2087 replies to this topic

[Slydog43]

totan plays buttery smooth on my system with vsync set to 1 with 9.9.  F11 show around 8.4 for the timing, so it looks like you need beefier hardware

[marco helmink]

my hardware is

 

Gt 630  2gb DDr 3 on my playfield

 

and

 

7600 GT 1 gb on my dmd and backglass

 

Running in win XP and hyperpin.

[randr]

I finally fully switch all tables to DX9 but do notice most tables seem to play very slow(ball speed). I adjust slope from lets say 6 to 6.5 and seems better but under a global setting can this be changed or is it on a table by table change? never noticed this before? I did match up all my settings as I see it uses new reg for official release.

[randr]

Ok. If I got to a table like Congo and check overwrite globle settings table plays good..is there a way to globally check this option for every table?

[boiydiego]

i didnt see that anywhere randr so think not i had the same problem like you ..

[mukuste]

Ok. If I got to a table like Congo and check overwrite globle settings table plays good..is there a way to globally check this option for every table?

 

Can you find out which physics parameter exactly causes the difference? We've had several people now report speed differences and I honestly have no idea what's going on. I think we should get rid of these global physics settings for VP10 because they cause nothing but trouble.

 

Also as usual, try messing with the frame limiter and see if it makes a difference.

[randr]

Strange. Earlier betas were fine.

Anyways ive tried just about all combinations but cannot seem to pinpoint it to one setting in the global physics panel. I will say checking overwrite global settings on each table does seem to fix the table I just wish we could globally check this for all tables.

 

I have tried all combinations and did try frame limiter also with no change to what i'll call "slow" play...like the table has very little slope

[mukuste]

Ok, can you link me to the exact table version this happens on (maybe also redownload it yourself, just in case something was changed) and I'll see if I can reproduce it when I have time.

 

The change in speed you notice is relative to 9.2.1? Did you compare 9.2.1 and 9.9.0 back to back?

 

Also please show me a screenshot of the global physics set that you used to override it. What exactly did you do to fix it, I think it's a textbox to override with global physics set rather than a checkbox?

[DJRobX]

Can you find out which physics parameter exactly causes the difference? We've had several people now report speed differences and I honestly have no idea what's going on. I think we should get rid of these global physics settings for VP10 because they cause nothing but trouble.

 

I had a similar problem with VP 9.12/9.15.    I had a bunch of tables that were not playable, most memorable being Diner, I couldn't make it up the ramps.     The problem was the Contact Friction value was clearly wrong for the way the table was designed.    If I play it today in VP 9.21 or VP 9.9, the original version is fine.   

Edited by DJRobX, 16 May 2014 - 02:51 PM.

[mukuste]

But then the behavior should be the same in 9.2.1 and in 9.9.0, right? I hope that people are crosschecking between these two versions because otherwise we're chasing ghosts here.

[Kernel]

Any advice to get less lag possible?

I've set fps limiter (vsync) to 0 (no vsync) and the "max prerendered frames" to 0. No AA, no anisotropic filtering, no ball trails, max textures to unlimited.

 

Thanks in advance

 

PS: I have an ati 6670 1gb

[mukuste]

0 disables the max prerendered frames setting, set it to 1.

 

I also recommend using some FPS limit to get smoother gameplay. I always use 120 but you can set it higher if your GPU is up to it (check on F11 display if the FPS limit is consistently reached, otherwise reduce the limit until it is).

[StevOz]

Here I've found the frame limiter introduces an ever so slight ball stutter and it plays perfectly stutter free without it and I've tried up to 360, I get over 500 on all tables, including AFM beta 5 and MB killer edition.

[garath]

Yes for me too, on a table like Medieval Madness (VP91x v2.4.3 by JPSalas, 2013), it's really better with Vsync to 0.

If using frame limiter, there will be a silght ball stutter. That's why I don't use it.

[Kernel]

0 disables the max prerendered frames setting, set it to 1.

I suspected it, thanks for the tip.

Is this setting also working for Ati cards?

 

Regarding the "max textures": default setting is "unlimited". For my 1Gb card is better to limit at 1Gb?

Edited by Kernel, 20 May 2014 - 01:02 PM.

[mukuste]

I haven't been able to test it on ATI cards, but I think it should work. Please report what you find out!

 

Yes, I would recommend limiting the texture size, although in general I think it will just error out if the textures are too large.

[Kernel]

Is there a "scientific" way to see if "prerendered frames" option is working or I can only "feel it"?

Which is the max value I can set (just to highlight the difference)?

[mukuste]

Is there a "scientific" way to see if "prerendered frames" option is working or I can only "feel it"?

Which is the max value I can set (just to highlight the difference)?

 

The "strongest" setting is 1, then as you increase the value, it does less and less. 0 means off, which is basically the same as setting a high number.

 

The "scientific" way is to use a program which can trace GPU activity, I used GPUView, but unless you are a developer, these kinds of tools are difficult to make sense of. Otherwise, only the perceived input lag will tell you if it's working or not. Also, the framerate should change slightly.

Edited by mukuste, 20 May 2014 - 03:00 PM.

[DJRobX]

But then the behavior should be the same in 9.2.1 and in 9.9.0, right? I hope that people are crosschecking between these two versions because otherwise we're chasing ghosts here.

 

In my case yes, the physics are the same between 9.2.1 and 9.9.0.    If you look through VP 9.2.1 release discussions you'll find people complaining about "ball too fast" issues there.    Point being, whatever is causing the physics to flip-flop between "faster" and "slower" didn't start with 9.9, it seems to be a much older legacy issue. 

 

In my case, I had the slower physics in 9.15.   These slower physics seemed incorrect as the Diner table is older than 9.15 and was unplayable.    The Diner table now seems to work as the author intended on my system in 9.2.1 and 9.9.0.

Edited by DJRobX, 20 May 2014 - 04:17 PM.

[Sunskift]

Thanks for the advice to try vsync set to 0! I have tried out a lot of different settings, but havn't tried 0 for qiute some time (guess that I forgot that that was an alternative). With vsync 0, the very faint stutter I have sometime noticed is gone. I know that vsync should bring screen updates to the same frequency as the monitor updates, but what is the "mechanics" behind NO vsync? Does that mean that the screen is updated as much as the GPU is capable off?