77 replies to this topic

[flpennstater]

I am an nVidia user as well. Wasn't there a problem with ATI cards until only recently which has been sorted out by the Catalyst drivers? Maybe now it is time to give ATI a go? Seeing some of the impressive FPSs quoted makes me wonder.

I know a lot of people in this thread are saying they run nVidia fine, and I did for many years on a low spec PC using desktop tables. It was only when I started introducing a 2nd screen and recently a 3rd screen that I have gone on a constant 'life-long' tweaking quest, disabling services, switching backglass to 16-bit, setting infinitty, overclocking the processor, standing on one leg, etc., etc. to get a stutter free game. The tables look so good you forget you are not playing a real table, until you get a stutter... and then you are reminded. I think most nVidia people here have had to tweak their PCs a fair bit to get where they are now - especially for the 2/3 screen setups.

I am not complaining about all this effort - in some ways it makes the hobby all that more enjoyable (). Also we are using the software in ways it was never designed to work.

I think the nvidia and ATI cards from the last 2/3 years are power monsters, and they both should be able to run a 2/3 screen setup with their hands tied behind their backs. All this CPU and GPU power and when you monitor the CPU and GPU they are hardly being utilised. There has to be some kind of communications bottleneck between the apps.

An nVidia and ATI side-by-side FPS comparison would be very interesting. It certainly sounds like ATI should be taken seriously from now on.

I am having issues where all of a sudden there is a delay in the time from when I press the flipper buttons and the flippers actually move. I have been running a 9800 nvidia card for a while with a 2 monitor setup and the 182.5 drivers. I got an ion nvidia card to run the backglass and to have the 9800 just do the 37" playfield. This didn't fix the flipper issue. Also, using the new driver makes it so the DMD doesn't stay where I position it. Should I be going back to the 182.5 drivers or is there a way to use the newest drivers and have the dmd positioning stay? Instead of these other improvements being done on VP, I'd like to see it stabilized so that you don't have to tweak so much just to get basic things like a dmd to stay in place.

A lot of people have responded with what nvidia or ati cards they are using but not what driver version. This was always the key in the past. Has something changed? There should be a sticky specific to what drivers to use with each card brand (nvida or ati) as to run vp at its best.

[lettuce]

Well i got an ATI 6970 2GB card about 2 months ago for my desktop pc, and can not for the life of me get it to work with VP correctly loads of gfx glitches all over the tables

[zebulon]

Ati 5770 running the playfield,5670 (or 5760...not sure atm) running dmd and uvp backglass,
vp908 (I know....but I've been busy building the cab and not playing much),
default settings on driver ver. 10.4,
phenom 555 x2 unlocked to x4 running at 3.6ghz
8gig ddr3 1066
win7 64bit
running the memory tweak that kills the offscreen error (don't remember the name)
running prifinity (doesn't seem to matter much)

...no issues with stutter or screen drawing, can almost play monopoly stutter free.


Maybe the extra apps you run in the background force vp off to a different core and solve the fps issue?

[bitupset]

What are you guys using to check the FPS of VPinball tables? i assume its something overlayed the table while playing...is it built in VPinball or an external utility?

[Bob5453]

What are you guys using to check the FPS of VPinball tables? i assume its something overlayed the table while playing...is it built in VPinball or an external utility?

Press F11

If Wizards_Hat reads this post, I didn't see that in the VP9 Guide.

[bitupset]

Thanks Bob. this is my desktop machine, win7 x64, 270.61 drivers, dual 9800gt SLI'd.
Cueball Wizard is the table...so framerate is 680? having a little difficuly deciphering that number...

[sleepy]

I am an nVidia user as well. Wasn't there a problem with ATI cards until only recently which has been sorted out by the Catalyst drivers? Maybe now it is time to give ATI a go? Seeing some of the impressive FPSs quoted makes me wonder.

I know a lot of people in this thread are saying they run nVidia fine, and I did for many years on a low spec PC using desktop tables. It was only when I started introducing a 2nd screen and recently a 3rd screen that I have gone on a constant 'life-long' tweaking quest, disabling services, switching backglass to 16-bit, setting infinitty, overclocking the processor, standing on one leg, etc., etc. to get a stutter free game. The tables look so good you forget you are not playing a real table, until you get a stutter... and then you are reminded. I think most nVidia people here have had to tweak their PCs a fair bit to get where they are now - especially for the 2/3 screen setups.

