Posts for Kirkq


1 2
17 18 19
31 32
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Aglar wrote:
You're marketing the run as your own while it to looks to me like the only thing you did was changing the last 17 frames or so.
If this run has plagiarized all of the input but 17 frames, I question whether or not the rerecord count is actually 449, or if he just edited your movie and applied 449 - 436 = 13 rerecords. Side Note: Does the game end on the exact same frame, or is it a bit later? While you seem to be taking this in a light-hearted manner, I would consider this pretty questionable. I'm quite sure I could grab numerous movies on the site, spend half an hour, hex a small improvement in, and submit it for my own. (Why bother hexing when you can just improve the ending by one frame?) Being that the submitter doesn't exactly seem to have a good reputation, I am assuming the worst. Based on the fact that this movie is so short, I don't know where we draw the line. If this was a 5 minute movie without extremely tight frame wars, I would give a one frame improvement a definite no. If I was convinced that significant effort was applied and further improvements greater than 1 frame had been thoroughly examined, I would reconsider. The following assumes that the run truly is plagiarized to the frame, which I did not check. I assume this isn't at all necessary for optimal input in this game. The author did not even give credit to the authors who he copied from. I would say one should typically request permission to simply copy the input, but in this case even minimal credit isn't given. Unless I am stating something incorrectly, my vote would be no on this basis. I will not apply it to the tally.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
It seems to me that this run wasn't actually improved. I have 3864 frames on mat89's run and 3893 frames on this run.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
The ending here looks like it may be faster. Maybe not. http://www.youtube.com/watch?v=DSs-pSopMrY Also, an elevator BLJ may be about as fast since this route gets to the elevator a fair amount earlier. It depends how fast the elevator actually gets to the top though. Maybe it has already been proven slower. Probably stating obvious facts.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
http://dehacked.2y.net/microstorage.php Select your file. Leave a comment. Hit upload. Copy the url on the next page to us. Uh, yeah that's about it. No file managing or anything.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Your inclusion of the words "in real time" confuses me.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
andymac wrote:
As far as I know, the NES at least has no non volatile memory. Which would mean that if it is turned on as soon as it is plugged in, then the same demo would be played.
Would unplugging the NES have the same effect?
Dwedit wrote:
The thing about the DMC channel is that its counters are always running no matter what. It's always ticking down so it knows when to fetch the next byte, even when it's turned off.
You say this runs even when the game is off. Are you assuming that or is it known? It could just be left in the state it was in when the game is turned off for when it is next turned on. Essentially, if we unplugged the NES every time, would the same demo play upon turning on the game's first play?
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
It is possible to do scripting in Mupen64 with L.Spiro's Memory Hacking Software (MHS) available at http://memoryhacking.com/ There is plenty of documentation available to make a script interact with the game and a few users here have learned to do it. Most of us don't know how it works, so you can either learn yourself or be patient and wait for lua scripting in an undetermined amount of time.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Captain Forehead wrote:
So my question is this: Should I get all of the remaining items to get my score even higher, or should I just get the minimum amount needed?
You should definitely get the minimal score necessary for a 5, I don't see any point in getting more. The goal of the run is to clear world 6. For most stages the score for a 5 is all permanent objects collected/destroyed. (Adok bombs are an extra 50 points, any enemies that suicide are also extra points.) I'd also like to see your progress before you go much further. You can PM it to me if you don't want to release it publicly.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
mklip2001 wrote:
Out of curiosity, what is the C meter? And how do you get the final boss to disappear like that?
