Forum Settings
       
This Forum is Read Only

Official FFXIV Benchmarking program.Follow

#552 Jul 20 2010 at 4:36 AM Rating: Decent
Scholar
Avatar
***
2,536 posts
RSquires wrote:
Ok, so I did some more searching and changed somethings. Lowered my Ram and increased my graphics card. My new computer specs will be;

Operating system Genuine Windows 7 Home Premium 64-bit edit
Processor AMD Athlon(TM) X4 635 quad-core processor [2.9GHz, 2MB L2, up to 4000MHz bus]
Memory 6GB DDR3-1333MHz SDRAM [3 DIMMs]
Hard drive 750GB 7200 rpm SATA 3Gb/s hard drive
Graphics card 1GB ATI Radeon HD 5770 [dual-link DVI, HDMI, DP, VGA adapter]

Think this will run the benchmark within reason??? Thanks for any and all help.


You should be fine.
____________________________
FF11 Server: Caitsith
Kalyna (retired, 2008)
100 Goldsmith
75 Rng, Brd
Main/Acc
Exp/Hybrid
Str/Attk
Spam/Others
#553 Jul 20 2010 at 4:38 AM Rating: Decent
Scholar
Avatar
***
2,536 posts
SundayMoney wrote:
Still trying to put together a rig but I'm gonna build from scratch but think I got it this time.

AMD Athlon II X3 440 Rana 3.0GHz 3 x 512KB L2 Cache Socket AM3 95W Triple-Core Desktop Processor ADX440WFGIBOX

I apologize if this seems kinda jumbled up but I'm having it built and before I give the go ahead I was hoping for some feed back. I don't want overkill but I'm looking to score decent on the benchmark and be able to play at normal setting without to many problems.


Upgrade that if possible, and you'll be fine on normal settings. Preferably a Phenom II x4 955 or better if you can afford it.
____________________________
FF11 Server: Caitsith
Kalyna (retired, 2008)
100 Goldsmith
75 Rng, Brd
Main/Acc
Exp/Hybrid
Str/Attk
Spam/Others
#554 Jul 20 2010 at 11:36 AM Rating: Decent
26 posts
score low (720p): 2940
score high (1080p): haven't tested yet

Windows vista 64bit
AMD Phenom X4 9950 2.6 ghz
2 x 1 GB ddr2 memory kingston hyperx 1066 mhz
ECS gf8200a MOBO(dirt cheap)
ELSA 9800gtx+ 512 mb
250gb SATA 1 hdd.

haven't done any overclocking except set the memory voltage to what the manufacturer recommends.


Edited, Jul 20th 2010 12:37pm by Juanmiguel
#555 Jul 20 2010 at 4:10 PM Rating: Decent
22 posts
Began with

WIN7 64 bit Home Premium
CPU: AMD X2 7750 @ 2.7Ghz
Integrated: GF 8300
2Gb RAM
--- > LOW: 252 points http://yfrog.com/2offxivbenchmarklowj

Bought MSI 260 GTX 768Mb (used card, 100eur)
--- > LOW: 2162 points
http://yfrog.com/5effxivlow260j

And OC'd CPU 10% (3.0Ghz):
--- -> LOW: 2374
http://yfrog.com/1rffxivlowcpuocj

Good enough for now..


Edited, Jul 20th 2010 6:11pm by Keitto

Edited, Jul 20th 2010 6:12pm by Keitto
____________________________
Linkshell Journal
#556 Jul 20 2010 at 11:48 PM Rating: Good
*
126 posts
I finished building my rig last week (first one..yay!!) and pulled 2700 on high. It's a phenom II 965 with a 5770. I was thinking of adding maybe another 5770 someime before the release. Would I be better off throwing another 5770 in there, or selling the 5770 I just bought and upgrading to say 1 5850?
____________________________
Carbuncle
75 Pld/Bst
#557 Jul 22 2010 at 1:29 PM Rating: Good
***
3,178 posts
I'd really like to see the benchmark spreadsheet that Mikhalia compiled stickied, but as an OP. I keep digging for it from time to time, and I doubt I'm the only one.

It's perhaps the single most useful tool for deciding what to expect from a given configuration without running the benchmark yourself.
#558 Jul 23 2010 at 9:15 AM Rating: Good
***
3,825 posts
Playing around a bit I took 2 sticks of RAM out of my system. So went from 4x2GB to 2x2GB, haven't changed drivers or overclocked... RAM timings were 1066 both setups. My score increased by 90ish to 2975 on High.

