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 EST
The run up to Computex has been insane. Kabini, Haswell and Iris hit us back to back to back, not to mention all of the travel before receiving those products to get briefed on everything. Needless to say, we're in major catchup mode. There's a lot more that I wanted to do with Haswell desktop that got cut out due to Iris, and much more I wanted to do with Iris that I had to scrap in order to fly out to Computex. I will be picking up where I left off later this month, but with WWDC, Samsung and a couple of NDA'd events later this month, it's not going to be as quick as I'd like.
One part that arrived while I was in the middle of launch central was AMD's Richland for desktop. Effectively a refresh of Trinity with slightly higher clocks, a software bundle and more sophisticated/aggressive turbo. Richland maintains socket compatibility with Trinity (FM2), so all you should need is a BIOS update to enable support for the chip. AMD sent over two Richland parts just before I left for Computex: the 100W flagship A10-6800K and the 65W A10-6700. I didn't have time to do Richland justice before I left, however I did make sure to test the 6800K in tandem with Haswell's GPU just so I had an idea of how things would stack up going forward as I was writing my Iris Pro conclusion.
For all intents and purposes, Iris Pro doesn't exist in the desktop space, making Haswell GT2 (HD 4600) the fastest socketed part with discrete graphics that Intel ships today. In our Haswell desktop review I didn't get a chance to really analyze HD 4600 performance, so I thought I'd take this opportunity to refresh the current state of desktop integrated processor graphics. Unlike the staggered CPU/GPU launch of Trinity on the desktop, the situation with Richland is purely a time limitation on my end. This was all I could put together before I left for Computex.
Although Richland comes with a generational increase in model numbers, the underlying architecture is the same as Trinity. We're still talking about Piledriver modules and a Cayman derived GPU. It won't be until Kaveri that we see GCN based processor graphics from AMD at this price segment (Kabini is already there).
As Jarred outlined in his launch post on Richland, the 6800K features 4 - 8% higher CPU clocks and a 5% increase in GPU clocks compared to its predecessor. With improved Turbo Core management, AMD expects longer residency at max turbo frequencies but you shouldn't expect substantial differences in performance on the GPU side. The A10-6800K also includes official support for DDR3-2133. AMD is proud of its valiation on the A10-6800K, any parts that won't pass at DDR3-2133 are demoted to lower end SKUs. I never spent a ton of time testing memory overclocking with Trinity, but my A10-5800K sample had no issues running at DDR3-2133 either. I couldn't get DDR3-2400 working reliably however.
| AMD Elite A-Series Desktop APUs, aka Richland | ||||||
| Model | A10-6800K | A10-6700 | A8-6600K | A8-6500 | A6-6400K | A4-4000 |
| Modules/Cores | 2/4 | 2/4 | 2/4 | 2/4 | 1/2 | 1/2 |
| CPU Base Freq | 4.1 | 3.7 | 3.9 | 3.5 | 3.9 | 3.0 |
| Max Turbo | 4.4 | 4.3 | 4.2 | 4.1 | 4.1 | 3.2 |
| TDP | 100W | 65W | 100W | 65W | 65W | 65W |
| Graphics | HD 8670D | HD 8670D | HD 8570D | HD 8570D | HD 8470D | ? |
| GPU Cores | 384 | 384 | 256 | 256 | 192 | 128 |
| GPU Clock | 844 | 844 | 844 | 800 | 800 | 724 |
| L2 Cache | 2x2MB | 2x2MB | 2x2MB | 2x2MB | 1MB | 1MB |
| Max DDR3 | 2133 | 1866 | 1866 | 1866 | ||
| Price (MSRP) | $150 ($142) | $149 ($142) | $120 ($112) | $119 ($112) | $80 | $46 |
Just to put things in perspective, here are the previous generation Trinity desktop APUs:
| AMD Trinity Desktop APUs | ||||||
| Model | A10-5800K | A10-5700 | A8-5600K | A8-5500 | A6-5400K | A4-5300 |
| Modules/Cores | 2/4 | 2/4 | 2/4 | 2/4 | 1/2 | 1/2 |
| CPU Base Freq | 3.8 | 3.4 | 3.6 | 3.2 | 3.6 | 3.4 |
| Max Turbo | 4.2 | 4.0 | 3.9 | 3.7 | 3.8 | 3.6 |
| TDP | 100W | 65W | 100W | 65W | 65W | 65W |
| Graphics | HD 7660D | HD 7660D | HD 7560D | HD 7560D | HD 7540D | HD 7480D |
| GPU Cores | 384 | 384 | 256 | 256 | 192 | 128 |
| GPU Clock | 800 | 760 | 760 | 760 | 760 | 723 |
| L2 Cache | 2x2MB | 2x2MB | 2x2MB | 2x2MB | 1MB | 1MB |
| Max DDR3 | 2133 | 1866 | 1866 | 1866 | ||
| Current Price | $130 | $129 | $110 | $105 | $70 | $55 |
For my Richland test platform I used the same Gigabyte UD4 Socket-FM2 motherboard I used for our desktop Trinity review, simply updated to the latest firmware release. I ran both AMD platforms using the same Catalyst 13.6 driver with the same DDR3-2133 memory frequency. AMD was quick to point out that only the A10-6800K ships with official DDR3-2133 support, so the gap in performance between it and Trinity may be even larger if the latter tops out at DDR3-1866. The HD 4000/4600 numbers are borrowed from my Iris Pro review using DDR3-2400, however I didn't notice scaling on Haswell GT2 beyond DDR3-1866.