I am not complaining about all this effort - in some ways it makes the hobby all that more enjoyable (). Also we are using the software in ways it was never designed to work.

I think the nvidia and ATI cards from the last 2/3 years are power monsters, and they both should be able to run a 2/3 screen setup with their hands tied behind their backs. All this CPU and GPU power and when you monitor the CPU and GPU they are hardly being utilised. There has to be some kind of communications bottleneck between the apps.

An nVidia and ATI side-by-side FPS comparison would be very interesting. It certainly sounds like ATI should be taken seriously from now on.

I am having issues where all of a sudden there is a delay in the time from when I press the flipper buttons and the flippers actually move. I have been running a 9800 nvidia card for a while with a 2 monitor setup and the 182.5 drivers. I got an ion nvidia card to run the backglass and to have the 9800 just do the 37" playfield. This didn't fix the flipper issue. Also, using the new driver makes it so the DMD doesn't stay where I position it. Should I be going back to the 182.5 drivers or is there a way to use the newest drivers and have the dmd positioning stay? Instead of these other improvements being done on VP, I'd like to see it stabilized so that you don't have to tweak so much just to get basic things like a dmd to stay in place.

A lot of people have responded with what nvidia or ati cards they are using but not what driver version. This was always the key in the past. Has something changed? There should be a sticky specific to what drivers to use with each card brand (nvida or ati) as to run vp at its best.

Have you tried deleting the \vpinmame\nvram\ .nvram files.
If you have old .nvram files from before a hardware or a software upgrade, including newer versions of VPinMame, sometimes the old .nvam file for a table will no longer be compatible.

Do you have Windows Accessibility Sticky Keys or a joy-to-key program installed or enabled. Sticky Keys can cause flipper slowdowns.

If you have a keylogger virus, that can also cause keyboard slowdowns.

[Zablon]

Anyone tried taking these FPS with V-Sync on? It gives you the true 60/60 FPS as opposed to the insane number that isn't real. I don't notice any difference in visual. One thing that also needs to be stated just because I'm not sure if it's obvious, but if you aren't running a "cab" setting (read FS versions) with multiple monitors, then please state this. Just saying "it runs smooth" can be misleading.

I've posted about this in an other forum but I'll reiterate here.

My test set up was\
Win7 32-bit
Pentium D 3.2ghz,
2gb ram
Radeon 4870
1920x1080(PF), 1024/1024 (BG) - Using UVP

Cab Setup:
Win 7 64-bit
AMD Phenom II X2 555 BE @ 4.0ghz
4gb @ 1666
EVGA GTX 460 1GB Superclocked (oc'd to 880/1890)
ASUS GTX 460 768MB (oc'd)
Running latest drivers
1920x1080(PF), 1920x1600(16-bit-BG), 640x480(16bit-DMD) - Using UVP

This isn't an apples to apples comparison but overall I would say the old/slower set up runs almost as good as the new setup (granted 1 less screen).

Initially everything ran bad on the cab so I did all the CPU affinity, patching to use 4gb etc. I watched Prefinity as I played tables and this is what I saw:
Most the tables I've tried so far play flawless, granted I've only done around 15 of them.

CV stutters for the first minute or so pretty badly, but if you wait it out, it clears up and plays as good as any other and I never see it stutter again for 30 minutes of play. If I close the table and reload it, it stutters again.
What I noticed was I had (in this case 1800-2300 FPS) high fps, but right as a stutter hit it would drop to 100FPS. CPU utilization at that point went up, but it did not go anywhere near 100% on any core.

I see this same stutter on my ATI test system, but FPS (capped at 60) only drops to 59 - this being a slower system sees stutters on more tables, but usually just at the beginning after a ball launch as with the cab table.

My question is, those seeing stutter, does it ever clear for you if you just keep playing? This would point to something other than CPU/GPU/Drivers. In some other threads it is looking more and more like simply 32 to 64bit is contributing to the issue.

If I get some time I will try throwing my 4870 in the new system and see what the outcome is.

