92 replies to this topic

[DKong]

Thanks vamp, I'll give that a try when I'm back at my cab.

[jbg4208]

I must have a setting wrong somewhere.  I see a lot of people are getting FPS rates between 50-100.  Mine are much higher, so maybe this is my stutter issue?

 

Do I want to set VPX to lock in the FPS at 60?  How/where do I do this?

 

My current FPS averages on a couple of tables:

 

TOTAN - VPX

140-160 fps (with B2S enabled)

150-170 fps (withOUT B2s enabled)

 

Pirates of the Caribbean - VPX

340-360 fps (with B2S enabled)

350-410 fps (withOUT B2s enabled)

 

I have read this entire thread, but still am a bit lost on how to correct the VPX stuttering.  

 

I would greatly appreciate some help in clarifying this issue (and if it should help correct the stuttering).

 

Thanks guys.

I was having abou the same issue. Maybe not getting that high of FPS but close. Yes enabling Vsync = 1 fixed stuttering for me. The setting is in the video option menu. Be sure to reopen VPX after you make this setting adjustment. Also keep an eye out for some tables that have Vsync disabled, in the options menu, that will override the global options in VPX.  

 

Edit - Vamp beat me to it..

Edited by jbg4208, 09 January 2016 - 04:02 AM.

[jimmyfingers]

Statistics / FPS values are not very meaningful and essentially incomparable without the following information:

 

1) General / main PC specs as to GPU and CPU

2) Screen resolution

3) Main VP performance affecting options used: 4xAA; FXAA levels (off, Fast, Quality, Extreme); Dynamic AO

4) Table specific main options such as playfield and ball reflections as the reflection rendering is costly to performance as well

 

Default settings for all of the VP settings above except screen resolution can be set per table and override the default VP settings so one needs to both confirm their global settings and if any changes from a default download of a table being compared has been made

 

Maybe even a separate thread would be in order for people to start listing statistics vs. information about optimizing as this could start a trend of pages of people just posting FPS "scores".  In any case, the 4 areas of background information is pretty necessary for anything useful to be determined by comparing FPS.

[PinballShawn]

Any idea of why Aero would cause me to have up and down frame rates? With it on my frame rate jumps from 150 to 33 to 153 to 30 to 152 to 33, making the game unplayable, yet the frame rate text is solid. I turn it off (desktop window manager session) and the frame rate is rock stable and the game play is smooth, but the Frame rate text is flickering.

I've run every conceivable program to check for problems and none of them say there is a problem, on or off. I've run a new windows experience test and came out with a 7.9 out of a possible 7.9. New graphics card, new drivers, game plays fine, should I care?

[PinballShawn]

The more I research, the more I find it seems to be a problem with Nvidia cards/drivers. So if anyone out there has jumping frame rates or stutter they can't figure out, and you are using an Nvidia card, Aero could very well be your problem. There are hundreds of threads out there about the problem with no real solutions.

[Aubrel]

Statistics / FPS values are not very meaningful and essentially incomparable without the following information:

 

1) General / main PC specs as to GPU and CPU

2) Screen resolution

3) Main VP performance affecting options used: 4xAA; FXAA levels (off, Fast, Quality, Extreme); Dynamic AO

4) Table specific main options such as playfield and ball reflections as the reflection rendering is costly to performance as well

 

Default settings for all of the VP settings above except screen resolution can be set per table and override the default VP settings so one needs to both confirm their global settings and if any changes from a default download of a table being compared has been made

 

Maybe even a separate thread would be in order for people to start listing statistics vs. information about optimizing as this could start a trend of pages of people just posting FPS "scores".  In any case, the 4 areas of background information is pretty necessary for anything useful to be determined by comparing FPS.

