Confused about 7970 World of Warcraft performance

GotNoRice

Senior member
Aug 14, 2000
329
5
81
I play other games like BF3 and Shogun 2: Total War but the vast majority of the time I spend playing games is spent playing World of Warcraft.

My 2x 4870x2 setup has served me very well but I am looking to upgrade at some point and I'm wondering if the 7970 is it. Obviously the performance in the main game I play is of great importance.

So far i've only been able to find WoW 7970 benchmarks on 3 websites:

http://www.techpowerup.com/reviews/AMD/HD_7970/23.html
http://www.tomshardware.com/reviews/radeon-hd-7970-benchmark-tahiti-gcn,3104-11.html
http://www.pcgameshardware.de/aid,8...xpress-30-und-28-nm/Grafikkarte/Test/?page=16

The TechPowerUp numbers for the 7970 in WoW are terrible. It struggles just to match the performance of a GTX580 and at 1920x1200 it is even beat by it's predecessor, the 6970.

The TomsHardware numbers are totally different. At 1920x1080 it is delivering performance better than both the GTX580 and 6990 by a significant margin, at least until you enable some advanced AA then it is back at GTX580 speeds. It also seems to fall pretty flat at eyefinity resolutions (though I don't run eyefinity).

The PCGamesHardware numbers are more like the TechPowerUp numbers but even worse, as now the 7970 is slower than the GTX580 and struggling to keep up with a GTX570...

So far only one benchmark (TomsHardware) at specific AA settings seems to show the 7970 in a positive light in World of Warcraft, but given how different the results are between the three sites, I'm really not sure what to think.

It's a shame Anandtech didn't bench WoW as Anandtech always seems to have the most reliable benchmark numbers.

If not the 7970, what other card would you all recommend for WoW? My requirements are only that it's faster than what I already have in WoW, though my 2x 4870x2 are still pretty fast for their age and WoW has excellent quad crossfire utilization.
 

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
Here is a 4th website that tested WOW. Now you have 3 that show that it runs faster on NV hardware and 1 that shows it runs better on AMD hardware.







Source

At 1080P:

A single HD5770 ~ HD4870 in DX9 mode gets 31 fps average in WOW. If you get 80% GPU scaling for all 4 GPUs, you'd get 31*4 HD4870s * 80% scaling = 99 fps in that benchmark.

A single GTX570 in DX11 mode gets 95 fps in the same benchmark.

A GTX570 can be bought for $270 or so. Sell those 4870x2s and get 2x GTX570s and overclock those. You'll probably do far better off than getting a single $550 HD7970 for this specific game.

And actually for BF3, GTX570s in SLI will clean up the HD7970 too at 1080P with 4x MSAA.
 
Last edited:

VulgarDisplay

Diamond Member
Apr 3, 2009
6,193
2
76
I personally think a lot of the hd7970 reviews showed that there are obvious driver issues with DX9 games that will most likely be fixed in future updates. The drivers for these cards are still very new, and this is a completely different architecture than what AMD has been using. I expect to see some pretty impressive performance increases from driver updates for GCN hardware.
 

GotNoRice

Senior member
Aug 14, 2000
329
5
81
Here is a 4th website that tested WOW. Now you have 3 that show that it runs faster on NV hardware and 1 that shows it runs better on AMD hardware.

Yeah, I'm pretty familiar with the performance of existing/older cards but i'm mainly interested in seeing how the 7970 compares. I'm not upgrading only for WoW, but since it is my main game, it is what I consider the most important.

At 1080P:

A single HD5770 ~ HD4870 in DX9 mode gets 31 fps average in WOW. If you get 80% GPU scaling for all 4 GPUs, you'd get 31*4 HD4870s * 80% scaling = 99 fps in that benchmark.

A single GTX570 in DX11 mode gets 95 fps in the same benchmark.

A common misconception about older cards such as the 4870x2 is that they cannot run games in DirectX11. In fact they can run games in DirectX11 just fine, due to backward compatibility built into DirectX11 itself. When you launch a DirectX11 game it will make a quick determination regarding which feature level your card supports, after which the game will run normally using the DirectX10.1 feature level. This is possible because DirectX11 is a strict super-set of DirectX10 and 10.1, so compatibility is as simple as shedding the features that are not supported. With DX10.1 hardware you are mainly talking about tessellation, something WoW doesn't use anyway. Even my 9800GT can run WoW in DX11 mode.

