Star Wars Trilogy

[gubbin]

5 minutes ago, slippifishi said:

With the greatest respect to jedimastermatt, LCM is probably a good way to solve this, but I personally think you should focus on color mask and replace sequences only for now (and palette switches, if you are using them) - understanding those two types correctly is key to fully exploiting the other scene types. Your not understanding why Yoda's hand is breaking the display suggests to me the penny of hashes and d-masks hasn't quite dropped yet, but you are really close man, stick at it!

 

Here is a post I made that discusses "replace sequence" vs "replace". Neither of those types is suitable for the Yoda scene, but I think it does give some good examples of how hashes work and how you need to think about triggering them when they change, maybe. At the very least it's got some diagrams!

 

 

Absolutely d-masks can have moving content, before long you will be cursing the fact that you made your d-mask that size because there is some movement you didn't realise  They can have dynamic content too, if you want, and sometimes the scene may even require it.

 

 

Your assessment on the penny dropping is 100% accurate.  I really appreciate everyone's patience here with helping me learn this 

[slippifishi]

Let's try this: do you know what a checksum is? Or a CRC? We keep talking about "hashes" but we could also call them Checksums or CRCs - they are a unique number that represent the EXACT state of ALL the dots in the display at any given moment in time. When you create a d-mask you are selecting an area to include/exclude in that CRC calculation.

 

I'm over simplifying here for the sake of discussion, but - every single frame the pin2dmd displays from the game - EVERY SINGLE ONE - will cause the PIN2DMD to calculate a hash/checksum/CRC for the full screen, and for all the d-masks you have defined, using the dots it receives from the DMD controller in the machine. It then looks through all your keyframes and sees if there is a matching hash. If it finds one, it will play that sequence, or display the frame associated to that keyframe, or switch to the palette you selected; if not, then it won't do anything - if there was a sequence triggered on the previous frame, then it will be allowed to continue playing, or if it was a single frame with a delay set, then it will let the delay "play out" and then revert back to the default palette. Then the next frame displays and the whole process starts again - EVERY SINGLE FRAME. * 

 

So if the hash never changes and there is a keyframe associated to that hash, it will keep retriggering the same sequence from frame 1 - it won't let the sequence "play" until the hash detected for the next frame of content doesn't have a matching keyframe.

 

I could probably do a better discussion of this but I want to draw some pictures to help describe it but it is a bit late so might save it for tomorrow or the weekend if you don't have any joy by then. One last note for today if you are trying to fix this, make sure you delete your previous triggers relating to this scene before trying new d-masks or anything - it would be very easy to do everything right here, but your earlier d-mask might keep triggering and covering up your success.

 

* The rules change in other modes, but again, understanding Replace and Color Mask sequence types is key to making good use of those modes.

 

[jedimastermatt]

2 hours ago, slippifishi said:

With the greatest respect to jedimastermatt, LCM is probably a good way to solve this, but I personally think you should focus on color mask and replace sequences only for now (and palette switches, if you are using them) - understanding those two types correctly is key to fully exploiting the other scene types. Your not understanding why Yoda's hand is breaking the display suggests to me the penny of hashes and d-masks hasn't quite dropped yet, but you are really close man, stick at it!

 

Here is a post I made that discusses "replace sequence" vs "replace". Neither of those types is suitable for the Yoda scene, but I think it does give some good examples of how hashes work and how you need to think about triggering them when they change, maybe. At the very least it's got some diagrams!

 

 

Absolutely d-masks can have moving content, before long you will be cursing the fact that you made your d-mask that size because there is some movement you didn't realise  They can have dynamic content too, if you want, and sometimes the scene may even require it.

 

No worries on that suggestion.  LCM is the deep end of the pool; but, there is a way you can use it without getting into the palette priority stuff and, at least for me, helped figure out really complex scenes without creating many masks.  
 

The way I thought of it is like this (using the Yoda scene above as the example):

 

- Think of the animations as a house with many, many rooms.  
- You initially chop up animations into these rooms.  
- The Yoda scene would be one of those rooms.  
- Now, when you build the house, you could build a room for Yoda and Yoda with flashing targets, and Yoda with Jackpot, and Yoda with hand, and Yoda  with hand and and jackpot, etc.  and then build doors from the main hallway to each of these individual rooms.  
- the doors are the d-masks or key frames you make.  
- doing this can make it a challenging and time consuming trying to figure out how many “doors” you need to make and keep track on if you’ve made a door that overlaps to similar but different rooms. 
 

