3889 replies to this topic

[freneticamnesic]

 

 

 

visible variable doesn't work on ramps?

It works for me.
 
Do you go to Ground Kontrol much?[/size]

 

 
If I set objects to be hidden based on ShowDT, most of them work except ramps... however, if I use Camera Mode it works fine. ie. If Table.ShowDT = true then ramp.visible = true else ramp.visible = false 
If I uncheck Show DT in backdrop and start in cabinet mode, I can see the ramps when I shouldn't, but if I start in Camera Mode I can't see the ramps. rev 1675

 

I had the exact same problem using ramps as the cabinet rails in DT mode and they wouldn't hide in FS mode. I made them out of primitives instead.

 

 

That's exactly it heh. Primitives hide just fine! so weird that it works in camera mode though

[toxie]

The thing is that all the static objects cannot be changed during gameplay anymore, as these are still prerendered because of performance.

Testing MB was the first table that I tried that had a DMD.  I noticed that the DMD looks different than under vp99 (same rom).  Like it's trying to simulate lighting on a DMD (antialiasing maybe?) Is that a setting in the table or a setting in VP that is doing that to the DMD display?

 

-Mike

 

MB uses the new builtin DMD of VP10. Take a look at the backdrop, there you have a textbox with the magic text "DMD" to trigger the connection to VPM. In addition you also need to set 'UseVPMDMD = true' in the script before loading VPM.

With the new parameter behind the color one can steer the brightness of the textbox, which can then lead to the bloom kickin in.

 

btw: all interested should also take a look into the new core.vbs, there is also a changelist at the top!

[fuzzel]

 

 

 

 

visible variable doesn't work on ramps?

It works for me.
 
Do you go to Ground Kontrol much?[/size]

 

 
If I set objects to be hidden based on ShowDT, most of them work except ramps... however, if I use Camera Mode it works fine. ie. If Table.ShowDT = true then ramp.visible = true else ramp.visible = false 
If I uncheck Show DT in backdrop and start in cabinet mode, I can see the ramps when I shouldn't, but if I start in Camera Mode I can't see the ramps. rev 1675

 

I had the exact same problem using ramps as the cabinet rails in DT mode and they wouldn't hide in FS mode. I made them out of primitives instead.

 

 

That's exactly it heh. Primitives hide just fine! so weird that it works in camera mode though

 

a ramp is only a dynamic object (where you can hide/unhide from the script) if you checked the opacity active checkbox of the ramp material. Otherwise the ramp is a static rendered element.

[fuzzel]

rev1684 is up:

 

- fix decals in BG mode & add night/day checkbox for BG image

- have more ball variations, and enable new low poly variant (good compromise for HD rendering)

- fix undo for builtin primitives

- reworked ball anti-stretch feature for cabinets (egg-shaped ball fix)

 

from the changelog:

 

- for cabinets: in the video options select "Cabinet anti-stretch the ball" and depending on you screen you can enter a correction offset for the
                aspect ratio. If you use a 16:9 screen use a correction offset for X=0.0 and Y=1.5. For a 16:10 screen use X=0.0 and Y=0.0 for best results.
                This option only applies to rotated and fullscreen rendered tables (cabinet mode)!
                Even with this option you might need to adjust the backdrop settings. The biggest problem for an egg-shaped ball is the Layback value.
                Values over 70-80 are much too high and should be lowered. The scaling can be a problem too.
                To find a better POV with a round ball over the playfield do the following:
                - set the min/max slope to 0 and start table in camera mode (menu Table -> Camera Mode)
                - enter debug mode and activate "Throw balls in player"
                - drop balls all over the table and adjust the POV settings
 

[BananaBoat]

on my system with bodydumps test table i get the following results

quality fxaa ON

4xAA ON

Force AT Filtering ON

Occlusion OFF

130 FPS - table plays smooth

 

with Occlusion ON

60 FPS - ball seems to skip frames, jerky

 

with everything OFF but Force AT Filtering ON, 290-320 FPS, smooth.