Another thing about most WoW benchmarks is the methodology used to generate those numbers. There are few things in an MMO that are truly repeatable, and so in most cases they use the in-game transit system (Wind-Rider or Gryphon on a flight-path). This is about as close to repeatable as you can get, but at the same time flying through the air on a wind-rider or Gryphon is not really all that strenuous for your GPU. Because of that the numbers in most benchmarks seem artificially high. The numbers you will see in various real-world situations such as a 25-man raid, or a 40-man Alterac Valley, or even running around in a populated city, will be much lower.

A GTX570 can be bought for $270 or so. Sell those 4870x2s and get 2x GTX570s and overclock those. You'll probably do far better off than getting a single $550 HD7970 for this specific game.

If I went Nvidia I'd probably try to go for a pair of 580s, though it's a shame prices have not come down much on those since their release. Alternatively I've also been looking at getting a used 6990 or maybe a pair of 5970s. 5970s in particular seem to be available pretty cheap used. The 570s are interesting, but for the performance they offer I could probably get a pair of used GTX480s for considerably less.

Another thing that has me interested in the 7970 is it's low power consumption. My computer at full load currently pulls over 1000w from the wall with probably half or more of that coming just from my videocards and most of that being converted into heat. It's not so bad in winter but extended gaming sessions on hot summer days can get a bit sweltering. A pair of 570s would obviously not suck as juice as my current cards but they are still pretty power hungry.
 
Last edited:
Feb 19, 2009
10,457
10
76
I used to play WoW, that game is severely CPU limited and depending on the scene tested, cpu bound may alter the results.

4xMSAA vs 8xMSAA in WoW, i saw no difference. Its just not a graphically demanding game.
 

GotNoRice

Senior member
Aug 14, 2000
329
5
81
I used to play WoW, that game is severely CPU limited and depending on the scene tested, cpu bound may alter the results.

Yeah I agree. I upgraded from a Q9650 @ 4.4Ghz to an i5-2500K @ 5Ghz this last September mainly just for WoW. I plan to upgrade again to Ivy Bridge some time next year but until then I don't think there is much more I can do on the CPU front.
 
Feb 19, 2009
10,457
10
76
But you should really get rid of those 4870x2, its such a power hog for the performance. A single 7970 3gb will run WoW on eyeinfinity no sweat.

And yes, WoW is awesome on 3 screens, i've seen it in action on a buddy's rig.
 

Arkadrel

Diamond Member
Oct 19, 2010
3,681
2
0
sorry for this but.... why?

What is the point if even 2560x1600 with AA, your getting like 70fps with it?
Any resolution (single screen) this thing will give fluent gameplay.

Do you use 3 monitors? ei. AMDs eyefinity/Nvidia surround?


The TechPowerUp numbers for the 7970 in WoW are terrible. It struggles just to match the performance of a GTX580 and at 1920x1200 it is even beat by it's predecessor, the 6970.



Yes it looks like its getting 99.5 fps, while the 6970 is getting 104.3fps, both are ahead of the 580.
Terrible terrible.... however at 100+ fps, you probably will get a good gameing experiance.

Its probably just driver work that needs doing, im guessing AMD doesnt weight the need for driver optimistions for WOW that heavly. Their probably putting most of their work into newer DX games, that are more requireing.
 
Last edited:

railven

Diamond Member
Mar 25, 2010
6,604
561
126
You need to ask yourself what situations each review site is benching.

25man Raids in ICC my 5870 would buckle for Marrowgar (or whatever the first boss' name was) to 20FPS - being a tank I had my back against that ice wall.

As a DPS, same fight no where near that ice wall I'd be locked at 60 FPS (I use vsync.)

Now, in Firelands 25man I never dropped below 50 FPS and that was only during Alysrazor's fire tornado phase.

My 5870 scores higher than these benches so I don't even know how to interpret them.
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
Here's my take. Who the hell cares about WoW performance? The game looks like garbage, 10 year old engine, is clunky as heck and slows down in 25 man raids on SLI GTX 580's. 200 fps in durotar but 30 fps in 25 man raids, yeah does that make sense? Hell no. Now that last part doesn't make sense to me and goes to show what a piece of crap engine it is. Glad I quit playing the game, everyone should.

