734 replies to this topic

[parabolic]

Sorry to be a pain in the ass about this but im having trouble getting this to work correctly. I have it calibrated and I can see it move the red bars in the windows controller properties. I go into a table using the vp9 accell mod - seems to work ok...but the Twilight zone table the ball just keeps bouncing off the tip of the plunger if its at idle - it wont just sit on it. If I use the physicsmod version it does the same thing but my flippers move WAY slow and the ball can be nudged completely from side to side of the table... 

So my question is - is it the table? is it my calibration? Is it the mods? Is it something else im doing?

Thanks!

Jeff

 - EDIT:

I re calibrated the plunger I think im having light source issues - I havent tried the EL solution yet - I jst want to get it working perfectly THEN make the changes - so I have something to compare it to. The light source I used to calibrate must have been too bright - but now its better but is not as fluent. I need to try to get the perfect light and distance. What color LEDS ? Does it matter? and what size? I have 10mm leds in various colors to try..

Edited by parabolic, 19 August 2014 - 11:40 PM.

[mjr]

Sorry to be a pain in the ass about this but im having trouble getting this to work correctly. I have it calibrated and I can see it move the red bars in the windows controller properties. I go into a table using the vp9 accell mod - seems to work ok...but the Twilight zone table the ball just keeps bouncing off the tip of the plunger if its at idle - it wont just sit on it. If I use the physicsmod version it does the same thing but my flippers move WAY slow and the ball can be nudged completely from side to side of the table... 

So my question is - is it the table? is it my calibration? Is it the mods? Is it something else im doing?

 

If it looks like it's working in the Windows control panel, then your device setup is probably good - the problem is probably the table.  I haven't tried TZ yet with it myself, but it's a good bet that it'll need adjustments, because about 80% of the tables I've tried don't work out of the box.  

 

Try Space Station and Tales of the Arabian Nights - I think both of those worked right off the bat for me.  

 

I'll try to take a look at TZ later tonight or tomorrow and see if I can get it working.  I've tweaked enough tables now that I can usually figure out a new one without too much work.  I tried to outline the procedure I used in the build guide, but I know it's a little vague; it's hard to get it down to a recipe with so much variation from table to table.  Maybe we should figure out a way to share tables we fix, so we can leverage each other's work on that.  Anyone using the VirtuaPin or NanoTech plunger kits would need exactly the same mods, so it's more than just us that would benefit.  It doesn't look like this kind of small fix usually merits an upload here - not sure if it's just that no one is doing that kind of fixing or if it's not kosher to upload such a small change.  (And I know some authors allow mods to be released and others don't, so I'd only do this for tables where mods are allowed.)

 

One more thing - if possible, try my modified version of VP when testing the plunger - the one I posted for the nudge mods (at https://www.dropbox....2HtOshcJWkIe6ha).  That has another change I made specifically for the plunger.  The standard version of VP has a way of handling calibration that I think is problematic, so I changed it in my version to do something more sensible (in my opinion).  The behavior you're seeing with the bouncing at idle could be related to that.

 

The modified version of physmod5 also has that plunger fix.  But physmod5 can only be used with tables specifically designed for it - you can't run the stock TZ on physmod5.  That's why the flippers aren't working.  I think there are only three or four physmod5 tables out there right now.

 

 

I re calibrated the plunger I think im having light source issues - I havent tried the EL solution yet - I jst want to get it working perfectly THEN make the changes - so I have something to compare it to. The light source I used to calibrate must have been too bright - but now its better but is not as fluent. I need to try to get the perfect light and distance. What color LEDS ? Does it matter? and what size? I have 10mm leds in various colors to try..

 

I don't think the color matters, but if it does, I'm using blue - two 5mm, 20mA blue LEDs, at about 8 inches, located roughly in line with the tip of the plunger at rest and directly below, facing up.  More than that seemed too bright.  If it's too bright, you'll probably see no movement on the control panel when you move the plunger.  If it's borderline too bright or too dim, it'll probably be jumpy.  If you think it's too bright, try partially blocking the light with something, like a little electrical tape.

