FFXI Bench 2 Score / Specs

[Thelonious]

Highest I"ve achieved with my setup:
High: 5720

---------------------------
Athlon XP 2400+ running at 2244 MHz (12 x 187)
2x512MB Corsair RAM
A7N8X non-deluxe
9800 bios flashed to 9800 pro @ 415/370


While it may seem like a good score, I read about people running equivalent setups but in higher resolutions. :confused: I can only stand running in 1024x768 and 1024x1024 background resolution.

I'm thinking about building a new A64 machine, the game is even more beautiful at higher resolutions.

[kira_blade]

What score would this get? (Don't own it yet):

Laptop

P4 2.4GHz
512 DDR PC2700 RAM
ATI Mobility Radeon 9000 Shared

Since the GFX is shared, I might get a 256MB upgrade...What do you guys think?

[Custos]

Just did the benchmark, testing to see if my pc will run the game when it comes out in Europe my stats are:

L. 4300

H.3064

-----------------

PC spec is

2.4+ AMD Athlon 266Mhz FSB
A7N8X Deluxe
512 (256*2) Mb DDR running at (333mhz, with dual architechture)
Geforce Ti4200 64mb DDR
80gb Maxtor 2mb cache bla bla bla

I feel that i can get more out of my pc, but dont know how! Any ideas to what i should do, well the cheapest to the most ridiculous will be nice.

[]satsujinsha[]

lets dream a bit of...

i was at the dell hp and look what nice pc
( when i had the money ) i coud buy.. and what u think woud
this pc, get at the benchmark test :

Dual Intelｮ Xeon・Prozessor (3,60 GHz, 1 MB Cache, 800 FSB)
8 GB ECC Dual Channel DDR2 SDRAM, PC400 (4x 1 GB & 2x 2 GB)
256 MB nVidia QuadroFX 3400 PCI-Express grafikcard


well okay togehter with some other nice stuff id woud cost
16000€ (euro) ... but well...
once again..
what u think woud this baby get on the benchmark test? :D

ya mata...

[]satsujinsha[]

gnaaa xD

sorry for the doublepost...

well anyone has a idea what i can do with my notebook xD?

the only thing wich really suck is the damn grafikcard
an Intel(r) 82852/82855 GM/MGE Graphics Controller

the rest is 512 mb ram
1,4ghz centrino ... >_>

anyone a idea what can i do?

i even can´t run the benchmark test with this
$%&/(&%$%&/%$§"$%&/( grafikcard.... >_>

ya mata...

[Ceelron]

Re: lets dream a bit of...

Originally posted by ]satsujinsha[
i was at the dell hp and look what nice pc
( when i had the money ) i coud buy.. and what u think woud
this pc, get at the benchmark test :

Dual Intelｮ Xeon・Prozessor (3,60 GHz, 1 MB Cache, 800 FSB)
8 GB ECC Dual Channel DDR2 SDRAM, PC400 (4x 1 GB & 2x 2 GB)
256 MB nVidia QuadroFX 3400 PCI-Express grafikcard


well okay togehter with some other nice stuff id woud cost
16000€ (euro) ... but well...
once again..
what u think woud this baby get on the benchmark test? :D

ya mata...

That's not a computer built for gaming, You could get a much higher score with something a ton cheaper.:

(1) Processor is made for servers. Xeon and Opteron will slow down games, in comparison to Athlon and Pentium 4 / Pentium M

2) That RAM is also made for servers and databases. Not only that, but latency will be horrible. But FFXI doesn't rely on RAM a whole lot...So this choose isn't that bad. You don't need anymore than 1 Gig of RAM if you are gaming.

(3) That video is absolute garbage. It's meant to be used in a workstation for artwork rendering, but even Quadro isn't a good brand for workstations. That card is optimized for performance; not speed. And Speed is what you need in gaming.

[fastcart]

Recently upgraded my pc's video card from 5600 (non-ultra) to a 6600 GT.

