involved was to simply pick up the emulator and start working, which might be difficult for someone so young. If you truly are interested, the best way to get started is to just dive right in. Most of us simply started out by download the emulators and recording some input. Our site generally tries to be supportive of constructive questions, but we hope you look through the documentation and have at least tried to fix your own issues before deferring to us. The documentation doesn't always cover everything, but we would hope that it is tried before asking someone to go out of their way to help. Start here: http://tasvideos.org/TASHowTo.html Of particular note: This site will not supply roms (games), or tell you where to get the roms. We only supply the emulators used to play the roms. Downloading games is a questionably legal activity, but many of the games we play are old and many of us own copies of them anyway. It is not illegal to rip your own copy of your own game to your computer, but some of us don't have the device to do that, so we get them online. This site also does not support many of the newer generation systems (PS2,PS3, Gamecube, Wii, Xbox, Xbox 360) yet because the emulators are not advanced enough. With regard to coding: Coding is something that takes a while to learn, and we probably can't offer the effort necessary to help someone to learn from scratch. Coding isn't terribly essential to many games. Many games can be performed well simply by monitoring the character X position, which is pretty easy to access through the emulator. Just to give some perspective on the tutoring. We currently have around 1700 published movies on this site, produced by around 300-400 individuals. Most of the non-Japanese population that produces TASes get them published here, so it is probably safe to assume that the number of people in the United States and capable of helping is less than that. A large portion of the skilled TASers are over 18 and either in college or holding a full time job. The odds of someone being nearby and interested are extremely low. This is probably more of a hobby than a learning experience until one starts advanced coding to perform more difficult runs. Hope I was able to supply some useful information. - Kirk
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Yeah now just persuade him to update the Individual Level records. :p He beat every single individual level record in a day at the SDA marathon, but he's too lazy to record and submit them.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Last I heard, emulators may not be emulating lag correctly for this game. Blast Corps is terribly laggy, and the in-game timer keeps running while the game lags. This could easily lead to emulators producing console impossible times. The camera angle is ridiculously important on certain laggy levels. The right camera angles can take off more than a second per lap in Mercury. Certain buildings will also lag the game much less upon exploding when switched to carrier view.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
I prefer seeing Aztec and Egypt included. Since cradle is so short, I don't think running it twice is bad. Is there any silly, quick novel glitch you can throw in the second time for not too much time expenditure?
1 2
7 8 9 31 32