 

What's the luminous intensity rating on your 10mm LEDs?  Mine are 5000mcd at 25 degrees.

 

Also, try to be sure the top/back side of the sensor is covered with something opaque.  It seems that the little printed circuit board is translucent enough that ambient light from the back of the board registers on the pixels.

 

Definitely start with the Windows control panel to judge if things are working, then you can try, say, Space Station.

 

Keep me posted - it sounds like you're really close to getting it working.

Edited by mjr, 20 August 2014 - 12:52 AM.

[slashbot]

Mjr Thanks for doing this. Will have give it a spin.

Can you change the registry location for your physmod build so we can run it easily side by side with the other vp builds without having to export import reg files?

Ive got a .bat file that i use in pinballx to change the registery so u can run it side by side with good nudge x-y settings

Someone intrested?


Sent from my iPad using Tapatalk HD

[RYSr]

 

 Anyone using the VirtuaPin or NanoTech plunger kits would need exactly the same mods, so it's more than just us that would benefit.  It doesn't look like this kind of small fix usually merits an upload here - not sure if it's just that no one is doing that kind of fixing or if it's not kosher to upload such a small change.  (And I know some authors allow mods to be released and others don't, so I'd only do this for tables where mods are allowed.)

 

One more thing - if possible, try my modified version of VP when testing the plunger - the one I posted for the nudge mods (at https://www.dropbox....2HtOshcJWkIe6ha).  That has another change I made specifically for the plunger.  The standard version of VP has a way of handling calibration that I think is problematic, so I changed it in my version to do something more sensible (in my opinion).  The behavior you're seeing with the bouncing at idle could be related to that.

 

The modified version of physmod5 also has that plunger fix.  But physmod5 can only be used with tables specifically designed for it - you can't run the stock TZ on physmod5.  That's why the flippers aren't working.  I think there are only three or four physmod5 tables out there right now.

 

MJ

 

I tried your latest modified VP 99 and Physmod5 on my VirtuaPin ver 2 controller last night. The nudge modification is very realistic and a large improvement over the old code. However I have plunger problems that didn't exist with standard release version of VP (9,9a or physmod5). The plunger was already working correctly with them.

 

I noticed above that you stated you modified the plunger code and it could cause bouncing at idle.

 

What I experienced was that if I used the plunger it would launch the ball with the slightest movement even before you released it, The graphic of the plunger tip on the table then moves up beyond where the ball normally rests. Then when the next ball come up after a drain it would auto-fire without me even touching the plunger.

 

If I only use the plunger button it works fine, but as soon as I make the slightest movement of the physical plunger it auto-fires.

The test table was UW's latest Bride of Pinbot.

 

There were a few tables that worked OK, one of them was the latest release of TOM, I looked at the Script and seen code for a Alpha Ramp impulse plunger.

It seems to work with or without the enable mechanical plunger option checked.

 

The majority of the tables I tried had the autofire problem.

 

However your nudge code combined with physmod 5 is fantastic and the secondary registry location solves the problem of having to change the nudge sensitivity settings when switching between 99 and pm5.

 

Could I twist your arm to add your version of PM5 without the plunger code changes on you dropbox site? Also if you had a Physmod 2 version I (and others) could use it with the physmod2 tables BigBoss is sharing if you PM him.

 

Thanks,

Rich

Edited by RYSr, 20 August 2014 - 10:42 PM.

[mjr]

I noticed above that you stated you modified the plunger code and it could cause bouncing at idle.

 

What I experienced was that if I used the plunger it would launch the ball with the slightest movement even before you released it, The graphic of the plunger tip on the table then moves up beyond where the ball normally rests. Then when the next ball come up after a drain it would auto-fire without me even touching the plunger.

 

Yep, this is probably my fault.

 

My changes in the plunger code are actually meant to *fix* the problem you're seeing, but I think they work at cross purposes with the other plunger kits.