C is candle. It is used by pressing B. It lights up the one dark portion in room 2, and if used in the right places it gives me hidden power ups. (The gray boxes that seem to appear out of nowhere.) The final boss is deleted by spawning the lizard king very near him. The lizard king deletes enemies on screen. This probably happened to people on console and confused them very much. It may be possible to force better lizard king spawns in room 1 to delete the rock throwing prisoners that held me up, but I couldn't find any way to do this.
mklip2001 wrote:
I'm just curious: is it possible to get the Lizard King to steal your gold before the third boss?
The lizard king does not appear in that room at all.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Uploaded to youtube. http://www.youtube.com/watch?v=1RySTbjhJQY (Can't seem to edit the submission at the moment. Not important.)
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
That would be awesome if it could be incorporated into the run, but I doubt it is able to be faster.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
If you want to link directly to a picture like that, you should right click on the picture and 'Copy Image Location'. That link as is won't work for anyone but you.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Goma Cave Strategy: -Enter the Cave -Go through the southwest door in the first room --(optional, move the log for the Mars Djinni) -Go to roughly 24 14 in the same room (a more optimal spot probably exists) -Retreat Save Glitch -Walk north through the door --(optional, go back and get Mars Djinni, it is a room away.) -Exit the Cave Originally off map you will find numerous doors that lead to the cave in Vale. Note that if you exit the cave your retreat glitch location gets reassigned to the exit as opposed to the entrance. This setting allows you to find off map doors to various places in the cave. (Seems useless currently.) Bilibin Cave: It may be possible to glitch to an area in the cave currently inaccessible at this point. It is possible to get off the map and cause some abnormal exits, but none are too notable. Mercury Lighthouse: a normal retreat glitch after going up the stairs in the room filled with water will take you to the aerie. Required Fights: Lizard Guy, Saturos More thoughts: We arrive in Imil just a few minutes after exiting Vault. Djinn count will play a factor in the routing. Unfortunately this run is going to have a huge cutscene density. Most of it will probably be fast-forwardable. EDIT: It seems that most of the bosses are weak to either Fire or Water, further research suggests that 2 Mercury Djinn are fairly essential to fight Saturos effectively. One possible strategy: recruit Mia, go down the stairs you came from, retreat glitch up the stairs, backtrack ~3 rooms, fight the Djinn, Retreat glitch out of the Djinn room. This still seems pretty mediocre, I can experiment with save glitching later, there are multiple retreat points in the lighthouse.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
You have to find bytes in memory that represent things you need to overwrite. (I say byte, but it could be a word or a float as well.) If you don't understand anything about video game memory, you probably won't be able to do this yourself. 1.- Ultimate cancel hack Just send Jpleal a message and I'm sure he can tell you what byte needs set to what value. 2.- Replace a character/player in vs made into: polygon mario/yoshi/etc, giant dk, metal mario. You need to find values that represent Player Character and change the values to get characters wanted. I'm sure there is a byte for 'always giant' or something. 3.- Replace a/all stage(s) into a beta stage or battle field or Final destination. Find values that represent which stage is being played on, and change them until you get these stages (if they exist in memory). Looking up bytes in an N64 game is much harder than looking them up in a SNES or NES game. It takes a lot of time and trial and error to pinpoint exactly how the byte is varying. Once you find the applicable byte, setting the value is very simple. We don't go making cheat codes out of nowhere; we spend plenty of time finding places in memory that we can change to make things happen. Unfortunately, no one here wants to spend a few hours looking for bytes for you. You will have to experiment yourself.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
What seems to be happening in Sol Sanctum is that room 3 is on a separate floor from most of the other rooms. When walking around off map you can run into the boundaries of most of the other rooms, but particularly you cannot run into the rooms you can retreat glitch out of. It is really strange that when I underflow Y coordinates I can suddenly get inside room boundaries or something. I may make a camera hack soon. Use this new script to observe and record coordinates.
local AD1 = 0x0200053a --Isaac Cur PP 
local AD1Value = 18
local AD2 = 0x0200047a -- Step Counter
local AD2Value = 1

