WoW Benchmarks

[Wreckage]

I've noticed a lot of people ask for World of Warcraft benchmarks. Xbit just did a big MMORPG round up including WoW using more affordable cards. (HD4670-GTX260)

Also the next WoW patch is supposed to tune up the shadows setting that's been such a FPS killer.

http://www.xbitlabs.com/articl...s-mmorpg_14.html#sect0

 

[TheMeanestGuest]

Well, I suppose that's fairly informative for people who enjoy MMOs.

I read through most of it but I must have missed it, what was the reasoning for the exclusion of the HD 4870?

Edit: I feel like it would have made more sense to exclude the 4770 or 4830 if it was simply a matter of having an even number of cards.

 

[crisium]

Yeah, the reviewer is a little foolish in the card choices. The 4870 is actually even cheaper than the GTX 260 in most cases, so to exclude it but include the 260 makes absolutely no sense. There's also no need for the 4830 as it's purchasing power is all but completely defunct because the 4850 is only $10 more and is modestly more powerful.

Even if they are all MMOs, it is nice to see some benchmark charts with some cards that people a generation or two back might consider buying.

 

[brblx]

wow seems like a silly gpu benchmark, shouldn't they be testing cpu's?

8800gt or better should be fine for that game unless you want to crank up ridiculous amounts of AA and AF at high resolution.

speaking of which, the 9800gt seems unusually strong there. does wow favor nvidia cards or is it the high AF?

 

[yh125d]

Yeah wow favors nVidia. The 9800GT is usually slower than 4850's but in that benchmark it's ~20% faster

 

[soccerballtux]

The WoW engine is such a bloated, ancient piece of garbage. I remember I would turn down all my graphics to minimum settings, 800x600 resolution, on my current card; and my FPS in Dalaran would be low 20s with processor at less than 45%.

Any time there were lots of characters on screen, even once the textures for armor were all loaded up, FPS was still terrible and CPU was not being utilized at all.

 

[Qbah]

Read the review, very informative The HD4670 isn't powerful enough to enjoy most MMOs to the fullest. However the omission of HD4870 is strange imo, as it's cheaper than a GTX260 and at least as fast. Not to mention they used all the slower ATi cards. Still, a nice review

 

[BenSkywalker]

wow seems like a silly gpu benchmark, shouldn't they be testing cpu's?

They are showing more then 100% scaling going from a 4670 to a GTX260, that doesn't tend to indicate the game is terribly processor limited.

Side note- those screenshots for WoW aren't maxed out details, textures are well below max level, AF looks like it is shut off(or on an ATi board) and shadow details are set to one notch above lowest.

Any time there were lots of characters on screen, even once the textures for armor were all loaded up, FPS was still terrible and CPU was not being utilized at all.

512MB board? One of the kids comps has a 1GB 9500GT in it(got it for like $20 ) and it smokes the 9600GT 512MB in certain spots in WoW, Dal certainly being one of them.

 

[Mango1970]

Now if they would go back... really back and do one for Everquest 1!!

 

[ShadowOfMyself]

Originally posted by: Mango1970
Now if they would go back... really back and do one for Everquest 1!!

Yeah such disappointing reviews... I hope they benchmark Diablo when the new DX11 cards get released

Also:

Originally posted by: Wreckage
I've noticed a lot of people ask for World of Warcraft benchmarks.

Originally posted by: yh125d
Yeah wow favors nVidia.

Reason why this thread exists

 

[Leyawiin]

I appreciate them doing this - they are some of the most played PC games out there - but its a shame they didn't have a direct competitor for the GTX 260 in the reviews (the HD 4870). I've gleaned over the years that WoW is more Nvidia friendly, but I would like to see confirmation across the board in all card catagories.

 

[Hardlin]

Does anyone know what they specifically did in the games? I play LOTRO and have an i7 920, 6GB, 4850 with Vista 64bit and I get better performance than they show. I only get as slow as they show in areas where everyone gets bogged down (compared to a friends 285). I only play in DX10 mode at 1920x1200 with the same AA and AF they tested with. Generally, I never drop below 56fps and go as high as 88fps.

Just thought it was strange that with a faster processor, they get slower results.

 

[vj8usa]

Originally posted by: Hardlin
Does anyone know what they specifically did in the games? I play LOTRO and have an i7 920, 6GB, 4850 with Vista 64bit and I get better performance than they show. I only get as slow as they show in areas where everyone gets bogged down (compared to a friends 285). I only play in DX10 mode at 1920x1200 with the same AA and AF they tested with. Generally, I never drop below 56fps and go as high as 88fps.

