Forum Settings
       
This Forum is Read Only

Official FFXIV Benchmarking program.Follow

#202 Jun 16 2010 at 2:40 PM Rating: Decent
28 posts
1242 XP Home AMD Athlon 64 X2 4400+ GF9600 GSO

This is mine so far being the best. I doubt I'm getting anything better in ways of the OS system I have. I'm just not exactly sure where I should start when updating my computer.
#203 Jun 16 2010 at 2:51 PM Rating: Good
Sage
****
5,587 posts
I wonder if once the actual game is released if we'll be able to adjust the resolution down lower than the two options in the benchmark, or at least do it through the registry. Realize the "low" setting in the benchmark is still higher than what options were available in FFXI's stock config without hacking the registry. I don't care if it ever runs on high on my machine since it's beyond my laptop screen's native resolution anyway.
____________________________
Harri
80BLU/80BST/76RNG/75THF/75WHM/60SCH
100+3 Bonecraft
#204 Jun 16 2010 at 3:05 PM Rating: Good
**
592 posts
Chances are, if you can run DA:O, Bioshock 2, and other titles also on the PS3 at medium/high settings then you *should* be able to run XIV without the system going to total ****. Crowded areas will be an issue, it may detract from the gaming experience, but many of the posters here seem to be in a state of panic about upgrading or building their systems right now over the results of a benchmark. Let me point out again that even if you disabled Aero, older systems are not going to run as well windowed as they do full screen, so your game play in full screen should be better than indicated.

There is slated to be a new generation of ATI cards out before the end of the year. AMD just came out with a 6-core processor that tests at ~70% of the $1000 Intel processor and only costs $300. There will be price drops for the holidays on all other hardware prices. There is no need to be overly concerned about (unoptimized) benchmark results and spend money now that could be used to pay off credit card debt or thrown into a 6 month CD on hardware that will drop in price before the game is released. I'm not trying to convince anyone with a <1500 benchmark that you won't need to upgrade, but it makes more sense to save some money and wait to see than ordering hardware that will be dated before the game is released.

In my experience with other games, the best drivers for a game aren't released for a solid month or two after the game goes on sale. Using the most up to date driver before release, and then waiting a couple months can make a frame rate difference of 10-20 fps. Even more on poorly programmed games like AoC at release. DA:O was especially challenging to get to run at 1080 with a single 8600M GT, but that was primarily because it liked to force-on several pieces of eye candy that I'd rather live without (are water reflections *really* necessary?). I expect there will be ways to dumb down XIV, as well.

tl;dr
Save $$$; try XIV with current rig; Use saved $$$ to upgrade if/when necessary.
____________________________
Inralkil-Seraph 75NIN/75SAM/68BST

Retired: Inra-Dark Crag 40/40 Witch Elf
Retired: Hollow-Thunderlord 70 Warlock S1/S3 T4 SL/SL
Retired: Horknee-Thunderlord 70 Druid T4/T5 Feral
#205 Jun 16 2010 at 3:15 PM Rating: Good
***
2,614 posts
I've been hearing a lot about scores, but what kinds of framerates are people seeing? Have they finally lifted the 30fps cap at least?

I'd also like to hear more about what kind of CPU usage people are seeing from multi-core processors, especially hexacores.
#206 Jun 16 2010 at 3:16 PM Rating: Good
Sage
****
6,470 posts
PLDXavier wrote:
Eske wrote:
I got the same .dll error that an earlier poster mentioned. I updated my drivers, and started to download the DirectX SDK, before I realized that it would take me about 4 hours on the horrific connection at my girlfriend's house.

@#%^ it, I'm sure my computer can run the game fine enough.


Try going to Microsoft's website and do a search for "DirectX End-User Runtime". The first result should be what you need for that DX9 dll (the page's section heading should read, "DirectX End-User Runtime Web Installer").

Edited, Jun 16th 2010 4:11pm by PLDXavier


Thanks, for the help : )

I'm not gonna bother with it though...I was sort of just checking out the benchmark to procrastinate on the work that I really ought to do be doing. I know my comp will run FFXIV fine enough.
____________________________
Latest Articles:
Monaco: What's Yours is Mine Review

Follow me on Twitter!
#207 Jun 16 2010 at 3:20 PM Rating: Good
***
1,159 posts
Hulan wrote:
Yog, perhaps I should go revise my original post as you do bring up a good point. The i5 is somewhat of a special case, as it is a core that is below 2.8GHz and still good for high preformance. This is a mixture of it being a Quadcore and having excellent design and emulation technology (basically, it's technically not faster in terms of operations per second, but it does those operations more efficiently than other processors). This also holds true for the i7 that the OP is using. I'm a bit tired so I was thinking faster than I could keep myself straight. I was writing the post to help people who weren't sure what they should do to get higher scores, and made the silly assumption that they would not likely have cores as new as the i5 and i7s.

In your specific case, not really, you would see a little bit of an improvement, since you are getting bottlenecks during load time, but nothing drastic. This is just a guess, as I'd have to look at your readings individually, and maybe not even then, but I'd say your i5 is probably past the point where it's slowing down your GPU. Once you pass that point, it's better to upgrade your GPU, although there are always shades of gray. At speeds like that (4k+) you're going to be running into RAM I/O issues, so you could possibly increase your score by getting RAM with faster timings. Or even things like how quickly your Hard Drive pages memory (this happens less often today since there's just so much space). So to conclude, you're computer is fast enough that any one thing is probably not going to make a huge difference, you need to tweek this and that to find what works best.

I apologize if some of this information is a little jumbled or misstated, I'm getting a little tired, and I do not do my best thinking this late.