while true do
gui.text(0,20,"Isaac PP " .. memory.readword(AD1))
gui.text(0,30,"Step Counter" .. memory.readbyte(AD2))
memory.writeword(AD1, AD1Value)
memory.writebyte(AD2, AD2Value)

gui.text(0,0,"X: " .. math.floor((memory.readdword(0x02030ec4))/1000000))
gui.text(0,10,"Y: " .. math.floor((memory.readdword(0x02030ecc))/1000000))

    vba.frameadvance();
end
If you find new doors please give coordinates to describe them, in addition to which room you left. Doors I found off screen: 1: 30, 4282 2: 28, 4278 3: 4265, 4278 4: 4263, 4289 5: 50, 4280 6: 4289, 4284 (The maximum coordinate is 4294.) Note that doors 1 and 2 compared to doors 3 and 4 seem to be mirrored coordinates of the same staircases. I have yet to figure out exactly what is happening, this is why a camera hack may or may not help. Also a friend of mind mentioned trying this twice in a row. Retreat glitch warp off map, retreat glitch warp while off map. I may try to find a 'floor number' address in memory. I really want to get to the upper floor of the sol/luna room and possibly skip the ridiculous Kraden cutscene. EDIT:: Some relevant camera values, I assume these are merely reflections from somewhere else in memory, as manually setting them only messes things up. I may try to see if there is a byte representing "center camera on Isaac." I'm not sure if I'll be able to make this part work.
--Differentials from X and Y position
local dxHigh = 43515904
local dxLow = 7864320
local dyHigh = 39845888
local dyLow = 6291456