[rob046]

Actually the 60 fps number with v-sync on is the number that isn't real, you have it backwards. The higher number without v-sync, that is your true fps. V-sync limits fps to 60 fps (or whatever your refresh rate is), so to get an accurate benchmark with VP or any other program, v-sync needs to be off.
I don't have v-sync or triple buffering enabled with VP because it doesn't seem to make use of it. I never used v-sync & I've never seen any screen tearing in VP, so it makes sense to disable all that stuff, save some video ram.

Trying to decipher what the FPS number means in VP could be confusing. After all, with most games if you are hitting 60 fps, things will be smooth. But in VP, if you are getting 60 fps (without v-sync), then with VP9 at least it will likely mean you have a lot of stuttering.
For whatever reason, the magic number that I like to see in VP is about 300 fps or higher. Start dipping below that & I start to see slight stuttering, though I am extra sensitive to stuttering & ghosting, so I am super picky. Others might think what I find unplayable is actually pretty smooth.

VP's fps are still useful for various reasons even if the numbers it gives aren't like what other PC games give. I use VP fps all the time to see if tweaks I do are helping a certain table perform better. Often times I'll take one very intensive table & use that as my test table for performance, my benchmark. & you suspect a table is playing poorly for you, you can just go on these forums & ask around & see how your fps stacks up to others fps for that table. This can tell you if the table actually is playing OK, or if you got issues you need to work out.

Some notes on glitches & performance issues... There is such thing as bad cards (some cards just suck right out of the box, have defects & may have artifacting or overheat too easily). Always make sure you card is good when you get a new one by running some intesive benchmarks, things like furmark or 3Dmark, & monitor temps as you run them as well.
I believe some people just have bad cards that cause some odd issues.

It can also be a bad driver install. It used to be good practice that when installing a new driver, you first uninstall the old driver, reboot PC, then install new driver, reboot again. To be safe sometimes I uninstall the old driver, reboot, run a cleanup tool like ccleaner (to remove old driver junk & registry entries), then install the new driver. But these days that doesn't seem to be recommended anymore unless you notice that you have issues. Not sure why this is.
Also, keep that computer case cleaned out, you get too much dust in your pc fans, or worse your gpu fan(s), that also can cause weird graphics stuff to happen due to overheating. I highly suggest a program like MSI Afterburner as a gpu monitoring software. Monitors temps, clocks, & does a ton of other stuff. & I was able to use that for both my nvidia & ATI cards. I think it is based on the old rivatuner software, which used to be super popular, now it just has a makeover & a new name.

Then let's not forget how crappy A/V software can cause issues, or spyware/malware/viruses can cause performance issues. Or just too much crap running in the background. As stated earlier in this thread, I certainly had programs that actually boosted VP's performance for odd reasons we are still unsure about. Fixed in the latest driver release, but it goes to show that background programs can have a positive or negative effect. I always take the time to go into "msconfig" & disable all the startup crap except a couple things I actually need. Then after an OS install, I disable a number of services I don't need (google "black viper windows services" for a guide on that).

Point is, we shouldn't always be too quick to blame the gpu or drivers themselves, when it could be a bad install, bad card, or the other things mentioned.

I also agree that with all the cab & desktop users out there it is more important that we list our specs, putting them in your sig would be a good idea, just the basics (cab or desktop, what OS?, cpu, gpu, & 32 or 64 bit). There seem to be a lot of issues that cab users have that desktop users don't, & even 1 or 2 that might be desktop only issues. Though of course when you are a cab user & dealing with multiple monitors and/or multiple GPU's, there is always the liklihood for more issues, especially when trying to get all that to work with software built in dx7 like VP, or even FP which never got fully optimized before Black gave up on it. I will say that for the most part, both nvidia & ATI seem to be doing a much better job with drivers these days. I'm amazed by that because their focus is on much newer, higher end PC games, yet they still manage to fix the occasional bug for some old games or 2D games, whether by on purpose or by accident!

Btw, something that I noticed on my PC, Zablon you mentioned above about setting affinity for VP.
I'm a desktop user & I know cab users might have more benefit from assigning cores to different things, but I tried setting VP to 1 core, 2 cores, & 4 cores. None of those performed as well as me leaving affinity alone, having all 8 of my cores enabled. I have an i7-930, so I got 4 real cores & 4 hyperthreading cores (virtual cores) for 8 total.