This is with an LGA775 chipset so for those clinging to 2 generations ago until the last minute you might want to fiddle with your RAM setups if you have more than 2 sticks.
____________________________
FFXI:Sylph - Perrin 75 Hume THF; Retired (At least from my use any way)
EVE Online:ScraperX; Retired
WAR:IronClaw- Peryn SW;SkullThrone- Grymloc BO; Retired


#559 Jul 23 2010 at 9:19 AM Rating: Good
Anterograde Amnesia
Avatar
*****
12,363 posts
I've heard that full RAM can ***** things up, I've been thinking about lowering my setup to 4gb from 8gb to see what happens. My memory is unlinked from my OC'd processor, running at 533mhz. I unlinked it so I didn't have to worry about RAM in the overclock, keeps it nice and stable.
____________________________
"Choosy MMO's choose Wint." - Louiscool
The greatest trick the devil ever pulled was to convince the world he didn't exist.
Keyser Soze - Ultros
Guide to Setting Up Mumble on a Raspberry Pi
#560 Jul 23 2010 at 12:18 PM Rating: Decent
Scholar
*
163 posts
Ok, This will be my final question when it comes to a computer for running 14. Which processor is better when dealing with running 14. This computer will pretty much be for 14 only as I'm normally not a big computer person anyways.

- AMD Phenom x4 995 quad-core. 3.2GHz, 2MB L2 + 6MB L3 shared, up to 4000MHz.
- Intel Core i5-750 quad-core. 2.66GHz, 1MB L2 + 8MB shared L3 cache.

Thanks to all for the help.
____________________________
So I rated you up for no good reason, big deal. Wanna fight about it?
#561 Jul 23 2010 at 1:21 PM Rating: Decent
Scholar
***
1,089 posts
Alright new computer!

My i5-750 got a little over 4,000 on 720p @ stock speeds.

I just OC'ed to 3.6ghz and it increased the score to ~5,100.

On 1080p I got a 750 increase or so from 2,000 to 2,750.
#562 Jul 23 2010 at 1:45 PM Rating: Decent
Scholar
Avatar
***
2,536 posts
RSquires wrote:
Ok, This will be my final question when it comes to a computer for running 14. Which processor is better when dealing with running 14. This computer will pretty much be for 14 only as I'm normally not a big computer person anyways.

- AMD Phenom x4 995 quad-core. 3.2GHz, 2MB L2 + 6MB L3 shared, up to 4000MHz.
- Intel Core i5-750 quad-core. 2.66GHz, 1MB L2 + 8MB shared L3 cache.

Thanks to all for the help.


I'm assuming you mean Phenom II x4 955?

I'd prefer the i5 750...unless you're thinking about upgrading later on (then go with the AMD).
____________________________
FF11 Server: Caitsith
Kalyna (retired, 2008)
100 Goldsmith
75 Rng, Brd
Main/Acc
Exp/Hybrid
Str/Attk
Spam/Others
#563 Jul 23 2010 at 1:46 PM Rating: Decent
Scholar
Avatar
***
2,536 posts
mpmaley wrote:

On 1080p I got a 750 increase or so from 2,000 to 2,750.


For high res, OC your GPU. It will give you way better results than OCing your CPU.
____________________________
FF11 Server: Caitsith
Kalyna (retired, 2008)
100 Goldsmith
75 Rng, Brd
Main/Acc
Exp/Hybrid
Str/Attk
Spam/Others
#564 Jul 23 2010 at 1:48 PM Rating: Decent
Scholar
21 posts
Hello, everyone, im new to the forums... I played FFXI bit on ps2 and 360.. Now i want to get a pc to play FFXIV. my budget is around 1200 - 1500. I dont have much space for a desktop and I usually take my laptop around, better for me. My current laptop cant handle this. Ive been reading all the posts and I think this laptop might be able to handle it. What do u think?

Asus G73JH-RBBX05
Intel Core i7 I7-720QM / 1.6 GHz Max Turbo Speed 2.8 GHz
6 GB / 8 GB (max) Technology DDR3 SDRAM - 1333 MHz
ATI Mobility Radeon HD 5870 Video Memory 1 GB GDDR5 SDRAM
Microsoft Windows 7 Home Premium 64-bit Edition

Thanks!
____________________________
PS3, XBOX 360 and Gaming ring owner..
#565 Jul 23 2010 at 8:55 PM Rating: Decent
Avatar
*
246 posts
I am currently running:

Win7 64-bit Ultimate
Asus P6T v2 Deluxe motherboard
IntelCore i7 920@2.67 GHz
6 GB Corsair Dominator (12800) DDR3 Ram
1 TB Western Digital Caviar Black HD
Sapphire ATI Radeon HD4890 Vapor-X 2GB Video Card

My low benchmark tested at 4256, but my high benchmark test was only 2476.

I know that this is "good enough" to run the game, but I was a hardcore FFXI player and would like to push that score up a bit higher because I anticipate playing this game at a high level for many years. My DirectX is updated, but I performed the test while my Sophos antivirus was still active.