local AD2 = 0x0200047a -- Step Counter
local AD2Value = 1

while true do

gui.text(0,10,"1: " .. memory.readdword(0x02030db0))  --LowX
gui.text(0,20,"2: " .. memory.readdword(0x02030dd0))  --LowX
gui.text(0,30,"3: " .. memory.readdword(0x02030e00))  --LowX
gui.text(0,40,"4: " .. memory.readdword(0x02030e30))  --HighX
gui.text(0,50,"5: " .. memory.readdword(0x02030db4))  --LowY
gui.text(0,60,"6: " .. memory.readdword(0x02030dd4))  --LowY
gui.text(0,70,"7: " .. memory.readdword(0x02030e04))  --HighY
gui.text(0,80,"8: " .. memory.readdword(0x02030e34))  --LowY
gui.text(0,90,"9: " .. memory.readdword(0x0203108c))  --EqualY
gui.text(0,100,"X: " .. memory.readdword(0x02030ec4))
gui.text(0,110,"Y: " .. memory.readdword(0x02030ecc))

memory.writebyte(AD2, AD2Value)  --Remove Encounters

--0x03001ad4 , 0x03001ad8 , and 0x03001adc (words) 
--seem to contain an x coordinate for the background layers.


vba.frameadvance();
end
EDIT: New idea, try glitching in room 9 once glitching to room 9 from room 3 and see if it puts you on a different floor. The goal is to skip the Kraden cutscenes by getting onto the top floor, or by some random chance get into inner sol sanctum.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
http://www.youtube.com/watch?v=4i6c6BfK_u0 Here's the results of today. There are 3 rooms in Sol Sanctum that point to the entrance when retreat glitching. If you save while retreat glitched, your coordinates are preserved, and then upon reloading you end up in the room pointed to, but at the preserved coordinates. It is possible to get inside an overlapping room on the first floor by retreat glitching. The room I glitch out of is not actually on the first floor. We'll call it the second floor. By glitching out of the second floor, my coordinates get translated to the first floor and I am either actually in a room, or outside all rooms. There are two rooms on the first floor with coordinates that match positions in the room I glitch out of. I can physically get into these two rooms, and they act normally. If I am outside of any rooms I cannot get into any rooms on the normal map. I then experimented walking around while examining my coordinates in memory. When I passed the y coordinate boundary, (it underflowed), I found strange walls that didn't seem to represent rooms I had mapped. I randomly found this door, it randomly put me in the final room. More investigation is needed to find out exactly what is happening when I cross the map boundaries. Sol Sanctum now takes roughly 1 minute to traverse. This saves around 2-3 minutes of gameplay.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Here's a thread on Global Assault: http://tasvideos.org/forum/viewtopic.php?t=6360 I remember the WIP in that thread was pretty entertaining. Don't use a password, there's no necessary reason to.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
I agree that the 120 star run will be slaughtered enough by BLJs that a 70 star no BLJ run will most probably be accepted as a category. The run has a clearly defined goal, and will most probably have support. The only thing left is to lock someone in a room with a computer and force them to do it so we can see the final product. So, if you're saying you'd like to see a 70 star run, you agree with most of us. If you're saying HurRy uP aNd F1n15h 7eh ruN pl0x, you have the wrong impression. Everyone here sets their own TASing goals and work pace. Since your argument seemed to have the wrong impression, I'll go ahead and assume you meant you'd like to see a finished run someday. EDIT: I'll add that I'd personally rather see the 120 star run updated first.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
funnyhair wrote:
How is this game very grind heavy? is it that there are powerful bosses that you need to level massively to beat? or am i completely off?
I'll try to give a more in-depth response. Currently in the run Lux is level 21. 165 HP 55 MP 112 Power 111 Guard 64 Magic 92 Speed Lux's average stats would be: -X- = per level gains -7- 158 HP -2- 43 MP -4- 85 Power (+4*2 Power Seeds) = (93) -5- 106 Guard (+4*1 Protection Seed) = (110) -3- 63 Magic -3- 63 Speed (+4*3 Agility Seeds) = (75) Magic attacks by bosses almost kill us currently at this average rate. This is why many of them are dodged in the TAS. Our dodge rates are in the 10% range or so. Also consider that you can only buy/carry 9 Potion 2's. On the way to a boss you may fight 30 or more fights, so this supply would deplete very quickly. We would rely on Kamil's Heal 2 much more, but half the enemies have MP Catcher to steal his MP. We can carry 9 M Herbs to try to boost his magic when healing is necessary. Power and speed are the most relevant stats. Our power is 19 over the average, and our speed is 17 over the average. This gives that we have a power stat of an average Lux 5 levels higher, and a speed stat of an average Lux 6 levels higher. In a normal game you would need to be roughly 5 levels higher to generate the same hit power and hit rate. We would need to be level 26 to match our stat manipulation. Level 26 requires 103870 experience. Upon gaining level 21, we had 53549 experience. This equates to 31 more S Brain kills. Note: S Brains were being killed on a roughly 20% chance of a vacuum kill. Even at level 21, they could kill me in 2 hits. They can also vacuum me. Grinding them alone as Lux, I would get killed by a vacuum frequently, and when I don't die, I'd have to heal after essentially every hit. So even to have a chance at level 21 I would need a partner, which halves my net experience gain. I now need to kill 62 of them to get from 21 to 26. (This assumes I have killed over 200 battles of suboptimal monsters to get from 13-21.) Add in the fact that the best S Brain area in the game has a 1/4 fight rate, you will engage roughly 248 battles before you kill 62 S Brains. Also you will need to stop in at the inn and/or buy more items at least every 10 fights, probably every 5. Add this to the fact that one would need to account for how much luck manipulation the TAS is doing to not die. Level 26-30 is another 53k experience, which is essentially another 248 battles to kill 62 S Brains. This might be good for a while, but at the end of the game you'll still be getting destroyed pretty badly at level 30. You see where this is going. This is the nature of 7th Saga.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
More reading. This particular post references Lost Age locations, but it gives you a better idea of how to use the glitching. http://speeddemosarchive.com/forum/index.php/topic,8737.15.html Also I should have provided this earlier. This lua script should remove random battles and allow you to manually set Isaac's PP for more convenient testing.
local AD1 = 0x0200053a --Isaac Cur PP 
local AD1Value = 0  --Value to set PP to
local AD2 = 0x0200047a -- Step Counter High Byte
local AD2Value = 1  --Removes battles

