FFXIV Benchmark

Pretty low compared to most of you.. Of course at only 1280x720 with no apparent FSAA, you aren't really stressing the GPU so you're mostly CPU/API bound.

 
what kind of score should I be expecting with a 3.0ghz intel quad duo with 6gig of ram and gtx 470?
 
I get the feeling this benchmark seems to be fairly useless in terms of telling people what kind of actual game play experience they're going to have with their particular setup.
 
Rig in Sig. Benchmark at 1080p:

ffxivtest.jpg
 
720: 4690
1080: 2970

Have an excuse to upgrade my rig now: Priceless...
 
IRSmurf said:
i7 920 @ 4GHz, GTX 480 @ 835MHz core / 4.05GHz RAM, 3GB DDR3 1600 @ 8-8-8-24, Intel X25-M G2 80gb SSD

Results:
Score: 4555
Load time: 11738ms

ffxivbench.png


I am disappointed. I expected a higher score. I am running the latest NVIDIA 3D driver version 1.29 which include their 257.21 driver. I don't suspect the 3D driver is hurting performance since the benchmark runs windowed... meaning no 3D.
Click to expand...

sorry
that is 720 P?
thx~~
 
Core i7 3.6GHz + 4850X2. I used a modified DLL to run in full screen and renamed the application to 3DMark06.exe to get a true result which would make more sense as the game would be able to run in full screen and should support multiple GPUs.

Without renaming it, it will ignore anything but your first GPU.

720p Low

1080p High

I apolgize for the funny looking images, this was done on my Sony GDM-FW900, which needs to be recalibrated which I'll end up doing next week once I get paid to buy the cable.
 
Reanimation_LP said:
Core i7 3.6GHz + 4850X2. I used a modified DLL to run in full screen and renamed the application to 3DMark06.exe to get a true result which would make more sense as the game would be able to run in full screen and should support multiple GPUs.

Without renaming it, it will ignore anything but your first GPU.

720p Low

1080p High

I apolgize for the funny looking images, this was done on my Sony GDM-FW900, which needs to be recalibrated which I'll end up doing next week once I get paid to buy the cable.
Click to expand...
Why in the hell would you not just print screen?
 
I've been using the FFXIV benchmark to test some hardware & software upgrades. I've recorded all of those results and am summarizing them below.

Before the results, a few things about my process: I tested each setting and test resolution a total of three times to weed out any anomalies or bogus results. The average of these results are what I am posting below.

Throughout all tests, I am using 8 gig of G.Skill DDR2 RAM a Q6600 overclocked to 3.3GHz.

The variables in my test were:

1) Vista 64bit Ultimate vs Windows 7 64bit Professional (note: I didn't test vista 64 on the SSD as I didn't see the need for it, based on my other results - I know, bad science).
2) EVGA 8800GTX vs EVGA 460GXT (1Gig Ram) stock setting vs EVGA 460GTX (1Gig Ram) over-clocked to 825Mhz.
3) Seagate 180gig 7200rpm HDD (SATA) vs Kingston 64Gig SSD (SATA II)

Now for the results (first number is score, second is Load Time):

Vista 64, Seagate 7200, 8800GTX:
Low: 2678, 11664
High: 1061, 16689

Vista 64, Seagate 7200, 460GTX (stock):
Low: 4160, 14453
High: 1412, 14765

Vista 64, Seagate 7200, 460GTX (OCed):
Low: 4548, 12437
High: 1598, 14678

W7, Seagate 7200, 8800GTX:
Low: 3319, 12481
High: 1866, 11432

W7, Seagate 7200, 460GTX (stock):
Low: 3978, 12124
High: 2337, 12089

W7. Seagate 7200, 460GTX (OCed):
Low: 4355, 12098
High: 2786, 12088

W7, SSD, 8800GTX:
Low: 3150, 11291
High: 1796, 11444

W7, SSD, 460GTX (stock):
Low: 3928, 13019
High: 2288, 12668

W7, SSD, 460GTX (OCed):
Low: 4245, 12184
High: 2724, 12230

Results summarized:

My Best Low Score: 4548 (Vista64bit, Seagate 7200, 460GTX OCed)
My Best High Score: 2786 (W7, Seagate 7200, 460GTX OCed)
My Best Low Load Time: 11291 (W7, SSD, 8800GTX)
My Best High Load Time: 11432 (W7, Seagate 7200, 8800GTX)

Observations:

SCORE:
1) There was a very expected increase in results from the 8800GTX to the 460GTX to the 460GTX OC'ed.
My low score went from a 2678 to a 4548 (Vista, Seagate, 460 OC'ed), a 69% increase.
The high score went from a 1061 to a 2786 (W7, Seagate, 460GTX OCed),163% increase!
2) Windows 7 didn't contribute much to the low results, and in fact, the low results were generally better on Vista64. However W7 had a substantial impact on the high scores. Just implementing W7 on my Seagate, 8800GTX, I saw an increase of 76%.
3) At least on my set-up, I received higher scores on the Seagate than I did on the Kingston 64Gig SSD.

LOAD TIMES
1) Just looking over the results, load times stayed roughly the same. There may be a decrease in the overall average load times with the SSD, but nothing drastic. I assume this is due to my older setup.

FINAL NOTES:

I was able to push my 460GTX a bit further to 850MHz. (the tests above were done at 825MHz OCed). My card wouldn't maintain a stable 875MHz. I've only been using the EVGA Precision tools, and the stock card cooler, so with better cooling and a different app, I could perhaps get a few more MHz out of the card.