What do you recommend would be the best way to achieve better performance results?

Should I get a new video card such as an ATI HD5800 or HD5900 series? Should I get a 2nd video card and run them in crossfire mode? Should I keep my existing hardware and attempt to overclock my system (any recommended speeds not to push the 920 past?) Any other tips?
#566 Jul 23 2010 at 11:28 PM Rating: Default
15 posts
To the guy above me:

New GPU is what I'd recommend. Everything else is fine. 5850 or better or it won't really be worth it. SLI/Crossfire is crap in my opinion... Oh, and you can get that 920 up to 4.0ghz easy with an aftermarket cooler.

Stock speeds, rig is in my sig:

1080: 5043
720 : 6043


Feels good man.

Edited, Jul 24th 2010 1:31am by MrSchwarz

Edited, Jul 24th 2010 1:31am by MrSchwarz
____________________________
i7 930 - ASUS P6X58D Premium
ATI 5870 - 6GB DDR3 (1600mhz,tri-channel)
Corsair 750TX PSU - 1TB HD - Antec 1200 Full Tower
5043 High Benchmark Score (No overclocking)
#567 Jul 24 2010 at 12:46 AM Rating: Decent
Avatar
*
246 posts
Thanks.

What OC settings are you using?
#568 Jul 24 2010 at 1:02 AM Rating: Default
15 posts
LordMethos wrote:
Thanks.

What OC settings are you using?


For my scores? None. I don't intend to oveclock until I get a new heatsink in the mail.

If you mean you are interested in OCing your 920, google it along with your mobo and you'll get what you need. Read a few guides on general OCing until you understand the finer points of OCing (to get a stable OC, requires lots of testing bclk, vcore, etc)

EDIT: If you just want a higher 1080 score you need a better GPU. Your high score is so low due to your graphics card. (low is more cpu based, high is graphics)

Edited, Jul 24th 2010 3:04am by MrSchwarz
____________________________
i7 930 - ASUS P6X58D Premium
ATI 5870 - 6GB DDR3 (1600mhz,tri-channel)
Corsair 750TX PSU - 1TB HD - Antec 1200 Full Tower
5043 High Benchmark Score (No overclocking)
#569 Jul 24 2010 at 4:13 AM Rating: Default
Thief's Knife
*****
15,053 posts
FYI the heatsinks that come with Intel CPU's suck. when I got my i7 860 it was hitting 90c on Prime 95 at stock speeds. Replaced the crappy stock cooler with a Xigmatek Dark Knight and it cuts my temps in half.

Also, push pins suck, I don't trust a large cooler hanging off my MB supported only by dinky pieces of plastic. Always, ALWAYS make sure that the cooler you are getting has a good bracket.


Edited, Jul 24th 2010 7:14am by Lobivopis
____________________________
Final Fantasy XI 12-14-11 Update wrote:
Adjust the resolution of menus.
The main screen resolution for "FINAL FANTASY XI" is dependent on the "Overlay Graphics Resolution" setting.
If the Overlay Graphics Resolution is set higher than the Menu Resolution, menus will be automatically resized.


I thought of it first:

http://ffxi.allakhazam.com/forum.html?forum=10&mid=130073657654872218#20
#570 Jul 24 2010 at 10:59 AM Rating: Decent
Scholar
*
89 posts
Ok, so I'm not sure if it was me renamed the .exe to crysis or updating to the new nVidia drivers, but my score literally doubled to 5300(it was 2500ish before I did any of this) on hi-rez.


http://i.imgur.com/1y55n.jpg

Edited, Jul 24th 2010 1:00pm by Coh
____________________________
The average man thinks he isn't.
#571 Jul 24 2010 at 1:38 PM Rating: Decent
Scholar
Avatar
***
2,536 posts
Coh wrote:
Ok, so I'm not sure if it was me renamed the .exe to crysis or updating to the new nVidia drivers, but my score literally doubled to 5300(it was 2500ish before I did any of this) on hi-rez.