while true do
--gui.text(0,0,"Isaac PP " .. memory.readword(AD1))
--gui.text(0,10,"Step Counter" .. memory.readbyte(AD2))
memory.writeword(AD1, AD1Value)
memory.writebyte(AD2, AD2Value)
    vba.frameadvance();
end
Copy to notepad, save as " Something.lua ", run in emulator.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
In the current WIP: -Gain control in Hyrule Field first time to Kakariko: 29 seconds -From beginning of transition into Kakariko to completion of transition out of Kakariko: 159 seconds -Kakariko to Lake Hylia: 55 seconds (Night time occurs 37 seconds into this.) -Enter Lake Hylia until talking to owl at dawn: 65 seconds. This would suggest that there are roughly 149 seconds to get to the next daytime. These timings are accurate within a second or so. ============ -Consider that you can probably get to Zora's River in about 20 seconds from Kokiri. -Consider it takes 15 seconds from entering Zora's River to get to the magic bean guy. -Consider it takes 35 seconds to get to Zora's Domain from there. Get the Silver Scale. The diving game takes about 20 seconds, and it takes about 25 more seconds for dialogs and traversing back up to him. 70 seconds of game time have elapsed and we are in Zora's Domain. -Estimate 10 seconds of game time to grab the bottle in Lake Hylia. In Lake Hylia it takes about 15 seconds to get from the warp to the magic bean spot, and probably 10 seconds to get back. -Estimate about 85 seconds (*real time) for talking to the king and getting back to the warp. (This conversation/moving takes almost 60 seconds.) 105 game seconds have elapsed and we are in Zora's Domain. -Exit Zora's Domain to Zora's River, estimate about 25 seconds to go down Zora's River. (May be a high estimate.) -Estimate 20 seconds to get to the castle from Zora's River with some sort of water/skeleton superslide. Exactly 150 game time seconds have passed with these estimates, the bridge should just be coming down. Any overestimating I have done can be accounted for by collecting rupees. I now believe this is the fastest route, and it seems significantly faster than the Cucco Bottle route. We save 55 seconds of owl flight time, 159 seconds from Kakariko Village time, and most of 20 seconds of wasted time waiting for the owl. This itself is 234 seconds. This is compared to 45 seconds of getting the silver scale, 85 seconds of talking to the king, and 15 or so seconds of getting the Ruto Bottle. That adds up to 150 seconds. This doesn't account for some various ~10 second travel times, but most of that can be accounted for by not traversing Hyrule Field again. ======== If getting out of the warp, getting the gold scale, getting the bottle, and getting back in the warp is faster than getting the silver scale, it can be considered. I have a feeling that it is over 60 seconds though. Especially considering swimming. It also adds an extra path to a warp which probably adds about 10 seconds. I'd estimate it takes the silver scale route about 65 seconds of real time to make it back to the castle from Zora's Domain. This is directly weighed against the 55 second owl flight time. The gold scale would have to be equally as fast including an extra trip warping from Zora's Domain and pass enough game time to be on time for the owl in order to be better than the silver scale. Also, considering a route that traverses from Zora's River to Lake Hylia through Hyrule Field, it would take roughly the same amount of time, if not more to get to the fishing pond. Going through Zora's Domain would be almost equally fast if not faster. The silver scale route also allows us to collect rupees on our way. This paragraph was a bit of a ramble, but I think it presents considerations that reject the gold scale as a viable route. ======= I hope this was a useful analysis, if anyone has better estimates feel free to correct mine.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Slowking wrote:
So the golden scale route should look like this: - get sword, shield and stick (-50r) - old forest escape - peahat SS to ZR - buy magic bean (-60r) - through hyrule field to lake hylia (I hope you can make it to the peahat before night) - get golden scale (-80r) - get ruto bottle - get real bottle from king zora - go back to lake hylia through the warp - plant bean - fly with the owl to castle town - buy the hylia shield (-160r) and travel foreward in time
Comparing this to the current WIP, I think it leaves you quite late for the earliest possible owl by a significant margin. Crossing Hyrule Field matches the current WIP, and adding all the errands and the fishing pond almost certainly makes you significantly late. I think going into ZD and grabbing the Silver scale will make up for time spent unnecessarily crossing Hyrule Field. I'm thinking: - get sword, shield and stick (-50r) - old forest escape - peahat SS to ZR - buy magic bean (-60r) - through ZR to ZD - Silver Scale - Warp to Lake Hylia - get ruto bottle - plant bean(*2:) - Warp to ZD - get real bottle from king zora 1: - Warp to Lake Hylia - plant bean(*1:) - fly with the owl to castle town - buy the hylia shield (-160r) and travel foreward in time 2: - make way towards Castle manually. - buy the hylia shield (-160r) and travel foreward in time I'm beginning to think that this spends enough game time. Since it is claimed that it takes a significant amount of time to go up Zora's River, perhaps this approximately matches the Hyrule Field crossing time in the current WIP. What is the approximate time difference for Silver Scale vs Gold Scale? It seems that crossing Hyrule Field the extra time and the Gold Scale are associated events. It seems that the benefits of each cancel each other out, as both together advance game time to the point that you are quite late for the earliest owl. (This is merely wasted real time.)
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
That is a phenomenal note, I didn't think of that. It is very possible that 3 sturdy characters is better than 4 weak ones.
Experienced Forum User, Published Author, Player (208)
Joined: 7/7/2006
Posts: 798
Location: US
We are weighing game time spent in one trip across Hyrule Field, vs. one trip up Zora's river and the amount of game time spent grabbing the Ruto Bottle and getting back to Zora's Domain. The trip across Hyrule field is 55 seconds. Weigh that against the real time spent going up Zora's River and getting the bottle. Though if you're planting the magic bean that may add 10 seconds in your favor. If going up Zora's River and grabbing the bottle takes 45 seconds, you're still just as early as the current WIP, which is about 20 seconds too early. I didn't really say it was entirely game time that matters. It's just one of the limiting factors. If a real time route is much faster, it is still worth considering, you just have to be aware of how much game time you will have to waste in exchange.
Slowking wrote:
That means we have to avoid areas where time doesn't move at all costs, since those would prolong the wait. ZD is one of these areas and getting the silver scale would keep us quite a while in it. The golden scale on the other hand only keeps us in the fishing pond for a short time and for most of the part in lake hylia, where time actually progresses.
It depends very highly on how much time actually needs to be wasted. I believe Silver Scale is still a legitimate possibility.
1 2
17 18 19
31 32