4027 replies to this topic

[freezy]

@toxie your b) would mean another kd-tree for nearby DTs? Or how would you find the other DTs? Maybe a) could be simplified by cloning the ball on DT collision, don't render it, don't apply the collision and only make it live a few ms (or until next hit)? That way you just need to handle creation/destruction of the phantom ball and not deal with DT-specific stuff, it would just hit whatever the next object is.

[toxie]

I'd make it even simpler: brute-force loop over all other hittargets on a hittarget-hit, then just compare distance ball to target-bbox and take direction/length of ball velocity into account to fake the bending effect. After all the ball must be very very close already to the drop bank and a table usually does not feature more than one or max 2 handful of targets anyway.

[toxie]

- cleanup internal resource handling

- cleanup script files

- fix recent table file menu issue

 

Some comments and especially HOMEWORK for you guys 

 

1)

Please do the following: Save copies(!) of a couple of your tables with this build, then close them, and reload and play them with this build. Do they still load/look/react the same (i.e. no missing images or materials or sound effects).

Bonuspoints: Also load these copies up in 10.6.X and test the same quickly in there.

 

2)

Use the new scripts. I mainly just threw out all the changelog and lengthy description blocks and repackaged these into the new Changelog_Scripts.txt and Scripts.txt (@fuzzel: the latter is unfortunately missing in that zip) in the hope of getting a tiny bit faster startup time, maybe even improving the AMSI scanning, who knows (depends on how that works under the hood). Also its now cleaner and less obfuscated this way IMHO.

[toxie]

(and something unrelated: anybody up to modernize this exotic one for VPX? https://aksioma.org/...s_vtc/index.htm )

Edited by toxie, 16 July 2020 - 08:18 AM.

[GInsonic]

FYIO: The "Table xxx has changed" request also appears in new rev4224 (e.g. in Batman (Data East)), VPX 10.6.1 works flawlessly.

BTW, thanks a lot for your work guys

[fourbanks]

 

Black Belt

 

Crashes on loading - did not on previous 10.7 release (4154).

https://vpinball.com...y-1986-v1-0-sp/

 

D&D

Very low and muffled sound from 7.1 setup - works perfectly in 10.6.1 and older.
Can switch in between 10.7 beta and 10.6 series, always same issue on 10.7.
Happens on all 10.7 I've tested (many of the later - not everyone tested)

Sometime still just crashes on exit from latest (4194).

https://vpinball.com...ons-bally-1987/


 

same here with this table only 

table now working no further problems 

[The Loafer]

(and something unrelated: anybody up to modernize this exotic one for VPX? https://aksioma.org/...s_vtc/index.htm )

 

Holy crap, I hadn't seen that in over 15 years!!!

[fuzzel]

btw here is a small teaser what will come in one of the next versions. It's a bit rough at the moment and needs a bit fine tuning.

 

https://youtu.be/L1lw-Q0oTTc

[The Loafer]

Well those should be cool tools for the table builders!

[fuzzel]

It will be not only heplful for table authors. We plan to add a better control UI for the camera mode. The only problem at the moment is that you can't rotate the UI if you are in fullscreen mode, so the windows will be always in desktop orientation. It's possible to rotate the windows but the mouse cursor and click events can't be translated afaik. But maybe we find a solution later.

Edited by fuzzel, 16 July 2020 - 07:18 PM.

[jpsalas]

 

(and something unrelated: anybody up to modernize this exotic one for VPX? https://aksioma.org/...s_vtc/index.htm )

 

Holy crap, I hadn't seen that in over 15 years!!!

 

 

Same here  

 

- cleanup internal resource handling

- cleanup script files

- fix recent table file menu issue

 

Some comments and especially HOMEWORK for you guys 

 

1)

Please do the following: Save copies(!) of a couple of your tables with this build, then close them, and reload and play them with this build. Do they still load/look/react the same (i.e. no missing images or materials or sound effects).

Bonuspoints: Also load these copies up in 10.6.X and test the same quickly in there.

 

2)

Use the new scripts. I mainly just threw out all the changelog and lengthy description blocks and repackaged these into the new Changelog_Scripts.txt and Scripts.txt (@fuzzel: the latter is unfortunately missing in that zip) in the hope of getting a tiny bit faster startup time, maybe even improving the AMSI scanning, who knows (depends on how that works under the hood). Also its now cleaner and less obfuscated this way IMHO.

 

 

1) I tried several of my tables and they all work fine, graphics, sounds and materials all are still there.

2) I used the new scripts without any problem.

[toxie]

Perfect, thanks a lot JP!!