 

That's what I've done: I gave my full specs in my previous post, and as said, excepting 4xAA, the other settings are mostly changing the visual aspect but the overral performance remains approx the same on my system. For example, with a table running at 50fps with most viual settings in high level, I can't reach 60fps even with the lowest setting 

 

But it's not a complaint because my specs are quite low (I don't have any GPU card !) and most of the tables are still running smooth with stable 60fps using adaptative vsync and a high quality level

Edited by Aubrel, 09 January 2016 - 10:53 AM.

[BigBoss]

A $100 i3 CPU and $120 gtx750ti will do fine.

That's interesting Bigboss. Is that your setup?
 
I find it curious because a search on the comparative specs of the GTX480 against the GTX750ti have them at level in terms of real world performance. Sure there are differences in the specs but some specs actually favor the GTX480 and from a real world perspective they seem to balance out. Given that this is the card I have with a higher i7 processor than the i3 you suggest I'm not sure how the suggested minimum specs could handle VPX?
 
Perhaps it relates to some specific aspects of a GTX750ti that has the benefit over the GTX480 that VPX takes advantage of. If that's the case it would be good to know what those aspects are. I would hate for someone to go buy that setup and have comparable results to what I am having.
 
However if this is indeed your setup and you are able to smoothly run every VPX table released so far at least 60 FPS then it would be great if we can understand why that is.

I haven't tried amh but all others are running great on that setup. The graphics card is running two monitors.

[StevOz]

For what it's worth I have core i3, 3.4GHz haswell and a GTX660, running W10, there is no aero, totally updated with all W10 spyware turned off it runs VPX without any stutter synced at 60 FPS, all VP settings at max, including 4xAA.

 

https://github.com/1...ableWinTracking

 

I have also over clocked my monitor for testing to 66Hz and still no stutter. Also tried it with no sync enabled and was getting over 200 FPS and with B2S running in the background still around 180FPS and over 160FPS, all be this in desktop mode with no frontend.

 

I've little to no doubt this PC could power a VP cab, issue free, the W10 upgrade is worthwhile as with W7 prior to my tweaked W10 set-up VPX always seem to miss a beat despite reporting similar frame rates.

[ringorian]

 

 

A $100 i3 CPU and $120 gtx750ti will do fine.

That's interesting Bigboss. Is that your setup?
 
I find it curious because a search on the comparative specs of the GTX480 against the GTX750ti have them at level in terms of real world performance. Sure there are differences in the specs but some specs actually favor the GTX480 and from a real world perspective they seem to balance out. Given that this is the card I have with a higher i7 processor than the i3 you suggest I'm not sure how the suggested minimum specs could handle VPX?
 
Perhaps it relates to some specific aspects of a GTX750ti that has the benefit over the GTX480 that VPX takes advantage of. If that's the case it would be good to know what those aspects are. I would hate for someone to go buy that setup and have comparable results to what I am having.
 
However if this is indeed your setup and you are able to smoothly run every VPX table released so far at least 60 FPS then it would be great if we can understand why that is.

I haven't tried amh but all others are running great on that setup. The graphics card is running two monitors.

 

 

I had the same setup with an i5 , no chance with the other pins from fren available at vpinball for vpx ...

Edited by ringorian, 09 January 2016 - 11:50 AM.

[jimmyfingers]

There is some misunderstanding about Windows 10 and "aero" / desktop composition that is throwing off reports / results and may mislead people to upgrade based on those findings.  "Aero" is commonly used to described DWM (Desktop Windows Manager) / Desktop Composition as it enables / needs that underlying windows component / rendering method to be running.  The "aero" theme and wording may be missing from Windows 10 but the underlying aspect of DWM / Desktop composition is still there.  The problem is it doesn't seem to have an option to turn off anymore so people seem to think it's not active or part of Windows 10 anymore.   One effectively disabled DWM by choosing the non-aero themes, which was one method to disable DWM / desktop composition, but another was on the .exe itself to choose "Disable Desktop Composition" and one could see when an application ran with that option that if they had "aero" as their desktop theme it would switch from transparent to opaque windows, which relates again to the perception that it's the theme controlling things vs. the other way around.

