4670k vs 3570k

[NaturalViolence]

AnyOldName3 Wrote:Also it wouldn't explain why I can do SMG at stock clocks.

Use dolphin 4.0 x64 at stock and give me your settings. I'll test it on my 3570K at stock and we can confirm this.

Also you need to provide me with the save data and the location you tested. And screenshots would be helpful too.

[xemnas]

Great. I look forward to the result.

[AnyOldName3]

(10-07-2013, 09:20 AM)NaturalViolence Wrote:

AnyOldName3 Wrote:Also it wouldn't explain why I can do SMG at stock clocks.

Use dolphin 4.0 x64 at stock and give me your settings. I'll test it on my 3570K at stock and we can confirm this.

Also you need to provide me with the save data and the location you tested. And screenshots would be helpful too.

It's the PAL version, so my saves won't help unless you've got that (which isn't tremendously likely when you're in the US). Either way, I'm not 100% sure where Wii saves actually live.

[Venomx1]

right click game & click
go to save location

---

also haswell has shown in emulation benchmarks (dolphin & pcsx2) that it is far superior to ivy.

if a 4.2 haswell can match my chip at its best then SMG should easily be playable at stock speeds which would easily be equv to a ivy @4to 4.2 range.

[xemnas]

(10-08-2013, 03:44 AM)Venomx1 Wrote: also haswell has shown in emulation benchmarks (dolphin & pcsx2) that it is far superior to ivy.

Do you have a link for the PCSX2 benchmark?

[Venomx1]

(10-08-2013, 04:48 AM)xemnas Wrote:

(10-08-2013, 03:44 AM)Venomx1 Wrote: also haswell has shown in emulation benchmarks (dolphin & pcsx2) that it is far superior to ivy.

Do you have a link for the PCSX2 benchmark?

http://forums.pcsx2.net/Thread-CPU-Benchmark-designed-for-PCSX2-based-on-FFX-2

and specifically this Post. http://forums.pcsx2.net/Thread-CPU-Benchmark-designed-for-PCSX2-based-on-FFX-2?pid=319389#pid319389

has his Shown and the other guys in a quote.

they havent updated the master list in a while.

still at 4.1ghz haswell is roughly as fast as my ivy at 5.0 (Teufel9000) while the guy at 4.7 is quite a big higher than mine.

[NaturalViolence]

PAL vs. NTSC won't matter if we go by fps instead of gamespeed.

Here is the benchmark he was referring to: http://forums.pcsx2.net/Thread-CPU-Benchmark-designed-for-PCSX2-based-on-FFX-2

The first page hasn't been updated in awhile so you'll have to read through the results on the last few pages of the thread manually. Just like with dolphin the 4670K leads the 3570K here by nearly 30% at the same clock rate. It is far more likely that for whatever reason haswell is just a good microarchitecture for CPU emulation in general.

[xemnas]

Thanks for the link. For those who might be interested:

Quote:81.22 FPS - SLUS 20672 - Intel Core i5 2500K - 4.0 GHz OC - Rezard
86.72 FPS - SLUS 20672 - Intel Core i5 3570K - 4.2 GHz OC (DDR3 1600) - rama
99.19 FPS - SLPM 65478 - Intel Core i5 4670k - 4.1 GHz OC - masterotaku

[Venomx1]

you missed the two score's that's closest in score to the 4670k fixed^^

(10-09-2013, 01:50 AM)xemnas Wrote: Thanks for the link. For those who might be interested:

Quote:81.22 FPS - SLUS 20672 - Intel Core i5 2500K - 4.0 GHz OC - Rezard
86.72 FPS - SLUS 20672 - Intel Core i5 3570K - 4.2 GHz OC (DDR3 1600) - rama
96.97 FPS - SLUS 20672 - Intel Core i5 2500K - 4.9 GHz OC - Rezard
99.19 FPS - SLPM 65478 - Intel Core i5 4670k - 4.1 GHz OC - masterotaku
99.79 FPS - SLUS 20672 - Intel Core i5 3570K - 5.0 GHz OC - Teufel9000

i would need to clock down my 3570k to roughly 4.85ghz i suspect to match that 4.1 oc (without overclocking my ram. pcsx2 likes overclocked ram lol).

but the numbers show that haswell is a big improvement

[pauldacheez]

The numbers also show that he's benching on a different region variant of the game, which may make a difference. Unless PS2 games do PAL60, which I hope they do.