My completely homemade arcade machine

[hypernova]

Looking good.

One suggestion.  Stick with one or two colors (one for each player) for your buttons.  The rainbow effect always ends up looking gaudy.

First was the hard drive and now the tv set
I am very unlucky 

It was working fine until now

Better get used to it!  My arcade's been down for over a year, because I'm too cheap to diagnose the offending part.  (HDD or Mobo)

[arximidis]

Looking good.

One suggestion.  Stick with one or two colors (one for each player) for your buttons.  The rainbow effect always ends up looking gaudy.

First was the hard drive and now the tv set
I am very unlucky 

It was working fine until now

Better get used to it!  My arcade's been down for over a year, because I'm too cheap to diagnose the offending part.  (HDD or Mobo)

I agree. I don't like it either. But I already bought the buttons

I don't mind if something electronic fails. I can fix it. After all the television was problematic from the beggining (that's why I got it for free)
The problem is that the ferrite of the transformer is broken (I don't know how. The pcb was well placed and it was working). I glue it back together, but it doesn't work (probably the magnetic field is not circulating correctly)

[Unstupid]

I agree. I don't like it either. But I already bought the buttons

Just call it a Neo Geo inspired button configuration!   

[arximidis]

I have some good and some bad news

The bad news is that I did fix the television but I broke it again


The good new is that I got for free a new television (SONY) that it was broken, but I fixed it

Now, the only problem is that the monitor is Trinitron and I must change the frame a lititle bit (and it's position). So what I have to do is unscrew the frame and place it to the new position

[newmanfamilyvlogs]

does it have those annoying lines where the shadowmask is held in place with wires in the middle of the picture?
i had a trinitron crt in a cab at one time and it was rather distracting.

[arximidis]

does it have those annoying lines where the shadowmask is held in place with wires in the middle of the picture?
i had a trinitron crt in a cab at one time and it was rather distracting.

I know that all SONY trinitron monitors do have these lines (I have some in my lab, and I have already seen these lines)

However, this is a television CRT tube. I don't know if they are exactly the same technology
I tested it with the tuner and I didn't notice any lines.
Maybe they are there... but unnoticable.

I haven't tested the monitor with the computer connected to the s-video input.
When I do, I will look closer...

[arximidis]

And this is the result after the modification
(ok I need to re-paint it)

[arximidis]

[dextercf]

use a roller dude!

[arximidis]

use a roller dude!

This is not paint!

I did use a roller for the painting

[Unstupid]

Would probably be easier if you paint before you install your buttons, joysticks and speakers.   Are your pictures coming out mirrored or are your buttons on the left side of the joysticks?

[arximidis]

Would probably be easier if you paint before you install your buttons, joysticks and speakers.   Are your pictures coming out mirrored or are your buttons on the left side of the joysticks?

Yes, I did remove the buttons for the paint job (This is not paint what you see in the picture)


If you are playing alone, you can change the "buttons - joystick" orientation on the fly (and by that, I mean without terminating the game).
Pause the game, select the new orientation, unpause and play

The only reason that the buttons are on the left side is because that was exactly the "orientation" on the original cab

[dextercf]

Ah, so it's a primer of some sort?

[arximidis]

Ah, so it's a primer of some sort?

It is Alkyd-based undercoat to be more exact
It's for wooden surfaces and walls


Now, I am doing my tests... Of course the game controller didn't work at first (after all, I hadn't tested in "full scale")
After some bug fixes, I managed to make it work with the frontend and nestopia
It's not working with mame yet (there are still some bugs left)
I know what is wrong, but I can't understand why it is not working. I thought I had fixed it. However, I did a lot of changes to the code, from the time of the first succesfull test with mame (you can see the succesfull test with mame in one of the videos in the first post of this thread)


In the picture you see the installation of the operating system

[SammyWI]

how is that gun sensor going, would be cool to see if this could be an alternative to the current solutions from aimtrak and ems?

Very interested in this myself.  I have an Aimtrak setup that I like, but it has it's limitations.  The requirement to calibrate for each shooter and inaccuracy if the shooter moves drives me kinda nuts  Especially since one of my other hobbies is shooting. 

I'm thinking about going to a laser + webcam tracking system for a front projection shooting setup.  From what I can find, commercial video training systems for the military and LEOs use this approach.  Seems like people have used this for basic mouse tracking in DIY setups like this:

http://channel9.msdn.com/coding4fun/articles/Webcam-Based-Laser-Tracking-for-Human-Computer-Interaction

or using software like this:

http://www.roborealm.com/

I'd love to see what you come up with.  Will you be making your tracking software available under GPL, etc?

