318 replies to this topic

[htamas]

Wow. This table is definitely going to push your video card.

 

Actually, I'm surprised that on my measly system, running on just a (not overclocked) GTX 750 Ti and Phenom II X4 965, how well this table plays without any stutter or problem - even with the modulated flashers enabled and during multiball.

I only have Quality SMAA enabled though, no brute force SSAA.

[The Loafer]

This has been talked about before, the flipper physics issue is all about trade offs. Yes you can get better back handed shots with some tweaks but then other parts of the games are way worse and the balance is shot. It is true that personal preferences and play style may factor into how we believe flipper settings should be done but personally , on my PCs, this is beautiful work

Something to remember but lag wise Right off the bat, VPX and us the players are negatively affected by the monitors we use. There is lag there that just isn’t there in real life, and it can’t be fixed by any script, it’s a hardware issue. It’s one of the reasons why VR has such an advantage, no lag. Hopefully someday soon lag-less large screens are coming for pincabs.

Anyway back to the physics, to me Roth, these are the best yet so thanks for these but also thanks for being open minded in understanding that if there are ways to improve them through constructive criticism, you are more than happy to explore alternate ideas. This is how flipper settings have improved over the years and this is how we will get from A+ to A++

[TNT2]

 

Wow. This table is definitely going to push your video card.

 

Actually, I'm surprised that on my measly system, running on just a (not overclocked) GTX 750 Ti and Phenom II X4 965, how well this table plays without any stutter or problem - even with the modulated flashers enabled and during multiball.

I only have Quality SMAA enabled though, no brute force SSAA.

 

I think you are right. I think it may be some of my settings. The first couple of times I played it no stutter, but later after reloading it some stuttering. Or maybe my cache.

Edited by TNT2, 09 April 2020 - 05:01 AM.

[Greynurse]

Great work, Fleep. The sounds are awesome. No problems running on my GT9600, but I've tweaked card settings and VP settings to within an inch of their lives. I really need to upgrade to a mid level gaming machine, but I must sell the house and move first. (Silly proirity, really) Any other sound improved tables in progress?

 

Cheers

 

Nursie

[Fleep]

I could tell you, but then I would have to kill you

I’m looking for someone who owns Whirlwind real table that can help with some recordings I need. That’s regarding flupper rebuild project of that table.

Other than that - I do have another table up my sleeve but I think I will prioritize flupper rebuild project first.


Sent from my iPhone using Tapatalk

[Fleep]

I just thought I’d add what may seem obvious regarding Roth’s work here. He has so much dedication to the physics it’s really unbelievable. One thing I can say without a doubt about Roth is that he is always willing to improve things, very open minded for constructive feedback, and would not give up on going challenges.

Yup. people opinions vary and I know for sure we could never satisfy everyone and these physics scripts are very deep and share various trade offs.


Sent from my iPhone using Tapatalk

Edited by Fleep, 09 April 2020 - 07:41 AM.

[ClarkKent]

I think you should simply add a switch at the beginning of the script to enable or disable the physics tweaking. I think a lot of script tweaking  doesn’t make the physics more realistic but only different - some people like it very much others not that much. The switch would make both happy...

[JLouLoulou]

I just thought I’d add what may seem obvious regarding Roth’s work here. He has so much dedication to the physics it’s really unbelievable. One thing I can say without a doubt about Roth is that he is always willing to improve things, very open minded for constructive feedback, and would not give up on going challenges.

Yup. people opinions vary and I know for sure we could never satisfy everyone and these physics scripts are very deep and share various trade offs.


Sent from my iPhone using Tapatalk

we have never says that Roth's made bad work.. We are also very impress by is work..

But I have an idea of the problem.. First, i'm totaly agree with the nfozzy physics and Roth's tuning... As he says for the script, it need to make compromise to be near the reality and what vpx it offer.

Some here talk about lag.. I have "no" lag, like we imagine it, i have no problem on any other table or game. But the reality with our cab, is there some little lag due to software and hardware compare to real pinball. This is why there is fast flip or other tweak to try to compasate it. So, if the nfozzy script need more time to calculate flipper bats physics, it's add calulation lag to lag already here. And i think i don't know why, the script add more lag in some player than other, even with good config.

I made a test at my workplace, with a poor PC, i3 550 3.2GHZ, 8Go RAM, GT 610, all graphics option of or at minimum, by the way it work like it could be!!!! But it work at 100fps... And to compare, if i play at 60fps, it's like if i play at 30fps for flipper bats physics, it's horrible. I think there is a glitch somewhere, where nfozzy calculation not work correctly on some config. I will do the same test on my cab when i return to home, to test it without any graphical option, or without vsync to have 110/120 fps.. 