I'll be following up with a more thorough look at Richland once I'm back from my current bout of traveling.
Facebook
Twitter
RSS
102 Comments
View All Comments
mrSmigs - Friday, June 7, 2013 - link
It still amazes me that websites bench the i7s vs the amd a10 platforms. Whole a10 systems can be purchased for the price of an i7 CPU. Why dont you test a $200 graphics card in an a10 system vs the i7 integrated graphics - if this is the case and show people what they can get (for their $350 dollars spent on cpu + graphics)???? The closest price processor from intel vs 6800k that i found locally was an i3 3240. Why dont you use these in the comparisons?????? Why use a Gt640 only with an i7 & not on the amd system to show the cpu bias in the benches???DeviousOrange - Friday, June 7, 2013 - link
While yes the i7 being the fastest mainline part available and x86 will help distort its numbers a bit nevertheless the issue is not x86, the issue is iGPU performance only. That being said the i7 4770K and A10 6800K are both the top line parts in review so there is no ambiguity as to one being entry level and the other top line, these are both flagships so the test is top product on top product iGPU showdown.What is disappointing is Anandtech still don't have any workable Frame Scaling tool to asses performance as Frames Per Second means diddly squat where Frame Latency is the true indicator. As before the A10 is always below 10ms and often below 1ms while the i7's HD4600 often hits 60ms+ latencies which is basically a microstut.tut.tut.tut.tut.ter.
I haven't personally tested HD4600 but I have been told it has boosted FPS but in terms of frame transistions which are often a combination of hardware and driver support that HD4600 is not much better so again while HD4600 is close in FPS in some instance, its very far off in latency. In short I would rather have a iGPU average 27FPS but have 0.1ms latencies opposed to 35FPS with 50ms latencies.
mrSmigs - Friday, June 7, 2013 - link
The comparison is amd's budget apu line vs intels top desktop line. You should throw a radeon 7850 in the amd system to even up the specs (at least dollars wise) and then run a few graphics benches... please someone do a comparison i7-4770k with its iGPU vs a10-6800k+7850hd in gaming. im pretty sure i could guess the winner here...sireangelus - Friday, June 7, 2013 - link
Just think about the performance of a i74770KR... 128 mb of L4 cache, an amount to put to shame any serverPhiro69 - Friday, June 7, 2013 - link
" AMD is proud of its valiation on the A10-6800K" - you mean validation, right?halbhh2 - Friday, June 7, 2013 - link
While it is interesting to see gaming benches, since I might game on an HTPC, I'm actually thinking more on how perfectly the chips can display a 4K video, *and* I'm interested in whether I can run 2 HD movies (two at once) on 2 displays without any dropped frames or stutters (this actually matters a bit at the moment for me). I know there is one article here for the Intel 4600 graphics re 4K.hobagman - Saturday, June 8, 2013 - link
Anand, I think this is a waste of time. I don't know anybody who buys desktops anymore, unless they intend to use it as a workstation or as a gaming platform. In the first case, they don't usually care about graphics, and in the second case, they will absolutely have discrete graphics and these graphics benchmarks are utterly irrelevant. I like this website and appreciate the work, but I would rather you spend your time on something more useful for us -- for example, comparing the notebook platform integrated graphics would be reasonable. This article puzzles me.fteoath64 - Saturday, June 8, 2013 - link
"comparing the notebook platform integrated graphics would be reasonable.". There are many reviews about IGP for the APUs for specific users or gamers. This article highlights how much AMD has arrived in terms of gpu and cpu balance in their desktop and notebook parts (ie APUs specifically) which will pose a serious challenge to Intel's dominance. To many, Intel based cpu with IGP is clearly not the way to go only Intel cpu plus Nvidia discrete gpu or go the APU route and compromise cpu somewhat gaining close to discretely gpu performance for way less money.Also shows that Intel 4600 IGP has gone a long way to within striking distance of AMD APU gpus but not good enough as the sliding scale of reference MOVES each time Intel approaches. The GT3e potentially can match a NV 650M discrete but at the cost of Intel $$$ to the user. Most manufacturers rather go Nvidia discrete which is cheaper and better as well. So unless Intel goes into heterogeneous core architecture chips for cpus, there is nothing really new in their offering.
zyky - Saturday, June 8, 2013 - link
With the dozens of different configurations that call themselves "GT 640" It's pretty important to specify which one was used in these tests. GF116? GK107? GK208? GDDR3? GDDR5?Ryan Smith - Tuesday, June 18, 2013 - link
There's only one retail GT 640 (as of this article); the GK107 based DDR3 version.