Just thought it was strange that with a faster processor, they get slower results.

The linked benchmarks are for WoW, not LOTRO.

edit: whoops, guess you were referring to the LOTRO benchmarks in the article. I got caught up in the whole WoW discussion, my bad.

 

[Hardlin]

If you read the whole article you will see the other games benchmarks.

edit: http://www.xbitlabs.com/articl...ds-mmorpg_9.html#sect0

 

[Wuzup101]

Hum... that was actually pretty useful. I don't play wow anymore; however, I have a younger brother that does. I am probably going to be building him a new computer for christmas, as his current athlon 64 and ati x800 based system sucks balls for pretty much anything at this point in time. He generally still manages to play wow at 1920x1200 though (with pretty much everything on low settings).

 

[nitromullet]

Originally posted by: ShadowOfMyself

Originally posted by: Mango1970
Now if they would go back... really back and do one for Everquest 1!!

Yeah such disappointing reviews... I hope they benchmark Diablo when the new DX11 cards get released

There are tons of reviews of these cards in more demanding DX10 titles, but not many reviews in MMOs. Nice to see someone doing something a little different. Although, the 4870 should have been included.

 

[Wreckage]

Patch 3.2 should be an improvement I hope....
http://www.worldofwarcraft.com...-realm-patchnotes.html

The frame rate for those with Shadow Effects enabled under the Ultra Video settings has been improved.

 

[BFG10K]

Originally posted by: TheMeanestGuest

Well, I suppose that's fairly informative for people who enjoy MMOs.

This is more than just about MMOs - it's about legacy games being tested.

 

[Denithor]

I'd like to see a mid-range GPU tied to a range of CPUs to see how much impact CPU power has on performance in these games.

 

[AyashiKaibutsu]

I'd like to see a similar article for CPUs too.

 

[lopri]

Correct me if I'm wrong - I think the reason why we don't see many GPU benches of MMO games is because it's difficult to get consistent test results. Since it is not possible to test MMO's off-line, there will be variables that cannot be effectively controlled. And the huge size of a game's world is also cutting against the usefulness of such tests.

It is good to see the review, however. MMO's and RTS's are the last (and strongest) strongholds of PC gaming, and the PC gaming industry needs all the help it can get right now.

 

[Denithor]

Very true.

WoW is the only MMO I have personal experience with and no two loggins are the same. However, some things hold true - the larger cities (formerly Shattrath, now Dalaran) suffer from lag & generally poor fps at peak times (mid to late evening) simply due to the number of people present. Large (25 man) end-game raids and the Wintergrasp battles (where you can have >100 people on each side duking it out in a confined space) also tend to slow appreciably.

One question I have - how do you tell the difference between network lag and poor fps due to your rig? When I play on my e8400/GTX 260 setup even 25-man raids with plenty going on onscreen don't slow me down. On my e7200/8800GTS rig these can slow (not quite into a slideshow but close) when there's a lot of AoE & large groups on screen. Now, a high count WG battle - in the fortress - even my GTX setup slows to a crawling slideshow. But I think that's more network/server lag than fps problems.

 

[AyashiKaibutsu]

Originally posted by: Denithor
Very true.

WoW is the only MMO I have personal experience with and no two loggins are the same. However, some things hold true - the larger cities (formerly Shattrath, now Dalaran) suffer from lag & generally poor fps at peak times (mid to late evening) simply due to the number of people present. Large (25 man) end-game raids and the Wintergrasp battles (where you can have >100 people on each side duking it out in a confined space) also tend to slow appreciably.

One question I have - how do you tell the difference between network lag and poor fps due to your rig? When I play on my e8400/GTX 260 setup even 25-man raids with plenty going on onscreen don't slow me down. On my e7200/8800GTS rig these can slow (not quite into a slideshow but close) when there's a lot of AoE & large groups on screen. Now, a high count WG battle - in the fortress - even my GTX setup slows to a crawling slideshow. But I think that's more network/server lag than fps problems.

Laggy network will make objects in the game jump around and be delayed/inconsistent, but you'll still have high FPS. If your games having trouble rendering everything will be up to date, but the graphics will be choppy.

 

[*kjm]

I would like to see that ran with the high end cards.......