Russian, please buy some 580s to end the misery


This is textbook thread-crapping. Please do not come into a thread with the intent to crap on it.

Administrator Idontcare
 
Last edited by a moderator:

GotNoRice

Senior member
Aug 14, 2000
329
5
81
You don't need a 7970 to run WoW.

sorry for this but.... why?

What is the point if even 2560x1600 with AA, your getting like 70fps with it?
Any resolution (single screen) this thing will give fluent gameplay.

I explained why the numbers in most WoW benchmarks are artificially high in an earlier post, so I'll just quote it:
Another thing about most WoW benchmarks is the methodology used to generate those numbers. There are few things in an MMO that are truly repeatable, and so in most cases they use the in-game transit system (Wind-Rider or Gryphon on a flight-path). This is about as close to repeatable as you can get, but at the same time flying through the air on a wind-rider or Gryphon is not really all that strenuous for your GPU. Because of that the numbers in most benchmarks seem artificially high. The numbers you will see in various real-world situations such as a 25-man raid, or a 40-man Alterac Valley, or even running around in a populated city, will be much lower.

Here's my take. Who the hell cares about WoW performance?

Well obviously I care as it is the main game I play. The 10 million other people who play might also care.

200 fps in durotar but 30 fps in 25 man raids, yeah does that make sense? Hell no.

Durotar is a starting area and mostly a barren desert. In 25-man raids you're talking about a large amount of people in close proximity, most with their own spell effects and trinket proc effects on top of the spell effects from the boss and the fight itself. There is a lot going on in a 25-man raid and it makes perfect sense to me that it would slow down in that situation. I see tons of people who talk about how their setup is overkill because they get 100+fps in some random area and in a raid they are sitting at ~17 FPS. Raids is where the FPS matters most and that is the main reason I want to upgrade.

Glad I quit playing the game, everyone should.

I don't really see what the point is for you to come in here just to drop a whole bunch of negative opinions about the game. You don't care about WoW and that's fine, but I obviously do and this is a legit question about hardware.
 
Last edited:

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
I explained why the numbers in most WoW benchmarks are artificially high in an earlier post, so I'll just quote it:




Well obviously I care as it is the main game I play. The 10 million other people who play might also care.



Durotar is a starting area and mostly a barren desert. In 25-man raids you're talking about a large amount of people in close proximity, most with their own spell effects and trinket proc effects on top of the spell effects from the boss and the fight itself. There is a lot going on in a 25-man raid and it makes perfect sense to me that it would slow down in that situation. I see tons of people who talk about how their setup is overkill because they get 100+fps in some random area and in a raid they are sitting at ~17 FPS. Raids is where the FPS matters most and that is the main reason I want to upgrade.



I don't really see what the point is for you to come in here just to drop a whole bunch of negative opinions about the game. You don't care about WoW and that's fine, but I obviously do and this is a legit question about hardware.

Sorry, but anyway poor WoW performance in 25 mans is because of the old clunky engine. What will help immensely is if you have any addons that interact with the combat log, disable them. Stuff like power auras, recount, etc, etc. to disable them.

I noticed it as well, and it doesn't make much sense since the engine is so dated.
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0


Seems fine to me, although some websites went on to test driver AA which is a flawed methodology -- I've posted about it before but nvidia ignores override settings for the majority of games in the CP. AMD CCC is much more reliable in that respect, it does not ignore override settings most of the time (as nvidia does) I wanna say that WoW doesn't have a AA compatibility bit in nvidia inspector so any driver AA settings you input will be ignored. Which is par for the course for nvidia CP.
 
Last edited:

GotNoRice

Senior member
Aug 14, 2000
329
5
81
Sorry, but anyway poor WoW performance in 25 mans is because of the old clunky engine. What will help immensely is if you have any addons that interact with the combat log, disable them. Stuff like power auras, recount, etc, etc. to disable them.

I noticed it as well, and it doesn't make much sense since the engine is so dated.

Well I do use addons such as the ones you mentioned but the whole point here is to improve my gaming experience and removing addons that I depend on would be a step in the wrong direction.