 

In Windows 7 it was fairly clear how to disable the "aero" theme and effectively disable globally desktop composition and using .exe compatibility options one could disable the underlying Desktop Composition per application - we've been needing to leave that on to get smooth VP play.  In Windows 8 and 8.1, from reports and reasonable research, although not running myself, it became more difficult to disable.  Now it seems with Windows 10 the option is basically gone, but it doesn't mean the key DWM services / Desktop Composition / desktop rendering aren't still using it, it just means it's no longer a choice to disable.  

 

I have not tested Windows 10 yet but certainly advise anyone upgrading to Windows 10 thinking that it doesn't use "Aero" to do more research into the DWM / Desktop Composition as it will likely yield the same result as far as stutter fixes as simply having it allowed / enabled in Windows 7.  I can't say about general FPS improvements but be careful of the "aero" talk and what really is going on under the surface with Windows 10.

[PinballShawn]

I'm curious if my processor affinity settings (realtime) are what's conflicting with Aero? I'll have to do a few tests to see if that's the problem. 

[TT11]

Very slight, but noticable ball stutter - I altered the Max Prerendered Frames from 2 down to 1.  It was marvellous the difference it made to the playability of almost every table.  No ball stutter now.

[Mirkin]

Hi Folks

 

Check this thread out for the settings that worked for me and the insane amount of time I spent finding them lol

 

http://www.vpforums....e=2#entry330001

Attached Files

•  VPX SETTINGS .jpg   160.44KB   25 downloads

Edited by Mirkin, 12 January 2016 - 08:27 PM.

[markrock76]

VPX plays absolutely great when I set vsync to 1.  But on some tables with a dmd the fps drop from 60 to 30 periodically.  I have the dmd on a 2nd screen.  When I move the dmd to the 1st screen I do not get the fps drop.  It stays at 60.  Anyone know why that is and what I might do to get the 60fps consistent but with dmd on the 2nd screen.  Thanks

[BigBoss]

VPX plays absolutely great when I set vsync to 1.  But on some tables with a dmd the fps drop from 60 to 30 periodically.  I have the dmd on a 2nd screen.  When I move the dmd to the 1st screen I do not get the fps drop.  It stays at 60.  Anyone know why that is and what I might do to get the 60fps consistent but with dmd on the 2nd screen.  Thanks

Well we aren't psychics here . What's your video card set up?

[markrock76]

 

VPX plays absolutely great when I set vsync to 1.  But on some tables with a dmd the fps drop from 60 to 30 periodically.  I have the dmd on a 2nd screen.  When I move the dmd to the 1st screen I do not get the fps drop.  It stays at 60.  Anyone know why that is and what I might do to get the 60fps consistent but with dmd on the 2nd screen.  Thanks

Well we aren't psychics here . What's your video card set up?

 

I have Windows 10

Nvidia GeForce GTX 750 ti

No Antialiasing in card

Maximum Pre Rendered frames - 0

Prefer Maximum Performance

Vsync set in VP

 

Extreme AA in VP

Tried 0,1,2,3 Maximum Prerendered frames

 

I've tried about every combo of settings with same result.  Great gameplay with vsync set to 1 with occasional drop to 30 fps on tables with dmd on 2nd screen.  It is something I can deal with but thought if anyone had suggestion I would be happy to try it.  

 

Thanks

[xzotic]

Thought I'd better update this thread as I've managed to solve my VPX/B2S Server performance problem (somewhat inadvertently while working through Windows 10 issues)...

 

First of all ignore that i've gone to win 10 (32bit). My initial testing under Win 10 gave me exactly the same results as windows 7.

 

I suspected that maybe my GFX card was at fault (GTX 480) so I decided to swap in my GTX780 from my gaming rig. Imagine my surprise when this ended up doing absolutely nothing to improve the performance! It was exactly the same. I was perplexed. Surely some sort of increase should have presented itself. But instead no change. So while I was refitting back my GTX480 I noticed that there was only 3gig of system RAM installed on the motherboard! I always thought this box had 4gigs. At some stage I must have removed one memory stick that may have been failing....not sure, but only 3 gigs were there. So I took a gig from my Mame box which has exactly the same RAM and quite frankly will run fine with just 3gigs anyway.