[arximidis]

how is that gun sensor going, would be cool to see if this could be an alternative to the current solutions from aimtrak and ems?

Very interested in this myself.  I have an Aimtrak setup that I like, but it has it's limitations.  The requirement to calibrate for each shooter and inaccuracy if the shooter moves drives me kinda nuts  Especially since one of my other hobbies is shooting. 

I'm thinking about going to a laser + webcam tracking system for a front projection shooting setup.  From what I can find, commercial video training systems for the military and LEOs use this approach.  Seems like people have used this for basic mouse tracking in DIY setups like this:

http://channel9.msdn.com/coding4fun/articles/Webcam-Based-Laser-Tracking-for-Human-Computer-Interaction

or using software like this:

http://www.roborealm.com/

I'd love to see what you come up with.  Will you be making your tracking software available under GPL, etc?

Well my first thought was a Laser + webcam system (I already wrote the code. The fellow of the first link had the same idea with me ).
But after a better thinking and some first results from simple tests, I don't think that it will work (at least not quite well).
If you carefully see the fellow's videos, you will notice that he is not aiming directly on the screen, but on a white surface. This way it's very easy to track the laser with a webcam. If you try to aim on the monitor, then it's a whole another thing. It will be very difficult to track the laser, due to all the collors of the screen (and don't forget that the colors change in a video game. Some colors will be close with the color of the laser, unless you have a very powerfull laser.
Also in order to focus the camera to view the whole monitor, you will be forced to place it far enough from the screen (even if you put it in an angle), meaning outside from the cab (if your cab is like mine). Otherwise on top, but that depends on your cab's design

Another thing is light. All this stuff (even the Aimtrak setup system) work quit well only in a dim enviroment. My cabinet is not in a closed room and the light from the sun is interfering. Reflections is another problem too. Even a system like Johnny Lee's will not work well enough (nottice the "darkness" in his videos) http://johnnylee.net/projects/wii/

Thinking all that, I am trying a whole different approach. I don't have results yet, because I haven't build the electronics yet (however, I already have the nessesary components in my lab)

[SammyWI]

I totally agree that the laser + webcam is unlikely to work with a monitor + cab.  I'm looking at a front projection setup for this approach.  I'm just at the researching point for now, but if the commercial systems work like this, there has to be a way.

Looking forward to see what you are coming up with.

[arximidis]

I remember such a system when I was in the army. We were firing with it. They had connected G3A3 machine guns and it was extremely accurate. I don't know how it was working (probably with lasers), but again I remember a dim environment...
The image was projected on the front wall... with the combination of a dark environment it's not so hard to track a laser beam accurately

[Jammin0]

what about IR lasers?  shouldn't be anything coming out the screen that is nearly as bright as a dot from an IR laser.  Just a thought.  I use the wiimote whiteboard setup all the time and have no problems with line of sight tracking.

[newmanfamilyvlogs]

what about IR lasers?  shouldn't be anything coming out the screen that is nearly as bright as a dot from an IR laser.  Just a thought.  I use the wiimote whiteboard setup all the time and have no problems with line of sight tracking.

hell no he doesn't want to use an IR laser! What the hell are you thinking?

[SammyWI]

what about IR lasers?  shouldn't be anything coming out the screen that is nearly as bright as a dot from an IR laser.  Just a thought.  I use the wiimote whiteboard setup all the time and have no problems with line of sight tracking.

hell no he doesn't want to use an IR laser! What the hell are you thinking?

Hmmmm.  No smiley face.  Not sure if serious.  

In case you are serious, an IR laser may be just the thing.  I believe commercial setups often use them and a web cam will often see IR just fine or can be modified to.  Wouldn't want to leave kids unsupervised with an IR laser system (too easy to burn retinas).  But then, that could happen even with visible lasers.  

[newmanfamilyvlogs]

IR lasers are bad news. Not to mention you'd be shining them onto a glass surface, where reflections, that you can't see, are going to be playing all over the room.

You can get Class I IR Lasers (called "eye safe") for things like night-vision gun-sights, but they are expensive, upwards of $900-1200USD. I'm sure you can source the modules themselves for cheaper, but I've never looked.

Don't look at IR lasers with your remaining eye.