Thanks for the clarification. I think I'll try to plug in that extra 4gigs of ram I have and overclock my cpu a bit to see if I can squeeze some more juice out of my rig. As it stands though, I'm doing better than most people so I suppose I should be content with the performance I have but it would be nice to reach at least 4500.
#208 Jun 16 2010 at 3:22 PM Rating: Decent
*
134 posts
The One and Only Aurelius wrote:
Zemzelette wrote:
2,126 on Low.
1,202 on High.

Windows 7 (64 bit)
Intel Duo Core 2 Duo CPU P8600 @ 2.40 GHz
NVIDIA GE Force GTX 260 M
4.00 GB Ram

Auuughhhh! Not low enough to justify buying a whole new desktop rig. Not high enough to be satisfactory.
Don't suppose any techsavvy gent out there might know if throwing more RAM at the problem would do any good? (that's about all I can do to a do laptop, ne?)



You'd probably see an improvement if you bumped it up to 6 or 8GB, but the question is whether or not the improvement would justify the cost. Something tells me that it wouldn't. Even if the extra RAM is super cheap, I'm not sure you'd see that much of a difference. You should be squared away to get yourself started, but I'd seriously look at moving over to a desktop system.


As far as the benchmark goes, it uses like 600MB of RAM at 1920x1080, so unless you are already using over 3.5GB of your RAM, I wouldn't think you would see a higher benchmark score.

I was using an average of 20% of my CPU, all cores were pretty even. During the loading sequences I was seeing loads of 85-90% on all CPU cores. My GPU usage was pegged at 99%, I don't have anything that monitors VRAM usage. My scores were:
High: 2,524 | 2,657 OC
Low: 4,134

AMD PhenomII x4 955 @ 3.2ghz.
Gigabyte HD4890 1GB 975mhz VRAM, 850mhz GPU | OC: 1100mhz VRAM, 900mhz GPU.
Mushkin 1600mhz 4GB RAM.

Edited, Jun 16th 2010 4:41pm by xXMalevolenceXx
#209 Jun 16 2010 at 3:24 PM Rating: Decent
10 posts
Score 1080p: 4388
Score 720p: 7117

Desktop PC (custom build)
Core i5 @ 3.65 Ghz OC
ATI 5850
6GB Corsair Dominator RAM DDR3
SATA HD @ 7200 RPM

img: http://img35.imageshack.us/img35/963/benchffxivoc.jpg



Edited, Jun 16th 2010 5:31pm by Tokion
#210 Jun 16 2010 at 3:38 PM Rating: Good
*****
11,539 posts
BluemageOfDoom wrote:
Quote:
Despite GTAIV clearly stating that the requirements are a "Core 2 Duo 1.8 GHz", I tried to run it on my P4HT 3.0. Spoiler: It didn't run very well. I got under 2 FPS, and it would usually force me to manually kill the system because I couldn't even force quit the program.


Depends on your gfx card to.
Try performance mods for GTAIV from Enbdev, that should boost FPS to decent levels.

hopefully people will be working on texture/shader mods for FFXIV to improve performance for lowend PCs.


The card in the system in question was a Radeon 3650. I've since replaced everything in that system besides the case, HD, and optical drive though; and it runs like a charm now :)


Yogtheterrible wrote:
Don't want to be vain here (I totally am) but my score was a high res score, not a low res.



I assumed a score was low res if someone only mentioned one score and didn't specify. My apologies.

Maybe once this thread hits 8-10 pages, I might go back and add to the list further.

Edited, Jun 16th 2010 5:39pm by Mikhalia
____________________________
[ffxisig]55836[/ffxisig]

Mikhalia: and FWIW, my posts are 95% helpful, informative, or funny.
Mikhalia: only 5% or less of my posts are utter crap.
Tyapex: 393 posts of utter crap...
Mikhalia: Sounds about right.
#211 Jun 16 2010 at 3:56 PM Rating: Good
*
134 posts
Mikhalia wrote:
I'm noticing that some Radeons under 5k are doing well and some over 5k have done less than stellar, which leads me back to the thought that this game may rely more on CPU than GPU.


The 5000 series ATI cards are 128-bit (even a few lower end 5000 series are 64-bit) until you get into the 5800 series cards, which are 256-bit. Those 4000 series cards are 256-bit.
#212 Jun 16 2010 at 4:02 PM Rating: Good
*****
11,539 posts
xXMalevolenceXx wrote:
Mikhalia wrote:
I'm noticing that some Radeons under 5k are doing well and some over 5k have done less than stellar, which leads me back to the thought that this game may rely more on CPU than GPU.


The 5000 series ATI cards are 128-bit (even a few lower end 5000 series are 64-bit) until you get into the 5800 series cards, which are 256-bit. Those 4000 series cards are 256-bit.


That would definitely explain it then.
____________________________
[ffxisig]55836[/ffxisig]

Mikhalia: and FWIW, my posts are 95% helpful, informative, or funny.
Mikhalia: only 5% or less of my posts are utter crap.
Tyapex: 393 posts of utter crap...
Mikhalia: Sounds about right.
#213 Jun 16 2010 at 4:24 PM Rating: Decent
**
473 posts
Hey I own an Alienware M17x will be doing the benchmark tonight

My config is intel core 2 quad Q9000 me thinks
with 4GB memory
dual 260M
and SSD for HD running RAID 0


I think I should be good will need to upgrade ram
and possibly the video cards in the near future or try OC'ing
the 260's

As far as buying alienware yes some of the cost is the brand
however I have not had a single issue with the laptop
all games run crisp and clean and the upgraded screen, alienware software,
and support are top notch. Now that Dell own alienware support is better
and as are the drivers.

If you have the budget I so go with it.
If not this is a really nice system to look at for mobile gaming

http://ca.asus.com/product.aspx?P_ID=8je5Ot4HBnKOdT81

its comparable and from what I've seen a bit cheaper than alienware.
But Asus support is not as good as dell. And you are limited with ram speeds
compared to alienware if you go with an i5 proc
i7 you have way more choice.

