Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
I made this simple COMPARISON.
You can easily see why I can hit enemies and no need to care about the lower speed.
If you think I should stop movie later like Aglar, don't worry, it can be edited easily. But I don't think this is very important or necesssary.
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
I'm mean we should change the subtitle in the movie and submission title's playing time.
For example, http://tasvideos.org/1767M.html
Arcade Metal Slug (World) in 12:10.23 by x2poet.
But the real playing time is 12:20.35.
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
I think all of the arcade games should be re-encoded, because of the playing time. Different hardware work on different fps. For example, Capcom's CPS2 works on 59.630 fps, SNKp's neogeo works on 59.180 fps, and BanPresto is 57.550 fps.
We can't use 60 fps as default, it's not the real playing time. But the .fbm didn't give the identifier to distinguish this, so when fbm movies be upload to tasvideos, how to solve this problem?
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
No need to count, because U version must slower than the J's, and I can't believe this game need jack-in more than 100 times.
And I can't find anything different between Erope version and U version. The movie file can be sync perfectly to E version.
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
Did you have encoded?
xipo has improved this run 138 frames before 2 years ago.
http://bbs.tasvg.com/thread-1459-1-1.html
But I don't know he's willing to submit again or not.
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
I'm thinking of the Bule versions. This is my test.
http://tasvideos.org/userfiles/info/2292230436886964
Except for flashman, Bule versions's boss battle is faster and easier than the white versions. Because I can buy
Twister C(3000Z, power 20, hit 8 times) from SciLab Square's shop. But the virus battles are harder than the white versions, I have to face some Momogra virus which always hide under ground. I can use IceBall M, Burner Q(bad choice), Knight U to beat Momogra virus. But every battle with Momogra virus will cost more than 2 seconds(more than 4 battles I know).
So it's seemed blue versions is hard to say better than white versions.
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
I have compared 3/4 between 2 runs, and I think no need to go on comparing.
This is the avisynth file, and there're more expatiation in it. Someone can output 2 movies with frame count first, and then using AvsPmod or VirtualDub to watch.
Now McBobX's run is 620 frames slower than the published run during the gameplay. In only 1 mission, McBobX's run is 87 frames faster than the published run. I can't believe this run has chance to catch up with the hellagels's run.
The same mistake http://tasvideos.org/forum/viewtopic.php?p=329286#329286
McBobX lost too much frames in the same place. And the lag count is added up to 183 from 41.
So I vote no.
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
TAS COMPARISON: Part1Part2
Sorry, I have to vote no. In fact, this run is slower than the published run except text, though McBobX improved in some place. In too many place, McBobX, you should do better indeed. For example, you lost too much time in this.
Did J rom's enemies are more than U rom's?
I'm interesting in this place.
Is it some thing different between U and J? Zero is nearer the door in J rom. But I only found it happened twice in the mission. Or is it means there are some faster way to through the doors?
This is the comparison avisynth file what I made. I hope it can help the judge, because of more info. And I hope this comparison can help someone to improve in the future.
I haven't made another comparison. Maybe wait 1 or 2 weeks later. The GBA's comparison is very hard to make, because VBA always output 1 more frames about per 200 frames. Which version I used, VBA-rr v23.5 svn421.
edit: I forget to say maybe there are 1 or 2 frames faster or slower in the middle of this comparison, because VBA always output 1 more frames about per 200 frames.
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
Hi, marx. I have watched you wip, there are 2 problems.
1. you used the wrong rom. Your rom is Ninja Gaiden (U) [b1].nes. U should use the rom like this: Ninja Gaiden (U) [!].nes.
[!] means perfect dump. The rule said the rom must be good.
2. what's wrong with this? |2|R.......|........||
In your fm2 file's line 2871. The flag 2 means reset the game. Is it a bug in fceux?