Using Benchmark2:
Low: 4704
High: 4336

Using Benchmark3:
Low: 5128 (Note: I got a 5369 using a SB Live sound card, but I'm force to use my mobo's sound because the SB Live causes my pc to Blue Screen when I play FFXI)

High: 3607

5600 score using Benchmark3 (56.64 drivers):
Low: 3567

The 6600GT is really nice! An 1800 (when using a SB Live) point increase from a 5600 on Bench3. I was able to go into dynamis last night at 1024x768 w/o shadows and bump mapping but everything else on and play w/o any major frame drops. I'll try bump mapping next time. Outside dynamis, I'm enjoying 1280x1024 w/ bump mapping. I had to keep my 5600 at 800x600 w/o the bells and whistles most of the time.

System Specs:
P4 2.4 HT/800 FSB
1 Gig RAM
MSI NX6600GT-VTD128 (AGP)
71.22 MSI_ForceWare_Release_70_Graphics_Drivers
Win XP Pro

[gaidensensei]

I did a test as the following:

AMD Athlon 64 3000+ (socket 754) vs Sempron 3100+ (socket 754)
Radeon 9800pro R350core 128MB
Asus K8V-SE DX 1005 Bios
ATi Catalyst 5.3
1GB (512x2) CL3 Kingston KVR400X64C3A

it's a full benchmark between the two cpu's, but you can see the results of FFXI Benchmark 3 on the last page.

click here to see it

Soon I am going to be getting a MSI 6600GT 128mb AGP, and will see if there is an increase from that.

[Serantos]

Ran it on my new computer, latest Vid drivers.

ASUS K8N Mobo
Athlon64 3200+
1 gig RAM 2700 (slow) >.<
NVIDIA GeForce 6800 128 MB
Audigy Soundblaster 2ZS
300 watt Power supply (Also horrible)

High resolution - 4709
Low Resolution - 6514

Both ran with 4x Antialiasing

Not to shabby by any means

[GSDragoon]

Originally posted by Serantos
Ran it on my new computer, latest Vid drivers.

ASUS K8N Mobo
Athlon64 3200+
1 gig RAM 2700 (slow) >.<
NVIDIA GeForce 6800 128 MB
Audigy Soundblaster 2ZS
300 watt Power supply (Also horrible)

High resolution - 4709
Low Resolution - 6514

Both ran with 4x Antialiasing

Not to shabby by any means

Just so you know, Antialiasing doesn't work in FFXI.

[gaidensensei]

Alright, have some new numbers.

I tried Benchmark 2,and Benchmark 3.
Apparantly, it seems that ATI is better suited for FFXI at 1024x768 (high mode) benchmark. I spent much of time trying to tweak my PC's getting it to work top notch.

But if you read a lot of other benchmarking sites, you'll see that the Nvidia Geforce 6600/6800 series is literally head to toe with the ATI Radeon 9800/x800 series in 1024x768 resolution. Sometimes the ATI comes out faster though.

When you start looking at the resolution being cranked up - such as 1280x1024, 1600x1200, ATi cards start falling behind, and the Nvidia cards perform better.

So what's the point of this? My point in this is being that although ATI almost all the time beats Nvidia everywhere from left to right with the FFXI Benchmark 2/3, ingame it may differ, because some players set their resolution to 1280x1024 or even higher. The benchmark doesn't have a higher resolution mode, which makes it misleading.

Also, the 6600/6800 graphics card series line seems to may be having problems with FFXI. It's as if there is a driver or setting incompatibility with the cards, that doesn't allow them to run at the best possible. Hopefully SE will fix this in the future, but for now, there is not much that can be done to tweak the FFXI benchmarks. One thing I have found that bumped it up significantly is the 'Fastwrites' options. Please set this in your bios to OFF to see if it makes a difference. If not, turn it back on. Also your AGP Aperature size should be set to 128/256 if you have a really good graphics card.

And now without further ado, I've tested about 4-6 different setups. Some were just video cards changed, so you can get an idea of how much it can change the scores.

If you look at the previous post I made, I had some small benchmarking results posted on my website.