My average results for this final 850MHz OC push were:
Low: 4393, 12176 (my highest score of 4548 was on Vista64, Seagate, 460GTX OCed)
High: 3011, 12360 (This is another 8% over highest score of the 825MHz OC, a total of 171% increase from where I started).

I have screenshots of all of my tests. Let me know if you had any questions! Hope this is helpful to someone out there.

D
 
Last edited:
This is a very sad day for me as I want to play FFXIV so much!! My score makes me want to cry....... :(

 
720p res, score is 4291/32928

FF14.jpg


Q6600 B3 @ 3.1 ghz, 4gb DDR2 800 RAM, Radeon 5850 @775/1125
 
What's your high score with that setup? I'm about to have a really similar setup and I'm curious.
 
Intel Core 2 Quad Q9450 @ 3.2ghz w/ Thermalright TRUE
Gigabyte GA-EP45-UD3P
4x2GB G'Skill DDR2 1066mhz
EVGA Geforce GTX280 @ 648/1404/1161 (Driver v. 258.96)
Windows 7 Ultimate x64
2xWD 640GB AAKS in RAID0
1xWD 640GB AAKS (Storage)
Samsung SH-203B SATA 20X DVD-R/RW

1080p

ffxiv_bench_thegeekredneck.jpg
 
Last edited:
1080p score 5238:



Uploaded with ImageShack.us

This game really heats up the video card...was surprised. (950/1300 OC)
 
Last edited:
Ok, so I have new scores

I was able to locate a modified .DLL file that allows for full screen action. Along with the file rename, scored the following - all High Res

3450 - Quad Fire - Full Screen
3391 - Crossfire - Full screen (Unchecked the Crossfire X box in CCC)
2204 - Single GPU Full Screen - (Unchecked Crossfire X box, and Checked "Disable Catalyst AI")

To confirm my results, I pulled the .DLL file from the benchmark folder to run in Windowed mode again

1488 - Quad Fire - Windows Mode
1490 - Cross Fire - Windows Mode
1481 - Single GPU Windows Mode

This was on my C2D OC'd to 3.6 along with running the bench from my SSD.

While the scores improved, the display was awful. Screen tearing in QuadFire is putting it lightly, more like screen shredding. The little blue bar looked like a seismograph the entire time.

Crossfire improved on the screen tearing, but still hard to ignore.

Single GPU was more to the norm...a smoother blue bar, but low FPS symptoms.

From what I've seen a Q9550 with a 5850 hits just over 3600 in windows mode, so I'm not sure if a Q9550 is a good idea, or simply dump the X2's fora 5xxx, or an upcoming 6xxx card.

I'll bench hers tonight and see if I can get a comparison.
 
HammerAZ said:
Ok, so I have new scores

I was able to locate a modified .DLL file that allows for full screen action. Along with the file rename, scored the following - all High Res

3450 - Quad Fire - Full Screen
3391 - Crossfire - Full screen (Unchecked the Crossfire X box in CCC)
2204 - Single GPU Full Screen - (Unchecked Crossfire X box, and Checked "Disable Catalyst AI")

To confirm my results, I pulled the .DLL file from the benchmark folder to run in Windowed mode again

1488 - Quad Fire - Windows Mode
1490 - Cross Fire - Windows Mode
1481 - Single GPU Windows Mode

This was on my C2D OC'd to 3.6 along with running the bench from my SSD.

While the scores improved, the display was awful. Screen tearing in QuadFire is putting it lightly, more like screen shredding. The little blue bar looked like a seismograph the entire time.

Crossfire improved on the screen tearing, but still hard to ignore.

Single GPU was more to the norm...a smoother blue bar, but low FPS symptoms.

From what I've seen a Q9550 with a 5850 hits just over 3600 in windows mode, so I'm not sure if a Q9550 is a good idea, or simply dump the X2's fora 5xxx, or an upcoming 6xxx card.

I'll bench hers tonight and see if I can get a comparison.
Click to expand...

Mind sharing where you get this dll?
 
pumi0.jpg


Q6600 @ 3.4Ghz
GTX 280 @ 680/1463/1300 (driver 258.96)
Velociraptor 150g
6 gig ram, ddr2 @ 755
 
I was able to bench my GF's rig last night. Rig in sig

Baseline test - No rename, no DLL - 3306
Full screen DLL with rename - 3255
Removed DLL with rename - 3304

While my scores compare to hers, and we fall into the playable category - there is a night and day difference between what we see.

Hers is very cinematic, very smooth, very little tearing - fun to watch.
Mine - terrible to watch...screen shredding, hitching, stuttering - unless I run windows mode, then its just poor FPS

the only conclusion I can draw is that the 4xxx series cards, no matter how many I have just wont give the visuals that a 5xxx can. Even with my OC'd CPU, my SSD, and my choice of 1, 2 or 4 GPU's...I'd rather game on her rig.

I guess I'm just bummed my X2's burned out so fast. Makes me hesitant to buy a high end card.

Hope the benches help :)
 
Specs in sig. Note i am running a 4890 for a historic benchmark :p This card still has the stuff i think...Only ran at High def... because thats all i run games at....

benchmark.jpg


BTW: To the guy above me, taking a picture of the monitor, you fail.
 
leehom said:
hi rez...
[IMggG]http://i204.photobucket.com/albums/bb41/humadragon/wrong.jpg[/IMG]
Click to expand...

im pretty sure you didnt change this from low to high, as your high score is better then your low res score...
 
Too lazy to post to photobucket but I scored 4446 with the system in my sig.
 
BTW: To the guy above me, taking a picture of the monitor, you fail.
Click to expand...
Nobody asked for your opinion, sir. Mind your own business and the world will be a better place.
 
You must log in or register to reply here.
Back
Top