http://i.imgur.com/1y55n.jpg

Edited, Jul 24th 2010 1:00pm by Coh


If I'm not mistaken, your video card is a dual GPU. The benchmark only supports both GPUs when in full screen mode. That's why when you activated full screen by changing the name to crysis, the second GPU on your card got activated and your scores shot up.
____________________________
FF11 Server: Caitsith
Kalyna (retired, 2008)
100 Goldsmith
75 Rng, Brd
Main/Acc
Exp/Hybrid
Str/Attk
Spam/Others
#572 Jul 24 2010 at 1:49 PM Rating: Decent
Scholar
*
89 posts
If that's the case, aren't the new 400 series dual gpus too? Wouldn't the score difference be much greater than what I'm witnessing right now between the 400 and the 200 series?
____________________________
The average man thinks he isn't.
#573 Jul 24 2010 at 1:54 PM Rating: Decent
Scholar
Avatar
***
2,536 posts
If you're talking about the new GTX 460, it's a single GPU card.
____________________________
FF11 Server: Caitsith
Kalyna (retired, 2008)
100 Goldsmith
75 Rng, Brd
Main/Acc
Exp/Hybrid
Str/Attk
Spam/Others
#574 Jul 24 2010 at 1:57 PM Rating: Decent
Scholar
*
89 posts
what about the 470 and the 5800 series? Aren't those dual gpus, too? I'm looking through the bg forums, and these scores are not as high as I thought they would be compared to a year-old card. Could it be that both gpus aren't activated?
____________________________
The average man thinks he isn't.
#575 Jul 24 2010 at 2:31 PM Rating: Decent
Avatar
**
312 posts
Coh wrote:
Ok, so I'm not sure if it was me renamed the .exe to crysis or updating to the new nVidia drivers, but my score literally doubled to 5300(it was 2500ish before I did any of this) on hi-rez.

http://i.imgur.com/1y55n.jpg


Very nice! Other than enabling multi-GPU setups, I wonder if renaming the executable has any other consequences?
____________________________
Steam: Xavier1216
FFXI: Astyanax (Bismarck)


#576 Jul 24 2010 at 2:39 PM Rating: Decent
15 posts
Coh wrote:
what about the 470 and the 5800 series? Aren't those dual gpus, too? I'm looking through the bg forums, and these scores are not as high as I thought they would be compared to a year-old card. Could it be that both gpus aren't activated?


5850/5870 are single gpu. I think you may be getting confused with a dual gpu vs a SLI/crossfire configuration(two non-dual graphics cards).

470/480 are single gpu.

5970 is a dual gpu. Your GTX 295 is a dual GPU, however a single 5870 performs better versus your single GTX 295 (perhaps not in this benchmark, but on paper/various tests). You didn't mention if you have a single GTX 295 or two?

____________________________
i7 930 - ASUS P6X58D Premium
ATI 5870 - 6GB DDR3 (1600mhz,tri-channel)
Corsair 750TX PSU - 1TB HD - Antec 1200 Full Tower
5043 High Benchmark Score (No overclocking)
#577 Jul 24 2010 at 2:52 PM Rating: Decent
Scholar
*
89 posts
I only have one 295. Ah ok, that makes much more sense then. What about the other 200 series though? Are those also single gpu or dual-gpu?

I am still quite shocked though that enabling my 2nd gpu would give me over 100% increase in performance.

Edited, Jul 24th 2010 4:53pm by Coh
____________________________
The average man thinks he isn't.
#578 Jul 24 2010 at 3:02 PM Rating: Decent
15 posts
Coh wrote:
I only have one 295. Ah ok, that makes much more sense then. What about the other 200 series though? Are those also single gpu or dual-gpu?

I am still quite shocked though that enabling my 2nd gpu would give me over 100% increase in performance.

Edited, Jul 24th 2010 4:53pm by Coh


Your GTX 295 is the last in the 200 series from nvidia, and the only dual-chip 2xx.

I have a feeling this benchmark is skewed and not balanced for adding additional cards/chips (wasn't intended anyway) and the scores aren't reflected properly.

For instance scores of 11,000+ with a tri-card setup. (5970 and x2 5870's)

I have doubts your GTX 295 will actually be performing better than a single 5870. The 5870's average score is 5000 on high with a decent quad core processor.
____________________________
i7 930 - ASUS P6X58D Premium
ATI 5870 - 6GB DDR3 (1600mhz,tri-channel)
Corsair 750TX PSU - 1TB HD - Antec 1200 Full Tower
5043 High Benchmark Score (No overclocking)
#579 Jul 24 2010 at 3:09 PM Rating: Decent
Scholar
*
89 posts
Quote:
I have doubts your GTX 295 will actually be performing better than a single 5870. The 5870's average score is 5000 on high with a decent quad core processor.


You might be right. I noticed that there was flickering during the meteor shower scene for a couple of seconds until it zoomed in to the character's face. I'm not sure what caused this, but it probably had to do something with force-enabling sli. But with that said, wouldn't it be a bit silly on SE's part to make sli/crossfire compatibility a hindsight during their development cycle?


Here is a more recent benchmark screenshot. Managed to squeeze out 5400 on this one, heh.

http://i.imgur.com/vLpfW.jpg
____________________________
The average man thinks he isn't.
#580 Jul 24 2010 at 3:17 PM Rating: Default
15 posts
Coh wrote:
Quote:
I have doubts your GTX 295 will actually be performing better than a single 5870. The 5870's average score is 5000 on high with a decent quad core processor.