Hope that helps
if you're going mobile gaming laptop
go alienware or asus if budget is an issue

I wouldn't trust anyone else

but for the money build a desktop a week or so before release
now that's bang for buck gaming in regards to ffxiv
#214 Jun 16 2010 at 4:29 PM Rating: Decent
Osarion, Goblin in Disguise wrote:
Quote:
My PC is 6 months old, and scored at highest a 533 on the benchmark. I'm a gamer, so i already realized that this piece of crap probably wouldn't be good enough, but most people aren't computer savvy enough to come to the same realization.



A 6 month gaming PC scoring a 533? My PC is like 3 years old and I'm around 2100 on low. Sure you have the newest drivers?

Actually, my gaming rig died, so i had to buy this piece of **** as a temp replacement. Money is a little tight right now. It's a 700 dollar wal-mart pre-made. That's how i knew it was crap, but i need to have access to the internet. I don't really like browsing on my PS3, so i bought this. I'll Get FF for PS3 if my new rig isn't done by then.
#215 Jun 16 2010 at 4:37 PM Rating: Decent
*
88 posts
so you can only select from 2 resolutions - so if you have say a 1680-1050 monitor and you select 1080 will it just downscale abit? or just go right down to 720 anyway?
#216 Jun 16 2010 at 4:50 PM Rating: Good
Tenfooterten wrote:

I understand what you are saying here, but please try to understand this:
SE saying that the graphic demands would be high is somewhat vague. To me, you are pretty much going to need a high end gaming machine to run FF XIV the way it's meant to be played. They never really gave us anything to compare in their statement. My PC is 6 months old, and scored at highest a 533 on the benchmark. I'm a gamer, so i already realized that this piece of crap probably wouldn't be good enough, but most people aren't computer savvy enough to come to the same realization.


The first part of the statement...that XIV's requirements would be quite high at release...could be construed as vague. The second part of the statement...that a good PC 5 years from now would be able to run it with no problems...made it a lot less vague. Think of computer hardware as functioning on the 2 year rule. If you buy the absolute best, cutting edge delicious goodies for a new PC today, 2 years from now your PC will still be considered "good" relative to whatever new technology becomes available. Not best, not awesome, but good. 2 years from that, the uber hardware you dropped your hard earned clams on will be considered "adequate" for most tasks and you'll still be able to run all the newest games and other miscellaneous shinies but you're going to start to see detail settings that you can't enable without a noticeable decrease in performance.

To consider that in the context of XIV and what SE has said, the "good" computer that will run XIV easily with maxxed settings isn't even available today. I would expect PLDXavier's rig to push somewhere in the 6-7000 range when he adds his second 480, which is still a ways from the 8k you'd need to meet the requirements of "max settings easily". That "good" PC SE is refering to will likely consist of hrrdware that is released as cutting edge technology two years from now (give or take).

So then take that and hold it against the denial. People buying rigs with components based on two year old technology were uninformed to think that it would hold a candle to what XIV is going to require. But when confronted with this reality based on what SE has said in conjunction with informed opinion, there's all this argument based more on hope than any meaningful substance. Which is fine in a sense, but after a while it also starts to get old. And when it reaches the point where people start pointing elsewhere to explin why their dated hardware isn't performing as they'd hoped it might, it starts to cross a line. I never in a million years thought I'd find myself defending SE after my last few miserable months in XI, yet here I am. People need to set aside the denial and start examining things in the context of what we're being told. Until they do, we're going to continue to have people sh*tting up threads with nonsense that only degrades the quality of the information to be found here. Hope is good. Optimism is good. Ignoring reality for the sake of either of those two things? Not so good.

Edited, Jun 16th 2010 3:52pm by Aurelius
#217 Jun 16 2010 at 5:10 PM Rating: Decent
I never said you were wrong. :P I actually always defend SE. People complain about CS, yet i have never had a single problem with CS *customer service. SE has always done what they said they would do. They told you to keep all your information, in case you need to contact CS, yet people would complain, and not have all of the information they needed.
SE said you will need High end graphics to run FF 14 on PC, and people are complaining that their store bought notebook's aren't good enough.
A lot of people are like this. Some are young, and others just use their comps to check e-mail (as a poster said earlier). They just don't understand.

#218 Jun 16 2010 at 5:17 PM Rating: Decent
Sage
**
770 posts
Tenfooterten wrote:
I never said you were wrong. :P I actually always defend SE. People complain about CS, yet i have never had a single problem with CS *customer service. SE has always done what they said they would do. They told you to keep all your information, in case you need to contact CS, yet people would complain, and not have all of the information they needed.
SE said you will need High end graphics to run FF 14 on PC, and people are complaining that their store bought notebook's aren't good enough.
A lot of people are like this. Some are young, and others just use their comps to check e-mail (as a poster said earlier). They just don't understand.



Just to add, not supportign any argument sides, For thoes who dont know laptops with built in graphic cards are not the best for gaming, they work great with videos but lack the required specifications for game input to output. If possible, some laptop cant do this, you can upgrade your graphics card to one that is not built in and run it off that. Personally my laptop SUCKS horribly lol, I can use a HDMI output to my tv to play videos works great, but for video games its lacking. PS3 has i think 4 graphic processers, most games dont use it to its full potential, from what im seeing from SE ffxiv might acually push thoes limits.
____________________________
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
#219 Jun 16 2010 at 5:45 PM Rating: Good
***
1,025 posts
Windows 7 Ultimate 64 bit / i7 930 / Geforce GTS 250 / 4 GB DDR3

3290 on Low. 1872 on High.