[arximidis]

If you watch the video I have uploaded (showing the software)... somewhere I write that probably I will use infrared light and an ir webcam (meaning a normal webcam hacked to see in infrared http://users.ntua.gr/dpiperid/MyWebPage/Contructions/TVSignal/irwebcamEN.htm)
However, I never thought using an IR laser (I don't have one, and I don't think that the lasers extracted from cd burners or normal cd roms would work)
My thought was in IR LEDs (excactly like the aimtrak system)

To tell you the truth, I didn't tried it yet (I have only tried the laser setup. I did some tests with it and it didn't work as I would liked to)
Probably I will test the infrared set up, but first I need to extract my homemade ir webcam, from my homemade security system (after all is a low budjet webcam and I want to replace it with a better one). I don't want to hack my second webcam...

As I said before, I already thought of another system. I did a first test and things are looking great so far... (at least for the signal transmision part). The only test I have, is if I can succesfully track the direction of the signal and with what accurancy (In fact it's possible to track signal's direction, but I don't no the accurancy yet)

[SammyWI]

cotmm68030 - We will have to agree to disagree for now.  arximidis is not using an IR laser, so I will not drag his thread off topic further.  Another thread and time.

arximidis - Pretty cool that you are making as many parts as you can yourself on this build.  Please keep us updated!

[arximidis]

Still, I am not saying it's a bad idea. The lasers from cd roms etc I think it's a bad idea. If you already have an IR laser, maybe it's worth the try

It is possible for me to end up with a combination system, meaning a system that combines both IR light and sound.
But first I want to try the sound system I am working on... The good thing is that you can measure the distance of the gun from the monitor (which is an importand information). This info may save you from recalibrating


Ah, I forgot to tell you that I am planning to make the gun wireless!

[arximidis]

Update

Well, here what is going on...

I am having a problem with some games (like newer versions of MAME) not accepting virtual keystrokes. Fisrt I thought it was a problem with directX input. But it's not completely true. You can send a keystroke (with a special API function of windows, called SendInput -> This is the function I am using) only to the foreground process.
I am guessing some games handle direct input, with a different background process and that way the game seems not accepting the strokes.
Now, I can think three ways bypassing this unexpected problem

a) Use an older version of MAME (like mame0117). Everything is working as it is.
b) Leave the hardware of the game controller as it is... and write a special driver (a fake keyboard driver). Well this is not easy... unless you can find such a driver for free (I searched with no luck)
c) Change everything... meaning change the hardware and the necessary software... It's a lot of work... but I think it's easier than option b).

Well, I selected option c) and I wrote the NEW basic code and it works. What I am doing is creating a hardware that makes the PC "think" it has an AT keyboard connected (exactly like ipac or other controllers do)...
Of course it's not a simple keyboard. It has the ability changing the keystrokes combination by the frontend.

The only problem is that all the code I wrote for the mapper... it's useless now

[drventure]

I believe there's an API function to retrieve the "foreground" window, whatever process it belongs to, so that you can always be sending keys to the right place.

Regardless, sounds like you got something working.

Just curious, though. Mame already had pretty good keyboard remapping capabilities. Was there a specific reason for not using that? or maybe you were looking for something more generic to work across a bunch of emulators?

[arximidis]

I believe there's an API function to retrieve the "foreground" window, whatever process it belongs to, so that you can always be sending keys to the right place.

Regardless, sounds like you got something working.

Just curious, though. Mame already had pretty good keyboard remapping capabilities. Was there a specific reason for not using that? or maybe you were looking for something more generic to work across a bunch of emulators?

Yes, I already use this function in the code... Still, it doesn't work.
But it does work with other programs - games (e.g. nestopia). With Pro evolution soccer...no. Same thing with most modern games. It doesn't work with MAME too (it works only with version up to 0117).

I can explain it only like this:
You can send keystrokes only to the foreground window... Ok, this is what SendInput does (you may need to bring the application to foreground, if it's not).
SendInput works with virtual keycodes (e.g VK_RIGHT, which is the right arrow key and so on). Direct Input works in low level. So SendInput. Why it's not working then?
I wrote a small program, that used direct input, and then I tried to send keystrokes with SendInput, from another application. It did work!!!
However it's not working with games. It doesn't work even when you change the VK_ constants, with direct input DIK constants!
I am guessing, because direct input works at low level, it can scan for keystrokes from a different (and background) process (like windows hooks can do). Because that background proccess has no window handle, it's impossible to bring it at foreground (foreground has no meaning). SendInput is sending the keystrokes to the foreground window (in our case the game you are playing) successfully... but there is no code there!!!