You might be right. I noticed that there was flickering during the meteor shower scene for a couple of seconds until it zoomed in to the character's face. I'm not sure what caused this, but it probably had to do something with force-enabling sli. But with that said, wouldn't it be a bit silly on SE's part to make sli/crossfire compatibility a hindsight during their development cycle?


Here is a more recent benchmark screenshot. Managed to squeeze out 5400 on this one, heh.

http://i.imgur.com/vLpfW.jpg


SLi/crossfire is still a relatively new technology and most all games currently out don't properly utilize the second graphics card if at all. It's something like only a 25% increase in performance when adding a second card. When you're shelling out $500+ for a new card, tacking on another $500 for a 25% boost isn't very cost effective...

Who knows though, this benchmark wasn't made for SLI/crossfire... The performance gains from the actual retail game is left to be seen.

I'm interested in seeing your low res scores with the crysis.exe fix. That'll most likely give you a more accurate guess to how well your computer will perform for now at least.
____________________________
i7 930 - ASUS P6X58D Premium
ATI 5870 - 6GB DDR3 (1600mhz,tri-channel)
Corsair 750TX PSU - 1TB HD - Antec 1200 Full Tower
5043 High Benchmark Score (No overclocking)
#581 Jul 24 2010 at 3:32 PM Rating: Decent
Scholar
*
89 posts
Here is the lo-res benchmark score:

http://i.imgur.com/TmWEw.jpg

About a thousand point increase. Not as much as I hoped for, but it still quite a large gain.
____________________________
The average man thinks he isn't.
#582 Jul 24 2010 at 4:38 PM Rating: Good
***
3,825 posts
MrSchwarz wrote:
SLi/crossfire is still a relatively new technology and most all games currently out don't properly utilize the second graphics card if at all. It's something like only a 25% increase in performance when adding a second card. When you're shelling out $500+ for a new card, tacking on another $500 for a 25% boost isn't very cost effective...


Um SLI has been around for over 10 years. 3DFX cards were stackable with I think the voodoo2 series? Nvidia bought them out and I think started SLI around 94 or 95. Games don't have to support SLI, they just won't be optimized if they don't support it. Considering almost every PC game in the last 4ish years has some sort of support for SLI it's pretty odd that any one would think XIV wouldn't have it.

Also of note is that with the GTX4__ series SLI has become extremely effiecient with it's scaling. Hopefully Crossfire will catch up pretty soon. Any way the SLI percentages are more like 60-80% with the new GTXs in most games.
____________________________
FFXI:Sylph - Perrin 75 Hume THF; Retired (At least from my use any way)
EVE Online:ScraperX; Retired
WAR:IronClaw- Peryn SW;SkullThrone- Grymloc BO; Retired


#583 Jul 24 2010 at 4:40 PM Rating: Decent
***
3,825 posts
Coh wrote:
You might be right. I noticed that there was flickering during the meteor shower scene for a couple of seconds until it zoomed in to the character's face. I'm not sure what caused this, but it probably had to do something with force-enabling sli.


I think it's the new Nvidia drivers, mine did it when I ran the test with the most recent drivers and it didn't with the drivers I had when the bench first was released.
____________________________
FFXI:Sylph - Perrin 75 Hume THF; Retired (At least from my use any way)
EVE Online:ScraperX; Retired
WAR:IronClaw- Peryn SW;SkullThrone- Grymloc BO; Retired


#584 Jul 24 2010 at 4:44 PM Rating: Good
Anterograde Amnesia
Avatar
*****
12,363 posts
PerrinofSylph wrote:
Coh wrote:
You might be right. I noticed that there was flickering during the meteor shower scene for a couple of seconds until it zoomed in to the character's face. I'm not sure what caused this, but it probably had to do something with force-enabling sli.


I think it's the new Nvidia drivers, mine did it when I ran the test with the most recent drivers and it didn't with the drivers I had when the bench first was released.


I noticed the flickering when enabling sli, and it went away after the newest whql drivers were installed. Fun fact, installing the new drivers caused my sli to become disabled, and so I was getting lower performance after updating. I went in to the config, and the sli option was checked but it said there were changes detected so I clicked apply and my scores shot back up afterwards.
____________________________
"Choosy MMO's choose Wint." - Louiscool
The greatest trick the devil ever pulled was to convince the world he didn't exist.
Keyser Soze - Ultros
Guide to Setting Up Mumble on a Raspberry Pi
#585 Jul 24 2010 at 4:52 PM Rating: Decent
*
149 posts
Aww no one kept the spreadsheet from page 4 up to date??