Somehow, even though VP is a single core app, my having hyperthreading & all my cores enabled gives a nice boost in performance. Same with compatibility mode, I just think some things should be left alone unless absolutely needed. I get best performance by not using any compatibility modes.
The only exe setting that gives me a little boost is "disable desktop composition".
What works for me may not work on other machines (I'm on win7 x64), but this is just an example of things to try, stuff you can mess with to see what works best for you. If you get too crazy with trying to tweak though, it can actually hurt performance.

For the most part VP should work fine just as it is, as long as you have a somewhat clean system, whether you are running 32 bit XP or 64 bit win7. If you are having some issues & its been 2+ years since you've last had a fresh OS install, that could certainly cause problems. I use my PC everyday, & I find it absolutely essential (even with the best A/V & cleanup tools run regularly) to get on a fresh OS install at least once a year. These days it isn't even that much of a headache if you have a good backup drive & software.
I've never had a PC get faster after a year+ of use, always gets slower & wonkier! But then I do a ton of things on my PC & use a lot of different software & fill up a ton of HD space. Nothing breathes new life into a PC like a fresh OS install.

[cupid]

Actually the 60 fps number with v-sync on is the number that isn't real, you have it backwards. The higher number without v-sync, that is your true fps. V-sync limits fps to 60 fps (or whatever your refresh rate is), so to get an accurate benchmark with VP or any other program, v-sync needs to be off.
I don't have v-sync or triple buffering enabled with VP because it doesn't seem to make use of it. I never used v-sync & I've never seen any screen tearing in VP, so it makes sense to disable all that stuff, save some video ram.

Trying to decipher what the FPS number means in VP could be confusing. After all, with most games if you are hitting 60 fps, things will be smooth. But in VP, if you are getting 60 fps (without v-sync), then with VP9 at least it will likely mean you have a lot of stuttering.
For whatever reason, the magic number that I like to see in VP is about 300 fps or higher. Start dipping below that & I start to see slight stuttering, though I am extra sensitive to stuttering & ghosting, so I am super picky. Others might think what I find unplayable is actually pretty smooth.

good written!

I have read something about fps counters in a programmers book. If I'm back at home, I'll search for that and try to summarize it. But you are right: FPS-Counters can be confusing, since there are several ways to calculate FPS and there are several implementations of FPS Counters outputting different things, which can be very confusing. The main Game loop in VP is several years old and i'm sure that a Gameloop would be implemented differently today -> not meaning that it would for sure give more smoothness, but it may be easier to maintain and optimize.

Cupid

[cupid]

OK, i'll try to summarize the chapter from the book, but the book is for XNA-GameProgramming and no easy stuff.

Since i don't want to get in trouble with copyright, i will say it in my own words:

Graphics performance is often misunderstood: People never think it could be physics performance or audio performance. Many do not understand what graphics perfomance is. The problem is that a GPU does often work in parallel to the CPU. Due to parallelisation of tasks inside a GPU, drawing twice as much on screen may have no impact in FPS. Newest GC do 1000s of things in parallel.

If you measure it, you may notice that the actual draw calls are fast (in some millisecs done), and the game should be running like hell, but it stutters (and has bad FPS). In a modern game loop the program is executing 4 things per loop -> 1. Update (physics and game logic), 2. Draw (calculate the graphics), 3. Present (present the graphics) and 4. Drawing.

While a frame is calculated (first three things), the frame before that frame is drawn on screen.

That is for modern game loops. maybe this is the same with our gameloop and DirectX 7 - i don't know, but i hope ;-). So lets assume this.

If now update, draw and present takes more time as the Drawing, you will have lower FPS. You will also get lower FPS, if the drawing take more time to complete. The problem is: how can we find out where the problem lies? The Book clearly states: "Unfortunately, it's not as easy to answer as you might think.".

Then the book becomes very XNA centric, not helping us in DirectX 7. To measure correctly, you have to turn off VertRetrace (which you did) and set the timestep for the game loop to as short as possible (i think this is done automatically, but only partially understood that in code).

They implement some measuring classes in the chapter of the book, but the author also advises to use special performance monitor programs (which one dev has done some days ago).

All that is from a very new XNA book... there are serveral years between VP and XNA. We might have problems that newer games do not have. Maybe we have heavy traffic on the Graphics bus... Higher as newer games. Maybe this whole 2d->3d emulation thing, VP is built on top, does give us more problems as we might think. Someone wrote that with a newer graphic card he gets less performance. Also the gap between ATI and NVIDIA should not be so high. Maybe this whole 2d copying (blitting) can not be optimized as much as 3D Stuff today (due to parallelisation (maybe) not possible).