Once it was explained to me that the LCM will continue to run until it gets an external trigger, e.g. another use of a frame pulling it out of the animation sequence, it for me made it much easier to keep scenes from overlapping with each other.  
 

For Yoda, if his mug on the left side is the only time it’s used, I would put all of the frames that comprise the Yoda suite into the LCM.  
 

Use the LCM’s dmasks to then handle all the key frames and then you can be certain that you don’t leave the room until Yoda has truly left the building. 
 

Using the LCMs color priority for dynamic scenes is much more complex. 

[gubbin]

I understand what a checksum is.  It makes sense about the dots/pixels positions reflects that unique hash.  For some reason I'm struggling to wrap my head around applying it with the Detection mask.

 

This is where Yoda's hand starts to raise and dropping a mask there would also interfere with the score and swapping message.  

 

I think maybe I still have a little covid brain fog and will focus on the easier replace scenes and come back after I've had some time to process this.  I really appreciate all the input. 

 

[jedimastermatt]

1 hour ago, gubbin said:

I understand what a checksum is.  It makes sense about the dots/pixels positions reflects that unique hash.  For some reason I'm struggling to wrap my head around applying it with the Detection mask.

 

This is where Yoda's hand starts to raise and dropping a mask there would also interfere with the score and swapping message.  

 

I think maybe I still have a little covid brain fog and will focus on the easier replace scenes and come back after I've had some time to process this.  I really appreciate all the input. 

 

I think the LCM would be perfect for this.   
 

- dump all the Yoda crap into one scene and make it an LCM If all of the variations have Yoda on the left.  
- once you have the scene formatted as an LCM, you can make as many masks as you need (LCM masks don’t count towards the overall total but do have a total that is really high for that specific LCM).

- I’m assuming Yoda’s hand animates up from the bottom, so make a mask each for the frames you need that cover the motion.  You might be able to use some of the same masks or make individual ones.  
- I don’t know this game well enough, but, I’m assuming that second line of text is also something you need to capture so make that a part as well.   Only the score part would be the part unmasked since it’s  the only true dynamic part.  
- make the mask and set the hashes and voila. 
 

Use Yoda’s face area to trigger the Lcm and the pin let the lcms dmasks do the heavy lifting. 

[gubbin]

6 minutes ago, jedimastermatt said:

I think the LCM would be perfect for this.   
 

- dump all the Yoda crap into one scene and make it an LCM If all of the variations have Yoda on the left.  
- once you have the scene formatted as an LCM, you can make as many masks as you need (LCM masks don’t count towards the overall total but do have a total that is really high for that specific LCM).

- I’m assuming Yoda’s hand animates up from the bottom, so make a mask each for the frames you need that cover the motion.  You might be able to use some of the same masks or make individual ones.  
- I don’t know this game well enough, but, I’m assuming that second line of text is also something you need to capture so make that a part as well.   Only the score part would be the part unmasked since it’s  the only true dynamic part.  
- make the mask and set the hashes and voila. 
 

Use Yoda’s face area to trigger the Lcm and the pin let the lcms dmasks do the heavy lifting. 

Would I need to catch every possibility of FORCE lit up?  for instance.. FORC/  FOR/ FO / F/ ORC/ F  C... etc?

[jedimastermatt]

12 minutes ago, gubbin said:

Would I need to catch every possibility of FORCE lit up?  for instance.. FORC/  FOR/ FO / F/ ORC/ F  C... etc?

You shouldn't need to if they are in the same places at all times and their apperance doesn't need to be used to trigger another action.  

 

If they don't, then leave them out of the mask; but, make sure you colorize the space they occupy with the way you want it to appear when they pop in.  

 

Think of the mask as an overlay that is already there and all you need to do is get the right overlay pulled up at the right time.  

 

Also, while one of the powers of the LCM is being able to use it as a container for many masks - you can also have multiple key frames working as multiple doors to jump into the LCM.   So if the game sometimes launches into the Yoda sequence and his hand is already up or moving up - you can have masks at the root level pointing to the same LCM.  

 

Once I figured out this stuff with the LCM, I end up using LCM or LRM for virtually everything that has multiple frames of animation.