Thats such a shame it would save so much browsing through such a long thread!

(Any takers?) lol
____________________________
Former THF, BST, PLD, NIN 75
#586 Jul 24 2010 at 5:00 PM Rating: Default
15 posts
PerrinofSylph wrote:
MrSchwarz wrote:
SLi/crossfire is still a relatively new technology and most all games currently out don't properly utilize the second graphics card if at all. It's something like only a 25% increase in performance when adding a second card. When you're shelling out $500+ for a new card, tacking on another $500 for a 25% boost isn't very cost effective...


Um SLI has been around for over 10 years. 3DFX cards were stackable with I think the voodoo2 series? Nvidia bought them out and I think started SLI around 94 or 95. Games don't have to support SLI, they just won't be optimized if they don't support it. Considering almost every PC game in the last 4ish years has some sort of support for SLI it's pretty odd that any one would think XIV wouldn't have it.

Also of note is that with the GTX4__ series SLI has become extremely effiecient with it's scaling. Hopefully Crossfire will catch up pretty soon. Any way the SLI percentages are more like 60-80% with the new GTXs in most games.


Sorry meant it's still new for modern PC gamers, most developers don't care to optimize their games for something that only a small percentage of players will have.

Give me some links backing up your "60-80%" boost in performance from at least 5 games (not some graphics instensive tech demo) where a single version of the card is performing 80% worse than x2...

Hopefully when FFXIV launches it'll have good SLI/Crossfire support, if it does perhaps I'll crossfire and buy another 5870 :)

____________________________
i7 930 - ASUS P6X58D Premium
ATI 5870 - 6GB DDR3 (1600mhz,tri-channel)
Corsair 750TX PSU - 1TB HD - Antec 1200 Full Tower
5043 High Benchmark Score (No overclocking)
#587 Jul 24 2010 at 7:04 PM Rating: Decent
1 post
I'm getting a new rig, this is the best i can afford:

Windows 7 (64bit)
Intel Pentium Dual Core E5400 (2.7Ghz)
ATI Radeon 5750 1024mb
4 GB RAM

What kinda score do you think that'd get (before I buy it)

I currently have:

Windows XP SP3 (32bit)
Intel Core 2 Duo (2.8Ghz)
ATI Radeon HD 2600 256mb
2 GB RAM

I scored 735 on benchmark. I would simply upgrade my graphics card however it is an imac (some idiot at apple soldered the graphics card to the motherboard), so i'm gonna have build this new one from scratch.
#588 Jul 25 2010 at 5:54 AM Rating: Excellent
10 posts
MrSchwarz wrote:
PerrinofSylph wrote:
MrSchwarz wrote:
SLi/crossfire is still a relatively new technology and most all games currently out don't properly utilize the second graphics card if at all. It's something like only a 25% increase in performance when adding a second card. When you're shelling out $500+ for a new card, tacking on another $500 for a 25% boost isn't very cost effective...


Um SLI has been around for over 10 years. 3DFX cards were stackable with I think the voodoo2 series? Nvidia bought them out and I think started SLI around 94 or 95. Games don't have to support SLI, they just won't be optimized if they don't support it. Considering almost every PC game in the last 4ish years has some sort of support for SLI it's pretty odd that any one would think XIV wouldn't have it.

Also of note is that with the GTX4__ series SLI has become extremely effiecient with it's scaling. Hopefully Crossfire will catch up pretty soon. Any way the SLI percentages are more like 60-80% with the new GTXs in most games.


Sorry meant it's still new for modern PC gamers, most developers don't care to optimize their games for something that only a small percentage of players will have.

Give me some links backing up your "60-80%" boost in performance from at least 5 games (not some graphics instensive tech demo) where a single version of the card is performing 80% worse than x2...

Hopefully when FFXIV launches it'll have good SLI/Crossfire support, if it does perhaps I'll crossfire and buy another 5870 :)


http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/30999-nvidia-geforce-gtx-470-sli-review.html

Use the top right menu where it says "Review Contents:" to view the different parts of the review.

Some good comparisons of the benchmarks on non sli and sli results for the 470 and 5870.

I'm waiting on release to see what kind of upgrades are really needed to run the ingame portion of it all if SLI even works. Personally I'm liking the High FPS marks on both the 470 and 5870 with SLI/CF. Better than owning a single 480 or 5870. Suffice to say a 5870 is beast alone. But seeing the scores battle it through for a little bit less in budget, I'm definitely going with the 470 SLI setup if it works for FFXIV.

Cheers :]


Edited, Jul 25th 2010 8:08am by AkkiZan
#589 Jul 25 2010 at 1:21 PM Rating: Good
***
3,825 posts
MrSchwarz wrote:
Sorry meant it's still new for modern PC gamers, most developers don't care to optimize their games for something that only a small percentage of players will have.