I know that mame has remapping capabilities. I am trying to create a more generic controller (like you said) and I don't want to use commercial controllers nor thirdparty remappers (freeware or shareware) and frontends. It's all about the "completely homemade" title of the thread

PS: I found out what to do with the old code... I will publish it for free (I mean, open it). After all, it's a working piece of code (with some limitations regarding the games only)

[bkenobi]

This is in no way intended to cut your project down, but I don't understand where you are drawing the line on "completely homemade".  You are certainly making the controller interface and interface software essentially from components, but you are using commercial joysticks, buttons, monitor, coin slot, etc.  I think what you are doing is awesome, but if you use MAME and Windows, then why is using a function that someone else bad?  Perhaps it just comes down to "I did everything I am capable of doing myself"?

Anyway, it's coming along really nice.  Good job!

[arximidis]

This is in no way intended to cut your project down, but I don't understand where you are drawing the line on "completely homemade".  You are certainly making the controller interface and interface software essentially from components, but you are using commercial joysticks, buttons, monitor, coin slot, etc.  I think what you are doing is awesome, but if you use MAME and Windows, then why is using a function that someone else bad?  Perhaps it just comes down to "I did everything I am capable of doing myself"?

Anyway, it's coming along really nice.  Good job!

Hello, bkenobi and thank you for your nice words

I think you misunderstood the title of the thread and it's not your fault. I have a difficulty on expressing and writting my thoughts in English and I apologise for that

Every homemade machine I have seen here or over the internet, are completely homemade too. I have seen implementations that are simply fantastic. I could never build such machines, even if I was working for years.

Joysticks, buttons, monitors, computers etc, are just components for me (at least for this project).
You connect the componets all together, but you still need something to control them. You need a controller. You can buy one, or you can make one.
I decided to build one myself, not because most people are buying them and not because I wanted to make a difference. I am building one myself, because this is what I am interested in (and saving money of course)

To answer your question the "completely homemade" title mostly goes to the controller (meaning the system that puts them all together, including the gun).
So, for me it just comes down to "I did everything I am interested in doing myself" (hardware and software)

At the end, it's not going to be better than the commercial ones, but it will work and that's enough for me!
After all it's just a suggestion (among others).

[newmanfamilyvlogs]

I think the difficulty people are understanding your intention, and the difficulty you're having in expressing it, comes down to expectations in the way things are phrased.

By titling this thread "my completely homemade arcade machine", I believe this is being interpreted by native English speakers to mean "completely" in a very literal sense. By choosing the phrase "completely homemade", an English reader understand this to mean that every single component should be manufactured by you. Likewise there is an understanding by English natives that by choosing this phrase as part of the title of your thread, you are emphasizing this fact as a defining feature of your build.

The problem with phrasing this in this way is that it gives the impression that you are attempting to call attention specifically to the home making of components that others would simply buy. And while you are attempting to build some components that most people simply purchase, there are still those components that you are not home making, such as the operating system, the emulator itself, the buttons, the sticks. By emphasizing the homemade nature of your cab, an english native probably assumes some degree of 'comparison' between your cab and any other cab here on the forums. From that view point, ANY cab on this site, even those made from a kit, is still a homemade cab, and thus to call yours 'completely homemade' while simply buying components like the buttons and sticks comes off as dishonest, even if that is not your intention.

Translation is hard.

[arximidis]

I think the difficulty people are understanding your intention, and the difficulty you're having in expressing it, comes down to expectations in the way things are phrased.

By titling this thread "my completely homemade arcade machine", I believe this is being interpreted by native English speakers to mean "completely" in a very literal sense. By choosing the phrase "completely homemade", an English reader understand this to mean that every single component should be manufactured by you. Likewise there is an understanding by English natives that by choosing this phrase as part of the title of your thread, you are emphasizing this fact as a defining feature of your build.

The problem with phrasing this in this way is that it gives the impression that you are attempting to call attention specifically to the home making of components that others would simply buy. And while you are attempting to build some components that most people simply purchase, there are still those components that you are not home making, such as the operating system, the emulator itself, the buttons, the sticks. By emphasizing the homemade nature of your cab, an english native probably assumes some degree of 'comparison' between your cab and any other cab here on the forums. From that view point, ANY cab on this site, even those made from a kit, is still a homemade cab, and thus to call yours 'completely homemade' while simply buying components like the buttons and sticks comes off as dishonest, even if that is not your intention.

Translation is hard.

I agree...