I'll likely be running it on Low for the most part anyhow. I don't have regular access to any displays with 1080p.
____________________________
日本台湾友好!
#220 Jun 16 2010 at 5:49 PM Rating: Decent
*
88 posts
got 4695 - high performance on 1080 mode on a 1080 tv.
running a core i7 920 at 3.2 ghz, a radeon 5870 and 6 gigs of ram. my score seems quite low considering it didnt drop below 60 frames per second once throughout the whole benchmark - i was running fraps - allot of the time it was around 100 fps.

Edited, Jun 16th 2010 7:49pm by jamiehavok
#221 Jun 16 2010 at 5:51 PM Rating: Excellent
**
473 posts
I remember on the crysis forums the same conversation.

Crysis is/was a very graphic intensive game. Now look almost any current gen system can run
it at almost full go. FFXIV will be the same deal I feel.

Quote:
Tenfooterten wrote:

I understand what you are saying here, but please try to understand this:
SE saying that the graphic demands would be high is somewhat vague. To me, you are pretty much going to need a high end gaming machine to run FF XIV the way it's meant to be played. They never really gave us anything to compare in their statement. My PC is 6 months old, and scored at highest a 533 on the benchmark. I'm a gamer, so i already realized that this piece of crap probably wouldn't be good enough, but most people aren't computer savvy enough to come to the same realization.


The first part of the statement...that XIV's requirements would be quite high at release...could be construed as vague. The second part of the statement...that a good PC 5 years from now would be able to run it with no problems...made it a lot less vague. Think of computer hardware as functioning on the 2 year rule. If you buy the absolute best, cutting edge delicious goodies for a new PC today, 2 years from now your PC will still be considered "good" relative to whatever new technology becomes available. Not best, not awesome, but good. 2 years from that, the uber hardware you dropped your hard earned clams on will be considered "adequate" for most tasks and you'll still be able to run all the newest games and other miscellaneous shinies but you're going to start to see detail settings that you can't enable without a noticeable decrease in performance.

To consider that in the context of XIV and what SE has said, the "good" computer that will run XIV easily with maxxed settings isn't even available today. I would expect PLDXavier's rig to push somewhere in the 6-7000 range when he adds his second 480, which is still a ways from the 8k you'd need to meet the requirements of "max settings easily". That "good" PC SE is refering to will likely consist of hrrdware that is released as cutting edge technology two years from now (give or take).

So then take that and hold it against the denial. People buying rigs with components based on two year old technology were uninformed to think that it would hold a candle to what XIV is going to require. But when confronted with this reality based on what SE has said in conjunction with informed opinion, there's all this argument based more on hope than any meaningful substance. Which is fine in a sense, but after a while it also starts to get old. And when it reaches the point where people start pointing elsewhere to explin why their dated hardware isn't performing as they'd hoped it might, it starts to cross a line. I never in a million years thought I'd find myself defending SE after my last few miserable months in XI, yet here I am. People need to set aside the denial and start examining things in the context of what we're being told. Until they do, we're going to continue to have people sh*tting up threads with nonsense that only degrades the quality of the information to be found here. Hope is good. Optimism is good. Ignoring reality for the sake of either of those two things? Not so good.


and tenfooterten is right in his logic. FFXIV probably will run fairly decent a mchine built
for gaming . . . not somehting you buy at bestbuy for 600-1k$
but on a decent system with quad or hex core proc and 4-8GB
ram and an up to date current architecture GPU. This will probably give you
the ability to run the game in 1080 with default or slightly improved settings.

as far as the diff between PS3 and PC there will be a marked difference I am calling that to
run 720P, if 1080 happens I would be impressed.

Mind you this is a benchmark test of your system. not the actual client having to load all the
textures 25-100 avatars and add in npc's . . . . .

even in ffxi in a highly camped HNM areas great rigs that time had a hard hard time
keeping up and lag was brutal. So please all of you with budget systems do not be deluded when
you're socres says the system won't run.

either upgrade or use PS3 if you're on a budget.
#222 Jun 16 2010 at 6:38 PM Rating: Good
***
1,978 posts
Windows 7 64 bit
intel i7 Q820
GeForce GTX 285m
4 GB RAM

High: 1042
Low: 2033

Does this seem about right? I bought this laptop pretty recently, kind of disappointed.
#223 Jun 16 2010 at 7:56 PM Rating: Good
***
1,159 posts
So I previously ran the benchmark a few times and averaged at around 4050. I then overclocked the cpu from 2.67 to 3.7 and added 4gigs of ram (to a total of 8) and it didn't change the average but very little(average was probably around 4070). I tried to overclock the gpu with the ati overclocking tool but it didn't actually work so I can't see if that would help any.
#224 Jun 16 2010 at 8:17 PM Rating: Decent
Scholar
*
117 posts
831 on low. I'm certain my GPU is bottlenecking.....at least I hope thats the case =/

Windows 7 Ultimate 64
MSI 790FX-GD70 mobo
AMD 956 @3.4
4 gb DDR3 1066
9500GT 1gb


I just built this this past weekend but had to use an old card. I will be upgrading to crossfire, but I'm not 100% on which cards just yet...any suggestions? lol (budget for cards will be approx. 250~) and I will be OC'ing the cpu and memory.
#225 Jun 16 2010 at 8:18 PM Rating: Decent
*
134 posts
Yogtheterrible wrote:
So I previously ran the benchmark a few times and averaged at around 4050. I then overclocked the cpu from 2.67 to 3.7 and added 4gigs of ram (to a total of 8) and it didn't change the average but very little(average was probably around 4070). I tried to overclock the gpu with the ati overclocking tool but it didn't actually work so I can't see if that would help any.


I put a mild OC on my video card... 50mhz gpu, 200mhz vram and scored 150 higher on 1920x1080. It is my guess that unless you have a horrible cpu with small cache, OC/upgrading your gpu will give you more performance than the cpu. I'll run the benchmark again with a small OC on my 3.2ghz x4 cpu to 3.4ghz with stock video card frequencies and compare. I might also underclock to 2.8ghz or so and see what happens, again leaving my video card stock.
#226 Jun 16 2010 at 8:22 PM Rating: Decent
*****
12,622 posts
I get 2700 in high resolution mode.