 

Monster bash on the other hand, Incan not get to play smoothly at all.

[randr]

 Screen Shot 2015-02-11 at 7.21.45 AM (2).png   953.26KB   27 downloadshere is what my FPS is on virtual os mac pro running parallels. notice backglass was even loaded up

[randr]

Fyi it's smooth and totally playable as is mb. Crazy?

[toxie]

on my system with bodydumps test table i get the following results

quality fxaa ON

4xAA ON

Force AT Filtering ON

Occlusion OFF

130 FPS - table plays smooth

 

with Occlusion ON

60 FPS - ball seems to skip frames, jerky

 

with everything OFF but Force AT Filtering ON, 290-320 FPS, smooth.

 

Monster bash on the other hand, Incan not get to play smoothly at all.

 

JSYK: FXAA and 4xAA together don't play well at the moment. So please only use 4xAA for now.

OR: Do not use FXAA and 4xAA in VP, but enable FXAA in the control panel of your gfx card.

 

And then enable AO.

 

Note that especially the combination of AO and 4xAA is really putting a lot of work to the GPU.

[ClarkKent]

If somebody is interested: I found out when switching off 4xAA but setting the FPS limiter to 120 the ball stutter that occured now and then even with vsync on is totally gone! Even on the right wire ramp in MB! 4xAA looks better but if I get constant framerate I prefer this more than visuals.

[LadyLuck]

Super! I like TOLEDO. Without any changing, it´s running cool.

[takut]

Here's a link for a new table I built in VP10.  I decided to scratch build a table in VP10 rather than convert one of my VP9 tables so I could get a better idea of the new features.  It's an EM table and I used much of Leon Spalding's script from his VP8 version.  I did the table without using Photoshop (except for the redraws of course).  I'm sure some things could look better with photoshop trickery but I wanted to see if you could build a realistic reproduction just in VP.  I think it came out OK.  I really like the ability to build in the desktop and FS together (thanks to the new visible option on em reels) and the day/night slider.  I changed up some lighting in the script based on the day/night value.  Kind of cool.  I figured I'd share it so everyone has another table to play around with and test for bugs.  If you mess around with the POV you'll probably need to adjust the pop bumper lights.  B2S is included and automically enables for FS and disabled for DT.  Sorry if you're a 2 screen desktop user

https://www.dropbox.... 1975).zip?dl=0

 

Thanks a lot for the dB2S table

 

For me it was necessary to adjust the Backdrop settings as shown below to get a round ball:

 

 

 toledo_backdrop.jpg   28.89KB   32 downloads   toledo.jpg   119.22KB   29 downloads

[sliderpoint]

The thing is that all the static objects cannot be changed during gameplay anymore, as these are still prerendered because of performance.

Testing MB was the first table that I tried that had a DMD.  I noticed that the DMD looks different than under vp99 (same rom).  Like it's trying to simulate lighting on a DMD (antialiasing maybe?) Is that a setting in the table or a setting in VP that is doing that to the DMD display?

 

-Mike

 

MB uses the new builtin DMD of VP10. Take a look at the backdrop, there you have a textbox with the magic text "DMD" to trigger the connection to VPM. In addition you also need to set 'UseVPMDMD = true' in the script before loading VPM.

With the new parameter behind the color one can steer the brightness of the textbox, which can then lead to the bloom kickin in.

 

btw: all interested should also take a look into the new core.vbs, there is also a changelist at the top!

 

 

OK, I'm kind of baffled here. I have tried setting the UseVPMDMD = False, I've tried commentting that entire set of code out of the table for it, tried deleting the textbox from the background and deleting the mb_106b registry key and it's still showing the bloom/brightness in the DMD in VPx.  Did it save these settings in a file or a different registry key somewhere?

 

-Mike

[jimmyfingers]

 

For me it was necessary to adjust the Backdrop settings as shown below to get a round ball:

 

 

toledo_backdrop.jpg  toledo.jpg

 

Did you try the updated anti-stretch code and options?  fuzzel lists them and the basics of working with them in his earlier post with the newest beta.