So i watch video from papa, for sure, there is lot of bump and rolling on flipper bats, it's hard to control the ball. So it's good for that. Maybe a little too much on vpx version . ( it's a joke ; )

But but i maintain some of us have responsive problem, because, even a flick pass not working correctly due to "lag".

So in the same way that say roth, we need to make compromise for physics to have a good deal, i prefer to have reactive bats like real without possibilty to make some trick, instead of you can make all trick but with latency far from reality. 

So i think it could be good to have the choice. I never, i pretend your physics is bad..; No, they are what we need. But Sorry, if that not work correctly for some players. It's not your fault, VPX is not perfect, It doing already more like any other pinball game, and that, it's perfect!

I'll will post my future test, i swear i can fin'd somthing. But keep working like that. We are here to push you to make the best and help you.. Not to descraditate you

Edited by JLouLoulou, 09 April 2020 - 11:49 AM.

[rothbauerw]

Here are my thoughts on flipper lag.

 

• In my experience, there isn't much flipper lag in VPX anymore.  Some, yes, but not much.
• A real flipper on this era machine takes 25-30ms for a full flip.  Determined through video analysis.
• The flippers with my current flipper settings take somewhere between 28-33ms for a full flip.
• In general, the current flipper physics shouldn't have any more lag than any other VPX flipper with one caveat.
  • Coil ramp up is a VPX flipper property that simulates the charging of the solenoid coil.  When set to a non-zero value, it takes "x" ms for the flipper strength to ramp up to it's full value.
  • This ramp up can make the flipper feel slower than a flipper without coil ramp up set.
  • However, a flipper without coil ramp up only takes about 10-15 ms for a full stroke, much faster than a real flipper.
  • A real flipper coil has two "charge" cycles, one before the solenoid plunger moves, which is slower, and one after the solenoid plunger moves, which is faster.
  • So I use two coil ramp up values, 8.5 before the flipper moves, and 2.5 as soon as the script detects motion in the flipper.
  • The script is set to run once every millisecond.  If for some reason, it's not running when expected, that larger coil ramp up value may be applied for too long.
• Other than that, there are no differences between a normal VPX flipper and the settings I use that would cause perceived flipper lag.
• As Loafer mentioned, input is limited by the monitor refresh rate.  I can't do anything about that.  But even so, with the current flipper settings, I can seamlessly move between VPX and real pinball without having to "recalibrate" my timing.  The timing is different, but it is so small, that I'm able to adjust without thinking about it.
• Prior to nfozzy physics, moving from VPX to a real machine was like learning a whole new game.  Shot trajectories and timing were completely different between VPX and a real machine.  After playing VPX for a few weeks, I could play a real machine for an entire night and never properly adjust to the differences in timing and trajectories.

One final note.  What I've done with flippers has all been based on video and data analysis of real flippers.  I look at trajectories, timings, and strength of shots.  I haven't done this willy nilly or based on what I think is right.  I base it on real world physics and video recorded results.  I've included a significant number of boundary cases, but not everyone.  I don't believe any other VPX flippers come close to where these currently are.  They aren't perfect, but they are pretty darn close.  And I'm willing to improve them if we can do it without sacrificing some of the advancements we've already made.

[Thalamus]

I find it interesting that you say that you can seamlessly move between VPX and a real machine. I surely can't. I need to spend quite a lot of time especially on shots where the ball comes down the inlane and are at speed. It would not surprise me much if the difference I'm experiencing is related to monitor refresh rate. On low speed, the scripted technique "rothfozzy" is way better than what I've seen on "regular" tables. I must admit though. Tip passing, either I'm really bad at it or there is still improvements that can be done. ( Typical having two balls on the flipper, the third comes down the inlane and then flicking it over to the other side)

[The Loafer]

"However, a flipper without coil ramp up only takes about 10-15 ms for a full stroke, much faster than a real flipper"

 

I wonder if because of monitor lag, a flipper without coil ramp up feels more like a real flipper to some because it ends up "compensating" for the monitor lag.  This could be how it feels.