 

What the standard VP does is kind of weird - logical in its own way, but still weird.  VP expects different (radically different) calibration slopes in front of and behind the rest position.  It does this because the distance you can push the plunger from rest is so much shorter than the distance you can pull it.  VP wants the rest position to be zero, the maximum pull direction to be +1000, and the maximum push direction to be -1000.  That's the sticking point: they want the logical maxima in the push and pull direction to be the same.  So 3 inches turns into +1000 on the pull side, but about 1/4" turns into -1000 on the push side.  So the unit systems on the two sides of the zero point are totally different.  

 

For my setup, the sensor is perfectly linear across its whole range, so it makes no sense to use different scales.  My change simply removes the knee and makes the whole range linear.

 

What's going on with your setup is that your plunger is still using the different coordinate system on the push side, so it's reporting gigantic numbers with very small movements - thus the jumpy behavior.

 

Here's how I plan to fix this.  I'll add code to detect if my plunger device is installed, and only enable the modified plunger code if so.  For all other devices I'll leave the original plunger code intact.  That'll let you use my modified version for the nudge features without breaking your plunger.  I'll try to get that into the code and post an update ASAP.

Edited by mjr, 20 August 2014 - 11:20 PM.

[RYSr]

 

I noticed above that you stated you modified the plunger code and it could cause bouncing at idle.

 

What I experienced was that if I used the plunger it would launch the ball with the slightest movement even before you released it, The graphic of the plunger tip on the table then moves up beyond where the ball normally rests. Then when the next ball come up after a drain it would auto-fire without me even touching the plunger.

 

Yep, this is probably my fault.

 

My changes in the plunger code are actually meant to *fix* the problem you're seeing, but I think they work at cross purposes with the other plunger kits.

 

What the standard VP does is kind of weird - logical in its own way, but still weird.  VP expects different (radically different) calibration slopes in front of and behind the rest position.  It does this because the distance you can push the plunger from rest is so much shorter than the distance you can pull it.  VP wants the rest position to be zero, the maximum pull direction to be +1000, and the maximum push direction to be -1000.  That's the sticking point: they want the logical maxima in the push and pull direction to be the same.  So 3 inches turns into +1000 on the pull side, but about 1/4" turns into -1000 on the push side.  So the unit systems on the two sides of the zero point are totally different.  

 

For my setup, the sensor is perfectly linear across its whole range, so it makes no sense to use different scales.  My change simply removes the knee and makes the whole range linear.

 

What's going on with your setup is that your plunger is still using the different coordinate system on the push side, so it's reporting gigantic numbers with very small movements - thus the jumpy behavior.

 

Here's how I plan to fix this.  I'll add code to detect if my plunger device is installed, and only enable the modified plunger code if so.  For all other devices I'll leave the original plunger code intact.  That'll let you use my modified version for the nudge features without breaking your plunger.  I'll try to get that into the code and post an update ASAP.

 

 

Sounds like a good plan...

 

Thanks for the consideration

Rich

[jkimbrell]

MJR, you are awesome.... that is all.

[mjr]

I just uploaded new builds of VP 9.9 and physmod5 with the plunger changes enabled only if you're using my plunger device.  If all goes well that should make other plunger kits work correctly again.  (The sensing is automatic based on the reported USB device info - no need for any config settings.)

 

https://www.dropbox....2HtOshcJWkIe6ha

 

RYSr, you requested a physmod2 build, but I don't have that one - I think I'm going to have to leave that to the physmod2 owner(s).  It'll get unwieldy for everyone if I keep proliferating the mod versions exponentially.  Hopefully the physmod2 people can merge the nudge changes into their branch now that the code is in the VP10 mainline.

Edited by mjr, 21 August 2014 - 07:02 AM.

[RYSr]

I just uploaded new builds of VP 9.9 and physmod5 with the plunger changes enabled only if you're using my plunger device.  If all goes well that should make other plunger kits work correctly again.  (The sensing is automatic based on the reported USB device info - no need for any config settings.)

 

https://www.dropbox....2HtOshcJWkIe6ha

 

RYSr, you requested a physmod2 build, but I don't have that one - I think I'm going to have to leave that to the physmod2 owner(s).  It'll get unwieldy for everyone if I keep proliferating the mod versions exponentially.  Hopefully the physmod2 people can merge the nudge changes into their branch now that the code is in the VP10 mainline.