Guess it's time to upgrade the box!
____________________________
Blah
#227 Jun 16 2010 at 8:24 PM Rating: Decent
*
134 posts
jhariya wrote:
I just built this this past weekend but had to use an old card. I will be upgrading to crossfire, but I'm not 100% on which cards just yet...any suggestions? lol (budget for cards will be approx. 250~) and I will be OC'ing the cpu and memory.


Just my opinion, but I would buy a single $250 card instead of two $125 cards. When I upgrade my video card I tend to stick to the $200 range. Anything cheaper and I'm probably not going to have satisfactory performance, anything more expensive is just too expensive for me, and I feel like I'm starting to pay for new technology at that point. If I was going to buy something today I would probably get a Sapphire HD5830.
#228 Jun 16 2010 at 8:46 PM Rating: Good
***
1,159 posts
xXMalevolenceXx wrote:
jhariya wrote:
I just built this this past weekend but had to use an old card. I will be upgrading to crossfire, but I'm not 100% on which cards just yet...any suggestions? lol (budget for cards will be approx. 250~) and I will be OC'ing the cpu and memory.


Just my opinion, but I would buy a single $250 card instead of two $125 cards.


I agree, single cards perform better and a lot of games don't have very good support for crossfire or sli.
#229 Jun 16 2010 at 8:48 PM Rating: Decent
I fully expected my laptop to land smack in the middle of average, based on FFXI performance and useless "Windows Experience Index" scores. It scored 568. Smiley: bah

I'm sure if I tried it, the game would probably still run better than my atrocious score would suggest, but it's looking like I'll be playing on PS3.
____________________________
[ffxisig]49934[/ffxisig]

SMN 90 | BLM 90 | WHM 83
twitter.com/nupinu
It rubs the fewell on its skin or else it gets the thwack again.

Thydonon wrote:
SMN is just epic in it's scope. You gain fame in all cities to have their avatars. When others finish ToAU, they get toys; you get gods.
#230 Jun 16 2010 at 8:56 PM Rating: Good
***
1,159 posts
Nupinu the Eccentric wrote:
I fully expected my laptop to land smack in the middle of average, based on FFXI performance


You could strap two hamsters to a series of wooden cogs and a tissue box with a candle in it and get it to run FFXI.
#231 Jun 16 2010 at 9:01 PM Rating: Decent
**
697 posts
Fun fact: After getting a 1434 on low on my rig, I tried it out on the wife's rig:
Windows 7 64
8 gb ram
Intel i7
...factory graphics card.

...and it got a 600. It bests my computer by 4x the ram (i only have 2 gigs!) and has a better processor (I'm using a newer AMD quadcore, forget exactly what), but man, the graphics card makes all the difference. Now i need to just "borrow" some of her RAM to get mine up to par.

BTW, if anyone knows of any good deals on RAM, i'm looking to get 2-4 more gigs, but don't want to spend over 100$.
____________________________
FFXI: Odin - Merylstryfe Summoner Woo Hoo!


#232 Jun 16 2010 at 9:06 PM Rating: Excellent
Scholar
**
602 posts
So I figure in the end everyone is asking themselves one question, Build a new custom PC or get a Playstation 3. If I were any of you, look at the card that is leading the chart right now, the card right before the latest nVidia/ATI does extremely well and you will probably have 0 troubles running the game in 30fps/60fps. If you wait until FFXIV's release you are probably going to be able to pick up a 5870 for 150.00 instead of the 400.00 it is now. Not to mention the latest i7 now will be cheaper then as well. Everything will be cheaper.

Honestly, I think you should just wait until just before the release and then start building your computer. Save yourself 500.00 to 1,000.00 by doing it this way. Of course you could build it with the latest technology too since you can just start saving money now. I have even been in contact with a few people that score low 0-1000 on the Benchmark but are in the Alpha. They say the game is barely able to run. Definite proof to me that I need to upgrade my system as well.

Will I play it on PS3 or a new PC? Is Final Fantasy XIV going to be worth investing 300.00+ into? Maybe I should just stick with Final Fantasy XI as my only/last MMO. I am in between. But I easily spend 1,000.00 easily every couple of months on retarded things like Blu-ray movies (I have Netflix so considering myself spending money on BR pointless but I do anyways), buying new Playstation 3 games (60.00 a pop), fast food (yeah, let's not go there), new toys (cameras, cellphone EVO/Iphone 4G *cough*).

The only thing I wonder, I have seen a 720p score reach that high threshold they have for the benchmark but there hasn't been a PC able to hit the highest score limit in 1080p. Since it seems more cores makes the game run way better, is the benchmark biased towards the Playstation 3 with it's 8 cores? Hopefully 8 core CPUs will be here by Christmas if they aren't already. And I know 8 core Server CPUs exist, I meant for the desktop user.

Edited, Jun 16th 2010 11:11pm by Excenmille
#233 Jun 16 2010 at 9:27 PM Rating: Decent
*
134 posts
xXMalevolenceXx wrote:
I put a mild OC on my video card... 50mhz gpu, 200mhz vram and scored 150 higher on 1920x1080. It is my guess that unless you have a horrible cpu with small cache, OC/upgrading your gpu will give you more performance than the cpu. I'll run the benchmark again with a small OC on my 3.2ghz x4 cpu to 3.4ghz with stock video card frequencies and compare. I might also underclock to 2.8ghz or so and see what happens, again leaving my video card stock.