Computer A.
Asus K8V SE Deluxe
AMD Athlon 64 3000+ Socket 754
1GB Kingston Valueram PC3200 DDR400 CL3
ATI Radeon 9800Pro 128MB AGP8X
Catalyst ver 5.3 / 5.5

Benchmark 3: ~4500 on high, ~6500 on low
Benchmark 2: didn't try / forgot

Computer B.
Asus K8V SE Deluxe
AMD Sempron 3100+ Socket 754
1GB Kingston Valueram PC3200 DDR400 CL3
ATI Radeon 9800Pro 128MB AGP8X
Catalyst ver 5.3 / 5.5

Benchmark 3: ~4000 on high, ~5800 on low
Benchmark 2: didn't try / forgot

Computer C.
Asus K8V SE Deluxe
AMD Athlon 64 3000+ Socket 754
1GB Kingston Valueram PC3200 DDR400 CL3
MSI Nvidia 6600GT NX6600GT-VTD128 128MB AGP8X
Nvidia Forceware ver 66.93, 71.25(MSI Optimized), 71.89

Benchmark 3: ~4050 on high, ~5900 on low
Benchmark 2: ~5000 on high, ~5650 on low

Computer D.
Asus A7N8X Rev 1.6 (1.008 bios)
AMD Athlon XP 3000+ Socket A (333fsb)
512MB Samsung PC2700 DDR333 CL2.5
ATI Radeon 9800Pro 128MB AGP 8X
Catalyst ver 5.3 / 5.5

Benchmark 3: ~4100 on high, ~5750 on low
Benchmark 2: ~5200 on high, ~5600 on low

Computer E.
Asus A7N8X Rev 1.6 (1.008 bios)
AMD Athlon XP 3000+ Socket A (333fsb)
512MB Samsung PC2700 DDR333 CL2.5
Nvidia GeForce 4 MX420 64MB AGP 4X
Nvidia Forceware ver 71.89

Benchmark 3: ~1900 on high, ~3000 on low
Benchmark 2: didn't try / forgot

Computer F.
MSI K8T Neo FSR
AMD Sempron 3100+ Socket 754
512MB PQI PC3200 DDR400 CL3
Visontek Nvidia GeForce 4 Ti4600 128MB AGP 4X
Nvidia Forceware ver 71.89

Benchmark 3: ~3000 on high, ~4750 on low
Benchmark 2: ~3600 on high, ~4800 on low

hope this helps some people out.

POWER SUPPLY INFO

For the people wanting to get the higher end graphics cards today - mainly above the 9700pro / GeForce 6XXX + series, you need a decent power supply. Having a crap or power supply that came with your computer case, I'm sure that over 90% of the time cannot cut and handle the good graphics cards. Worse comes to worse, you could even physically fry your own pc parts if your power is inadequate.

A good way to know if your powersupply is good or not: Check the weight. Cheap, general powersupplies that come with PC Cases usually weight around the 1-3LB area. This is because they use cheaper conductors and other internal parts. You do not want something like this if you plan to use your system long term.

Ccheck newegg.com for brands such as Antec, Fortron (aka FSP), Enermax, OCZ are a few good ones to name. A good powersupply will generally be VERY heavy. For example, my Antec Truepower 2.0 430w weights almost 6lbs or heavier, while the 350w that came with the case weights around a measly 1lb.

Also check the readings on the Power supply label. You will see that the power is seperated into 3.3V, 5V, and 12V. You want to check the 12V section, atleast have 18A or higher PS if you want to run today's video cards (year of 2004 and up). I recommend atleast +25A on 12V for 9800pro and Geforce 6xxx systems.

Wattage on the PSU means almost nothing. An Antec 300W psu could beat a standard case 500w psu easily. What you really want to know is your voltage numbers.