 

Thank you! ....

Rich

 

PS: I tested it and the plunger works fine now with my Virtupin2 controller..

Edited by RYSr, 22 August 2014 - 01:08 AM.

[mjr]

I've added a new feature to the Windows config tool that should be really helpful setting things up.  The tool can now display the raw pixel data off the sensor, so you can see exactly what the sensor sees in terms of exposure level.  The display updates continuously, so you can see the effects immediately as you adjust the light source brightness or position.

 

This helped confirm that the two-blue-LED light source I'm using is pretty ideal. The contrast level and sharpness of the shadow edge are both excellent.  It's also pretty neat that you can see a very faint image of the plunger spring coils - very slight gradations in the shadow region that get closer together as you pull the plunger.

 

Hopefully this will be helpful to anyone trying to set up the CCD.  Note that you'll need to install the latest version of the controller software (the .bin file) to use the new version of the config tool.  Everything's available on the mbed project page: http://mbed.org/user...ape_Controller/

Edited by mjr, 23 August 2014 - 01:33 AM.

[Gilrock]

I just got my Freescale board configured and installed in my cabinet.  I'm not trying to use the plunger yet...just installed it for nudge.  Nudge is working but it doesn't seem to nudge much at all.  To make it move any decent amount I'm afraid my backbox is going to shear off.  I've got the card hard mounted to the bottom of the cabinet.  I'm using the 8/20 build you provided of VP.  Boosting the nudge percentage in the settings doesn't seem to have any effect.  I moved it from 1000% to 3000% and then 8000% and nothing changed.  I also tried the 8/20 physmod5 version with the new Playboy table and the ball got stuck trying to get shot out of the left capture lane.

[mjr]

I just got my Freescale board configured and installed in my cabinet.  I'm not trying to use the plunger yet...just installed it for nudge.  Nudge is working but it doesn't seem to nudge much at all.  To make it move any decent amount I'm afraid my backbox is going to shear off.  I've got the card hard mounted to the bottom of the cabinet.  I'm using the 8/20 build you provided of VP.  Boosting the nudge percentage in the settings doesn't seem to have any effect.  I moved it from 1000% to 3000% and then 8000% and nothing changed.  I also tried the 8/20 physmod5 version with the new Playboy table and the ball got stuck trying to get shot out of the left capture lane.

 

Strange that there's no effect from changing the gain settings - my setup is quite sensitive to those.  One thing I've noticed is that it seems to take *two* restarts of a game before a change to any of the nudge settings sticks.  I haven't traced through that part of the code to see if there's some reason for that or if it's just my imagination, but it seems to be a real effect.  Also, try quitting out of VP entirely and starting it up again after a change - that guarantees that it's re-reading all of the registry data and there's nothing stuck in memory.

 

You might post a screen shot of your keyboard options screen - there might be some other setting that's the issue.

 

Also, have you fired up the Windows "Set up USB Game Controllers" control panel to make sure you're getting the expected input from the accelerometer?

[Gilrock]

Thanks for the suggestions...I'll get a chance to experiment more later today.  Thanks for your efforts in this project and sharing it with us. 

[jkimbrell]

Something I noticed a while back when trying to initially setup my sidewinder pro for nudge, you actually have to close the table down in VP (file->close) and re-open it for any changes to the nudge settings to be applied.  Make sure you try that.  You should see a really good and noticeable nudge at   1200 to 1500 gain.  

Also to MJR, I've been traveling a bunch for work over the last few weeks and just now got around to trying the physmod version of VP with your patches.  I had to dial the gain way down (like around 100 to 150) or the nudge was crazy-strong.  Did you notice that on your setup?

 

FYI - my current nudge settings are 1500,1200 for reg physics and 150,100 for physmod version.  For nudge I realize its kind of a "salt to taste" thing but I'm curious of what others have their settings at.

[The Loafer]

The physmod 2 people are the same as the physmod 5 people (same devs), physmod 2 has just been supplanted by the newer physmod 5. The reason why people are asking is there are several test tables that exist, unfortunately they use the standard VP9.9.0 registry path.

 

It's actually not a big deal because you gave vp9.9.0 and vp physmod 5 each their own registry paths different from the old one, so physmod 2 can use the old one and we have 3 different versions now (hmmm, maybe that leaves vp9.2.1 in the cold though. eheh)

 

I have yet to try this, thanks very much for putting these out, heard great things from improved nudging so looking forward to finding the time to test it out.  The separate registry is a huge bonus though, so extremely appreciated!

[RYSr]

I just got my Freescale board configured and installed in my cabinet.  I'm not trying to use the plunger yet...just installed it for nudge.  Nudge is working but it doesn't seem to nudge much at all.  To make it move any decent amount I'm afraid my backbox is going to shear off.  I've got the card hard mounted to the bottom of the cabinet.  I'm using the 8/20 build you provided of VP.  Boosting the nudge percentage in the settings doesn't seem to have any effect.  I moved it from 1000% to 3000% and then 8000% and nothing changed.  I also tried the 8/20 physmod5 version with the new Playboy table and the ball got stuck trying to get shot out of the left capture lane.

 

 You didn't mention it , so make sure the enable analog nudge box is also checked...

Rich

Something I noticed a while back when trying to initially setup my sidewinder pro for nudge, you actually have to close the table down in VP (file->close) and re-open it for any changes to the nudge settings to be applied.  Make sure you try that.  You should see a really good and noticeable nudge at   1200 to 1500 gain.  

Also to MJR, I've been traveling a bunch for work over the last few weeks and just now got around to trying the physmod version of VP with your patches.  I had to dial the gain way down (like around 100 to 150) or the nudge was crazy-strong.  Did you notice that on your setup?

 

FYI - my current nudge settings are 1500,1200 for reg physics and 150,100 for physmod version.  For nudge I realize its kind of a "salt to taste" thing but I'm curious of what others have their settings at.

 

The physmod nudge settings are normally 1/10 of the normal version's.

[toxie]

I just uploaded new builds of VP 9.9 and physmod5 with the plunger changes enabled only if you're using my plunger device.  If all goes well that should make other plunger kits work correctly again.  (The sensing is automatic based on the reported USB device info - no need for any config settings.)

 

https://www.dropbox....2HtOshcJWkIe6ha

 

RYSr, you requested a physmod2 build, but I don't have that one - I think I'm going to have to leave that to the physmod2 owner(s).  It'll get unwieldy for everyone if I keep proliferating the mod versions exponentially.  Hopefully the physmod2 people can merge the nudge changes into their branch now that the code is in the VP10 mainline.

 

 

I already PMed you, but maybe you didn't see it: Could you please send me the recent changes again, so i can add it to VP10?

[mrarcade]

I am starting my first machine build. I am very glad I read this post. I will be ordering the parts for this in the near future. Great addition, thanks!

[Gilrock]

May I ask what is the proper way to calibrate the X and Y nudge axes?  I think that's why mine wasn't working well.  I went through the Windows calibration but what am I supposed to do just try to hit the table from both sides to determine the limits?  Earlier I saw you mention a config tool but I didn't see any program in the dropbox so wasn't sure what you are talking about.  After I did the Windows calibration it was flying across the table when i nudged it so I think this will work if I figure out how to best calibrate it.

[mjr]

May I ask what is the proper way to calibrate the X and Y nudge axes?  I think that's why mine wasn't working well.  I went through the Windows calibration but what am I supposed to do just try to hit the table from both sides to determine the limits?  Earlier I saw you mention a config tool but I didn't see any program in the dropbox so wasn't sure what you are talking about.  After I did the Windows calibration it was flying across the table when i nudged it so I think this will work if I figure out how to best calibrate it.

 

Don't use the Windows calibration tool!  The nudge feature is self-calibrating.  The Windows calibration will only confuse things by changing the axis ranges to essentially random values, which is why you're getting the gigantic inputs in VP.  To delete the calibration you did: open the "Set up USB Game Controllers" control panel, double-click on Pinscape Controller, select the Settings tab at the top, and click Reset to default.