 

Oh, and I don't know if it's still a "bug"; but, I've found the editor really doesn't like switching modes once you've set up a few masks in the mode.  If you need to do that, I usually have to reinsert the frames to a new scene and then copy anything colored over to it.  The symptom of when this happens is you have things you know should be working; but, they aren't getting called up properly in the layered modes.  

[gubbin]

This came today so that's something 

[gubbin]

Haven't made a ton of progress with the project.   I had something strange going on with some of the scenes where there appeared to be boxes around some of the different colored objects.  turned out, I Chose a black early on that wasn't actually #000000 black and it was causing "gray boxes" 🤦‍♂️  Took a bit to fix several palettes but lesson learned!

 

I think I'm pretty close to having all the basic replace scenes done and will need to spend some time on masks scenes.  I have a lot of those working as well.  

 

Meanwhile, here's one I thought was pretty cool.

 

[gubbin]

Looking for some advice on this scene.  To the right of Chewie.  At first I thought this was a chair back, then maybe 3po.  Another member suggested a woman/tribute to the programmer.  It doesn't look like Leia's hairstyle.  So far, most of these scenes I can find nearly exact instances in the movies.  

 

Sure does look like a woman's face:

 

 

 

Here it is just gray tones:

[lucky1]

I think it is C3PO

[DanLShane]

Yep, C3PO

[gubbin]

We'll try this

 

[MCtrickyD]

@gubbin This looks so promising, I have a real pin of this game as well. If I can help please let me know. Any progress since last year?

[gubbin]

On 3/5/2023 at 9:37 PM, MCtrickyD said:

@gubbin This looks so promising, I have a real pin of this game as well. If I can help please let me know. Any progress since last year?

 

I got a little burnt out on the project and it's currently on the side.  Getting the Keyframing/LCM etc. to work was a little overwhelming.  A lot of the coloring is complete but I'm not sure I can handle the really technical scenes without some assistance.

[Spacegoogie]

Bummer, all the hard work that has been put into this and it goes dark.🥹

[gubbin]

18 hours ago, Spacegoogie said:

Bummer, all the hard work that has been put into this and it goes dark.🥹

 

Apologies for the disappointment.  I really enjoy the coloring aspect of this project but the key framing and getting all the timings to sync up ended up driving me mad.  I've watched Martin's videos maybe 1000 times but for whatever reason it never did fully click for me.  All of my struggles were based on VPX on a laptop.  I can't imagine getting this to work on a real pin. 

[ronald71]

WOW loved those coloring ..... why don't you upload a beta version (latest version please?)  to test for everyboddy. and then someone or many  will help you with the coding stuff.

it is only a Idee...

LOve what i see so far.... and MAY THE FORCE be with you.

 

[Dicentim]

Hi, a very underrated table that deserves a color DMD. Thanks for your work !

[gubbin]

I hope this is the right way to do this.  I've attached the project files here for anyone that might be interested in completing the project.    This should be where I left off with coloring, plus all the capture files from gameplay.    It's been about a year since I made any changes so trying to estimate "completion percentage" is difficult.  Most of the attract mode should be complete.  I didn't do anything with the scoring scenes (It felt like there was a lot going on with the random ships flying around)   A big thanks to @Terranigma for the encouragement and sorry it took so long for me to get these uploaded.   

SWT-Project.zip

[dragon44]

Thank you for your time and your great work mate!

[PIPAllah]

On 10/10/2023 at 4:05 PM, gubbin said:

I hope this is the right way to do this.  I've attached the project files here for anyone that might be interested in completing the project.    This should be where I left off with coloring, plus all the capture files from gameplay.    It's been about a year since I made any changes so trying to estimate "completion percentage" is difficult.  Most of the attract mode should be complete.  I didn't do anything with the scoring scenes (It felt like there was a lot going on with the random ships flying around)   A big thanks to @Terranigma for the encouragement and sorry it took so long for me to get these uploaded.   

SWT-Project.zip 12.61 MB · 33 downloads

Peace,

 

Thanks for the hard work.  I might have to learn this so I can finish off the complete package.  Considering this is my favorite table, I was really hoping someone else would do the color DMD to go along with the table and pup pack.

[Bertone]

Yes I hope this project will be finished soon. Siggis table with the Puppack is so cool. This table needs Altcolor and Altsound.