Basically, I go through a lot of graphics cards.
One thing I've noticed over the past few years is that NVIDIA SUCKS with VP.
I just installed my GTX 570, & in 10 minutes I've had more glitches & worse performance than I ever had with the 5870. I had the same problem with the GTX 280, which is why I got the 5870 to begin with. I thought the performance problem with NV & VP was a fluke, so I figured I'd go back to NV. Surely it was fixed by now.
Nope. With very latest drivers in all cases...
VP 9.12, 1600x1200, ball AA enabled, hardware rendering enabled, alpha slider set halfway...
With the 5870, a blank table runs about 6400 fps.
JP's Game Show, ran smooth usually near 800 fps.
With the GTX 570 OC (again a faster card with even more ram, 1280mb), a blank table only runs about half the fps of the 5870, maybe 3300 fps. Ugh.
JP's Game Show, stuttering in the 400's, again nearly half the performance of the ATI card.
Btw, I tried VP 9.08 as well, VP8 as well, same kind of results. ATI is blazing, Nv is not.
But you guys who know nothing else but Nvidia probably have no idea what you are missing out on & what a mistake it may be to be putting Nvidia into your desktops & cabinets.
& I AM A NVIDIA FANBOY SAYING THIS, & IT HURTS BUT ITS TRUE!
I have even messed with some settings, maximum performance is on. But this time around I may not bother.
As I mentioned, I went through this same thing with the GTX 280, so its not the new card or the latest drivers. VP just sucks with Nvidia for some reason. I also have this weird glitch with the ball drain on blank tables & probably others.
I wanted to mention this because people always say how you should get Nvidia for VP, & I never understood that.
The day I got my 5870, & saw my performance nearly double as well as not have a single glitch or crash since I owned the card... I should have never gone back to Nvidia. The truth is that from my experience, ATI's 5xxx series or later (& maybe the 4xxx series) absolutely dominates Nvidia in regards to VP.
I even saw a stutter or two on the blank table test. Ugh.
Btw, I got a 3.62ghz i7 930, & 6gb of DDR3. So it is surprising how VP isn't getting smoked right now.
Now what do I do. I did get the GTX 570 OC because I'm gearing up for games like Crysis 2 & other upcoming DX11 titles. So the 570 will give me a boost with those games, though VP is taking a major hit. This is especially upsetting because now tables once again can use all the FPS they can get, cuz like JP's Game Show, those alpha ramps need all the fps they can get. Thus such a huge performance drop is a pretty big deal.
I even had the 570 clocks on full blast (while ATI clocks were underclocked during VP, I believe, what is up with that? Why is Nvidia not going into a 2D mode or why is VP not letting it?). I miss the days where I had my best performance with onboard graphics, a single core cpu, & a CRT monitor (great for pinball, no ghosting). Maybe I should just build another tower, slap my old CRT on there, & put it somewhere else in the house & make it my VP station.
Have VP devs been made aware of this issue? I feel like I can't be the only one testing out high end cards from both manufacturers, so I'd be shocked if I'm the 1st to mention this. Though I have even mentioned this before when people say nvidia is the way to go, I've said "no it isn't", but I think I get outnumbered & ignored.
I feel like this is a VP problem, because from what I'm reading, other 2D benchmarking has both Nv & ATI handling anything 2D quite well. These cards are beasts, after all. So what is it with VP that is limiting Nvidia?
Plus I think its time people stop thinking ATI sucks with VP. Yeah I think that was true a few generations ago, but certainly not anymore.
While I still have both cards in my possession (if I don't sell one or the other), please let me know if I can do anything testing wise to help out to shed more light on what I'm talking about.
I'm also curious to hear the fps of other nvidia users on certain tables or on the blank table (note whether you use ball AA or not).
& from what I'm seeing the FPS are very accurate from what is shown to what I am seeing. If its a 40% performance drop (roughly) that I'm seeing, then the FPS is also down about 40% as well from my ATI card. It seems to correlate just fine.
I need to cut this post short but I didn't even get into the multiple table crashes I've had already as well. Oh, & noticeably longer loading times vs ATI, & I think I did mention the glitches as well.
Edited by rob046, 23 March 2011 - 08:56 AM.
Top
Contributor
are all trademarks of VPFORUMS.