I just ran the benchmark at 3.4ghz with stock video card frequencies. There was no difference. At stock I scored 2,524 and with a small 200mhz OC I scored 2,514. My CPU usage is around 20%, so it sort of makes sense a small OC won't do anything. I'm not going to bother underclocking to see if that makes a big difference, but I suspect I could underclock a fair bit and it wouldn't make any difference.

Compare that to OC my GPU a mere 50mhz and VRAM 200mhz and scoring 150 higher (2,657). It's my opinion for this benchmark, improving your video card performance is more beneficial than CPU performance, unless you have a terrible CPU. Of course, if you have a kinda bad CPU you might get good results by OC the CPU and GPU, but as far as one vs the other, GPU is more important IMO... again, assuming your CPU doesn't completely blow balls and isn't throttling your whole system.

I have not tried a significant CPU OC, but I suspect I won't gain any performance for this benchmark. I also believe (but can't confirm) increasing your RAM storage won't help unless the ~600MB of RAM this bench uses is a problem for you. If you have 4GB of RAM and only using 50%... another 4GB isn't going to help score higher in the benchmark. Getting faster RAM or tightning the timings might help slightly, but I still wouldn't expect much. I believe this bench is mostly about the GPU. That's why you see these guys with ATI 58xx and nVidia 480s scoring over 4,000.

Of course, I could be totally wrong about everything.

Edited, Jun 16th 2010 10:29pm by xXMalevolenceXx
#234 Jun 16 2010 at 9:40 PM Rating: Decent
Scholar
44 posts
Excenmille wrote:

The only thing I wonder, I have seen a 720p score reach that high threshold they have for the benchmark but there hasn't been a PC able to hit the highest score limit in 1080p. Since it seems more cores makes the game run way better, is the benchmark biased towards the Playstation 3 with it's 8 cores? Hopefully 8 core CPUs will be here by Christmas if they aren't already. And I know 8 core Server CPUs exist, I meant for the desktop user.

Edited, Jun 16th 2010 11:11pm by Excenmille


From what Ive seen its really not based on the number of cores but the clock speed. I was looking at other charts and the 4 core i7s that were overclocked to 4.5ghz were able preform very simliar to the 6 core i7. About the 8 core desktop/enthusiast CPUs, those won't be out until late 2011 from Intel, called Sandy Bridge. I had an 8 core server before but sold it, I'll ask some of my buddies who have dual xeons and 480's to run the bench. I think that if the benchmark program was SLI or Xfire enabled we could see quite a few people hit 8000 on 1080p.
#235 Jun 16 2010 at 9:41 PM Rating: Excellent
More important than any of this, did anyone leave it at the final screen where it posts your system specs and score and let the music run for a while? Nice twist on the old school FF main theme thar, Uematsu. Well played.
#236 Jun 16 2010 at 9:45 PM Rating: Decent
Scholar
**
384 posts
Why the @#%^ does this benchmark not support SLI? Is SLI not the best anymore? I know there is SLI 3-way and crossfire, but I thought SLI was supposed to be pretty **** good...

Edited, Jun 16th 2010 10:46pm by drunktexan
____________________________
Drunktexan of Phoenix ~ Mithra

Monsieur MojoVIII wrote:
I campaign like I make love.



Naked and bellowing while swinging my large weapon at ugly things.
#237 Jun 16 2010 at 9:49 PM Rating: Good
drunktexan wrote:
Why the @#%^ does this benchmark not support SLI? Is SLI not the best anymore? I know there is SLI 3-way and crossfire, but I thought SLI was supposed to be pretty **** good...

Edited, Jun 16th 2010 10:46pm by drunktexan


SLI and Crossfire are like Eyefinity...support comes from the driver side, not the application side. SLI/no SLI is completely transparent to the application. I don't know where people came up with the idea that it doesn't support SLI, but meh. The video of Tanaka showing off the 3D...did you see the three monitors it was running on? Not many people are running Eyefinity/nVidias triple monitor option on a single card. The required resolution is much too high to run on even a top end single card for a high res triple monitor setup.
#238 Jun 16 2010 at 10:09 PM Rating: Good
Scholar
**
384 posts
The One and Only Aurelius wrote:
drunktexan wrote:
Why the @#%^ does this benchmark not support SLI? Is SLI not the best anymore? I know there is SLI 3-way and crossfire, but I thought SLI was supposed to be pretty **** good...

Edited, Jun 16th 2010 10:46pm by drunktexan


SLI and Crossfire are like Eyefinity...support comes from the driver side, not the application side. SLI/no SLI is completely transparent to the application. I don't know where people came up with the idea that it doesn't support SLI, but meh. The video of Tanaka showing off the 3D...did you see the three monitors it was running on? Not many people are running Eyefinity/nVidias triple monitor option on a single card. The required resolution is much too high to run on even a top end single card for a high res triple monitor setup.


Fair enough, ignorance on my part...

Ok my current setup (I just ordered Windows 7 Ultimate):

WIndows XP Pro (5.1, Build 2600) Service Pack 3
INtel Core2 Extreme CPU Q6850 @ 3.00Ghz (4 cpu)
Nvidia GeForce GTS 250

I have 2 of the 250's set up in SLI...
Driver version is 196.21 (I seem to get better performance with FFXI on this driver)

So its a driver issue? If I install to the newest nvidia driver (257.21) would that help fix the SLI problem?

____________________________
Drunktexan of Phoenix ~ Mithra

Monsieur MojoVIII wrote:
I campaign like I make love.



Naked and bellowing while swinging my large weapon at ugly things.
#239 Jun 16 2010 at 10:13 PM Rating: Good
drunktexan wrote:
Fair enough, ignorance on my part...

