The OFFICIAL FFXIV Benchmark & Beta Performance Thread

[Gulkeeva]

Well after some searching/cpu cooler benchmark browsing etc decided to get a KINGWIN XT-1264 120mm Xtreme Direct H.T.C. cpu cooler since its low price/very good apparently.

4ghz 100% load at 60C isn't bad on a 930 from what most people post/benchmarks show so!

[Katlan]

Well after some searching/cpu cooler benchmark browsing etc decided to get a KINGWIN XT-1264 120mm Xtreme Direct H.T.C. cpu cooler since its low price/very good apparently.

4ghz 100% load at 60C isn't bad on a 930 from what most people post/benchmarks show so!

Mine doesn't break 90C on 100% load when left on all night lol, guess i shoulda gotten a water cooler

[Gulkeeva]

Its considered best to stay in the 70C range, going to 80+ is bad for your CPU. Its why I'm getting a new cooler since one I got going at 3.2ghz i7 930 100% load is 75C'ish

surprised your PC doesn't shutoff at near 90C, thats pretty dangerous temp to be running

[Eyedea-Seraph]

Stock videocard seems like the obvious answer here, CPU can only do so much on high. My 470 is clocked (way)higher than a stock 480SC, if you left that alone your score isn't gonna be where it could be. Its a simple process when compared to CPU oc'ing, and since you handled that just fine you shouldn't overlook it.

Cel when overclocking your GPU are you just using something like EVGA precision?

I tryed to O/C my scc 260 and found that it was very unstable when i went to use it on Heaven 2.0 i would get clipping and frame hopping if i went up with the clocks even just a little bit, or are you using an after market colling solution on your GTX470, if at all possible iwas thinking of getting a GTX 470 and O/Cing it to 480 specs but if i need a water block or something to achieve that i will just spend the extra and get a 480scc

[Katlan]

Its considered best to stay in the 70C range, going to 80+ is bad for your CPU. Its why I'm getting a new cooler since one I got going at 3.2ghz i7 930 100% load is 75C'ish

surprised your PC doesn't shutoff at near 90C, thats pretty dangerous temp to be running

Well 90C was full load over night stress testing, otherwise most games keep it around 60-75C (just cause 2 for example). I just bought the fan... When I do replace it I guess it's time to go water cooling. The TJmax is 100C, so I don't think 90C on full load is horrible.

To above: 480ssc is harder to get, it's why I just got the sc, and wasn't in the mood to spend 120$ more for the ftw edition

[Eyedea-Seraph]

Well 90C was full load over night stress testing, otherwise most games keep it around 60-75C (just cause 2 for example). I just bought the fan... When I do replace it I guess it's time to go water cooling. The TJmax is 100C, so I don't think 90C on full load is horrible.

To above: 480ssc is harder to get, it's why I just got the sc, and wasn't in the mood to spend 120$ more for the ftw edition

Kat can you PM me where you got your card and what company youo chose i am leaning to EVGA since i love the support they give ..but maybe i try someone new.

for that matter what GTX480 or 470 should i get ladies and gents and wher eis best price!

[Gulkeeva]

Well 90C was full load over night stress testing, otherwise most games keep it around 60-75C (just cause 2 for example). I just bought the fan... When I do replace it I guess it's time to go water cooling. The TJmax is 100C, so I don't think 90C on full load is horrible.

To above: 480ssc is harder to get, it's why I just got the sc, and wasn't in the mood to spend 120$ more for the ftw edition

You can basically put a pot of water on your cpu to boil for coffee at that temp... almost


I looked up some water cooling solutions myself but considering the space of my PC, I don't have room for it (hell I had to modify my case just to get a video card in). After checking around the one i mentioned was the best I could find (out performs some water cooling solutions also!) that and apparently its light weight enough I don't need to takeout my motherboard and put on a backplate etc (always a plus, would take me like 2 hours with all the crap in my PC the way it is).


edit
I found a nice little fan to add to the PC for extra cooling
1 of the fans
http://www.youtube.com/watch?v=--GeF...layer_embedded

don't think I could use 4...
http://www.youtube.com/watch?v=E8VLr...eature=related

[Cautery]

Firt time I did the bench I scored around 3700. Thats because my HD5970 was working with a single GPU.

Followed the guide for run the bench in full screen and scored around 2200 with the .exe renamed as Crysis.

Scored 5693 with the .exe renamed as 3DMark06 and heard the fan on the GPU increase speed during the bench. Seems this time both of my GPU worked?

Why that difference occured just renaming in 2 different way the .exe?

[Murureru]

Is the difference between low and high just the resolution?

When the game is released I'm sure I will be turning of shadows at 1920x1080 res as they generally hinder performance anyway, wish there were more options in the benchmark.

