[Hitokiri]

Hitokiri
the most striking thing is that the chun li stage had and they did not use it in the SF Zero 97 ( SF Zero )

Chun li stage ( hidden game code )

Amazing! I don't know why they didn't put on this Chun-li stage.

As I said, this game was very well done by NES standards.
It would be nice to make an improved version of this Street Fighter Zero 2 '97.

If I knew how to program and modify NES games I would do it, but unfortunately I don't know and I don't even find a tutorial teaching how to do it.

 So I can only congratulate and admire who can do these jobs.

[sics]

Hitokiri & Supergamerguy
When I started, the machine translation tools weren't that sophisticated and since I never got along with languages ​​other than mine, I didn't have the possibility of accessing any tutorial, but that was not a limitation for me.

And no, I did not have other knowledge related to the world of computing, less I am some kind of gifted, indeed I am the least intelligent of my brothers.

That's why I just loaded the game in the emulator and began to delete random data until I touched something that affected what I saw in the game, analyzing again and again what was happening, slowly I was taking more and more notion of why each thing happened.

Even for each graphic edit I wanted to do, at the beginning I had to rebuild each frame sprite by sprite in an image editor, which made for each small modification it was a torture to assemble and disassemble graphics.

Many still think that Sonic Improvement Vol.2 is just an appendix to Jabu's project, when in fact it dates from 2006

Nothing is easy, but at the same time nothing is impossible, it is just a matter of trying to find a way to do it:


It is obvious, the more experience, the easier each thing seems, but despite this it is always a good time to start



I finally had to start over, how ironic


Changing the subject, as someone will surely end up asking, Chun-Li's sprite is not a recreation, it just adapts the data of some sprites to the format that the NES Screen Tool can read, in theory this could be automated with a script, but neither It takes a long time to do it manually, as I said before I am not interested in doing this task, but it seems important to clarify that it is not something unfeasible, it just requires a lot of work.

[Hitokiri]

Hitokiri & Supergamerguy
When I started, the machine translation tools weren't that sophisticated and since I never got along with languages ​​other than mine, I didn't have the possibility of accessing any tutorial, but that was not a limitation for me.

And no, I did not have other knowledge related to the world of computing, less I am some kind of gifted, indeed I am the least intelligent of my brothers.

That's why I just loaded the game in the emulator and began to delete random data until I touched something that affected what I saw in the game, analyzing again and again what was happening, slowly I was taking more and more notion of why each thing happened.

Even for each graphic edit I wanted to do, at the beginning I had to rebuild each frame sprite by sprite in an image editor, which made for each small modification it was a torture to assemble and disassemble graphics.

Many still think that Sonic Improvement Vol.2 is just an appendix to Jabu's project, when in fact it dates from 2006

Nothing is easy, but at the same time nothing is impossible, it is just a matter of trying to find a way to do it:


It is obvious, the more experience, the easier each thing seems, but despite this it is always a good time to start



I finally had to start over


Changing the subject, as someone will surely end up asking, Chun-Li's sprite is not a recreation, it just adapts the data of some sprites to the format that the NES Screen Tool can read, in theory this could be automated with a script, but neither It takes a long time to do it manually, as I said before I am not interested in doing this task, but it seems important to clarify that it is not something unfeasible, it just requires a lot of work.

Thanks for the great master tips.

I will try to do that then.

Thanks.

[shakunetsu]

looks good the characters feels a bit smaller for there background but the new background are looks great

[gadesx]

Many hacks/bootleg games used a weird IA, I'm not sure about if this can be enhanced to work like the arcade version.
I know the characters tactics, if someone need help.

I suppose that this game have a lot of info from before.

[shakunetsu]

for AI insights

https://sf2platinum.wordpress.com/2017/01/20/the-ai-engine/

https://towardsdatascience.com/street-fighter-ii-is-hard-so-i-trained-an-ai-to-beat-it-for-me-891dd5fc05be

[sics]

sebastianangel
Sorry I can't answer your message, your inbox is full, I recommend you save everything as separate web pages and then delete the messages

gadesx & shakunetsu
I don't like fighting games, which makes my experience in this area nil, that's why I don't think I'm the one to carry out their proposals.

On the other hand, although the first article seemed interesting to me, in the second I don't think that artificial intelligence has given an optimal result, I think player 1 only won because of the way the game's AI works, but other challenges such as bonuses , the player was certainly lost. Anyway, I don't know how that information could benefit the development of this project...

I think it would have been better to use an emulator with saves states and save games before performing each action, then load state and compare the results taking the player's score as parameters.

In addition, it would also be a good idea to establish a different strategy for each stage, if it were me I would use the timer as a reference point, since at the beginning of each fight or each bonus it will always go through the same states. This could help the AI ​​link that sector to a strategy. Obviously I don't know anything about Artificial Intelligence, from what I say they are just speculations 

[sebastianangel]

sorry  deleted my tray and save everything!
On the other hand, I think that artificial intelligence would be like a game of chess and in fighting games everything is summed up in rock, paper or scissors ... I don't know about the specific topic of fighting games, I was only fond of double dragon and golden ax and the street figther 2, usually I'm more of the adventures and the role

[shakunetsu]

Thanks sic for the insights

I just posted it to inform that the AI in SF2 wasn't really an complicated AI or anything interesting.

So it doesn't really matter if AI is untouched and stay the same in this hack/mod SF2 nes bootleg.

[gadesx]

Sf2 AI use an easy learning system to try another way to hit you. Used in a lot of fighting games.
2 hits and change the way to fight.
For example:

Ryu vs E. Honda
You use Hadoken - Hit
Hadoken - Hit
Hadoken, he is going to jump the hadoken.
There's only some broken moves that the opponent can't do anything.

[Xander max]

Sebastian, very good images!  Just a detail, the structure in the center of the screen is a bell, in the original it is attached to the ceiling and has a shadow on the ground.  The rest is perfect.

[sebastianangel]

Sebastian, very good images!  Just a detail, the structure in the center of the screen is a bell, in the original it is attached to the ceiling and has a shadow on the ground.  The rest is perfect.

Thank you very much xander max! Very nice detail, I moved some things to add the ceiling

[shakunetsu]

nice update with the new stages

[bogaabogaa]

1. At a graphic level thanks to (tygerbug) he taught us to use nest screem tool. We achieved great graphic changes and title screen editing and new selection of characters, now we are dedicated to the wallpapers of the different scenarios.

2. Our problems focus specifically on very technical situations, for example (users ask us to add a new character or for example when we play versus we always have the same background scenario ... ect ... I thank you very much in advance.

3. Many recommended us to expand to MMC5 but we discarded all that when (TI)
It helped us by expanding the memory and adding new songs and then the rom was no longer viable to play in a real nes, it only worked in emulators there we decided to adjust to all the limitations!

4. These are some of our advances, there is still a lot to do, what draws our attention is that it has the best game engine seen in any nes game, it is far ahead of any official game!

1. I am glad that the NES screen tool works out well for this game and you can edit most of the levels tile mapping well with it. I appreciate your efforts to this bootlag and the effort to show how things can be done on a limited platform like the NES.

2. When people ask about adding new characters does not mean you need to follow the idea. Adding a character means to expand on many locations of the game what requires a large amount of code and data. How should this be archived when you fail on a easy task like expanding the ROM what only requires some understanding of the hardware. There is no documented source code so you need to spend a lot of time just revers-engineering the game. Changing the location used in two player mode should be a small change I can look into that. It might give you a idea how to go about to learn to code for this platform.

3. Why would you recommend to expand to MMC5 and what are the benefits? The current game is designed around the sprite animation. In order to have access to the needed amount of graphics the game uses CHR bank switching for Sprite banks. The MMC5 could give you a dictional 4kb sprite page. Sure this might help if the goal would be bigger sprites but this is a huge amount of work. The coding of the new bank swapping would be ok but then you need to redo every sprite assembly and adjust all the pointers to them along expanding for larger data tables. May be the current allocation of the code does not allow it and would require extra work to allocate the data to different new added banks. Beside from that change I don't see any benefit in a MMC5 for this project.

Ti is a great member of this community who did make many tools so people with less romhacking knowledge can change many things in there favorite games. You did not even share his work on this project page (donno why the expansion did not work out for you). He even added new music? I do hope you tread members like him with huge respect.

4. What do you mean with best game engine? Do you mean like the Nesmake engine that makes everything very modular? I bet you did look at the PPU sprite page and been amazed about CHR swapping for sprites. There are some games that do that. Sure engine can be used modular but I would like to hear how you describe a gameengine?

I did tell you that I do not have interest in the project. I also lost interest to explain hardware to graphic designer with no interest to learn how to code themselves. You can even find conversation where this escalates in one of my resend posts. I am no genius and struggle to learn things myself. I do explain things wrong and make mistakes but I keep learning and I do have fun with my projects. When you are frustrated there is not much I can do about. This has more to do with how you go about things as a person.
Is this because you tried to code or could not add the things you like? It is easy to get frustrated about that.. for example I did spend the last 4 evenings figuring out how to code a small change in a game and I get closer to finally solve how to do it properly. This is also a project for a game I don't even like to play.. I have many projects I like to go after. I did tell you that I have not interest in this project. Knowing the arcade and SNES ports it is hard for me to find any joy in this game. I do not wish further notifications about it.

[sebastianangel]

I am totally grateful for the help provided by (TI) when nobody had confidence in this project or in me and they gave me their help and I am sure that I would do it again, perhaps speaking another language give other interpretations of what I said, thank you For everything not I will make more comments because it seems unnecessary.
thank you very much 

[joanilson41]

It's looking amazing guys! Good job team!
There's no alpha version for us just to check the progress?
Like a rom hack lover, I'm very excited with this project!

[bogaabogaa]

Here is the promised patch for level select. Works for 1 and 2 player. Once you press select you can choose levels yourself. Else it will run the selection based on the opponent as usual. This is based on the ROM you shared with me.
Probably named as Street Fighter III (9 Fighter) (Unl).nes

File SHA-1: 2916D55BE157858EB377B4A22F070763D9B07AC1
File CRC32: 22CDD2F6
ROM SHA-1: 13FF5A62AFE5EE7B4EA5D537EE92BFA253EB98F6
ROM CRC32: A09AA82C

I provided a IPS and also the source with general info. It might be helpful to anyone who likes to start getting the hand dirty with code.

https://www.dropbox.com/s/ivov46z65hynovh/Streetfigher3LevelSelect.zip?dl=0

The research required and anything else it was a 5h task for me. I am aware that it might have bugs or might even crash. I did not have any as I tested but given how the game is build and how I had to implement the changes many things are possible.

Update: Do you think I been technical so far? This are general suggestions why to go for things. There are some technical info's in the readme. Goodbye and good luck. If you keep beeing frustrated since you don't like to make progress with learning technical stuff. As a artist you can go to many sprite resource pages and match yourself with other artist there. Way less frustrating as doing hackerinos with some old ROMz.

Sure I hope for you to find some programmer dude jump along so the project can be finished in a good state. Been reading up on some stuff and thinking about the task involved making the port (but it is technical..). If you find anyone interested to do the change and likes to talk a bit technical about it I don't mind contacting me. I do have general interest to bring people ahead and learn stuff.

[RealGaea]

Here is the promised patch for level select. Works for 1 and 2 player. Once you press select you can choose levels yourself. Else it will run the selection based on the opponent as usual. This is based on the ROM you shared with me.
Probably named as Street Fighter III (9 Fighter) (Unl).nes

File SHA-1: 2916D55BE157858EB377B4A22F070763D9B07AC1
File CRC32: 22CDD2F6
ROM SHA-1: 13FF5A62AFE5EE7B4EA5D537EE92BFA253EB98F6
ROM CRC32: A09AA82C

I provided a IPS and also the source with general info. It might be helpful to anyone who likes to start getting the hand dirty with code.

https://www.dropbox.com/s/ivov46z65hynovh/Streetfigher3LevelSelect.zip?dl=0

The research required and anything else it was a 5h task for me. I am aware that it might have bugs or might even crash. I did not have any as I tested but given how the game is build and how I had to implement the changes many things are possible.

Update: Do you think I been technical so far? This are general suggestions why to go for things. There are some technical info's in the readme. Goodbye and good luck. If you keep beeing frustrated since you don't like to make progress with learning technical stuff. As a artist you can go to many sprite resource pages and match yourself with other artist there. Way less frustrating as doing hackerinos with some old ROMz.

Sure I hope for you to find some programmer dude jump along so the project can be finished in a good state. Been reading up on some stuff and thinking about the task involved making the port (but it is technical..). If you find anyone interested to do the change and likes to talk a bit technical about it I don't mind contacting me. I do have general interest to bring people ahead and learn stuff.

Thank you very much for this patch.

[sebastianangel]

Here is the promised patch for level select. Works for 1 and 2 player. Once you press select you can choose levels yourself. Else it will run the selection based on the opponent as usual. This is based on the ROM you shared with me.
Probably named as Street Fighter III (9 Fighter) (Unl).nes

File SHA-1: 2916D55BE157858EB377B4A22F070763D9B07AC1
File CRC32: 22CDD2F6
ROM SHA-1: 13FF5A62AFE5EE7B4EA5D537EE92BFA253EB98F6
ROM CRC32: A09AA82C

I provided a IPS and also the source with general info. It might be helpful to anyone who likes to start getting the hand dirty with code.

https://www.dropbox.com/s/ivov46z65hynovh/Streetfigher3LevelSelect.zip?dl=0

The research required and anything else it was a 5h task for me. I am aware that it might have bugs or might even crash. I did not have any as I tested but given how the game is build and how I had to implement the changes many things are possible.

Update: Do you think I been technical so far? This are general suggestions why to go for things. There are some technical info's in the readme. Goodbye and good luck. If you keep beeing frustrated since you don't like to make progress with learning technical stuff. As a artist you can go to many sprite resource pages and match yourself with other artist there. Way less frustrating as doing hackerinos with some old ROMz.

Sure I hope for you to find some programmer dude jump along so the project can be finished in a good state. Been reading up on some stuff and thinking about the task involved making the port (but it is technical..). If you find anyone interested to do the change and likes to talk a bit technical about it I don't mind contacting me. I do have general interest to bring people ahead and learn stuff.

fantastic !!! thank you so much for the great work

[bogaabogaa]

You are welcome. When it works for your hack I could imagine that the visual appearance is something you don't like. This is also the part where I am the most worried it will break something. Overall I did ignore how the game handles sprites and hardcoded my stuff in after the game was done doing the sprite stuff for that screen.
Not done in a way where you just change some data table to make the sprite bigger but would be a easy thing to do in the code.. Still the sprite is metadata and I only process Xpos when selecting a different stage.

I did think it would be best to hack the plain routine to be triggered earlier and use that one to cruse around the map to indicate the stage. But this takes time to revers engineer. If you don't find any crash might be easier to code the plain logic yourself or change it to something that looks visual pleasing .