PCSX2 review : CPU power DOES matter.

[ssj4breakthru]

Gj mate . Very good in showing how mhz does matter . Anyhow , in some of the benches I can see that even 2.8Ghz @single core is not fast enough to match stock dual core (1.8ghz) , espescially in Phantom Brave SS1 . Indeed , as everyone hope, hopefully the emu will take more advantage of single core and the GPU

[Shin_Gouki]

any Core Duo Tests to compare?
wbr Shin Gouki

[RPGW1ZaRD]

Originally Posted by Shin_Gouki

any Core Duo Tests to compare?
wbr Shin Gouki

Sure if you send me one.

At least I'm gonna stay with this Opteron 165 at least for a year, but if I have the chance, I'd gladly test it out.

As for Conroe speeds in pc game speed it seems 300 - 350MHz clock for clock advantage for Conroe according to some benchmarks but no idea how it compares in PCSX2 as I haven't seen any tests really but I guess it's pretty equal results there.

If that's true then ~2.5GHz Conroe or E6600 and up should provide at least similiar speeds as my Opteron 165 @ 2.8GHz. But then I've seen lots of people already that overclocked their Conroe samples to 3.5GHz+ and I can only imagine what kinda speeds that would give. FFX would prolly run like 70 - 200 fps.

[Mystiq]

Originally Posted by paladin_le

i only wish pcsx2 will take more power from GPU in the future

Unlikely in the near future, unless somehow they can magically turn all the PS2 GPU calls directly into PC hardware calls.

[JKKDARK]

Great job, |Gand|Alf|

[mugener]

Originally Posted by Mystiq

Unlikely in the near future, unless somehow they can magically turn all the PS2 GPU calls directly into PC hardware calls.

Please tell me one thing:

Does piexl shader matter on pcsx2???

My card doesn't supoort ps at all, so I can only use gxsd9, but my cpu is 2.80ghz.

Can I get good speed even without ps???

.

[refraction]

I concur with the above, excellent review!

It certainly highlights the importance of processing power in emulation!

It would have been cool if you could have some other systems in there, single core (like you said) and maybe some Intel comparisons to show the differences, but i cant expect you to be rich enough to build them all

Maybe if somebody else could do the same comparison shots on say, conroe, P4, and single core AMD then pass them on? Obviously id expect folk to have the same sort of spec ;p

[ulwan25]

the review was great! to bad me CPU don't have dual core.

[k3N]

Yeah i saw a guy with a an overclocked e6300 running tekken 4 at 25 fps ingame.

[refraction]

Originally Posted by k3N

Yeah i saw a guy with a an overclocked e6300 running tekken 4 at 25 fps ingame.

yeh so did the rest of the forum

[RealOne]

Since that is the FPS obtained using GSdx9, the probability of getting full speed with ZeroGS is quite high, isnt it?

[RPGW1ZaRD]

Originally Posted by refraction

yeh so did the rest of the forum

Never tried out Tekken 4, I don't own it myself but I know some friend that does, perhaps I could try it out on this system as well with both GSdx9 as he used and ZeroGS.

I agree it would be nice with a P4, Pentium D and Conroe included in the test for example, too bad I don't have any friends either where I could put up some good tests, some have crappy P3 and most of them have AMD systems. lol

[Qwin]

nice review hmmm i wonder if AMD X2 would be a great difrents with opterons

[El_Diablos]

Yeah, somebody plz post the FPS results on AMD X2 ?

[refraction]

Originally Posted by El_Diablo

Yeah, somebody plz post the FPS results on AMD X2 ?

this is NOT a request thread, he did this off his own back, he doesnt have to do anything for you. The opteron 165 should be a close enough guideline for you.

Meh i knew that would start sooner or later..

Originally Posted by |Gand|Alf|

Never tried out Tekken 4, I don't have own it myself with both GSdx9 like he used and ZeroGS but I know some friend that does, perhaps I could try it out on this system as well.

