Diablo III Performance, Benchmarked

[superfula]

[citation][nom]Cleeve[/nom]I'm curious why you would think the final game will produce such a drastic difference in frame rates when it's half a month from the end of beta. Blizzard has been polishing this thing for months, they're not a second-rate developer.[/citation]

Blizzard said as much. Not to mention the countless threads in the beta forum saying parts of the game weren't smooth on even some of the best cards. The beta was nothing more than a infrastructure test so there's no reason to put time into how efficient it runs. The final will be much different.

 

[Guest]

I have a Notebook with windows 7 64bits ( i5 2410M // GTX 460M 1GB // 4GB DDR3 1333mhz // HD 7.200RPM )

Can i run the game in 1366x768 on HIGH ?

 

[al360ex]

I guess that for Blizzard, the accessibility of the game is more important than its visual quality...

 

[GNCD]

[citation][nom]TheHelix[/nom]To the people complaining about the RMAH. Diablo 2 was plagued with sites that would sell you items for money (no need to mention names.. everyone who has played diablo enough knows what i mean). I guess Blizzard thought that if they can't find an efficient way to stop that from happening , at least provide the players with an "official" alternative. So don't act like money for items is something new to the diablo community.[/citation]

exactly. and it seems to be safer in terms of scams and all that BS. won't be using it though.

 

[Guest]

While the article does appear to be thorough, one thing it misses completely is the performance drop in multiplayer. I've got an i5 @ 4.7GHz, 8GB memory, and a 560 Ti, and I need to drop shadows to medium level to not get minor choppiness even in town.

Granted, this might be un-optimized beta code, but if the low-end hardware is getting "passable" on low detail in single-player, it's nearly guaranteed to be unplayable in multi. There really SHOULD be a caveat posted to this effect.

 

[gogogadgetliver]

For future benches I would like to see 2560x1440 run with no AA. Jaggies aren't much of a problem at that resolution.

 

[AntaresX]

[citation][nom]MegaTheJohny[/nom]you have to test FPS rate with 4 players in party, with max details and spam some spells. Than you will see FPS drop till 20. (I have I7 2.7 Ghz, 6GB RAM, ATI Raedon HD 5900 series.)[/citation]

I agree with this, but as was mentioned in the article, this was the only real way they could think of to reproduce the exact same system strain on every piece of hardware they used. Battles in groups of 4 can never be repeated exactly, and this fact would likely skew the results.

 

[51l3n5t]

[citation][nom]jeffredo[/nom]I'm happy. My Phenom II X4 980 will run it as fast as an i5-2500k.[/citation]

this idiot...

 

[pcCodinFoo]

Seriously? 10 years go by and the graphics only advance by 2 years??? And they expect people to lay down $50 for the exact same game they bought 10 years ago??? Blizzard is worse than Apple... Dirty money grubbers!!! When WoW finally went free for the first 20 levels i said to myself, "ok, I should at least try it"... not even 5 minutes into it i had to stop... the 12 year old graphics almost made me puke... then i said to myself "11 million people play this game with graphics this bad???" Blizzard sucks!! And people who buy their pathetic games are ruining the market by sending the message that gamers don't mind that the graphics. are 12 years old... Shame on anyone who buys Blizzard games and dares call themselves a gamer.

 

[blazorthon]

[citation][nom]pcCodinFoo[/nom]Seriously? 10 years go by and the graphics only advance by 2 years??? And they expect people to lay down $50 for the exact same game they bought 10 years ago??? Blizzard is worse than Apple... Dirty money grubbers!!! When WoW finally went free for the first 20 levels i said to myself, "ok, I should at least try it"... not even 5 minutes into it i had to stop... the 12 year old graphics almost made me puke... then i said to myself "11 million people play this game with graphics this bad???" Blizzard sucks!! And people who buy their pathetic games are ruining the market by sending the message that gamers don't mind that the graphics. are 12 years old... Shame on anyone who buys Blizzard games and dares call themselves a gamer.[/citation]

Worse than Apple? Now that's an insult. I also happen to disagree. Just raise the quality settings and such. This game is only a little under half as heavy on the graphics card as some of the top games now. The 7850 is about as good in 2560x1600 as the 680 and 7970 are in the top games such as BF3 and Metro 2033.

Furthermore, if graphics were the only measure of a game, then a lot of bad games would not be considered bad anymore. Diablo 3 has the complexity and mechanics of a great game and that counts. That it also has a good story line also helps it. Graphics are not only not the only measure of a game's quality, but I don't think it's even close to being the most important. Besides that, there might be higher quality texture pack or something. Modders will probably do something like that if Bliz doesn't, so just give it time if you really only care about graphics quality.

 

[unluckyone]

Sort of disappointed with the benchmark hardware selection. Wanted to see what fps would an Ivy Bridge HD4000 would achieve. Most often, I would be playing this game on my laptop. Maybe Tomshardware.com will do a full benchmarking after the game is officially released and include the HD4000.

 

[gogogadgetliver]

Cleve can I get you to take one more pass for me?

I'm on a GTX 460 and will be running 2560x1440. I see in the bench that it is going to be around the 30-40fps range. I'm wanting a bit more breathing room on that framerate and will likely turn of AA (not badly needed at that res).

Will turning off AA at 2560x1440 make much of an improvement on a GTX 460? Could you do one more spin through for me?

Thanks for doing this by the way....there is a big demand for these numbers out here

 

[gogogadgetliver]

The 9600 GT is probably a little faster than the 6670 we benchmarked for this article, because we used the DDR3 version of the Radeon.

people waited this long for these mediocre graphics?....what a disappointment

No, we didn't.

Diablo II released with 800x600 graphics when most gamers were swinging 1600x1200. We didn't give a crap then and we don't now. The graphics are good enough for immersiveness. The shawshank redemption had crappy special effects too, know what I'm saying?

 

[cleeve]

[citation][nom]superfula[/nom]Blizzard said as much. [/citation]

Blizzard was talking about complaints of stuttering, something that a few people experience. We didn't experience stuttering, as far as raw frame rates they won't change.

Game's almost here, I guess we'll see who'se right in a couple days.

 

[simplesheep]

I've notice that TH's mostly testing games with 1-1.5yrs old CPU/GPUs, is there any particulate reason for this ?
"At the highest detail settings with AA enabled, even a 2 GHz Phenom II X4 achieves 80 FPS minimum. Clearly, the game doesn’t need an incredibly potent CPU."
That means if i have a quarcore it would be a "waste" ?
Can some one help me out here ?

 

[blazorthon]

[citation][nom]simplesheep[/nom]I've notice that TH's mostly testing games with 1-1.5yrs old CPU/GPUs, is there any particulate reason for this ?"At the highest detail settings with AA enabled, even a 2 GHz Phenom II X4 achieves 80 FPS minimum. Clearly, the game doesn’t need an incredibly potent CPU."That means if i have a quarcore it would be a "waste" ?Can some one help me out here ?[/citation]

Do you take offense at Tom's when they show us how having faster CPUs affects Diablo 3 performance? They just wanted to show us what level of performance that is required by Diablo 3.

 

[autoordnance]

a couple things I noticed in the first 2 - 3 paragraphs.

1. you can zoom in/out

2. d2 had crafting

 

[justoneaim]

yes ditto courtlandb a build for diablo 3 would be nice under 800 is ok for me
please do not include operating system cause i have a disk
and im opting for a SSD hard drive only 128 gig as to keep the cost down
i have been alerted to end game fps id like to finish the game in the 30 to 50fps in large battles
help

 

[blazorthon]

[citation][nom]inko221[/nom]i stopped following tom's hardware advice on games after it claimed SWTOR can be played well on any CPU with GPU being the important component. So i am not taking this advice with any credibility.[/citation]

They're light games and you question Tom's credibility when they tell us how light these games are... Play the game before being an asshole and insulting them.

 

[Guest]

AMD released a hotfix for the 12.4 catalyst driver to fix that issue, check this thread:
http://us.battle.net/d3/en/forum/topic/4876967911?page=1#0

 

[kulmnar]

So everyone and their parents will be able to play this game according to the benchmarks. Blizzard is maybe planning to release this to consoles as well? If so, it would be awesome if they allowed PS3 and 360 owners to play with PC owners.

 

[blazorthon]

[citation][nom]kulmnar[/nom]So everyone and their parents will be able to play this game according to the benchmarks. Blizzard is maybe planning to release this to consoles as well? If so, it would be awesome if they allowed PS3 and 360 owners to play with PC owners.[/citation]

That doesn't go to well, most of the time. The consoles have less accurate controls, so they're almost always slower and less responsive than the PC gamers. I'm pretty sure that something like this was already tried before.

 

[f-14]

i noticed that with a dual core e2200 clocked at 2.4 with a 3850 512mb when i was trying to see how low end you could go with the beta. when it releases i will try it again and then drop in a c2d 3ghz[citation][nom]geotek[/nom]Were any large fights present in the FPS recording? Because of the computers ive tested ive noticed computers with slower CPUs, especially dual cores, would get quite choppy during large fights and all the ragdolls flying everywhere. This problem seemed to either vanish or greatly diminish when physics were changed from high to low. Perhaps the physics are handled by the cpu? If so, this means the cpu testing results may not be a good measure of real world use of the game.[/citation]

 

[simplesheep]

[citation][nom]blazorthon[/nom]Do you take offense at Tom's when they show us how having faster CPUs affects Diablo 3 performance? They just wanted to show us what level of performance that is required by Diablo 3.[/citation]
No that doesnt worries me a tiny bit. I have a not so good laptop and thinking to get a desktop in 6months time. So im trying to gather as much information as possible.

 

[blazorthon]

[citation][nom]simplesheep[/nom]No that doesnt worries me a tiny bit. I have a not so good laptop and thinking to get a desktop in 6months time. So im trying to gather as much information as possible.[/citation]

The triple core Phenom II at 3GHz did far better than the quad at 2GHz (I don't think that there are 2GHz Phenom II quad core CPUs, so it was probably just an underclocked 9xx CPU in the review). Core count is generally less important than performance per core in games because performance per core scales gaming performance better than increased core counts does.

Regardless, the quad core did keep over 80FPS... That's more than 60 and you can't see more than 60 on a 60Hz display, so there's no point in having a faster CPU than even the quad core Phenom II at 2GHz. You can get a laptop that can play this game at the maximum settings for almost as cheap as a desktop because this game is just that light.