Code change to enable SAM LE driver board support

[Drybonz]

I wonder if that was consistent with the old VPM's prior to all the recent sound improvements, etc... if turning it down and back ironed something out... I don't think I ever thought to try it.

[DJRobX]

On 11/13/2016 at 11:53 PM, robertms said:

Another observation, the choppy audio on Scared Stiff occurs only when running the table directly from the editor (v2847) but seems fine when launched via Pinball X (v2.16). I'm not even sure how that makes any sense, am I doing something wrong?

I'm glad you posted that -  I've actually seen something similar for quite a while on my cab.  Had been going on well before I started messing around with VPM timing.  I noticed it when I was messing around with the modulated lights.    So, it's not really a new issue.    I've done some work to address the warbling music that I hope will also help with this.   I will post a build soon (need to back home to my cab to test it). 

 

 

 

[Dazz]

On 11/11/2016 at 9:08 AM, DJRobX said:

.NET exceptions are B2S.   If you dig through this thread (LOL) you'll find a post where I shared an updated B2S DLL that will stop intermittent crashing when modulated solenoids are on.   Here's the link to that download.    Really hope we can get all this new stuff stable so we can do a unified release. 

https://www.dropbox.com/s/5o61qx0gahc8r24/b2s102516b.zip?dl=0

 

Hey Guys,

I've been out of the loop for a little while dealing with real life issues.  Just downloaded the updated DLL and trying to get B2S working again.  I downloaded from the link above and replaced my files and re-registered them.  Every time I run a table B2S crashes.  Is there something I am missing?

[DJRobX]

"Trying to get it working again" sounds like you replaced the DLL because it was already crashing? 

The fix is a very small change intended to fix a very sporadic crash mid-game on tables using modulated solneoids.  If it was already crashing before in a different manner, this DLL wouldn't fix that.

Some details about the crash would help.   The exception details will tell us a lot.

 

 

 

[Drybonz]

Hey DJR, just a heads up there were some reports of a mini-freeze in Fozzy's new Scared Stiff over at "the other site"... I remembered that there was some recent talk about that one so just wanted to bring to your attention in case it was VPM related.  Thanks.

[Dazz]

7 hours ago, DJRobX said:

"Trying to get it working again" sounds like you replaced the DLL because it was already crashing? 

The fix is a very small change intended to fix a very sporadic crash mid-game on tables using modulated solneoids.  If it was already crashing before in a different manner, this DLL wouldn't fix that.

Some details about the crash would help.   The exception details will tell us a lot.

 

 

 

Hey Rob,

I found the crash that I was having was due to the ScreenRes.txt being overwritten and values were all showing 0.  I fixed the values and B2S is no longer crashing, but B2S is showing on the wrong monitor.  No matter what value I input for the screen I cannot get it back to my backglass monitor.

[DJRobX]

Here's a new VPM that attempts to address the WPC DCS sound issues.   It now uses the same strategy to keep the buffer full as SAM.   I've also enabled this for WhiteStar II. 

I was able to get the warble out of Indiana Jones:Pinball Adventure VP9, which was always a problem on my cab.    Hopefully it works as well for you guys.   These sound buffer issues are really tricky to deal with.  

I also marked the switches and solenoid state variables as "volatile" as they're being read/written to from separate threads.   Seems to improve flipper latency/consistency.  

 

vpm111916.zip

[benlogan]

Thanks for your continued work, DJRobX. Testing tonight. :)

 

[benlogan]

Did some testing tonight with Diner WIP (which was really stuttering and looping before), and Funhouse WIP. Over the course of about ten or twelve games, I only had once instance of looping: "Quit playing with the clock......ock!"

Major improvement!

[Thalamus]

Seems T2 still has the same problem. Since you mentioned WPC DCS, I though, lets try som STTNG. Sound is great, but, I didn't manage to play through a 4 player game one without ending up with indefinite ball hunts. It looks like there is something wrong with recognizing drained balls. On what version this started to happen - IDK. I tried 11316 before I went to non-sam official just to verify the problem.

To tired now to investigate this further - 06:28 in the morning here :-)

 

[ultimate11]

Speaking of STTNG, I noticed it has a bug I've noticed where if you perform many "Q's Challenge" bonuses (correct shots), the game freezes. Is it a bug on STTNG, or on PinMAME? I'm using the LX-7 Special ROM as a test.

[arngrim]

10 hours ago, DJRobX said:

Here's a new VPM that attempts to address the WPC DCS sound issues.   It now uses the same strategy to keep the buffer full as SAM.   I've also enabled this for WhiteStar II. 

I was able to get the warble out of Indiana Jones:Pinball Adventure VP9, which was always a problem on my cab.    Hopefully it works as well for you guys.   These sound buffer issues are really tricky to deal with.  

I also marked the switches and solenoid state variables as "volatile" as they're being read/written to from separate threads.   Seems to improve flipper latency/consistency.  

 

vpm111916.zip

tested Scared Stiff and it's much better, hard to tell if it totally fixed because the original sound is maybe not perfect, but i think it is fixed

[MrKnister]

Hey all,

i tried with that super marion table and there is still maaaasive sound echoing going on :). sounds pretty weired there. do you guys have the same or is it just something with my setup and totally unrelated to the changes ?

cheers

[Thalamus]

1 hour ago, MrKnister said:

Hey all,

i tried with that super marion table and there is still maaaasive sound echoing going on :). sounds pretty weired there. do you guys have the same or is it just something with my setup and totally unrelated to the changes ?

cheers

No, There is a lot of sound looping going on for me too for SMB-TT-NZ_1.1.vpx, on rom smb3. 