The engine is old but we are now at the tail end of WoW's 3rd expansion and every expansion has included substantial tweaks and improvements to WoW's engine. We now have a game that benefits from DX11, can make use of multiple CPU cores, and is Large Address Aware, among many other improvements. Raiding with large numbers of people is not something that is new and in-fact when the game first came out most raids were 40-man. The only thing that has really changed is the number of spell effects has increased. I'm not convinced that there isn't any benefit to be had by using faster hardware.



Seems fine to me

That benchmark does look promising, but it's also the ONLY benchmark from any site that shows the 7970 in a positive light running WoW, even when ignoring the nvidia cards and just comparing to other AMD cards.

I mean the TechPowerUp benchmarks don't appear to have used anything other than regular MSAA, but the 7970 is only 80% as fast as the 6990 and 95% as fast as a 6970. Then in the TomsHardware numbers the 7970 is virtually the same speed as the 6990 and is 113% faster than the 6970. Why the inconsistency?
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
Well I do use addons such as the ones you mentioned but the whole point here is to improve my gaming experience and removing addons that I depend on would be a step in the wrong direction.

The engine is old but we are now at the tail end of WoW's 3rd expansion and every expansion has included substantial tweaks and improvements to WoW's engine. We now have a game that benefits from DX11, can make use of multiple CPU cores, and is Large Address Aware, among many other improvements. Raiding with large numbers of people is not something that is new and in-fact when the game first came out most raids were 40-man. The only thing that has really changed is the number of spell effects has increased. I'm not convinced that there isn't any benefit to be had by using faster hardware.



That benchmark does look promising, but it's also the ONLY benchmark from any site that shows the 7970 in a positive light running WoW, even when ignoring the nvidia cards and just comparing to other AMD cards.

I mean the TechPowerUp benchmarks don't appear to have used anything other than regular MSAA, but the 7970 is only 80% as fast as the 6990 and 95% as fast as a 6970. Then in the TomsHardware numbers the 7970 is virtually the same speed as the 6990 and is 113% faster than the 6970. Why the inconsistency?

Does WoW support any specific types of AA? I can't remember. But if they're using driver AA thats the easy explanation, nvidia ignores override AA for WoW (while CCC obeys it) if I remember right. I'll double check on that when I get home though.
 

GotNoRice

Senior member
Aug 14, 2000
329
5
81
Does WoW support any specific types of AA? I can't remember. But if they're using driver AA thats the easy explanation, nvidia ignores override AA for WoW (while CCC obeys it) if I remember right. I'll double check on that when I get home though.

I'm not entirely sure. In my case the only options it gives me are 1x, 2x, 4x, and 8x. I have it set to 4x and in Catalyst Control Center I have "Use Application Settings" selected along with the Edge-Detect filter for what supposedly works out to ~12x AA.
 

kami

Lifer
Oct 9, 1999
17,627
5
81
Here's my take. Who the hell cares about WoW performance? The game looks like garbage, 10 year old engine, is clunky as heck and slows down in 25 man raids on SLI GTX 580's. 200 fps in durotar but 30 fps in 25 man raids, yeah does that make sense? Hell no. Now that last part doesn't make sense to me and goes to show what a piece of crap engine it is. Glad I quit playing the game, everyone should.

My rig runs at over 60 FPS in 25 man raids in 1080p/16xAF/4xAA. The key is to max out every setting except shadows. Put that 1 or 2 ticks below max.
 

GotNoRice

Senior member
Aug 14, 2000
329
5
81
The key is to max out every setting except shadows. Put that 1 or 2 ticks below max.

Good idea, but I think these sort of suggestions would apply regardless of what card I had. I have other reasons to upgrade beyond WoW (Power, Heat, BF3 performance, etc) and I'm actually already pretty satisfied with the performance i'm getting in WoW with current hardware. But even if i'm not upgrading my card(s) specifically for WoW, I still want to make sure a new card would give me at least the same performance if not more in WoW since that is still my main game, by far. It's difficult to determine that given the wide variation in performance shown in the available benchmarks.
 

Ovven

Member
Feb 13, 2005
75
0
66
7970 seems to maintain minimum FPS better than 5xxx-6xxx series, having less of a delta between min-max.
 

fourdegrees11

Senior member
Mar 9, 2009
441
1
81
Well I do use addons such as the ones you mentioned but the whole point here is to improve my gaming experience and removing addons that I depend on would be a step in the wrong direction.

