Dolphin, the GameCube and Wii emulator - Forums

i7 920 at 4ghz, no HT
6gb 1600mhz ddr3
gtx580

Dolphin x64, Windwaker PAL version:
test 1: 81 fps
test 2: 60 fps
test 3: 95 fps

Off Topic: There'd probably be more PAL benchmark/submittals if the save was added to the first post

@Garteal, which page is the chart on, lazy i am.

Here you go Zee. Now contribute to the benchmark.

With my weak ass system, well since there's an atom there, might as well, it would really be nice if someone could move it to the first page or at least make a link that redirects there.

One of these days I'm going to benchmark my AMD C series netbook that a friend gave me (screen is broken). So we'll have bobcat results too. I'm a bit surprised at just how slow the atom results were. The IPC of the architecture in this test is well below that of a pentium 4. And when compared to benchmarks of other applications it shouldn't be. It should be inbetween pentium 4 and K8 IPC. I am almost positive that the nvidia ion gpu is bottlenecking the game even at 1x IR. The system has 5.33 GB/s of memory bandwidth shared between the cpu and gpu. The IGP will be lucky if it gets more than 3 GB/s. I would love it if the test could somehow be conducted again with a 0.5x (or possibly even lower) IR to confirm my suspicions.

Luckily the C series APUs IGP should be able to handle 1x IR in this game.

nintendo maniac 64 Wrote:Lawl. XD I would totally run this on a Pentium 4 machine if I had a PCI (not PCIe) GPU with shader 3.0, but I'm not even sure those exist.

I'm positive that they do but they would be too slow even at 1x IR. If the GPU doesn't bottleneck it the PCI bus will.

Edit: Yup. http://www.newegg.com/Product/Product.aspx?Item=N82E16814139050

I'm also glad that we finally have a nehalem result. Too bad it's PAL. We should be able to adjust it (-16.66%) to predict the NTSC results with good accuracy though.

Zee530 Wrote:With my weak ass system, well since there's an atom there, might as well

Please do. We need as much variety is possible. So far half our results are from high end sandy/ivy bridge cpus.

Am I the only one surprised that the A10-5700 at 3.4GHz is ranked lower or equal to my A6-3400M at 2.3GHz in a few spots?

Quote:Am I the only one surprised that the A10-5700 at 3.4GHz is ranked lower

There is sth wrong with his system for sure . Should we remove that invalid result ?

I think maybe we should.

I have an 1.71GHz Athlon XP 2100+ and a 1.6GHz Atom netbook and can attest that their single-threaded performance is darned near identical. I also have a 2GHz Pentium 4 Northwood that also has darned near identical single-threaded performance compared to both the Athlon XP and the Atom.

The easiest reference is that all three systems are fast enough to run YouTube 720p WebM via VLC, but not YouTube 720p MP4 via VLC. By comparison, my Brisbane can do YouTube 720p WebM @ 1GHz via VLC but YouTube 720p MP4 via VLC requires 1.3GHz. This is a performance gap of only 33% which is pretty small considering that's only the difference between a 2GHz Northwood vs a 2.66GHz Northwood.

So therefore I think it is safe to say that Atom has an IPC extremely similar to K7. Also, Brazos in my experience has an IPC that's about 75% of K8, which means the upcoming Jaguar core should have K8-levels of IPC.

Fun fact: The Core Duo and Core2-based Celeron also have IPCs nearly identical to K8.

The strange A10-5700 result has been removed and the chart is updated once again.

Just benchmarked my laptop.

OS: Windows 7 x64 Ultimate
CPU: Intel Core-i5 2410M @ 2.3GHz (Turbo-boosted to 2.7GHz while emulating)
GPU: NVIDIA GT 540M
RAM: 4GB DDR3-1333