Diablo III Performance Tested, Benchmarked

Julio Franco

Posts: 9,099   +2,050
Staff member
Read the full article at:
[newwindow=https://www.techspot.com/review/532-diablo-3-performance/]https://www.techspot.com/review/532-diablo-3-performance/[/newwindow]

Please leave your feedback here.
 
No matter what the performance, the recent debacle is making people take a 50-50 approach towards Diablo.
[CENTER]50% [/CENTER]
[CENTER] It's a legend! Buy it![/CENTER]
[CENTER] [/CENTER]
[CENTER]50%[/CENTER]
[CENTER]Is this what I was waiting for?! WTF![/CENTER]
 
It's pretty much more of the same old Diablo so far for me. Thing that's annoying me is that WoW and other mmorpgs have conditioned me into ignoring the quest chat/text, so I'm having to learn not to skip them as I don't really want to miss the story. That being said I did play from 7 to 11 in co-op and never noticed where time went. Reasonably enjoyable but the weird two main ability mouse button setup is so restrictive after playing WoW. Will at least play through once with one character/class.
 
<p>It's pretty much more of the same old Diablo so far for me. Thing that's annoying me is that WoW and other mmorpgs have conditioned me into ignoring the quest chat/text, so I'm having to learn not to skip them as I don't really want to miss the story. That being said I did play from 7 to 11 in co-op and never noticed where time went. Reasonably enjoyable but the weird two main ability mouse button setup is so restrictive after playing WoW. Will at least play through once with one character/class.</p>

Why would you skip the quest text in one game and not in another? Because you've already leveled so many characters in WoW? Will you skip it in Diablo as well once you've played through one time? Also not sure why anyone would expect an arcade game to have more options or depth than an MMO. People seem to have been expecting Diablo 3 to be something more than Diablo II, even though SC2 showed that Blizzard isn't about pushing the envelope on anything. In truth, they've never been all that innovative. Their code is extremely solid but their hardly the cutting edge of game design.
 
Why would you skip the quest text in one game and not in another? Because you've already leveled so many characters in WoW?

Pretty much, even with new quests you can guarantee they will be "Help me gather 10 things from there and kill 20 things".
I know the Diablo voiced story and lore from books you pickup in game will be superior to this but it's habitual and not a fault of the game.

Also not sure why anyone would expect an arcade game to have more options or depth than an MMO.
Maybe because it's still within the RPG genre, and generally offerings which have single player content/storylines often offer more involving content(although SW:TOR might be an exception from the little I played of it) than that which is bolted onto WoW for instant. Less generic pointless quests that side track from the story than MMOs have to provide.

People seem to have been expecting Diablo 3 to be something more than Diablo II, even though SC2 showed that Blizzard isn't about pushing the envelope on anything. In truth, they've never been all that innovative. Their code is extremely solid but their hardly the cutting edge of game design.
Indeed, people are assuming that it's been in development since Diablo2 and will be disappointed by the lack of innovation in the 12 year span. When in fact we don't know how much (or little) development time was involved.
 
I would like to see some benchmarks for Diablo 3 on the Mac. The game is out on both PC and Mac and runs differently when bootcamped and when run natively in OSX. There are differing stories about performance between different models and graphics cards - a much more interesting story really than how stressed out (or not) your PC is going to get.

Benchmarks for the HD4000 on low settings would be nice too. I am sure some people will want to get a laptop sooner than later or an Ive bridge ultrabook - how about showing whether these can play Diablo 3?

Article seems a bit rushed in comparison to the thorough job that could have been done.
 
Yeah you are never going to see me testing a game on a Mac so let’s move on.

So we tested 29 graphics configurations, 14 different processors across half a dozen platforms as well as CPU scaling results for Intel and AMD processors and that seemed rushed to you.

Interesting take on the article, just for your information we only test games using the maximum in-game graphics and for a game that is not that demanding we saw no reason to change this.

If the demand is there we could publish a Diablo III article that focuses on low-quality visuals, kind of like a mobile gaming version.
 
The HD6670 that have used is GDDR3 or GDDR5?

Sorry for my english is not good. :)
 
Yes please do. There are also people that are using mid range setups, you know...
 
Yeah you are never going to see me testing a game on a Mac so let’s move on.

Shame but I can understand why given the different configurations and possibilities involved. Fair enough.

So we tested 29 graphics configurations, 14 different processors across half a dozen platforms as well as CPU scaling results for Intel and AMD processors and that seemed rushed to you. Interesting take on the article, just for your information we only test games using the maximum in-game graphics and for a game that is not that demanding we saw no reason to change this

When you tested Starcraft 2 you did it with 21 different graphics cards at medium/high and ultra settings with 3 different resolutions and with 12 different processors. Thats around 200 different results in the Starcraft 2 benches compared to about 100 for Diablo 3. It's your own damn fault, you set the bar too high!

If the demand is there we could publish a Diablo III article that focuses on low-quality visuals, kind of like a mobile gaming version.

Fair enough. Ultrabooks, tablets, Ivy Bridge, HD4000. Loads of people will want to play portable Diablo3 if its possible in Windows 8 on a HD4000 in the next couple of years. You don't buy a game like Diablo3 to play on a tablet and expect to play it at high frames on high quality but it would be really sweet to see the results for lower settings (and maybe some lower resolutions).
 
I think it would actually make sense to see how the HD 4000 and the AMD APUs perform at lower settings, given the target audience for this particular game. I have many friends who are playing D3 on laptops and low end graphics cards.
 
Interesting how StarCraft II utilizes the CPU a lot more than Diablo 3. I mean it uses the same game engine, right? Why would it make a difference?

And when is Blizzard finally going to grow up beyond DX 9??
 
And thank you very much for testing overclock scaling with both AMD and Intel. That was awesome.
 
Steve has done a wonderful job of covering all relevant GPU's on the discrete market, but I agree with some of the comments above, a single test at a modest resolution to see how AMD APU's and Intel's HD integrated graphics fare in Diablo 3 would seal the deal as the ultimate guide to Diablo's performance.
 
<p>Interesting how StarCraft II utilizes the CPU a lot more than Diablo 3. I mean it uses the same game engine, right? Why would it make a difference?</p>
<p><br /></p>
<p>And when is Blizzard finally going to grow up beyond DX 9??</p>
In their defence, DX11 is not really something that would benefit this genre of game...
 
SPOILER
Does it feature not standing in the fire?
:), the shadow mode is a rip just sayin.

I am going to take a shot in the dark and say the many units in sc2, yours and your allies is what makes sc2 cpu intensive compared to d3. Yes the graphics are archaic. Fear not there are alternative hack and slashes as many of you know. Coming soon path of exile, torch light 2 and another one I cant remember.
 
Long time fan of techspot here, but the CPU benchmarks are a really BAD measure of the game. In large battles, which happen all the time in this game, nothing AMD has can run the game smoothly, diablo appears to require at least a decent sandy bridge to maintain 60 fps in large fights. It appears to be because of the physics.

I guarantee if you guys were to look at the minimum fps in large fights with different CPUs you would see an AMD, even overclocked is will have dramatic fps drops when things start flying in large battles. A easy way to test is the part where the templar breaks the barrier before and after jondar, amd processors drop down as far as 20 fps and the ivy bridge processors ive tested with it dont even drop fps.

Not sure if its CPU or GPU? Here this, I ran my brother's new gtx 670 with an unlocked and overclocked amd 960t cpu and it still lagged the same as with my 6950. He used my 6950 and maintained a smooth 60 fps with his 3770k intel ivy bridge. Clearly AMD processors either need more optimization from blizzard or they cannot handle the physics diablo requires. Its also noticeable to the eyes, with both computers next to eachother its easy to tell which one has the AMD and which one has the new intel because of the huge drop in fps.
 
fwiw, I've logged 8 hours act I and act ii including boss - I'm using a 16GB i5-2500k, not OC and HD3000 GPU. I chose medium graphics at 1920 (Diablo III defaulted all to low).
I have not experienced any graphics issues at all - I am no expert, and not picky, so take with a grain of salt, but there you have it. ( I have experienced a game playing at 11fps and it is horrible. DIII's -internal- framerate-display shows me at 25, dropping to 21 in act I end battle)

hope this helps..
 
To be honest and given the title, I expected some more low spec hardware in here. Most of its fairly high end imo (even if its not to you, comparatively to most consumers that don't upgrade every year or less).

Playing the game on my laptop (i7, 6gb ram, 1gb 330GM) sees much lesser results. Be good if blizzard implemented a netbook mode like torchlight did.

After playing the beta, I wasn't wow'd enough by the game, even though I've been hyping it up for the past several months.

Still good overview, I tried not to look at the screenies in the event that I do play it.
 
With my bros machine Ive found that the amd 1090t with a 590gtx is alright with 1920x1080 120fps when there upto 8 monsters on screen. when it gets hectic (15-20 monsters. more exciting parts of the game) it nose dives to 15-25fps.

Same moments in game on my i5 2500k 580 gtx sli no fluctuation at all in the framerate.

I think you need to show some cpu apples to apples with demanding parts of the game. (which can crop up quite often later in the game)
 
Funny, I've been running the game on AMD's FX-8120 with 16GB DDR3 1600 and a Radeon HD6850 (1GB GDDR5) and I have not had any framerate drops or issues at all on the local machine. I've found that the problems I'm having is that the major ISPs out here have such poor latency between here and the servers stateside, that on occasion, my character's animations will continue and enemies will still damage me, but my attacks won't show up for several seconds.
 
Of course his 1090t w/ 590GTX is not going to out perform your rig. He's got an older processor and if I read correctly, only a single GPU. Your SLI'd 580's will massively outperform a single 590 no matter what processor.
 
For whatever reason, some people cant see fast drops in FPS, however I think you would be able to see the difference clearly if you played on a machine that had at least a sandy bridge. Ive tested that amd cpu with a gtx 670 and saw fps drops to 20-40 fps for brief moments in very large fights, the drops only occur when several monsters are killed at once (hence the processor having issues with the physics calulated for that moment.)
 
Back