I agree it would be nice with a P4, Pentium D and Conroe included in the test for example, too bad I don't have any friends either where I could put up some good tests, some have crappy P3 and most of them have AMD systems. lol

Maybe you should add P3 results, just to validate your point that if your CPU does suck ass, you arent going to go very quick ;p

[RPGW1ZaRD]

Originally Posted by Qwin

nice review hmmm i wonder if AMD X2 would be a great difrents with opterons

Originally Posted by El_Diablo

Yeah, somebody plz post the FPS results on AMD X2 ?

I can confirm that Opteron dual core performs exactly the same as AMD64 X2 CPUs; X2 4400+ 2.2GHz and 4800+ 2.4GHz at same clock speeds it's been tested many times but they are slightly faster than X2 3800+, X2 4200+ and X2 4600+ at same clock speed since these only have 2x512KB l2 cache instead of 2x1024KB, but they all support the same features. That's why I wrote Opteron 165 @ 2.2GHz should reflect Opteron 175 OR X2 4400+ performance as they both are 2.2GHz and have 2x1MB l2 cache and share same architecture. In other words it would be useless to compare a AMD64 X2 cpu as it would perform exactly same. This is also one reason why I choosed 2.2GHz instead of 2.3GHz for example.

Opterons overclocks better in general is why I choosed Opteron so I would be able to run at AMD64 FX62 2.8GHz speed.
ve taken all shots with AA4x instead.

Originally Posted by refraction

Maybe you should add P3 results, just to validate your point that if your CPU does suck ass, you arent going to go very quick ;p

LOL, yea maybe I should but I wouldn't even have the patience to take all those screenies with such slow speed.

I think the difference is rather huge between 1.8GHz single core and 2.8GHz dual core already...

EDIT: just found out that I'm able to run with either Bilinear filtering or AA4x only with a very minor performance loss, just not both combined at least in FFX...

Might add VM vs non VM build shots too.

[refraction]

yea thatd be good.

also some AA comparisons would be cool (maybe go over the bars so you have lines for each AA speed, rather than seperate ones) or just do 1 test.

for example, FFX has little to no impact on the different AA settings, but if you try Grandia 3, you get about a 50% performance hit between off and AAx4

[emwearz]

"for example, FFX has little to no impact on the different AA settings, but if you try Grandia 3, you get about a 50% performance hit between off and AAx4"

MY speed can drop from 1/3 to a 1/2 if I enable AAx4 in FFX.

[RPGW1ZaRD]

Originally Posted by refraction

yea thatd be good.

also some AA comparisons would be cool (maybe go over the bars so you have lines for each AA speed, rather than seperate ones) or just do 1 test.

for example, FFX has little to no impact on the different AA settings, but if you try Grandia 3, you get about a 50% performance hit between off and AAx4

yea and the prob is where and how many shots to take as it varies so much between areas. I just tried Grandia 3 and in one place I got even 18 fps increase (78 vs 60) with VM build while in another it added only ~2 fps increase.

And yea it does vary so much from game to game if AA and BF slows down or not, Grandia 3 does suffer so much more with AA and BF. But I love how sleek it looks during battles with max quality settings during spell animations.

Better than on a real ps2...

Won't promise anything that I will add things to this review, but if I do I'd prolly add max quality settings performance for at least 2.8GHz speed to the graphs and possibly some VM vs non VM build screenshots. I think my bro saved the MS Excel work but not sure, takes time to add all those 72 fps rates to the work sheet and create the graphs as well so wouldn't wanna do it again...

[shafeen]

wellfirst of all congrats for ur overwhelming research
and tahnk for review
but i know i am dumb asking this my computer has an amd athlon 64 3000+ proccesor and 512 mb ram with an nvidiageforce 440 card will pcxs2 games run at acceptable speeds on my system i (think proabably not but just i wanna know)...