[jpsalas]

I have been playing with the editor and I have found a few errors:

 

- on the Flashers: the mode names wrap and world are exchanged. If you choose wrap it shows world, and the same the other way around. The tables play fine, so the internal values are ok, but it is shown wrong in the properties. They should work like the ramps, which still are ok.

 

- if you select 2 or more lights and try to edit them together to change the colors, then 2 things may happen:

  - all the lights will get the same Falloff size even if you have not touched the falloff size.

  - and you need to change the colors light by light, with the color windows popping up for each light. If this was done intentionally then it is ok, it just takes longer to change the colors

 

- if you select 2 primitives or more and try to change their height or material you may end up with all them in the same x position. This doesn't happen all the time, so this may be difficult to track.

 

I must say this latest version is quite stable. It is not 100% as sometimes the editor just quit, mostly when working with the layers. But it is much more stable than 10.6.

[fuzzel]

I have been playing with the editor and I have found a few errors:

 

- on the Flashers: the mode names wrap and world are exchanged. If you choose wrap it shows world, and the same the other way around. The tables play fine, so the internal values are ok, but it is shown wrong in the properties. They should work like the ramps, which still are ok.

 

- if you select 2 or more lights and try to edit them together to change the colors, then 2 things may happen:

  - all the lights will get the same Falloff size even if you have not touched the falloff size.

  - and you need to change the colors light by light, with the color windows popping up for each light. If this was done intentionally then it is ok, it just takes longer to change the colors

 

- if you select 2 primitives or more and try to change their height or material you may end up with all them in the same x position. This doesn't happen all the time, so this may be difficult to track.

 

I must say this latest version is quite stable. It is not 100% as sometimes the editor just quit, mostly when working with the layers. But it is much more stable than 10.6.

Flasher issue is fixed in the next update. Changing the color of multiple lights will only show one color dialog for the selection in the next update. But I wasn't able to repro the rest of the seen behavior.

If you have some random crashes from time to time: do you have a crash.txt file for us?

[toxie]

and maybe an exact repro (i.e. with specific table) for the lights/primitives? thanks!

[Robsoie]

I tested most of my tables with the latest version without having any trouble, but it seems when i tested Seawitch it gave me a error undefined variable, just before the pinmame loading screen appears, but it does not tell me which variable unfortunately.

 

I redownloaded the table (it's the one from jpsalas https://www.vpforums...&showfile=14916 ) as i had possibly edited some part of the script, but same error, as the error happens just before the pinmame loading screen, i deleted the seawitch related entry in the HKEY_CURRENT_USER\Software\Freeware\Visual PinMame\ registry .

Then deleted the related nv file in ...\Visual Pinball\VPinMAME\nvram\ and just in case the related .cfg in ...\Visual Pinball\VPinMAME\cfg\

 

But still same error.

I have a backup of a previous Visual Pinball entire directory with a slightly older VPinballX.exe (it's one of the previous beta 10.7, not sure which one) , switched this executable to test, still same error.

 

Removed that whole visual pinball directory and restored the older one from my backup (it has older vbs scripts too, i think most are still from default 10.6) , and Seawitch has no error.

Maybe it is related to the new vbs scripts then ?

Edited by Robsoie, 18 July 2020 - 12:28 AM.

[webby]

Hi,

Just really like to say thanks for the latest beta as it fixed an ongoing issues with a couple of UltraDMD tables having the Bass/Music script loading error especially Kiss. No error anymore

Thanks again keep up the great work.

Webster

[jpsalas]

I tested most of my tables with the latest version without having any trouble, but it seems when i tested Seawitch it gave me a error undefined variable, just before the pinmame loading screen appears, but it does not tell me which variable unfortunately.

 

I redownloaded the table (it's the one from jpsalas https://www.vpforums...&showfile=14916 ) as i had possibly edited some part of the script, but same error, as the error happens just before the pinmame loading screen, i deleted the seawitch related entry in the HKEY_CURRENT_USER\Software\Freeware\Visual PinMame\ registry .

Then deleted the related nv file in ...\Visual Pinball\VPinMAME\nvram\ and just in case the related .cfg in ...\Visual Pinball\VPinMAME\cfg\

 

But still same error.

I have a backup of a previous Visual Pinball entire directory with a slightly older VPinballX.exe (it's one of the previous beta 10.7, not sure which one) , switched this executable to test, still same error.

 

Removed that whole visual pinball directory and restored the older one from my backup (it has older vbs scripts too, i think most are still from default 10.6) , and Seawitch has no error.

Maybe it is related to the new vbs scripts then ?

 

I see what I did wrong in the script I forgot to add     vpmInit me   after the Sub table1_Init Like this:

 

Sub Table1_Init

    vpmInit me

    With Controller

and maybe an exact repro (i.e. with specific table) for the lights/primitives? thanks!