There are many open questions.

I recently tried to play Diablo 2 on a modern computer, and could not get it to run as fast as several years ago. Do you think they will rewrite the engine for Diablo3? No, they will write a completely new engine based on modern standards, but they also have the programmers and (some) money to do that. We can try that, but we are not big business.

A thread like this really helps us devs to get an idea what could couse troubles on some machines. So thanks for this. But i don't think that there will be a "Hey that's it... i changed that and now we have 300% frames on all computers.". But maybe we find that switch in the code and turn it on, but i believe that there are several switches in the code and we have to try very long to get the best combination for all. We recently had a commit, that increased performance much at many computers, but we also had reports that reported a performance decrease. We could not yet find out why.


I hope this gives some background (although i know that i could not answer anything with this).

Cupid

[kruge99]

Thank you for the explanation Cupid, while I don't understand how to write software code, your description makes 'logical' sense to me.


Best Regards,
Todd.

[Shooby Doo]

I've been following this thread but didn't have anything to add until now. I was playing JP's new Flinstones for VP9.12 and it was playing a little choppy, so I pressed F11 and the frame rate was jumping between 60 to 200, and before this, everything had been playing real nice. So I thought, "why don't I try opening firefox like they said in that thread", and lo and behold, it jumped between 300 to 450 after that. It's a GTS450, detailed specs in sig. Just thought I'd throw that out there.

[Zablon]

Thanks Rob and Cupid,

Yes, I wouldn't be surprised at all if it is simply old code running on new code. I've often wondered if it isn't related directly with VP but with PinMame as emulation can be tempermental.

@Rob - Actually in terms of smoothness I did not notice any difference between having Vsync on or off. Actually I didn't even know VSync was on until I checked the FPS on that system.
Interestingly, I took a bunch of FPS last night to compare and my FPS on my cab system is much lower than it was prior.

In CV for instance, before was 1800-2300, now it was running 414. NOTHING CHANGED. The only thing I can think is prior I had IE open to a bunch of pages, and this time I had just rebooted. That is the only thing that was different. I know there's that whole "flash" thing, maybe it's true...

[rob046]

@ Cupid... Keep that brain churning! I'm really happy you & the other devs are taking the time to try to solve some of VP's odd issues it has probably had for years. & I know it can't be easy, trying to keep a DX7 program optimized with all this new tech & hardware. Beefy graphics cards with DX11, multi core cpu's, 64 bit OS's... all things that weren't common back when VP first started.
If you think you are onto something, & have had some good results, maybe don't be too quick to disregard these changes. After all, the few people reporting performance decreases might have other issues they need to deal with. Or maybe they need to update their drivers or perhaps had the odd issue that I had, & now all these other people are now noticing.

@ Everybody with this FPS issue where firefox or other things speed up peformance, have you tried the new nvidia drivers? Fixed it for me & others. Whether the new driver helps everybody out or not, I'd sure like to know one way or another. & it might help the devs figure out what is causing this weird conflict with other drivers & maybe even help us get VP more optimized in the future. At first when I posted my problem, I kinda knew people would think I was crazy but I wanted to mention it anyhow cuz it had happened before.

Now it turns out I'm far from the only one who has or had this issue, & its great that more people are noticing it or are hopefully knowing to check for this issue after reading this thread or the other threads this is mentioned in, cuz then when you get it fixed up you get a nice performance boost, & more importantly you get consistently good VP performance that doesn't depend on whether or not you have firefox or some odd program opened!
Sometimes if you post an issue, you just might be pointing something out that other people just aren't catching.

I keep strongly suggesting that everybody keep an eye on their performance & fps. On most systems these days, most tables should run nearly perfect. But please note on these forums if you have a hunch that things aren't running up to par.
One thing I noticed with this issue I had (where media players & certain browser pages would fix it), is that even well running tables might run mostly smooth, so the problem isn't super obvious. One thing to look for though is intermittent stuttering. Like things may be smooth for a few seconds, then you get a stutter regardless of what is going on with the table, even a slight stutter. I noticed I was even getting occasional small stutters on the blank "new table" in VP9 despite my fps still being thousands.

