AMD's A10-5750M Review, Part 1: The APU and Radeon HD 8650G Performance
by Dustin Sklavos on June 28, 2013 12:00 PM EST
Introducing the AMD A10-5750M and Mobile Richland
As an incremental release, AMD's Richland is a little bit hard to build up enthusiasm for. Architecturally almost identical to its predecessor, Trinity, Richland's chief refinement is a substantial improvement on its implementation of AMD's Turbo Core technology. Richland is able to manage its power states with finer granularity, which allows AMD to in some cases substantially beef up the clocks that the CPU and GPU halves of the chip can hit.
On the flipside, despite the branding shenanigans going on at AMD, we're still fundamentally dealing with the same architecture as Trinity. That means one or two Piledriver CPU modules paired up with VLIW4 GPU clusters. It's a little bizarre that we saw Kabini and Temash first, since those chips marry AMD's updated Jaguar low-power CPU architecture with their current generation GCN GPU architecture; Richland is essentially old technology that's seen a healthy refinement. The market segments that Kabini and Temash serve are the fastest growing, so it's understandable, but AMD's "high end" APU architecture looks a little antiquated by comparison.
While desktop Richland has been a little underwhelming, the mobile version should have a lot more teeth. Here's a comparison, generation by generation:
| AMD A-Series Mobile APUs (Mainstream TDP) | |||||||||||||
| Model | A10-5750M | A10-4600M | A8-5550M | A8-4500M | A6-5350M | A4-4000 | |||||||
| Modules/Cores | 2/4 | 2/4 | 2/4 | 2/4 | 1/2 | 1/2 | |||||||
| CPU Clock | 2.5 | 2.3 | 2.1 | 1.9 | 2.9 | 2.7 | |||||||
| CPU Turbo | 3.5 | 3.2 | 3.1 | 2.8 | 3.5 | 3.2 | |||||||
| Graphics | HD 8650G | HD 7660G | HD 8550G | HD 7640G | HD 8450G | HD 7520G | |||||||
| GPU Cores | 384 | 384 | 256 | 256 | 192 | 192 | |||||||
| GPU Clock | 533 | 496 | 515 | 496 | 533 | 496 | |||||||
| GPU Turbo | 720 | 685 | 720 | 685 | 720 | 685 | |||||||
| L2 Cache | 2x2MB | 2x2MB | 2x2MB | 2x2MB | 1MB | 1MB | |||||||
| Max DDR3 | 1866 | 1600 | 1600 | 1600 | 1600 | 1600 | |||||||
With the new generation, everyone gets a 200MHz bump in CPU base frequency and a much more modest 20-40MHz jump in base GPU clocks. Turbo clocks jump 35MHz on GPUs across the board, nothing to write home about, while CPU turbo clocks jump 300MHz across the board. Since Trinity was chiefly CPU limited, virtually any improvement in core clocks can be a big help. What we really need is for Richland to hit and sustain turbo clocks for longer periods of time, though, and hopefully AMD's improved Turbo Core technology can make up the difference.
Unlike with Trinity, AMD didn't seed Richland reference notebooks to reviewers, so our reference unit is the updated MSI GX60. Part 2 of my review will cover the MSI GX60 specifically, but for now, here's the spec table:
| MSI GX60 (2013) Specifications | |
| Processor |
AMD A10-5750M (4x2.5GHz, Turbo to 3.5GHz, 32nm, 4MB L2, 35W) |
| Chipset | AMD Hudson-3 |
| Memory | 2x8GB A-Data DDR3-1600 (originally 1x8GB) |
| Graphics |
AMD Radeon HD 8650G (VLIW4; 384 cores; 533/720MHz base/turbo frequencies) AMD Radeon HD 7970M 2GB GDDR5 (GCN; 1280 cores; 850MHz/4.8GHz core/memory; 256-bit memory bus) |
| Display |
15.6" LED Matte 16:9 1080p LGD0259 |
| Hard Drive(s) | Western Digital Scorpio Black 750GB 7200-RPM SATA 3Gbps HDD |
| Optical Drive | TSSTCorp SN-406AB BD-ROM/DVDRW |
| Networking |
Killer Networks e2200 PCIe Gigabit Ethernet Atheros AR9485WB-EG 2.4GHz 802.11b/g/n Bluetooth 4.0 |
| Audio |
Realtek ALC892 HD audio (THX TruStudio Pro) 2.1 speakers Mic, headphone, line-in, and line-out jacks |
| Battery | 9-cell, 87Wh |
| Front Side | - |
| Right Side |
Mic, headphone, line-in, and line-out jacks 1x USB 2.0 Optical drive |
| Left Side |
Vent 3x USB 3.0 SD card reader |
| Back Side |
Kensington lock AC adapter Ethernet D-SUB Mini-DisplayPort HDMI Vent |
| Operating System | Windows 8 64-bit |
| Dimensions |
14.97" x 10.24" x 1.77" 380mm x 260mm x 45mm |
| Weight |
7.7 lbs 3.49kg |
| Extras |
Webcam USB 3.0 Card reader THX TruStudio Pro audio Killer Networks wired networking SteelSeries keyboard |
| Warranty | 2-year parts and labor |
| Pricing | $1,199 |
Without getting into the details, relevant to testing is that the GX60 actually ships with only one DIMM channel populated. While the CPU isn't heavily affected by operating in single-channel mode, the IGP takes a nearly 50% hit to performance virtually across the board. It also doesn't ship with any solid state storage, so PCMark7 is going to be heavily impacted by the mechanical hard disk. In the second part of this review, when I tackle the GX60 specifically, you'll be able to get a better idea of what the loss of that second DIMM means.