[GSDragoon]

Also, the 6600/6800 graphics card series line seems to may be having problems with FFXI. It's as if there is a driver or setting incompatibility with the cards, that doesn't allow them to run at the best possible. Hopefully SE will fix this in the future, but for now, there is not much that can be done to tweak the FFXI benchmarks. One thing I have found that bumped it up significantly is the 'Fastwrites' options. Please set this in your bios to OFF to see if it makes a difference. If not, turn it back on. Also your AGP Aperature size should be set to 128/256 if you have a really good graphics card.

( I don't understand.) My 6800 runs it just fine.

[gaidensensei]

Originally posted by GSDragoon
( I don't understand.) My 6800 runs it just fine.

Check out these, there are a lot more, but I did not bookmark them.
http://www.hothardware.com/viewartic...05&page=2&va=y

http://www.hothardware.com/printarti...?articleid=677

http://www.xbitlabs.com/articles/vid...gt-agp_17.html

http://www.digital-daily.com/video/g...gt/index04.htm

You'll see that generally the 96XX and higher series ATI beats the 6XXX and equivalent Ti4XXX series almost everytime in FFXI usually, which the only explainable reasoning I can think of is some driver or setting incompatibility between the two. It's definately not Nvidia side, you notice that in other games, given proper settings, Nvidia beats the competition. Only in FFXI, there seems to be some sort of incompatibility, and that is something square should get around to fixing.

Granted, the difference is not much - usually around up to 5-15% difference. But that small difference, is what makes a pet peeve for people like me, who want top notch performance.

[fastcart]

Can you post low res scores? That's what the official benchark is suppose to use to tally scores.

"Note: The following scores are for low-resolution mode."

http://www.playonline.com/ff11us/mul...ownload/bench/

Using that guide you can better tell how well your pc does.

Btw, I run 6600 gt at 1280x1024 just fine I use the newer msi drivers tho (71.25), not the 66.93 as suggested by SE. Even so, I don't see how a pc with similar setup as mine (p4 2.4 cpu, 1 gig ram, msi 6600 gt as with the http://www.hothardware.com/viewartic...05&page=2&va=y link) can get a 4719 score on high res, where I get 4413 on ffxibench2 and 3635 on ffxibench3 w/o AA or AF. If they used low res scores then I'd be able to see what they really mean.

[gaidensensei]

alright, i just edited my post regarding the multiple benchmark tests to reflect the low settings. I didn't get to test them on a couple namely because they're already in use by the other PC's, and too lazy to take them off.

If you had noticed again, the 9800p takes over the 6600gt in bench2 as well. In the LOW benchmarks, they're almost identical, which leads me to think that there's something in high settings in ffxi (1024x768 resolution) that causes some sort of slowdown at a certain sequence or something.

I have 4 pc's to watch and test, two are of my own. I notice that in some areas, the ATI doesn't have a particular sluggish movement in some scenes that my 6600gt gives off.

The way I have scores of exact numbers, is because I round the numbers to the clostest hundredth / 50's.
for example, if i got
3010 > I put ~3000
3025 > I put ~3050
3051 > I put ~3050
3075 > I put ~3100
etc

etc, I mainly do this because sometimes it seems that there's inconsistency in the benchmarks - a few times I'll get scores some points higher, other times they'll be lower.

As for the post regarding that other P4 2.4, I have no clue about that. It does seem like the same Radeon > 6600GT is happening again.

I'm thinking the difference between yours and theirs, that it may have had something to do with the memory, if you see that they're using Kingston PC3500 with optimal settings, so it's likely that it can handle memory timings much quicker.

I think CL2 and CL3 timings can play a small difference in FFXI benchmark, since the game relies a lot on memory instead of just raw cpu and video. If just adding more ram just helps ingame with other characters (ex, rendering all the people in low jeuno)

I remember before, i had 512MB intially in my A64 3000+ and 9800p system, then I put in another 512mb of the same brand, ran the benchmark, my points only went up by around 10-20.

Also, some 6600GT's are faster than others.
check out:
http://www.pcstats.com/articleview.cfm?articleid=1794

some run at a slightly higher clockspeed from factory default. IIRC, XFX GF6600GT has a memory clock at factory 1000mhz while our MSI's only have it at 900. But the MSI still beats them in some games.