In fact, the new table isn't a bad universal VP benchmark table. For me when this issue was happening, I was in between 2k and 3k fps, then when the issue is cleared up I was closer to the 5k range. In fact, with my cheaper, slighly lower end ATI card, I was actually hitting well over 6k with the blank table even with ball AA on & maxed out driver settings (minus AA, which can drop fps a bit & also cause some glitching). There is another little tip for those wanting a bit more performance or you notice some minor (or major) visual glitches. Make sure you don't have AA forced on.

[gear323]

Here are my results for what they are worth.

PC:
Windows 7 X64 ULT
ATI 5870 1 GB
Ram 6GB
Core i7 Extreme OC 4 GHZ

VP Settings:
VP Version 912
Resolution - Full Screen 1920x1080 32
Alpha Ramp slider in middle
Hardware Render - enabled
Textures- unlimited
Draw Ball Shadows - enabled
Draw Ball Decals - enabled
Antialis Ball - enabled

Results wtih blank table:
Ball sitting in outlane waiting - 6457 FPS
Ball moving around table - 7357FPS

Any yes, I too found it strange that the FPS were better with the ball moving rather than just sitting there waiting to be hit with the plunger.

Results wtih JP Flintstones 1.0.2 FS:
Ball sitting in outlane waiting - 678 FPS
Ball moving around table - 592FPS

I tell ya, the ball moves like silk at 7000 FPS on blank table... Maybe in 5 years all tables will play at 7000 FPS ..one can dream...

Edited by gear323, 28 June 2011 - 09:06 PM.

[Centaur Machine]

All that is from a very new XNA book... there are serveral years between VP and XNA. We might have problems that newer games do not have. Maybe we have heavy traffic on the Graphics bus... Higher as newer games. Maybe this whole 2d->3d emulation thing, VP is built on top, does give us more problems as we might think. Someone wrote that with a newer graphic card he gets less performance. Also the gap between ATI and NVIDIA should not be so high. Maybe this whole 2d copying (blitting) can not be optimized as much as 3D Stuff today (due to parallelisation (maybe) not possible).

There are many open questions.

I recently tried to play Diablo 2 on a modern computer, and could not get it to run as fast as several years ago. Do you think they will rewrite the engine for Diablo3? No, they will write a completely new engine based on modern standards, but they also have the programmers and (some) money to do that. We can try that, but we are not big business.

A thread like this really helps us devs to get an idea what could couse troubles on some machines. So thanks for this. But i don't think that there will be a "Hey that's it... i changed that and now we have 300% frames on all computers.". But maybe we find that switch in the code and turn it on, but i believe that there are several switches in the code and we have to try very long to get the best combination for all. We recently had a commit, that increased performance much at many computers, but we also had reports that reported a performance decrease. We could not yet find out why.


I hope this gives some background (although i know that i could not answer anything with this).

Cupid

Yeah, I guess in about 2000, it was polygons, then in 2005, graphics cards started to become really shader heavy, because that's what the latest games required lots of. I guess more recently, it has been tessellation. These are things that aren't going to benefit VP. Now when you dedicate a large portion of a GPU to shaders, you have to sacrifice something else. So absolutely, an older card may have better performance in certain areas compared to newer cards. However, older cards weren't designed for high resolution monitors, and so this is sort of a barrier for people with ~HD monitors.

In any case, VP just seems so simple. But all of the features people add to their tables really adds up though, to be sure. Let's also not overlook the fact that PinMAME is emulation, and emulation is extremely CPU cycle hungry - and this goes for the emulated audio too. Physics is another thing that requires a lot of processing, so these things all add up.

But that said, I haven't had any issues with even though most demanding tables with my new GTX 460 while under-clocked. But I'm only running 1 monitor at 1080p, so who knows. Is there a particular table that is the most demanding that we can use as a benchmark, ALA Crysis? Wasn't there a recently Monopoly table or something here that was hardcore? Or maybe it was JP's Big Game, the one I keep hearing people refer to.

On viewing FPS: I pressed F11 while playing a table, but there are a lot of numbers there. I think what I'm seeing is ~370 FPS while playing Tales of the Arabian Nights, with 8X AF at 1440x1080 on VP 9.21

AM I DOING IT RITE?

Edited by Centaur Machine, 29 June 2011 - 08:26 AM.

