The OFFICIAL FFXIV Benchmark & Beta Performance Thread

[Gulkeeva]

2Gb edition eh, looks like he "other" bottleneck for ffxiv is video card memory then not just cpu+fast card

[viperlasson]

Got 2314 on HIGH :\ . Framerates looked steady the whole time, so looks like a good score.

Spoiler: show

http://i48.tinypic.com/20iuwlw.jpg

[Gustave]

For camparison what would a PS3 score on this? or assumed?

Wondering if I should just buy a PS3...

[Cahlum]

I'd assume average

[Xanthe]

For camparison what would a PS3 score on this? or assumed?

Wondering if I should just buy a PS3...

PS3 version will only run at 720 from what I remember.

[Jinko]

At a guess I would expect the PS3 to be around 3000-3500 at 720p.

But PS3 can be optimised in a way a PC can't.

[Epical]

Low

Spoiler: show

http://i46.tinypic.com/4iizic.jpg

High (I resized this one but it came out a little hard to read... it's 2146 though)

Spoiler: show

http://i47.tinypic.com/33usbcw.jpg

[Kurokikaze]

Noticed we're all getting ridiculously high load times. Is there anyone that got lower than 10k?

[alecrast]

Noticed we're all getting ridiculously high load times. Is there anyone that got lower than 10k?

I'm not sure what those load times are actually a reflection of. Let's assume a load time of somewhere between 10000 and 20000ms, that's only 10 to 20 seconds. I've got 3 different SSDs (Intel X25-M 80GB, OCZ Vertex 120GB Indilinx-based, and OWC Mercury Extreme Pro SF-1500 Sandforce-based). My load times are non-existent for pretty much everything out there and, the second time you run the benchmark (on a system with 8GB, anyway), the entirety of it is sitting in system cache so it's never loaded from disk and still the load times are in the 10-15 second range. I think it's reasonably safe to discount the load times as a nonsensical value.

[Don Katsu]

For those of you using ATI cards, there are new drivers out that can increase the performance of the 5800 series by up to 10%. Catalyst 10.6 was posted yesterday

[layte]

Nothing like the opportunity for some PC spec dickwaving!

1080 - 5256

Spoiler: show

http://homepages.nildram.co.uk/~layt...0-31-27-42.png

Specs: http://homepages.nildram.co.uk/~layt...pu-cpu-ram.JPG

[Arc07]

I suck at overclocking so I'm not gonna even bother. I did however update my graphics driver and it made a significant difference.
720p

Spoiler: show

http://img121.imageshack.us/img121/7015/52113277.jpg

1080p

Spoiler: show

http://img202.imageshack.us/img202/7083/36805046.jpg

I have no idea how to make the loading time any faster. I'm running this off one of the fastest SSDs money can buy and it still blows. ASUS 5870 Eyefinity6 Edition btw.

[alecrast]

I suck at overclocking so I'm not gonna even bother. I did however update my graphics driver and it made a significant difference.

Spoiler: show

http://img121.imageshack.us/img121/7015/52113277.jpg

I have no idea how to make the loading time any faster. I'm running this off one of the fastest SSDs money can buy and it still blows. ASUS 5870 Eyefinity6 Edition btw.

SSDs don't seem to make much difference here except in the case of a cold boot load. As for your score, I'll bet your video card comes in right around 4920 on high-res with the setup you have. The 1280x720 benchmark is CPU-bound. The 1920x1080 benchmark is GPU-bound. Overclocking your CPU shows an almost linear improvement in performance with the overclock on the low-res test.

[ClassicRed]

Oh man ._. This only reinforces my decision to probably play on PS3 for a while ~ Altho I haven't had terrible issue's with lag playing Alpha


Low ~
http://www.melodiesofdeath.com/classicred/low.jpg

High~
http://www.melodiesofdeath.com/classicred/high.jpg
ATI 4850~

Really thought I was doing good computer wise D: *snuffles*

Picked cat model D:~

[Sinzfeldt]

Not sure if this has been posted but the results are skewed, different race selection == higher/lower poly count. I selected lala and my score shot up at least 40 points.

[Kurokikaze]

That difference is so low its negligible that 40 points isn't gonna make or break anything.

[Zhane]

I just get a black screen but can hear the voices fine. Have Directx9 and newest Nvidia driver. What could this be from?

[Cahlum]

SSDs don't seem to make much difference here except in the case of a cold boot load. As for your score, I'll bet your video card comes in right around 4920 on high-res with the setup you have. The 1280x720 benchmark is CPU-bound. The 1920x1080 benchmark is GPU-bound. Overclocking your CPU shows an almost linear improvement in performance with the overclock on the low-res test.

If thats the case I don't get why I get such crappy scores in low

[alecrast]

If thats the case I don't get why I get such crappy scores in low

Okay, let me rephrase that. 5850/5870 cards... GTX480/470 cards have no problem tossing around the pixels necessary for the low-res demo. A GT 240 is a low-end card. Your CPU completely outclasses your GPU. You need an upgrade. 5770 or 5850 would do you wonders with that processor. A GT240 is worse than a 9600GT.

[burrito]

Low on top, high on bottom. 4870 card, nothing OC afaik (hesitant, I have the electronic black thumb of death).

http://img266.imageshack.us/img266/6505/benchmarkp.jpg