That's why I bought the GTS 250 and was dissapointed not seeing an improvement in the apps I was interested in. It makes it hard to justify the expense for a non-PC gamer.
A GTS 250 is a fine entry level card for any PC user in my opinion. I don't see a reason to exchange it again. Your CPU on the other hand is the real problem, specially with a cpu hog like Dolphin.
(04-26-2011, 03:05 AM)kernel64 Wrote: [ -> ]A GTS 250 is a fine entry level card for any PC user in my opinion. I don't see a reason to exchange it again. Your CPU on the other hand is the real problem, specially with a cpu hog like Dolphin.
Oh, the card itself is a fantastic value. I'm just thinking I don't even need that much of a card in my system where all I'll be running is MAME and a few other emulators, Dolphin included. It will just be a $55 accessory if I get my rebate money. I can live with SFIV running at lower quality levels, since I'll be displaying to a CRT TV.
I'm just getting a case of buyer's remorse.
One way to know which is the bottleneck is to watch your CPU and video card usage.
Get GPU-Z, it has a function that lets you watch your video card usage. Play a Wii game. Compare your CPU usage vs your GPU (Video card) usage. If your GPU usage is not near 100% that means your CPU is the culprit, and vice-versa.
(04-26-2011, 05:05 AM)Mereo Wrote: [ -> ]One way to know which is the bottleneck is to watch your CPU and video card usage.
Get GPU-Z, it has a function that lets you watch your video card usage. Play a Wii game. Compare your CPU usage vs your GPU (Video card) usage. If your GPU usage is not near 100% that means your CPU is the culprit, and vice-versa.
That's a great tip. I'll do that tonight with both cards. The result with the GTS-250 will be predictable, i.e., very little GPU usage. I'm pretty sure the 8500 GT will not show full usage, though, or I would have seen an difference by switching over to the more powerful card.
It will be a fun experiment.
Thanks
I'd keep the card and look for a second hand cheap dual core. Maybe a lowend i3 if you're on the right socket type.
I'm stuck in the C2D generation, in terms of memory and motherboard. Sounds like I might need something that overclocks to over 3.5GHz to begin to approach full speed. I've been looking at Soul Calibur II, mostly, since I want to play it in my cab.
Good luck on your experiment, but I'm pretty sure your CPU is the culprit. Emulation is very CPU intensive since it has to emulate the WII's CPU.
(04-26-2011, 08:33 AM)Mereo Wrote: [ -> ]Good luck on your experiment, but I'm pretty sure your CPU is the culprit. Emulation is very CPU intensive since it has to emulate the WII's CPU.
Good news and bad news, I guess.
The good news is that, even with the old card, I'm running SCII at 50-60 VPS. I was looking at FPS before, but after reading the forums found that the game wants to run at FPS during fights (unldess I'm misinterpreting). Apparently, if my VPS is close to 60, I'm approaching full speed. The %GS shows 90-100%. Better news is the fact that I haven't even bothered to optimize the settings. On PCSX2, however, no luck. That does run slow, period.
The bad news is that the GPU-Z experiment yielded confusing results. CPU-Z was showing the CPU going balls out, regarding of what card I was using (expected. However, GPU-Z reported ~20% GPU usage with the GTS 250 and closer to 16% with the 8500 GT. I guess I was expecting the 8500 GT to be put to work at a 100%, after seeing the 20% on the better card. I don't understand why the GTS 250 should be working harder than the 8500 GT. Truly bizarre. If it helps clarify anything, I'm running DX9 on my system because it is XP.
(04-26-2011, 12:30 PM)themanuel Wrote: [ -> ]However, GPU-Z reported ~20% GPU usage with the GTS 250 and closer to 16% with the 8500 GT. I guess I was expecting the 8500 GT to be put to work at a 100%, after seeing the 20% on the better card. I don't understand why the GTS 250 should be working harder than the 8500 GT. Truly bizarre. If it helps clarify anything, I'm running DX9 on my system because it is XP.
In my opinion, that means the CPU is the culprit since both cards revealed such a low usage.
One other way to test it is to play them in a low resolution, if high and low resolutions results in the same performance, the cpu is the culprit.
That's how I found that my CPU was holding back UT3 ( a PC game) even if I ran the game at a low res of 640x480, I had the same FPS. After I upgraded the CPU, the performance was much better.
Which PC game do you play? Perhaps you can also test those and see.