[R_L]

Here are my results for what they are worth.

PC:
Windows 7 X64 ULT
ATI 5870 1 GB
Ram 6GB
Core i7 Extreme OC 4 GHZ

VP Settings:
VP Version 912
Resolution - Full Screen 1920x1080 32
Alpha Ramp slider in middle
Hardware Render - enabled
Textures- unlimited
Draw Ball Shadows - enabled
Draw Ball Decals - enabled
Antialis Ball - enabled

Results wtih blank table:
Ball sitting in outlane waiting - 6457 FPS
Ball moving around table - 7357FPS

Any yes, I too found it strange that the FPS were better with the ball moving rather than just sitting there waiting to be hit with the plunger.

Results wtih JP Flintstones 1.0.2 FS:
Ball sitting in outlane waiting - 678 FPS
Ball moving around table - 592FPS

I tell ya, the ball moves like silk at 7000 FPS on blank table... Maybe in 5 years all tables will play at 7000 FPS ..one can dream...

What Catalyst driver do you use gear323?

I have a similar setup and having problem running the tables at 32-bit colour depth.

My spec

Catalyst ver. 11.6
2D Driver Version 8.01.01.1162
Direct3D Version 7.14.10.0841
AMD Radeon HD 6970 with 2048 MB GDDR5
CPU Intel® Core™ i7 CPU 920 @ 2.67GHz, 2668 Mhz, 4 core, 8 logic cpus
I get approx. 620 FPS sitting in the lane on JP´s Flintstone ver 1,02 at 1920x1200. There is no change in fps if you compare 16 or 32 bit but the 32 looks funny. Here is a description of the problem I have running 32bit colour depth. http://www.vpforums....mp;#entry104055

Only workaround to the 32 bit col problem I have found, are change to 16 save table exit and then fire it up again and it works fine but in 16 col depth ofc...

[Centaur Machine]

Here are my results for what they are worth.

PC:
Windows 7 X64 ULT
ATI 5870 1 GB
Ram 6GB
Core i7 Extreme OC 4 GHZ

VP Settings:
VP Version 912
Resolution - Full Screen 1920x1080 32
Alpha Ramp slider in middle
Hardware Render - enabled
Textures- unlimited
Draw Ball Shadows - enabled
Draw Ball Decals - enabled
Antialis Ball - enabled

Results wtih blank table:
Ball sitting in outlane waiting - 6457 FPS
Ball moving around table - 7357FPS

Any yes, I too found it strange that the FPS were better with the ball moving rather than just sitting there waiting to be hit with the plunger.

Results wtih JP Flintstones 1.0.2 FS:
Ball sitting in outlane waiting - 678 FPS
Ball moving around table - 592FPS

I tell ya, the ball moves like silk at 7000 FPS on blank table... Maybe in 5 years all tables will play at 7000 FPS ..one can dream...

What Catalyst driver do you use gear323?

I have a similar setup and having problem running the tables at 32-bit colour depth.

My spec

Catalyst ver. 11.6
2D Driver Version 8.01.01.1162
Direct3D Version 7.14.10.0841
AMD Radeon HD 6970 with 2048 MB GDDR5
CPU Intel® Core™ i7 CPU 920 @ 2.67GHz, 2668 Mhz, 4 core, 8 logic cpus
I get approx. 620 FPS sitting in the lane on JP´s Flintstone ver 1,02 at 1920x1200. There is no change in fps if you compare 16 or 32 bit but the 32 looks funny. Here is a description of the problem I have running 32bit colour depth. http://www.vpforums....mp;#entry104055

Only workaround to the 32 bit col problem I have found, are change to 16 save table exit and then fire it up again and it works fine but in 16 col depth ofc...

I'm surprised you even tried 16 bit. That's something that was a low setting on Windows 3. Hehe.

It may fix part of the problem, but I just don't think 'bit depth' is the cause of the actual problem. And the only thing I can suspect is the video drivers for your card. Maybe someone else has more ideas.

[bladexdsl]

i'm using gtx 460 for pf and gt 550 TI for bg and dmd on my cab. everything runs smoothly running windows 7 64 bit, 6gb ddr3 ram and phenom x4 3.2ghz as long as i keep UVP turned off. with uvp on though get lots of ball stutter on some tables CV being the worst.