I find it interesting that you say that you can seamlessly move between VPX and a real machine. I surely can't. I need to spend quite a lot of time especially on shots where the ball comes down the inlane and are at speed. It would not surprise me much if the difference I'm experiencing is related to monitor refresh rate. On low speed, the scripted technique "rothfozzy" is way better than what I've seen on "regular" tables. I must admit though. Tip passing, either I'm really bad at it or there is still improvements that can be done. ( Typical having two balls on the flipper, the third comes down the inlane and then flicking it over to the other side)

 

In my experience (very important distinction), I would have agreed with you previously. In fact, the difference between my older pincab and real pinball was responsible for rarily turning on the pincab during the course of a year or two.  I mean I could play it sometime but NEVER right after playing real pinball, it just didn't work.  Then I upgraded my LG 1080p playfield monitor for a 4K Samsung TV with lesser lag and it was night and day.  Although I sold all my real pinball games (sniff, goodbye Firepower, Goodbye Tron, Sinbad, Stern Star Trek, etc) I still play the odd real pinball and I find now I can go back and forth and it's near seemless.  To be clear, I don't doubt there is still some lag with the 4K screen, it's a hardware issue, there is going to be some lag here.  But the perceived lag is definitely improved.

 

Your TV/Monitor is going to impact the back/forth 100%.  Stick only to the TV/monitor, then you are ok the brain will compensate to some extent but back and forth is hard, especially if we have no lag.

 

Finally, perceived lag is such a personal thing.  At the last Ottawa Pinball and Gameroom show, there was a guy who brought his pincab and he said "no lag".  Jeesus, was there ever a lot of lag. To each is own, maybe some of us are more prone to noticing this, I dunno.

 

Bringing this back on topic, I haven't tried TOM yet in VR, will do so today, will comment back later if I find the flipper response even better.

[Genesis38]

I don't know that I can offer anything to the topic at hand, aside to say that we must not forget that VP will NEVER be like the real thing, but it sure is getting closer, this beautiful table a prime example. Why keep comparing real vs visual? Just enjoy the hardwork that many put forward in this hobby. I also can't speak for others, but I can't afford an $8,500 (CAN) for a Theatre of Magic pinball table, so I am quite happy playing it in VP form

 

Thank you again to Fleep, 3rdaxis, rothbauerw and many others for their incredible work on this. I can't stop playing!!

[rothbauerw]

"However, a flipper without coil ramp up only takes about 10-15 ms for a full stroke, much faster than a real flipper"

 

I wonder if because of monitor lag, a flipper without coil ramp up feels more like a real flipper to some because it ends up "compensating" for the monitor lag.  This could be how it feels.

 

 

 

No doubt in my mind this is true.  In the old days I would always set the coil ramp up to zero to make the flippers more responsive.  But then you give up other desired behaviors.  But playing side-by-side with a real machine, it became apparent the small sacrifice and was worth the improvement in create a more realistic experience.

 

I find it interesting that you say that you can seamlessly move between VPX and a real machine. I surely can't. I need to spend quite a lot of time especially on shots where the ball comes down the inlane and are at speed. It would not surprise me much if the difference I'm experiencing is related to monitor refresh rate. On low speed, the scripted technique "rothfozzy" is way better than what I've seen on "regular" tables. I must admit though. Tip passing, either I'm really bad at it or there is still improvements that can be done. ( Typical having two balls on the flipper, the third comes down the inlane and then flicking it over to the other side)

 

I can't either if I'm spending any time at all with standard VP flippers.  Of note though, I do find shots on the move to be easier on real machines.  Please make this change Thalamus and see if you feel like it improves "on the move" shots.

 

Const SOSRampup = 8.5 

 

to 

 

Const SOSRampup = 2.5

[onemanproject]

THX

[rothbauerw]

I think you should simply add a switch at the beginning of the script to enable or disable the physics tweaking. I think a lot of script tweaking  doesn’t make the physics more realistic but only different - some people like it very much others not that much. The switch would make both happy...

 

I'll make a deal with you.  I'll add an option to turn off my physics when you add an option to turn on nfozzy physics on your tables.  You first!

[bigus1]

3 beers and a few milliseconds don't matter so much

[3rdaxis]

3 beers and a few milliseconds don't matter so much

lol no doubt. 

[batch]

Thanks again for this fantastic table

 

Here is the backdrop I did for desktop users

 

[bolt]

Top as always, thank you batch.

[Thalamus]

@Rothbauerw : I just did a very short test game with SOSRampup = 2.5 - and I immediately feel more at home. Bare in mind though. I don't say that this is the correct setting. I say, it feels better being used to VP flippers. It might very well be that what you have as default is more correct to the real world. I really want to stress that I do - ALWAYS, need time to adjust going to real machines. I think you are doing god's work here.