It's not new, it's not widely unsupported. Please stop spreading false information.

Some people would rather buy 2 GTX 460s and some would rather buy a single GTX 480... but for budget minded folks a GTX 200 series in SLI might be the best option for them at this time. Or buying a single 460 (or ATI equivalent) now and splurging on a second card later... Almost every developer supports multi-GPU gaming. Usually the multi-GPU is handled in the drivers of the hardware, not the software.

However alot of EA games used to suffer from artifacts caused by SLI (Need for speed series had Akira like light trails for instance). They no longer do (at least with ME and DA:O). Ultimately it's just another option when budgeting and thinking about the future. Some people don't like the hassle of multi-GPU gaming, there can be artifacts, but usually (at least in the last few years) it's something that ATI/Nvidia, or the Developer, or even the community come up with a fix for IF there even are problems. And if a game isn't using your second card, you can force it to.
____________________________
FFXI:Sylph - Perrin 75 Hume THF; Retired (At least from my use any way)
EVE Online:ScraperX; Retired
WAR:IronClaw- Peryn SW;SkullThrone- Grymloc BO; Retired


#590 Jul 25 2010 at 1:23 PM Rating: Decent
***
3,825 posts
Wint wrote:
Fun fact, installing the new drivers caused my sli to become disabled, and so I was getting lower performance after updating. I went in to the config, and the sli option was checked but it said there were changes detected so I clicked apply and my scores shot back up afterwards.


I've always had to reset my SLI and Physx after each driver upgrade, or maybe I just double check I forget since I'm not running SLI for right now with my 8800 to 470 swap.
____________________________
FFXI:Sylph - Perrin 75 Hume THF; Retired (At least from my use any way)
EVE Online:ScraperX; Retired
WAR:IronClaw- Peryn SW;SkullThrone- Grymloc BO; Retired


#591 Jul 29 2010 at 12:43 AM Rating: Decent
Scholar
Avatar
**
600 posts
I have one question... how do I get my window to be a bit smaller while running benchmark in high? I can't see the score :/
____________________________

Quote:
Fiddle Faddle!

#592 Jul 29 2010 at 6:05 AM Rating: Decent
Sage
***
1,675 posts
jayfly wrote:
I have one question... how do I get my window to be a bit smaller while running benchmark in high? I can't see the score :/


The only way I've done it was to use Aero Flip in Windows 7 after the benchmark was over.
#593 Jul 29 2010 at 7:56 AM Rating: Decent
Scholar
*
73 posts
High: 2552
Low: 3883

Intel Core 2 Duo E8600 @ 4.0 Ghz
XFX 4890 1gb @ 900/1050
8.0gb DDR2 @ 533 Mhz (5-5-5-15)
Gigabyte EP45-UD3P Mobo
Creative Xi-FI Audio Processor
4X WDC 250 gb RAID 0+1
Windows 7 Home Premium 64-bit


Just ordered a Sapphire Vapor-X 5870 1 gb video card. Will post improved scores once available.



Edited, Jul 30th 2010 12:19am by lsykora
____________________________
#594 Jul 29 2010 at 2:26 PM Rating: Good
Scholar
*
61 posts
So I got my new computer built and benched. Scores were about where I thought, if maybe a bit lower than I'd hoped-

Old build:
Core 2 Duo e4300 @1.8GHz
Radeon HD4890
2GB RAM (DDR2 I think)
160GB HDD (probably not the best one... but don't know a speed)

Scores: High: ~1250 Low: ~1800

New build:
ATI Phenom X6 1055T @2.8GHz
Radeon HD4890 (only part I kept from the old build)
4GB DDR3 RAM @1333MHz
1TB 7200RPM HDD

Scores: High: 2430 Low: 4000

Like I said, maybe not quite what I hoped for, but I'll still take it for now until some point in the future when GPU prices go down. :)
#595 Jul 29 2010 at 6:58 PM Rating: Decent
Scholar
*
247 posts
Low: 1724-1751

Intel C2D E6600 OC'd to 3.0GHz
eVGA 8800 GTS 320MB OC'd to 576MHz
3GB PC2-6400 DDR2 RAM

The CPU usage never got over 70%, and hovered constantly around 55-60%. Does this mean the GPU is the bottleneck for now?

I will probably upgrade the GPU to a GTX 460 and see how much improvement I get. If the GPU upgrade alone doesn't get me well ovre 2000 on low I will just build a new PC with an X4 955.

Note: It looks like just running task manager to check your CPU usage reduces your score by about 50-75 points. The character model you select may also have a 10-20 point reduction as well.

