Ps3 rapid fire studying and understanding project open source

[hyper999]

and no real progress my way waiting for scope but i have the controller and dev board done so i can get straight into coding

@RDC are the COD demos on the playstation store?
Post Merge: September 04, 2010, 03:47:00 AMsmall update: my scope will be leaving hong kong tommorow   

[RDC]

Yes they are both on there.

I'm interested in seeing how well your scope works out. My 50MHz Fluke 97 ScopeMeter isn't the greatest for messing with the PS3 controllers, though it is almost 2 decades old now, and the 60MHz Analog one I'm using is much better for this, but they each have their strong points and I couldn't pick one over the other as the situation really dictates which will work best.

[hyper999]

yh well i read plenty of reviews before i bought it and nobody had a bad word to say about it performance wise and im unlikely to be doing any projects involving particulairly fancy or high frequencey signals so i think it will be ok

[RDC]

New one to test out if anyone likes.

Same wiring as before..

 PINOUT

1 - Vdd
2 - LED Cathode
3 - COM2
4 -
5 - Button (other side goes to GND)
6 -
7 -
8 - Vss

This one has 5 Modes, but only Mode 1 and Mode 2 have R/F on them for right now, which are the same as the 2 previous hex files, Modes 3, 4 and 5 are R/F off at the moment.

Tap the button to cycle thru Modes and the LED blinks to show which one is loaded. First tap turns the LED on and R/F is enabled in Mode 1, tap again and LED blinks twice (off/on/off/on) now it's in Mode 2, and so forth and so on. After Mode 5 is R/F Off and the LED turns off, then it starts all over. You can hold the button down to cycle thru all the Modes until the LED turns off, a little bit faster way of getting the R/F turned off for right now, though not really by much.

The only real issue at the moment is when the controller is turned off the PIC R/F still runs, this is because the point used for Vdd is always on and I haven't coded it up yet to know when the controller is off to turn it off, so you have to manually turn the R/F Off either before or after the controller is turned off or LED4 will just stay on.

[geraldrubalcava]

ok well i hope your scope works great hyper. and i want to see progress with the code! xD lol

RDC i tested your code of the battery/usb fix with led on/off and shots fast.
2N2 shoots fater on cod:mw2
2n3 shoots even faster ! i mean fast! wow! like amazing! lol better then the ebay chips. on mw2
The 5 modes one i still haven't tested. and i haven't tested on cod5 or world at war. lol

i will get back to you tomorrow to give the results in the other games too. lol awesome!

Edit: hey! hyper u stil wanna make a pretty inexpensive diy oscilloscope. you already have a pic programmer right? your halfway there! well if you want check this out. http://www.semifluid.com/?p=9

[hyper999]

yeah i saw that one but looks pretty naff and i didnt have a level shifter handy to test it but my scope should be here tomorow or the day after so i should have the majority of the code done before i start at sixth form on friday

[geraldrubalcava]

all right that is awesome! hope your scope works well i will be also here to test it out. 
Post Merge: September 06, 2010, 06:25:55 PMall right well ive tested RDC code with 5 modes. on mw2 and cod5. both work great! mode one is same as 2n2 and mode 3 is 2n3 i believe? well both work great with cod5 but keep the both because it one is on medium speed and another is fast speed. well that test was great! hum. i has to cycle though five modes to turn it off. well im not sure about he power but when the controller is off does the power to the analogs sticks turn off too? i mean on the newer version of the controller were it has polls. (or 4 pins) if the power to that also cuts off then you may not need to code to turn it off. well there are guns in mw2 who have burst mode. so im not sure if you want to add one of the other modes like that but wow. great work! also. If you press the button for 3 sec it turns of the rf.

well Great! i this is so awesome! well great work! and hyper i still look forward for your work too

[RDC]

Latest one to try out, just let me know what Mode does what best for what games/guns and such.

I'm not sure about the power on the newer controllers, honestly never checked to see if they stay powered on or not, but I will later this week as I have one on the way here that's going to be hacked up, so I'll check it before I lay into it, but the MSU PP 4.05 I'm using does, so I just coded it up now to still be able to cycle thru the Modes so you can get to R/F Off for now even if the controller is turned off first.

No idea what all I'll do with this as far as other modes and such go yet. Since I don't really have any games to test on and only these couple of Demos I can't see at all how it's doing for anything else, and that means more miss than hit and more error in the trials, so it's getting to be no fun at all at that point.

Most of these on this latest one are kinda close together, testing out what works best for what, so don't be surprised if they seem kind of similar speed wise in the same game, but they might work different in another one, but I can't tell that here for right now.