If I go back to VPinMAME.dll , release: 103016b, md5: 013A6B631E1B423EBFB5FA373AEEE4E9 - SMB seems to be running fine.

On this version, the STTNG problem described above is already happening. I went back all the way to vpm101416.zip and could reproduce the problem.

[DJRobX]

Which ST:TNG table are you using? 

I don't have the SMB problem, but I am able to simulate it pretty readily by making the DMD really big with ddraw=0.     If I do that the sounds starts looping and scratching.    So that will definitely happen if the CPU isn't keeping up.     So I've made a few changes here that will hopefully help.

1) De-tune the mid-sync throttling by a few MS, and only sync at 25% and 50% (where it previously synced to 25%, 50%, and 75%)..    Still should provide most of the benefit in terms of catching flipper presses earlier but gives the CPU more breathing room if the emulation's demands are unevenly distributed.  

2) Merge toxie's latest.  He has a optimized the ddraw=0 code pretty significantly.   The only negative here is now I can't as easily simulate the smb problem.    

Give this a try and let me know how it goes.  

 

vpm112016.zip

[DozerAUS]

Hi All, that STTNG bug has been around for a long time.  Many reported success way back when in preventing the ROM reboot / freeze etc. by turning off anti aliasing for that ROM entry in the registry.   I remember doing that but don't really play the game that much so can't confirm whether it worked or whether anti aliasing even does anything .  

[CarnyPriest]

Weird about that anti aliasing. I'm not sure why it was even termed that. I'd say it is more like a rudimentary bloom setting as it represents how much "light" bleeds though to the dark areas in between DMD pixels. It definitely does something. The default is 50%. Those in the know who have played with it put it around 25 to 40% as they say it appears more like the glow from a real DMD. Some just turn it off by setting to zero percent for no bleed over into the spaces in between DMD pixels. If you have a real DMD then you should definitely turn it off. I actually put it at 60% but I have a p2k-style cab and I lose something in projecting the image onto the playfield glass. 

[ultimate11]

53 minutes ago, DJRobX said:

Which ST:TNG table are you using? 

For me, I'm using PinMAME's internal simulator. The "Q's Challenge" freeze occurs when I use the simulator, but I haven't on regular tables.

[CarnyPriest]

2 minutes ago, ultimate11 said:

For me, I'm using PinMAME's internal simulator. The "Q's Challenge" freeze occurs when I use the simulator, but I haven't on regular tables.

Nothing is optimized in the PinMAME executables that I build. I figured they would only be used to get lixtxml outputs. They are not very clean builds is what I am saying. You may be better off using the official version.

[robertms]

On 11/15/2016 at 9:59 AM, DJRobX said:

I'm glad you posted that -  I've actually seen something similar for quite a while on my cab.  Had been going on well before I started messing around with VPM timing.  I noticed it when I was messing around with the modulated lights.    So, it's not really a new issue.    I've done some work to address the warbling music that I hope will also help with this.   I will post a build soon (need to back home to my cab to test it). 

 

 

 

Hey DJR, I'm happy to report vpm112016 fixed all my Scared Stiff sound issues, even when launching the table directly from the editor. Thank you so much for your hard work, you're unstoppable!

[ultimate11]

1 hour ago, CarnyPriest said:

Nothing is optimized in the PinMAME executables that I build. I figured they would only be used to get lixtxml outputs. They are not very clean builds is what I am saying. You may be better off using the official version.

I've tried the offical version (2.5), and the crash occurs.

[CarnyPriest]

8 hours ago, ultimate11 said:

I've tried the offical version (2.5), and the crash occurs.

I'm talking strictly the PinMAME executable from the first post in toxie's VPM beta thread at VPF. That would be the latest and greatest. Worth a try although I don't think that toxie does any more while compiling it than I do.

[Thalamus]

STTNG available here - https://www.dropbox.com/sh/t1wfzdge7mkcixx/AACR3ResI7FJWggR1oxyt-NIa?dl=0

These are both Physmod5. I don't have crashes or anything like that, the balls seems vanish either via locks or drains. Ending up in endless ball searches. Haven't had the time to investigate much yet. ( update :  Just tested STTNG for the roms below - not any issues. Got it to be stupid at 110716b. I don't know the history of this table. I cannot say that it is related at all to these updates. But, what I've seen suggests that it is related. Will have to switch a few times between the releases to verify I guess.

Another update on STTNG - It might not be related at all to updates, so ignore for now, managed to get it on a release I felt was always playing ok.

About Super Mario Brothers. Following this thread, works aye' ok up until vpm110216, then you did flipper-tweaks and released 110616b that I can't get this table to start with.

Next release 110716 has the stutter problem.

 

 

[DJRobX]

2 hours ago, Thalamus said:

Next release 110716 has the stutter problem.

 

Yes, now have you tried 112016? 

[Thalamus]

Yes, unfortunately. It was really bad lag/studder.

So, since SMB according to ipdb is a Gottlieb System 3, I just tried Surf'n Safari = studder, Freddy's Nightmare = studder. Class of 1812 as well.

But, not on Deadly Weapon ?!

Didn't try any more than these. Very noticeable.