Edited, Jul 29th 2010 9:42pm by Enscheff
#596 Jul 29 2010 at 7:40 PM Rating: Good
Sage
**
743 posts
Enscheff wrote:
Low: 1662

Intel C2D E6600 OC'd to 3.0GHz
eVGA 8800 GTS 320MB OC'd to 576MHz
3GB PC2-6400 DDR2 RAM

The CPU usage never got over 70%, and hovered constantly around 55-60%. Does this mean the GPU is the bottleneck for now?

I will probably upgrade the GPU to a GTX 460 and see how much improvement I get. If the GPU upgrade alone doesn't get me well ovre 2000 on low I will just build a new PC with an X4 955.

Edited, Jul 29th 2010 8:59pm by Enscheff


I've got an e7400 (not OC'ed) and added a HD5770 and got over 2900 for low res. FYI
____________________________
I think you've been smoking the Moko...
http://na.finalfantasyxiv.com/lodestone/character/350413/
http://na.finalfantasyxiv.com/lodestone/character/1628942/
http://www.nerdist.com/
Angus of Cerberus (retired)
#597 Jul 29 2010 at 7:42 PM Rating: Decent
21 posts
Thanks to everyone that helped me in making my decision on the parts that I bought from newegg.com. After taking my sweet time, cause it's been a while, finally got everything together and running.

Just a quick spec list:

AMD Phenom II X6 1055T Thuban 2.8GHz Socket AM3 125W Six-Core
G.SKILL Ripjaws Series 4GB (2 x 2GB) 240-Pin DDR3 SDRAM DDR3 1333
MSI 890GXM-G65 AM3 AMD 890GX SATA 6Gb/s USB 3.0 HDMI Micro ATX
CORSAIR CMPSU-750TX 750W ATX12V / EPS12V SLI Ready CrossFire
GIGABYTE GV-N460OC-1GI GeForce GTX 460 (Fermi) 1GB 256-bit GDDR5

Scores:
Low: 4331
High 2596

So much better than my low of 1598 on my laptop.... Glad I made this choice!

Cheers
#598 Jul 29 2010 at 7:59 PM Rating: Excellent
Scholar
*
151 posts
Just finished building my first PC!

i5 750 (OC to 3.6 ghz)
GTX 465
4GB DDR3 Ram
Cooler Master HyperTX 3
ASUS P7P55 LX Mobo
Antec Nine Hundred Case


Low: ~5400
High ~2800

Looks like I'm set for plenty of FFXIV goodness.
____________________________
Flotsam Jetsam - Lindblum
lodestone.finalfantasyxiv.com/rc/character/status?cicuid=1570183
#599 Jul 29 2010 at 8:42 PM Rating: Good
Sage
**
770 posts
Sefi wrote:
Just finished building my first PC!

i5 750 (OC to 3.6 ghz)
GTX 465
4GB DDR3 Ram
Cooler Master HyperTX 3
ASUS P7P55 LX Mobo
Antec Nine Hundred Case


Low: ~5400
High ~2800

Looks like I'm set for plenty of FFXIV goodness.


Sweet congrats :) jsut for they you get a rank up in congradulatory honor:)
____________________________
I do not suffer from insanity.. I rather enjoy it.

{retired} Devalynn Mithra WHM extrodinare -Garuda (gives everyone a high paw! yeah!)

Church OF Mikhalia
#600 Jul 29 2010 at 9:23 PM Rating: Decent
Scholar
Avatar
**
600 posts
Finished building my rig today:

OS: Windows 7 64bit
CPU: AMD Phenom II X2 555 3.2 ghz (Unlocked to AMD Phenom II x4 B55 & OC'ed to 3.6 ghz)
Memory: 2 x 2 GB ddr2 memory PNY 800 mhz
Motherboard: Gigabyte GA-MA785GM-US2H Motherboard
GPU: Radeon HD 5770 (OC'ed a small bit from 855 to 900 mhz)
HDD: 500gb SATA hdd.


Score on LOW: 4395

Still need to test High.

I'm pretty happy for about a $600.00 starter rig :)
____________________________

Quote:
Fiddle Faddle!

#601 Jul 29 2010 at 10:07 PM Rating: Decent
Sage
**
784 posts
From what I've been hearing, even though this game is an "nvidia the way it's meant to be played" game, 5870s are outperforming 480s. I find this disturbing.

I was going to try out a couple of GTX 480s in SLI, but now I'm considering sticking with my 5870 and getting a second one for crossfire. Anyone have personal experience comparing the two cards, in single and/or SLI/crossfire? Or just experience with 480s (in relation to ffxiv)?

Cheers
____________________________
Amazing linkshell/guild hosting

This forum is read only
This Forum is Read Only!
Recent Visitors: 19 All times are in CST
Anonymous Guests (19)