Ok my current setup (I just ordered Windows 7 Ultimate):

WIndows XP Pro (5.1, Build 2600) Service Pack 3
INtel Core2 Extreme CPU Q6850 @ 3.00Ghz (4 cpu)
Nvidia GeForce GTS 250

I have 2 of the 250's set up in SLI...
Driver version is 196.21 (I seem to get better performance with FFXI on this driver)

So its a driver issue? If I install to the newest nvidia driver (257.21) would that help fix the SLI problem?



You haven't explained what issue you're having, but as a general rule it's never advisable to run with anything less than the best drivers available at any given time. If it's a case of the benchmark not even running, I'd say it's highly likely that running ancient drivers is part of the problem. If it's a low performance issue, I'd say running dated video cards...even in SLI...is the problem.
#240 Jun 16 2010 at 10:29 PM Rating: Good
Scholar
**
384 posts
The One and Only Aurelius wrote:

You haven't explained what issue you're having, but as a general rule it's never advisable to run with anything less than the best drivers available at any given time. If it's a case of the benchmark not even running, I'd say it's highly likely that running ancient drivers is part of the problem. If it's a low performance issue, I'd say running dated video cards...even in SLI...is the problem.


Ok, I appreciate your patience.

1. The issue I am having is that the benchmark runs, but does not recognize that I have 2 video cards running in SLI (the end screen with your scores only shows one card).

2. I am running that old of driver because I am running FFXI on that computer and I seem (no benchmark) to get the best performance while running that program on my computer using that driver.

3. I originally had dual nvidia 8800GT 512 MB PCI-E cards installed. Alienware replaced them with the 250's when one of the 8800's died.

Again, I appreciate you help and patience.

P.S. I get a score of 2273 High and about 3900 Low. I just want the **** thing to use both cards.

Edited, Jun 16th 2010 11:30pm by drunktexan

Edited, Jun 16th 2010 11:31pm by drunktexan
____________________________
Drunktexan of Phoenix ~ Mithra

Monsieur MojoVIII wrote:
I campaign like I make love.



Naked and bellowing while swinging my large weapon at ugly things.
#241 Jun 16 2010 at 10:55 PM Rating: Excellent
Scholar
**
602 posts
Don't rate me down but I have never understood why people do the SLI/Crossfire thing. You're only paying for 30% more efficiency so in the end you're getting ripped off by nVidia/ATI. I would never run a dual setup unless I got the same performance of each card x2. It just seems a waste of money. Something always dies inside of me when people mention it. Not knocking anyone for wanting 30% more increase in graphics but you're just throwing money away, where's the other 70%?

Also the only 30% increase, is that a hardware limitation or software limitation based on a per game basis?

Edited, Jun 17th 2010 12:57am by Excenmille
#242 Jun 16 2010 at 11:01 PM Rating: Good
Scholar
**
384 posts
Excenmille wrote:
Don't rate me down but I have never understood why people do the SLI/Crossfire thing. You're only paying for 30% more efficiency so in the end you're getting ripped off by nVidia/ATI. I would never run a dual setup unless I got the same performance of each card x2. It just seems a waste of money. Something always dies inside of me when people mention it. Not knocking anyone for wanting 30% more increase in graphics but you're just throwing money away, where's the other 70%?

Also the only 30% increase, is that a hardware limitation or software limitation based on a per game basis?

Edited, Jun 17th 2010 12:57am by Excenmille


Same reason why people upgrade Hauby --> Hauby +1...

In reality it was because the computer was a gift to myself for finishing lawschool and getting a paying job. I actually built my own computers long ago when I had more time than money, now I have more money than time.
____________________________
Drunktexan of Phoenix ~ Mithra

Monsieur MojoVIII wrote:
I campaign like I make love.



Naked and bellowing while swinging my large weapon at ugly things.
#243 Jun 16 2010 at 11:08 PM Rating: Decent
9 posts
The benchmark seems to notice sli at the end it shows 2 video cards for people that have it,
I wonder if it would show a tri sli setup. I wouldn't dare spend the money on that, sli seems like a rip off, It only splits the screen in half, one video card takes the top and the other the bottom. There have also been very few games that take advantage of sli in game options. I doubt FFXIV will be a exception.
#244 Jun 16 2010 at 11:12 PM Rating: Good
Scholar
**
384 posts
Nosul wrote:
The benchmark seems to notice sli at the end it shows 2 video cards for people that have it,
I wonder if it would show a tri sli setup. I wouldn't dare spend the money on that, sli seems like a rip off, It only splits the screen in half, one video card takes the top and the other the bottom. There have also been very few games that take advantage of sli in game options. I doubt FFXIV will be a exception.


That was kind of my point. The benchmark does not seem to be recognizing the 2nd card, the end screen only shows one card.


Edit: and yes, I checked, I do have SLI enabled.


Edited, Jun 17th 2010 12:13am by drunktexan
____________________________
Drunktexan of Phoenix ~ Mithra

Monsieur MojoVIII wrote:
I campaign like I make love.



Naked and bellowing while swinging my large weapon at ugly things.
#245 Jun 16 2010 at 11:19 PM Rating: Excellent
Excenmille wrote:
Don't rate me down but I have never understood why people do the SLI/Crossfire thing. You're only paying for 30% more efficiency so in the end you're getting ripped off by nVidia/ATI. I would never run a dual setup unless I got the same performance of each card x2. It just seems a waste of money. Something always dies inside of me when people mention it. Not knocking anyone for wanting 30% more increase in graphics but you're just throwing money away, where's the other 70%?

Also the only 30% increase, is that a hardware limitation or software limitation based on a per game basis?