Because my implementation has something different, I wanted samehow to write it in the title
It's about a differnet approach (at least for some things). You can understand that it would be difficult to discribe them in a title like this:
"My homemade arcade cab,  with a homemage game controller and a homemade frontend (but all other components, programs, games etc, are not homemage)"
And it has some other homemade stuff too (like the tilt sensors and the gun)

Either way, I can not change the title

However, the moderators can change it to "my homemade game machine" or to "just a homemade machine".
I don't have any problem.

[bkenobi]

No need to change it.  As cotmm68030 said above, the problem is in translation.  Even as a native English speaker, I might choose a similar title for my build thread if I were to post a similar project.  I was simply trying to understand what the goal of the project was since the title and implementation were not exactly congruous.

 

[arximidis]

I can understand the confusion. And this is not only for native English speakers. Same thing in other languages too

Even when you say "My homemade machine" wou will encouter the same problems. What will be homemade. If your buttons, joystics are not homemade, then why you said the machine homemade...

Think a battery charger as a DIY project. You buy a ready chip for the job (see MAXIM). You build the pcb and you have a homemade charger (you didn't purchase a charger, you build it yourself)
Another fellow is doing the same thing, but instead of buying the chip, he is creating a circuit from other electronic componets. He is not creating all the other necessary parts, like the first fellow.
They both end up with a homemade charger.
The question is what's the difference? (if any)

[bkenobi]

Yup.  Noone expects someone building a "completely homemade cabinet" to develop and manufacture every IC in the thing.  Like I said, the only confusion was where the line was being drawn.  And you answered it nicely by saying that you are building everything that you have an interest in building.  Good enough for me!   

[Unstupid]

Yup.  Noone expects someone building a "completely homemade cabinet" to develop and manufacture every IC in the thing. 

But if someone actually did it, that would be awesome...  Even "god-like"!   

[arximidis]

Yup.  Noone expects someone building a "completely homemade cabinet" to develop and manufacture every IC in the thing.

But if someone actually did it, that would be awesome...  Even "god-like"!   

No one does that. Even the original arcade games were not completely homemade. Yes they created the cabins, the circuit boards and the code of the games, but they didn't build the buttons, the joystics, the coin acceptors nor the processors. Everybody works in this way. It's stupid to rediscover the wheel.

But it's not stupid trying to build a different controller or a homemade coin acceptor. The only differnece is that a coin acceptor can be a standalone project. We are not using coin acceptros only in game machines. Same thing with buttons and joystics. Buttons, are buttons. I can use them everywhere. Joystics too. For example, I can use a Joystic to move a robot (or a remote control vehicle)
However the controller is specified only for the game machine. There is no meaning using it somewhere else. Same thing with e.g. i-pac. It's only for game machines. But it's not homemade...

This was what I was thinking, when I wrote the word "completely" in the title of the thread.
I was refering to the hardware, that it can be used for the game machine only. And exactly this is what I am intersted in.

I wrote the necessary software, only because of the hardware. To make it work, with the other systems. Think of the code something like a driver for windows.
On the other hand, I wrote the frontend only because the commercial or freeware frontends, will not cooparate with my hardware (I didn't said that in previous posts, but that was the main reason I wrote the FrontEnd).
I would prefered to wrote only the code of the keyboard mapper, which is the only software that communicates with the hardware directly. Again, I had no luck

[arximidis]

Question

Hello again!
I have a simple question

Is there a way making a frontend (like hyperspin), executing an external program (taking command arquments) just before a game is loaded and just after the game terminates?

[newmanfamilyvlogs]

Mentioned here:
http://www.hyperspin-fe.com/forum/showthread.php?t=9010

The "Hyperlaunch" AutoHotKey script gets passed variables, and from there you can script it to do basically anything pre and post launch. It's a pretty modular platform.

[arximidis]

Mentioned here:
http://www.hyperspin-fe.com/forum/showthread.php?t=9010

The "Hyperlaunch" AutoHotKey script gets passed variables, and from there you can script it to do basically anything pre and post launch. It's a pretty modular platform.

Thanks for the reply cotmm68030

Do you mean it can do that without Autohotkey program present? I am asking because I don't want to run one million programs (so to speak) just to load a game.
I want to press one key and hyperspin to load the selected game... before that I want it to load another program (and hyperspin pass the name of the  game and emulator)
Same thing when game terminates

I tested hyperspin today and I have some other questions too

Why CPU load is going over 80% for a pentium 4 3GHz computer ? (due to that fact, video is not playing correctly... and frames are lost)
What if I setup hyperspin on an older computer?

Why AVG finds a trojan in hyperHQ?