[takut]

Did you try the updated anti-stretch code and options?  fuzzel lists them and the basics of working with them in his earlier post with the newest beta.

 

Yes, I try it, but the ball is always an egg and not round (see screenshot below). I set into the video options from the "Ball aspect ratio..." field to "Cabinet anti-stretch the ball" and the X value to 0 and the Y value to 1.5 (I have a 16:9 monitor).

 

For me the best results are to change the backdrop values.

 

 01.jpg   72.3KB   30 downloads   02.jpg   82.06KB   29 downloads

[atarian]

Many thanks fuzzel and toxie for the new builds. Like takut, I tried the cab anti stretch with the 16:9 settings takut mentioned and also get an egg ball near top of table. I played a bit with layback like fuzzel suggests and can improve it. I guess it isn't straightforward to come up with an algorithm which can take into account all the view settings, and the position of the ball on the table, but over time with experimentation we should be able to come up with nice params for various views/screen combos.

Edited by atarian, 11 February 2015 - 08:22 PM.

[jimmyfingers]

takut,

 

For FS try 6, 10, 60 for incline, FOV, Layback with a Y offset of 1.25:

 

Looks like this:

 

 Capture Toledo 6, 10, 60 with 1.25 for Y offset.PNG   965.13KB   31 downloads

 

Settings capture:

 

 Capture of Settings.PNG   69.24KB   31 downloads

 

 

Mind the 720p resolution - testing rig and I toggle between 1080p and 720p for readability / eye strain (projector on wall)

[bodydump]

takut,
 
For FS try 6, 10, 60 for incline, FOV, Layback with a Y offset of 1.25:
 
Looks like this:
 
Capture Toledo 6, 10, 60 with 1.25 for Y offset.PNG
 
Settings capture:
 
Capture of Settings.PNG
 
 
Mind the 720p resolution - testing rig and I toggle between 1080p and 720p for readability / eye strain (projector on wall)

Those settings look better. The problem with takut's settings are the ball maybe round but the light inserts aren't.

[atarian]

takut,

 

For FS try 6, 10, 60 for incline, FOV, Layback with a Y offset of 1.25:

 

Looks like this:

 

Capture Toledo 6, 10, 60 with 1.25 for Y offset.PNG

 

Settings capture:

 

Capture of Settings.PNG

 

 

Mind the 720p resolution - testing rig and I toggle between 1080p and 720p for readability / eye strain (projector on wall)

 

Thanks jimmy - a nice round ball and good table view with those settings.

 

Excellent EM bodybump.  The ball really shows off the ball spin nicely - cheers

Edited by atarian, 11 February 2015 - 08:55 PM.

[takut]

For FS try 6, 10, 60 for incline, FOV, Layback with a Y offset of 1.25:

 

Thanks for your help, but with your settings the ball is round, that's right, but the bumpers look like eggs

 

Those settings look better. The problem with takut's settings are the ball maybe round but the light inserts aren't.

 

For my opinion my modified backdrop settings are the best solution for the table. I only want to share my setting. Maybe somebody likes that also like I do. Otherwise, it doesn't matter. I don't care about the not round light inserts
 

[jimmyfingers]

 

For FS try 6, 10, 60 for incline, FOV, Layback with a Y offset of 1.25:

 

Thanks for your help, but with your settings the ball is round, that's right, but the bumpers look like eggs

 

The bumpers may look exaggerated because now you can see more of their base (a good thing I would say for the depth illusion).  In your previous / first screen shot they look actually a little compressed squat.  They may be a little elongated in the settings I suggest / showed but I'd say they're actually still closer to round but at least they're both out a little in different axis so it's almost 6 and one half dozen which slight distortion you choose.

 

None the less, I was only trying to initially just help you out (and possibly others) - not trying to make it a competition of it. So, no worries and choose whatever you like most.  I think everyone should be welcome to post settings and help find the best combination of view settings and offset values for the new anti-stretch options / coding.