GTX480 x 2 is more like a 70% increase according to early benchmarks. Also, 30% for adding a second card is still 30% and in the case of people buying top end cards, that's their only option to bump the performance even more. For those buying more dated cards, it's often a cheaper alternative to squueze a little extra performance out of a second, older card than it is to replace the first card with a newer model. Crossfire/SLI setups often produce better benchmark results than top end cards from the next generation. In my case, in a year or two if I want to squeeze a bit more performance out of my rig without going whole hog, I can throw in an a second 5870 for $150-200 and milk a little bit more performance out of my rig before jumping into another top-end single card build.
#246 Jun 16 2010 at 11:20 PM Rating: Good
Scholar
**
384 posts
Ok, update. I downloaded the most recent driver. I uninstalled the old driver and rebooted... or at least I tried. Now the computer will not get through the POST. It appears that perhaps the computer was on the verge of dying anyway. At least this **** thing is under warranty.

in before FFXIV killed my computer :P

Edited, Jun 17th 2010 12:25am by drunktexan
____________________________
Drunktexan of Phoenix ~ Mithra

Monsieur MojoVIII wrote:
I campaign like I make love.



Naked and bellowing while swinging my large weapon at ugly things.
#247 Jun 16 2010 at 11:21 PM Rating: Decent
9 posts
drunktexan wrote:


That was kind of my point. The benchmark does not seem to be recognizing the 2nd card, the end screen only shows one card.


Edit: and yes, I checked, I do have SLI enabled.


Edited, Jun 17th 2010 12:13am by drunktexan


that's odd, the op's video card shows 2 at the end.

http://img691.imageshack.us/img691/3821/ffxivbench.jpg

did you check to make sure everything is connected right?
#248 Jun 16 2010 at 11:22 PM Rating: Excellent
drunktexan wrote:
Nosul wrote:
The benchmark seems to notice sli at the end it shows 2 video cards for people that have it,
I wonder if it would show a tri sli setup. I wouldn't dare spend the money on that, sli seems like a rip off, It only splits the screen in half, one video card takes the top and the other the bottom. There have also been very few games that take advantage of sli in game options. I doubt FFXIV will be a exception.


That was kind of my point. The benchmark does not seem to be recognizing the 2nd card, the end screen only shows one card.


Edit: and yes, I checked, I do have SLI enabled.


I wouldn't worry about what the benchmark is telling you at the end. When I run it, it lists Radeon 5800 series twice, and I've only got one card. If you want to determine whether or not it's making use of the second card, write down your test results under SLI and then pull the second card and run the tests again. You're assuming it's not recognizing...I'm telling you the software doesn't know or care whether you're running SLI or not...it's all handled driver side.
#249 Jun 16 2010 at 11:28 PM Rating: Good
Scholar
**
384 posts
The One and Only Aurelius wrote:
I wouldn't worry about what the benchmark is telling you at the end. When I run it, it lists Radeon 5800 series twice, and I've only got one card. If you want to determine whether or not it's making use of the second card, write down your test results under SLI and then pull the second card and run the tests again. You're assuming it's not recognizing...I'm telling you the software doesn't know or care whether you're running SLI or not...it's all handled driver side.


Yeah, now it appears that I have bigger fish to fry... The motherboard seems to be having issues, and I will have to contact support :/ thank you for the help anyway.
____________________________
Drunktexan of Phoenix ~ Mithra

Monsieur MojoVIII wrote:
I campaign like I make love.



Naked and bellowing while swinging my large weapon at ugly things.
#250 Jun 16 2010 at 11:35 PM Rating: Good
Scholar
23 posts
Low: 3935
High: 2641

Intel core i3-530 2.93GHz OC@3.52GHz
Kingston HX1600 2x2GB
Radeon HD5770 128-bit 1GB
WD caviar SATA 7200RPM 16MB non-RAID config
Asus P7P55-M
Windows 7 64-bit

#251 Jun 17 2010 at 12:39 AM Rating: Good
Thief's Knife
*****
15,053 posts
xXMalevolenceXx wrote:
The One and Only Aurelius wrote:
Zemzelette wrote:
2,126 on Low.
1,202 on High.

Windows 7 (64 bit)
Intel Duo Core 2 Duo CPU P8600 @ 2.40 GHz
NVIDIA GE Force GTX 260 M
4.00 GB Ram

Auuughhhh! Not low enough to justify buying a whole new desktop rig. Not high enough to be satisfactory.
Don't suppose any techsavvy gent out there might know if throwing more RAM at the problem would do any good? (that's about all I can do to a do laptop, ne?)



You'd probably see an improvement if you bumped it up to 6 or 8GB, but the question is whether or not the improvement would justify the cost. Something tells me that it wouldn't. Even if the extra RAM is super cheap, I'm not sure you'd see that much of a difference. You should be squared away to get yourself started, but I'd seriously look at moving over to a desktop system.


As far as the benchmark goes, it uses like 600MB of RAM at 1920x1080, so unless you are already using over 3.5GB of your RAM, I wouldn't think you would see a higher benchmark score.

I was using an average of 20% of my CPU, all cores were pretty even. During the loading sequences I was seeing loads of 85-90% on all CPU cores. My GPU usage was pegged at 99%, I don't have anything that monitors VRAM usage. My scores were:
High: 2,524 | 2,657 OC
Low: 4,134

AMD PhenomII x4 955 @ 3.2ghz.
Gigabyte HD4890 1GB 975mhz VRAM, 850mhz GPU | OC: 1100mhz VRAM, 900mhz GPU.
Mushkin 1600mhz 4GB RAM.

Edited, Jun 16th 2010 4:41pm by xXMalevolenceXx


Just an FYI if you have 32-bit windows you probably don't have 3.5 gigs of RAM available. The maximum amount of RAM 32-bit windows can access is 4 gigs minus your video card's RAM and the adddress space used by your hardware. And yes this means if you have a video card with 2 gigs you now have less than 2 gigs of RAM available in 32-bit windows.


3990 at 1920x1080 with a Radeon 5850 btw.



Edited, Jun 17th 2010 4:01am 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
This forum is read only
This Forum is Read Only!
Recent Visitors: 13 All times are in CST
Anonymous Guests (13)