[Kryssan]

Ok, so here we go:

First settings (no overclock or registry tweaks, using Enzyme 1.4 Beta 2 Drivers):

Spoiler: show

http://www.aethernavale.net/Media/im...ettings_01.jpg

http://www.aethernavale.net/Media/im...ettings_02.jpg

http://www.aethernavale.net/Media/im...ettings_03.jpg

http://www.aethernavale.net/Media/im...ettings_04.jpg

Now the results on High Settings:

Spoiler: show

http://www.aethernavale.net/Media/im...CF_High_01.jpg

http://www.aethernavale.net/Media/im...CF_High_02.jpg

http://www.aethernavale.net/Media/im...CF_High_03.jpg

The last three were with the camera since taking with Print Screen resulted in a black output image. Probably as a result of the D3D file and the full screening or maybe the forcing crossfire, not sure.

Score improved 2~3k using these methods. Duplicated results after rebooting machine. Test has been run 5 times so far on high with comparable results.

The one thing I also noticed is that during the scene where the character walks up onto the deck and observes the meteors falling from the sky, several black bands appeared horizontally across the screen seemingly running up and down. That was the only texture corruption I noticed. Lowering/changing the settings has no effect, it always occurs when using the fullscreen force + ATICE programs.



Edit: Here are the results from running the benchmark on Low:

Spoiler: show

http://www.aethernavale.net/Media/im..._CF_Low_01.jpg

http://www.aethernavale.net/Media/im..._CF_Low_02.jpg

I'm still unable to hit above 8k regardless of what I do. Not sure how the video guy got 10k unless it really is just down to the difference in processors and overclock.

[Kincard]

Every time I try to run the benchmark, the program crashes. I don't even get an error message like a lot of other people seem to get. It just crashes with the whole "Encountered a Problem - Send Error Report?" thing.

[Celeras]

Cel when overclocking your GPU are you just using something like EVGA precision?

I tryed to O/C my scc 260 and found that it was very unstable when i went to use it on Heaven 2.0 i would get clipping and frame hopping if i went up with the clocks even just a little bit, or are you using an after market colling solution on your GTX470, if at all possible iwas thinking of getting a GTX 470 and O/Cing it to 480 specs but if i need a water block or something to achieve that i will just spend the extra and get a 480scc

I have an eVGA card, but use MSI Afterburner because they let you use a custom fan profile. I can go from 600core->700-725 on stock volts, and get the 825-850 core if I increase to 1.087V

[Yoli]

This is a dumb question (again), but is there anyway to tell what my PC's temperature is running at just from somewhere in windows, or do I need a special program/part for that?

Windows Vista right now, but should be 7 soon, for whatever that's worth.

[Kryssan]

You'll need a program. Some MB manufacturers will ship with said software, if not or if its bloatware grab something like CPUZ, GPUZ, RealTemp, or similar.

[Yoli]

Ok, thanks for the info.

[Celeras]

This is a dumb question (again), but is there anyway to tell what my PC's temperature is running at just from somewhere in windows, or do I need a special program/part for that?

Windows Vista right now, but should be 7 soon, for whatever that's worth.

http://www.cpuid.com/softwares/hwmonitor.html

Will tell you everything your bios can, though some boards will label things 'TmpIn1' etc and you'll have to figure out what corresponds to what. The other programs listed are only for specific pieces of hardware, that's more of a solution for everything.

[Deadgye]

Its considered best to stay in the 70C range, going to 80+ is bad for your CPU. Its why I'm getting a new cooler since one I got going at 3.2ghz i7 930 100% load is 75C'ish

surprised your PC doesn't shutoff at near 90C, thats pretty dangerous temp to be running

90C is fine for the i7 architecture. It's not somewhere you want to be, but as long as you don't approach 100C you're not going to run into any problems.

4ghz 100% load at 60C isn't bad on a 930 from what most people post/benchmarks show so!

My 920 at 3.6 hits about 65C on full load. I'd say that temperature is great. (Assuming air cooled.)

[Sonomaa]

http://sonomaa.com/random/FFXivWinBe...-49-05-81s.jpg


went up a bit on high from my 9800 GT. Im happy, it wont be perfect but once the game is properly optimized it should purr

[bluestar]

I'm still unable to hit above 8k regardless of what I do. Not sure how the video guy got 10k unless it really is just down to the difference in processors and overclock.

the guy over 10k was using 2x HD 5970x2's - so essentially 4 video cards

edit: they aren't actually called 5970x2, just 5970 but the 5970 is basically 2 hd 5870's on 1 card so he was running 4x hd 5870's in essence

[Yoli]

At some point these benchmarks become just numbers, right? I mean, is there actually going to be a noticable difference between 8k and 10k when you're playing?