The engine is old but we are now at the tail end of WoW's 3rd expansion and every expansion has included substantial tweaks and improvements to WoW's engine. We now have a game that benefits from DX11, can make use of multiple CPU cores, and is Large Address Aware, among many other improvements. Raiding with large numbers of people is not something that is new and in-fact when the game first came out most raids were 40-man. The only thing that has really changed is the number of spell effects has increased. I'm not convinced that there isn't any benefit to be had by using faster hardware.



That benchmark does look promising, but it's also the ONLY benchmark from any site that shows the 7970 in a positive light running WoW, even when ignoring the nvidia cards and just comparing to other AMD cards.

I mean the TechPowerUp benchmarks don't appear to have used anything other than regular MSAA, but the 7970 is only 80% as fast as the 6990 and 95% as fast as a 6970. Then in the TomsHardware numbers the 7970 is virtually the same speed as the 6990 and is 113% faster than the 6970. Why the inconsistency?

Drivers are not optimised yet, simple as that.
 

animekenji

Member
Aug 12, 2004
85
0
0
With DX10.1 hardware you are mainly talking about tessellation, something WoW doesn't use anyway. Even my 9800GT can run WoW in DX11 mode.

Sorry, but I'm calling you on this. First of all, AMD cards have been using tesselation since R600. Secondly, your 9800GT is NOT running in DX11 mode because it does not support the DX11 feature set. The 9800GT is a DX10 card, so you are playing in DX10, not DX11. Your 4870x2 is running in DX10.1, not DX11. You can't run features from a later DX version on a card that does not support those features. You can still run WoW on a Radeon 9700, and that's a DX9 card, but it doesn't mean you are going to be able to play in DX11. No feature incorporated into the DX specification after DX9 will be accessible, so you'll be playing in DX9.
 
Last edited:

railven

Diamond Member
Mar 25, 2010
6,604
561
126
Good idea, but I think these sort of suggestions would apply regardless of what card I had. I have other reasons to upgrade beyond WoW (Power, Heat, BF3 performance, etc) and I'm actually already pretty satisfied with the performance i'm getting in WoW with current hardware. But even if i'm not upgrading my card(s) specifically for WoW, I still want to make sure a new card would give me at least the same performance if not more in WoW since that is still my main game, by far. It's difficult to determine that given the wide variation in performance shown in the available benchmarks.

What are your current results? I personally don't know how to interpret their results either since my card scores higher than all the cards then benched in situations I assumed are much more intense (minus that ice wall in ICC.)

I have my settings on Ultra (including shadows) and I don't really drop below 40 FPS. I was watching my server's first legendary staff event and there were hundreds of players and I stayed hovering in the 50s.

I also run an RAID 0 and have my CPU clocked at 3.6ghz, not sure if that would help my performance more so than just a beefier GPU (but I assume they run their bench on a high clocked CPU too and possibly an SSD.)
 

GotNoRice

Senior member
Aug 14, 2000
329
5
81
Sorry, but I'm calling you on this.



First of all, AMD cards have been using tesselation since R600

I'm aware of AMD's past implementations of tessellation going all the way back to Truform, but the only one that matters is the one implemented in DirectX11 and AMD has only had support for that since the 5870.

Secondly, your 9800GT is NOT running in DX11 mode because it does not support the DX11 feature set. The 9800GT is a DX10 card, so you are playing in DX10, not DX11. Your 4870x2 is running in DX10.1, not DX11. You can't run features from a later DX version on a card that does not support those features. You can still run WoW on a Radeon 9700, and that's a DX9 card, but it doesn't mean you are going to be able to play in DX11. No feature incorporated into the DX specification after DX9 will be accessible, so you'll be playing in DX9.

I can and do play many different games in DirectX11 on my 4870x2. I never claimed it is able to use all of the DirectX11 features. It runs in DirectX11 using the DirectX10.1 feature level, just as the 9800GT can run in DirectX11 using the DirectX10.0 feature level. Again, this is possible because DirectX11 is a strict super-set of DirectX 10.1 (which itself is a strict super-set of DirectX10) and so compatibility is as simple as shedding the small subset of features that are not supported. Feel free to read more about it if you are still confused:

http://msdn.microsoft.com/en-us/library/windows/desktop/ff476872(v=vs.85).aspx
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |