AMD's Richland vs. Intel's Haswell GPU on the Desktop: Radeon HD 8670D vs. Intel HD 4600
by Anand Lal Shimpi on June 6, 2013 12:00 PM ESTMetro: Last Light
Metro: Last Light is the latest entry in the Metro series of post-apocalyptic shooters by developer 4A Games. Like its processor, Last Light is a game that sets a high bar for visual quality, and at its highest settings an equally high bar for system requirements thanks to its advanced lighting system. This doesn’t preclude it from running on iGPUs thanks to the fact that it scales down rather well, but it does mean that we have to run at fairly low resolutions to get a playable framerate.
Looking at desktop parts alone, Intel really suffers from not having a socketed GT3 SKU. Although HD 4600 is appreciably faster than HD 4000 (+30%), both Trinity and Richland are around 17% faster than it. As you'll see, Metro ends up being one of the smaller gaps between the two in our suite.
As memory bandwidth becomes the ultimate bounding condition, the gap between Richland and Haswell shrinks considerably. Note that on the HD 4600 side, the difference between DDR3-1333 and DDR3-2400 is only 10% here. Given the limited performance of the 20 EU Haswell GPU configuration, it doesn't seem like Intel is all that bandwidth limited here.
BioShock: Infinite
Bioshock Infinite is Irrational Games’ latest entry in the Bioshock franchise. Though it’s based on Unreal Engine 3 – making it our obligatory UE3 game – Irrational had added a number of effects that make the game rather GPU-intensive on its highest settings. As an added bonus it includes a built-in benchmark composed of several scenes, a rarity for UE3 engine games, so we can easily get a good representation of what Bioshock’s performance is like.
If Metro was an example of the worst case scenario for Richland, BioShock: Infinite is the best case scenario. Here the Radeon HD 8670D holds a 50% performance advantage over Intel's HD 4600 graphics.
The gap narrows a bit at higher resolution/quality settings, but it's still 39%.
Sleeping Dogs
A Square Enix game, Sleeping Dogs is one of the few open world games to be released with any kind of benchmark, giving us a unique opportunity to benchmark an open world game. Like most console ports, Sleeping Dogs’ base assets are not extremely demanding, but it makes up for it with its interesting anti-aliasing implementation, a mix of FXAA and SSAA that at its highest settings does an impeccable job of removing jaggies. However by effectively rendering the game world multiple times over, it can also require a very powerful video card to drive these high AA modes.
Richland is approaching 60 fps in our Sleeping Dogs benchmark at medium quality, definitely not bad at all. The advantage over Intel's HD 4600 is 34%.
The performance advantage grows a bit at the higher quality/resolution settings, however we drop below the line of playability. With most of these games, you can trade off image quality for resolution however.
Tomb Raider (2013)
The simply titled Tomb Raider is the latest entry in the Tomb Raider franchise, making a clean break from past titles in plot, gameplay, and technology. Tomb Raider games have traditionally been technical marvels and the 2013 iteration is no different. iGPUs aren’t going to have quite enough power to use its marquee feature – DirectCompute accelerated hair physics (TressFX) – however even without it the game still looks quite good at its lower settings, while providing a challenge for our iGPUs.
Tomb Raider is another title that doesn't put Richland in the best light, but it still ends up around 23% faster than Haswell GT2.
Battlefield 3
Our multiplayer action game benchmark of choice is Battlefield 3, DICE’s 2011 multiplayer military shooter. Its ability to pose a significant challenge to GPUs has been dulled some by time and drivers at the high-end, but it’s still a challenge for more entry-level GPUs such as the iGPUs found on Intel and AMD's latest parts. Our goal here is to crack 60fps in our benchmark, as our rule of thumb based on experience is that multiplayer framerates in intense firefights will bottom out at roughly half our benchmark average, so hitting medium-high framerates here is not necessarily high enough.
Richland's performance in Battlefield 3 climbs around 30% over the HD 4600 regardless of quality/resolution.
Crysis 3
With Crysis 3, Crytek has gone back to trying to kill computers, taking back the “most punishing game” title in our benchmark suite. Only in a handful of setups can we even run Crysis 3 at its highest (Very High) settings, and the situation isn't too much better for entry-level GPUs at its lowest quality setting. In any case Crysis 1 was an excellent template for the kind of performance required to drive games for the next few years, and Crysis 3 looks to be much the same for 2013.
Crysis is another benchmark where we see an increase in performance in the low 30% range.
Facebook
Twitter
RSS
102 Comments
View All Comments
jrs77 - Thursday, June 6, 2013 - link
Sorry to say, but if you wanna play games, then grab a dedicated GPU. Noone plays modern games with integrated GPUs. iGPUs are only good for some casual gaming that don't demand any powerful GPU.The only thing I'm interested in is the CPU-part and intel wins this comparison by a mile. AMD needs better CPU-performance or they'll never win me back as a customer for desktop-parts.
Voldenuit - Thursday, June 6, 2013 - link
Exactly. I'm mystified by Richland's existence. Exactly where does it make sense for desktop users?If you're gaming on a desktop and you're on a budget, it's more expensive than the outgoing A10 5800K and not much faster. Nor does it make any games playable that were unplayable before on the older APU.
If you're gaming on the desktop and budget is not the biggest factor, why even bother looking at AMD parts right now?
If you're planning on a HTPC build the 100W TDP is too high. Get a Haswell or an older (and lesser model number) Trinity APU.
If you plan to build your own all-in-one, again the TDP is too high.
So why would anyone buy Richland on the desktop? It's $20 (15%) more expensive than the A10 5800K but only ~5% faster. Is there +1000 model number simply there to justify the price hike?
silverblue - Friday, June 7, 2013 - link
It's slightly better silicon released to make AMD look a little better. There's also the much touted software bundle they've been mentioning - AMD seems to be more about the "experience" nowadays.As an upgrade to my old PII X3 710, it'd be significant... but the GPU wouldn't be better than my 4830. Kaveri, on the other hand, would likely improve on the latter as well as providing more than double my current CPU speed. In 2009, the CPU and GPU cost me approx. £180 (about $250?), but in 2013, I'd be surprised if I had to pay more than two thirds of that for something much better. :)
Calinou__ - Friday, June 7, 2013 - link
There are games that don't require a powerful GPU and are hard to play.kgh00007 - Thursday, June 6, 2013 - link
For **** sake please increase the size of the text on your site, it's too small and I'm getting eye strain reading all the articles on Computex, do you not listen to your readers, the text is too small!!!Sent from my nexus 7 with eye strain!
Gigaplex - Thursday, June 6, 2013 - link
It's too big for me, so I just use the zoom setting on my browser. I suggest you try the same, or stop using a 7" tablet.kgh00007 - Friday, June 7, 2013 - link
I suggest you stop giving useless advice, I'm not going to stop using my tablet. Even with text at 120% the text is too small and the text on other websites is huge @ 120%.The text on every other website I have ever visited on my nexus 7 is fine, so why can't anandtech be the same? There is an issue here that needs to be resolved.
bji - Friday, June 7, 2013 - link
Regardless of your feelings, posting about your issue in this discussion is off topic and pointless as Anand is not going to read and act on your post.I'd say to write to Anand directly but I've tried that and it seems the emails go ignored anyway. I think you either just have to live with the site as it is or stop visiting the site. But whatever you do, don't introduce more off topic posts please.
duploxxx - Friday, June 7, 2013 - link
by not running the trinity on supported max mem you effected the whole review, Richland ends up in that case being a very minor update....On top of that not a single word on the improved power.
But then again as you mention with all the things around with Intel and all the nice motherboards linup for hasswell etc why even bother its just AMD. pls continue the efforts like this, a few years from know you regret that you will need to pay double for the same cpu, dominiated predefined designs by marketing geeks etc... way to go
DeviousOrange - Friday, June 7, 2013 - link
Its pointless doing a iGPU review if you don't have frame metering factored in. Tomshardware, Techreport, PCPer all did frame transition ratings and this is where HD4600 took a massive beating, in most titles showed up Intel's iGPU to hit 60+ms frame lags while the APU is very low latencies for many titles at lower resolutions (sub HD) is under 1ms where Intel spikes will result in noticeable lags and stutters despite looking close on FPS which are basically not worth what they were.BF3 @ 1080 on low settings with DDR3 2400 on my 5800K manages about 30FPS but its almost lag free, tested on HD4000 was completely undesireable to even persist, basically playing a slideshow, since HD4600 doesn't fix this much it still puts AMD top in the iGPU stakes by a healthy margin irrespective of frames per second. Since we are comparing top vs top there is no ambiguity. THG showdown between a i3 and 6800K was interesting, a 6800K can beat a i3 + 6670 in a few titles so that is another testiment to the improvement of APU technology.