[orko007]

I got around to testing using the comparator method this weekend. for me it is more bad news than good.
Using the comparator I set it to look for every falling edge. But when looking at the common line it never sees a falling edge unless I press a button. so the buttons at rest do not create enough if a falling edge to trigger the interrupt. If I set it up on the R2 line using that as a trigger it works fine. But since I am looking for a solution to use only the common line it looks like the comparator method will not work and leaves only the ADC method.
The ADC method seems to be working very well but the comparator method would allow for a lot cleaner code and more exact changes.

Chris

[hyper999]

ok thanks chris when scope finally arrives im gna try interupt on change then if that fails i will be using adc
Post Merge: September 08, 2010, 12:27:18 PMok got bored this afternoon so wrote and adc code for you to test gerald
pin3 - COM2
pin5 - an led or preferabley your oscope
pin 5 should be low until everytime the r2 pulse falling edge is detected then it will switch high for 5ms

also please upload video if you can

[RDC]

@ hyper999 - Pin 5 stays Lo with that code, I scoped it as well.

[hyper999]

ok take two use exactly the same as the last code

[RDC]

You have output on Pin 5 now (see attached pic) but it doesn't matter if Pin 3 is connected or not, so your code is running regardless of the input. It's also way, way too fast (172Hz) for using an LED to tell if it's working or not. A human eye can't see the 'blinks' at that speed and with your code it just appears to be on all the time, but the scope can tell that's not the case (shown in the pic) hence another reason for needing one before really messing with this mess.

[geraldrubalcava]

i agree with RDC on the first led test i keep getting a constant low. and i dint check without connecting it to com2. so i guess RDC is right about it triggering without the R2 drop. and the thing with the led is true ( my eyes are not what they used to be) lol.

hum well i will test of the 5 mode one. im still testing with the two games and different guns lol so i will get back on you with the full report.

[hyper999]

ok thanks for that just spotted an error in the acquisition delay
so take 3

[SUBPLAY]

Hi guys, this is my first post. I am a reverse engineer for a well known console accessories manufacturer and I have just been reading some of these comments about making a controller rapidfire.

So can someone tell me how far you have got with all of this?

This latest attachment which has been included what does it do?

I am just trying to see where we stand with things before I add my bit to this project

[hyper999]

ok here we go
- orko and chives have both successfully written ps3 rapid fire codes professionally
- RDC has now done it successfully in basic
- I am in the process of doing the same in asm, however am struggling to debug my code as i am still waiting for my oscope to arrive in the mail so my only way to test is through feedback from gerald
-the above code repeatedly polls adc for the initial voltage drop then makes pin5 high for 5ms so that i can identify if i am detecting the falling edge

[RDC]

@ hyper999 - Pretty much the same deal on that one as well. The output has changed some in duration only, but the input makes no change to it at all.

If you can, code up something simple to check and make sure your PIC is configured correctly for the Analog input on pin 3. This isn't in any coding language at all, just what it should be doing for a test, something like..

START:
Get AN value
If AN value is higher than 512 turn LED on
If AN value is lower than 510 turn LED off
goto START

This way we can connect a POT up to the PIC (voltage divider between Vdd and Vss, Wiper to pin3) and see if the ADC and your code are working. Since it will be checking the AN value over and over, with the POT turned past half way up the LED should be on, turned past half down the LED should be off. If this doesn't work then something isn't configured correctly with the TRISIO and such.

[hyper999]

ok have another one to try have tested this on my pickit demo board so i know the adc is working
GP0 - COM2
GP2 - Oscope

[geraldrubalcava]

well my findings on hypers code is the same as RDC. subplay, at the end what matters is if it works well i can conform that in the axis controller on tp11 the RF turns off when controller is off. and in the newer models were the pot has Positive and negative. when controller is off the RF turns off. so coding for that may not be needed.

[hyper999]

ok for some reasone the atattchment didnt upload properly but never mind here it is and once again
GP0 - COM2
GP2 - Oscope

[RDC]

@ hyper999 - You're getting closer now with that one, but it's sporadic, like the timing is off somewhere with the ADC. The acquisition time or your wait time before starting the next ADC conversion, but the output is being controlled by the input now.

[hyper999]

i think now it is simply a question of finding the correct threshold for the adc heres the same again but with a higher threshold

[RDC]

That's a bit better, but something is causing a pretty big pause in there, see first attached pic, that's on USB power. On Battery power it's much different, second attached pic.

[hyper999]

ok im gna try shotening the delay time for how long the pin is high forr down to 3ms
also i now have my scope up and running but ive never used one before so i will continue posting them up here for you and gerald to test aswell as testing myself

ok same as before with shorter delay...

[RDC]

There's still a small glitch in it. I've no idea how your scope looks there, but on the Digital scope here it's a little hard to see it while it's running, and on a screen shot, which is held, you can't really see it at all, but on the Analog scope it's pretty obvious something isn't quite right, but I can't get a screen shot of that as you have to see it running to be able to 'catch' it.

[hyper999]

ive been fiddling and ive managed to get it pretty consistant but something like 1 in 10 pulses from pin 5 are missing and it averages out at 78hz on my scope
Post Merge: September 10, 2010, 06:14:25 PMok im gna call it a night now cos its 2.30 in the morning over here but do you think you could post a video of the glitch perhaps?

[geraldrubalcava]

all right finished testing fully RDCs code. its a pretty similar result overall. some for faster then others. but there isnt much difference in mw2 the speeds dependent of the gun. all that mattered was just the speed programed. But for cod5:waw it did matter the type of gun you would use. But it might be because of the patch the game has for rapidfire. its limit is like 10 sps? or something. but waw game kinda sucks, only ting good about it is National Socialist-zombies. (fuggen awesome game mode) lol. i don't own any other game first person shooting game with R1 so with testing that is as far as it goes. good code! awsome speeds! i should suggest Modder Matt to add this to the list of available codes. Because its awesome! if you need me to test out any other code i would be happy to give it a go.

Now @ hyper, Is com2 connected to pin 3? and were the oscope? sorry. hum i might not be able to help you out much with he o scope part but i can show you if you finalize one code with he rapid fire. ok. well i will still test your code so should i test it with usb or bettery? or is that currently fixed?

one more question... Now we know how The sequence of Right common works. But is t similar to Left common? just asking. and does the adc effect if there are two inputs/outputs? because one of the neet features that would be nice would be dual "trigger" but i guess that's still for the future, lol soon i will post a vid of all the working codes and there reactions if it would help

[RDC]

ive been fiddling and ive managed to get it pretty consistant but something like 1 in 10 pulses from pin 5 are missing and it averages out at 78hz on my scope

ok im gna call it a night now cos its 2.30 in the morning over here but do you think you could post a video of the glitch perhaps?

Sorry man, I can't upload video of it here.

I don't know how you're having the code work exactly, so the 'glitch' could be normal, but if you're making the output trigger from the Falling Edge of COM2 remember the button presses are 500us each and it's 6 buttons, so 3ms in total, but you'll want to wait say another 1ms or so before starting the next ADC conversion or you might catch the /\ button somewhere or the Rising Edge and that will cause a problem. Waiting 4ms from the Falling Edge will make sure that you're back into the 'Hi' part of the signal and the ADC then can wait for the next Falling Edge.

The way I have my R1 Disable code done is as soon as it detects the Falling Edge it waits a bit to skip the R2 button, then sets the output hi for the duration of R1, then waits 3ms to skip the rest of the buttons, then starts all over. It's technically only 2ms for the other 4 buttons, but I have it wait 3ms so it doesn't catch the Rising Edge or /\ button, which would start the next timing run all out of whack and cause issues.


@ geraldrubalcava - The COM1 and COM2 lines work pretty much the same way and are timed the same, just the order of the buttons are different is all. It shouldn't be too difficult to make L1 fire as well and is on my list of things to try out when I mess with this again.

[Modded Matt]

2 cents:

first I want to thank everyone who is a part of this project, everyone is working together in harmony and it is paying off. good job guys!!

secondly, I think the author of each code your developing needs to ask for it to actually be published once the author feels the code is COMPLETE. in order to publish it, you will need an actual install diagram (art team will help if needed, make a post in that section)

Finlay, I want to address the recent WAW comment. cod WAW is a aw some game. the rapid fire, once configured correctly fires much faster than anything allowed in mw2. I know that sounds hard to believe, but its not. mw2 the guns have diff limitations. which affects RF, in WAW the game its self is what is restricting RF...this means for mw2 you need diff codes/speeds for diff guns. but in WAW once you get around the game patch, all RF guns fire faster and they  are much cooler in doing so.

by the way, trearch made the waw and the patch for it. treacrh are also the ones making the new game, therefor its only logical they will implament a similar, if not the same patch WAW has. as a logical person, I would advise working with that info in preparance for the next game.


anyway, good job everybody. I will begin testing next week. please let me know when you guys feel your codes are COMPLETE so they can be added to the official list.

as an after thought, I dont play ps3 much, but what about your face buttons? reading through your work so far, and how they are on a common line, I dont think it would take much alteration to make a mode for the face buttons.. similar to instafire... but thats just my 2 cents