 

I'll see if I can reproduce the error easily. As I said it doesn't happen all the time.

 

The crash while editing is also rare, but next time I'll post the crash.text. 

Edited by jpsalas, 18 July 2020 - 03:58 AM.

[jpsalas]

Here it is an error, I was just selecting some objects with Shift mouse click. VPX quit without any messages-

 

Crash report VPX rev4224

============

Process: VPinballX.exe

Reason: 0xC0000005 - EXCEPTION_ACCESS_VIOLATION at 0023:004819F0

Attempt to read from 0x3D7A8B6F

Thread ID: 0x6C0 [1728]

 

Call stack

==========

004819F0 VPinballX.exe (0x3D7A8B6F 0x3E348830 0x74BD41F0 0x00000001)

0046C93B VPinballX.exe (0x0DB6B721 0x00160610 0x3E348830 0x00427158)

0046DBDE VPinballX.exe (0x00000110 0x0016063C 0x00000000 0xffffffff80006011)

004252E6 VPinballX.exe (0x00425290 0x00160610 0x00000110 0x0016063c)

74BD77B2 user32.dll CallWindowProcW + 0x1B22 (0x00160610 0x00000110 0x0016063C 0x00000000)

74BD6D81 user32.dll CallWindowProcW + 0x10F1 (0x03978E00 0x00000000 0x00000110 0x0016063c)

74BCC4B5 user32.dll CreateWindowExA + 0x65 (0x00160610 0x00000110 0x0016063C 0x00000000)

74BE5CAB user32.dll AddClipboardFormatListener + 0x4B (0x74BCC490 0x00160610 0x00000110 0x0016063c)

74BD67BC user32.dll CallWindowProcW + 0xB2C (0x1656F4E0 0x00007FFA 0x00000110 0x0016063c)

74BD5F6D user32.dll CallWindowProcW + 0x2DD (0x03978E00 0x00000000 0x00000000 0x00000000)

74BD34B6 user32.dll GetWindowRgnBox + 0xC86 (0x00400000 0x007CA598 0x00000644 0x000f0526)

74BD09A5 user32.dll CreateDialogIndirectParamAorW + 0x35 (0x00400000 0x007CA598 0x000F0526 0x00425290)

74BCF81C user32.dll CreateDialogParamA + 0x6C (0x00400000 0x00000225 0x000F0526 0x00425290)

0042538C VPinballX.exe (0x3E348830 0x03031B68 0x3E348830 0x00000000)

0044E5B5 VPinballX.exe (0x3E348830 0x00745128 0x00000000 0x00000000)

0044CAB4 VPinballX.exe (0x0DB6B5DC 0x03047C50 0x0019E760 0x0019dfb8)

00481CF6 VPinballX.exe (0x03049168 0x03047C78 0x0019E760 0x0053f29e)

005BB33A VPinballX.exe (0x00000268 0x00537470 0x0DB6B5DC 0x00000001)

00537470 VPinballX.exe (0x0DB6B5DC 0x00000001 0x00000001 0x00000000)

00537470 VPinballX.exe (0x74BD67E0 0x2229205C 0x00000000 0x0053922d)

74BD6A67 user32.dll CallWindowProcW + 0xDD7 (0x2229205C 0x00000000 0x0053922D 0x00000424)

74BD67E0 user32.dll CallWindowProcW + 0xB50 (0x00000424 0x00000268 0x00000001 0x000f05ae)

 

Environment

===========

Date/time: 18/7/2020, 08:37:31:967

Number of CPUs: 8

Processor type: 586

System: Windows 10 (10.0 17763)

 

Memory status

=============

Total Reserved: 120844K (118M) bytes

Total Commited: 1148824K (1121M) bytes

Total Free: 2924572K (2856M) bytes

Largest Free: 2095448K (2046M) bytes

 

Registers

=========

EAX=0019D51C EBX=3E348A74 ECX=3D7A8B70 EDX=3D7A8B6F

ESI=3D7A8B6F EDI=FFFFFFFF EBP=00000000 ESP=0019DFFC EIP=0053922D

FLG=00010206 CS=0023 DS=002B SS=002B ES=002B FS=0053 GS=002B

 

 

Mini dump saved successfully.

on the new blackbelt table, when i update the script files i get a small error, when i put the script files back to the older ones, its fine

 

This is the same error as in Seawitch,  vpmInit me is missing in the Table1_Init

 

Sub Table1_Init

    vpmInit me

    With Controller

 

Many tables do not have this  vpmInit me so maybe it is something missing in the new core.vbs.

[toxie]

JP, you have to help me a bit here. How does "vpmInit me" make a difference there? From my limited VBS/core script skills, the only thing that could matter in that function is the check for UseModSol.

Or how does that function change other things??