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
zk547, did you complete mission4? Something I forget to say, when you got L, you can pass once A every 24 frames. Because Laser gun power is 8*2, every 4 frames 1 shot, shot 5 times when pass A once. But the last shot power is 8*4 in 8 frames. So Laser's real arms is equal to 240. Before boss battle 4, if you can save more 1 or 2 Flameshot than x2poet's run, you didn't need to wait the second L from prisoner.
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
This run I used marco complete mission1, and 58 frames faster than my eri's run. But I don't want to go on working this, because I dislike no death or only use one character.
I hope this can help someone who want to improve the publish one. emulator: 0.9.7, Bus-Level Timing onhttp://dehacked.2y.net/microstorage.php/info/419663533/metalslug7marco.dsm
version 1
emuVersion 90700
rerecordCount 5983
romFilename Metal_Slug_7_USA.nds
romChecksum CF8D691B
romSerial NTR-YM7E-USA
guid 84BE2329-6CE1-AED6-9052-49F1F1BBE9EB
useExtBios 0
advancedTiming 1
encode:
http://www.mediafire.com/?73n2sdtiulo9z3j
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
The higher evasion is very useful, e.g. if be hit by MiniBomb, it will cost more than 2 normal chips be used, and Roll can easier avoid sword chips. In classE, different chars' opponents use different chips, and roll's opponents are easier than others.
http://www.gamefaqs.com/gba/918715-mega-man-battle-chip-challenge/faqs/29301
But after get gaiasword, Roll is the easiest Navi be deleted, his recovery chips can't work. The most important reason what I choose roll is that YoYo3 is very useful. Others can't get this chips early.
in fact, the more MB is useless except ClassE and D. Because the element Navi has the lower MB than main characters. But I have to admit Woodman(290MB) maybe better than Plantman(270MB), because when use woodman, I can equip gaiasword(70MB) Goldfist(50MB)x2 MegaCannon(40MB)x2 Dynawave(40MB). Don't forget the more chips in deck the higher levels and the more chance get chip, but can use slot-last glitch to reduce the level.
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
I have explained when you take part in Class at first time, Holding A can't save any frames. Sometimes because of the lag, you may lose 1 frames.
Thanks for your help, pass start at first frames and no need to holding. I will fix it. But this ClassS file is not equal to my full tas, I didn't manipulate luck for Gaiasword.
edit: thanks AKheon again, you watched very carefully. There is a mistake in 152230.
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
Thank you very much, and I had found how many frames be saved when holding A: be in pet, 17 frames be saved; out pet, 18; in shop, 19; out shop, 18; be in battle chip GP, 19; choose class(E, D...X,Z) first times, 0 or -1(-1 is the lag); choose class second, 11 or 12 or 13(because of the lag); before edit chips, 17.
Now, I have fixed during ClassS, and saved 2646 frames.
http://dehacked.2y.net/microstorage.php/info/783626564/1403MBCCclassS.vbmEdit: fix a mistake, now saved 2664 frames.
http://dehacked.2y.net/microstorage.php/info/501669231/1403MBCCclassS.vbm
I use CodeBreaker cheat to change the name, but after my retest, I found that change name can't effect RNG, only change the NAVI-code. I used to make NAVI-code by myself for the 128 tournament. The CodeBreaker you can see
http://www.gamefaqs.com/gba/918715-mega-man-battle-chip-challenge/faqs/33138
But this play get the different result. http://tasvideos.org/3620S.html
I angered that this should be avoided in the beginning, but he disappeared at that moment.
Another thing I have to say, my this tas's rerecords count is far more than it displayed. In most cases, I must use tas-edit again and again to adjust chips position for next 2 or more battles. If you watch it frame by frame, you would realize.
Experienced Forum User, Published Author, Skilled player
(1152)
Joined: 5/11/2011
Posts: 425
Location: China
No no no, you should post your proposal in three months ago, indeed. http://tasvideos.org/forum/viewtopic.php?p=313987#313987
This tas has wasted me too long time. All right, I will go on tasing megaman battle network 3, and please don't say this is a bad game choice again until I complete.