 

I fired up the cab with the extra gig and tested the table. Absolutely no change. So I thought that was disappointing. However I was looking at the F11 FPS and noticed it was not vsyncing again (going from 60 to 80 but stuttering which again shouldn't really show stuttering at those frames, but it does). So I checked my Vsync settings again and noticed I was on adaptive sync (ie '2') in VP which would have been how I last left it during my last tests. I put it back to force Vsync (ie: 1) and also added 1 to the prerendered frames (thinking maybe the extra RAM would assist this setting - if it does or not I don't know).

 

Anyway, fired up VP and low and behold, 60 FPS locked butter smooth. I couldn't believe it. I re downloaded the B2S backglasses that I had issues with before and again butter smooth. Absolutely no stutter what so ever. I had to boot it up again this morning to test to see I wasn't dreaming it all as it was a late finish last night!!! Yep all good.

 

So I don't know. I had tested vsync and pre-rendered frames before but that was without the extra gig of memory. Is it possible that the extra gig was all that was needed to perhaps solve a RAM resource contention issue between VP and B2S for some backglasses? Not sure. But I can be sure that the only thing that is really different between stuttering and smoothness was the addition of the extra gig of RAM and bringing the settings back to V/Sync.

 

So my settings/config are currently:

VPX

Core 2 Duo 2.3GHZ (Note I took OFF the overclock I previously had as I was trying to fault find some win 10 issues however I don't even need to overclock to get the smooth frames!)

FOUR Gb's of Ram!!! LOL

GTX 480

Windows 10 - 32 Bit (But remember has same issue on Win 7 and perhaps it would also be solved on Win 7 if I was still on it - Note however that 64bit windows is not required in my case.)

Enable Fast FXAA

Reflect ball on play field

Ambient Occlusion Enabled

FPS limiter/V-Sync 1

Max Pre-rendered Frames 1

Force Antistropic Texture Filtering (quality)

Max texture dimensions 2048

Object Detail on Maximum (to the right)

 

Result - Butter Smooth in all VPX tables I have tested so far.

 

I know this won't help everyone as there are always so many variable affecting everyone's performance however I thought I would share as I really thought the GTX480 was to blame initially (and was just about to go and buy a new card!!) however it looks to be the extra 1gig of RAM that was the decisive factor in my setup. 

[wrd1972]

Does V-sync need to be enabled in Nvidia control panel in order for it to function in VP?

Assuming yes.

[Mirkin]

 

 

VPX plays absolutely great when I set vsync to 1.  But on some tables with a dmd the fps drop from 60 to 30 periodically.  I have the dmd on a 2nd screen.  When I move the dmd to the 1st screen I do not get the fps drop.  It stays at 60.  Anyone know why that is and what I might do to get the 60fps consistent but with dmd on the 2nd screen.  Thanks

Well we aren't psychics here . What's your video card set up?

 

I have Windows 10

Nvidia GeForce GTX 750 ti

No Antialiasing in card

Maximum Pre Rendered frames - 0

Prefer Maximum Performance

Vsync set in VP

 

Extreme AA in VP

Tried 0,1,2,3 Maximum Prerendered frames

 

I've tried about every combo of settings with same result.  Great gameplay with vsync set to 1 with occasional drop to 30 fps on tables with dmd on 2nd screen.  It is something I can deal with but thought if anyone had suggestion I would be happy to try it.  

 

Thanks

 

Have you tried turning all AA off.

I was getting the same thing, great play with drops to 30FPS.

It was much better with AA off.

[kixfan]

I'm not going to pretend I know what I am talking about but...

 

In regards to the 3gb or 4gb memory, don't some motherboards require paired sticks of memory?   Maybe that was part of the issue?