Facebook
Twitter
RSS
82 Comments
View All Comments
monstercameron - Friday, June 28, 2013 - link
don't think it is fair comparing TDPs, we all know that they mean slightly different things for the two, intel 17W parts blow past 30+ watts from time to time.andrewaggb - Friday, June 28, 2013 - link
Ultimately the power measurement's that matter are battery life in various use cases and how much heat/active cooling is required. 35W under full gpu load is probably acceptable as long as the idle power usage is competitive with haswell. (I'm doubtful that it is though)monstercameron - Friday, June 28, 2013 - link
yeah idle is where haswell has richland and kabini beat...Khato - Friday, June 28, 2013 - link
While I doubt anyone would dispute the fact that Intel allows its processors to turbo past their specified TDP (considering that, ya know, it's kinda part of Intel's specifications) your argument would be better served by providing actual numbers instead of inflating them into the realm of fantasy. Specifically, Notebookcheck's power numbers for systems using the i7-3667U show a delta of roughly 23W between maximum idle and maximum load. Meanwhile their review of the A10-4600M shows a delta of 44W between idle and load.Regardless, going over specified TDP isn't really an issue, in fact it's typically a good thing. But it definitely means that actual power draw must be considered when comparing performance.
monstercameron - Friday, June 28, 2013 - link
really bro? http://www.notebookcheck.net/Review-Fujitsu-LifeBo...idle 9W, maxed out 40W...and that is only an i5.
Darkstone - Friday, June 28, 2013 - link
Download throttlestop. Click on the button that says 'tpl'. Expected output: power limit #1: 17W. power limit #2: 21W. Turbo time: 28s.Review the stresstest screenshot in this review:
http://www.notebookcheck.net/Review-Acer-Aspire-M3...
max
Max. package power, 19w, current package power, 16.5w.
Although the power usage can definitely exceed the TDP, it only does so for a short time. The effect on games or benchmarks taking >5 minutes is too small to make a difference. A processor with a TDP of 45w can be cooled by a heatsink rated at exactly 45w, but it might not always enter turbo mode for the full amount of time.
But that's not all, this review compares the graphics performance of an 17W tdp CPU with an 35W AMD part. The intel ULV parts are heavy power constrained. An 35W i5 part would probably be ~ 30% faster in games. Review notebookcheck, for example, starcraft 2 17W TDP ivy bridge: 31 fps. 45W part: 41 fps. Starcraft 2 does not scale above 2 cores. Check Dead Space or Hitman Absolution for more numbers that tell the same story.
Khato - Friday, June 28, 2013 - link
Bravo at finding a single example of a 17W SKU behaving abnormally in their review database. If you notice I specified a particular part which outperforms the one included in your linked review, and all reviews notebookcheck has done with that part show a delta of roughly 23W between maximum idle and maximum load.Just because there's an outlying result doesn't mean that it's the norm.
wumpus - Friday, June 28, 2013 - link
Are you seriously suggesting that Intel draws more power than TDP for long enough to soak the thermal mass of the CPU+recommend heat sink? The whole point of TDP means "thermal design power", and needs to be taken into account only for heatsink and mobile manufacturers.If you want to know what maximum current an Intel chip will draw, pull the datasheet (not the stuff made for marketing or consumers: it should be a pdf with several hundred pages. If you really want to build a motherboard; sign the NDA for the real one with the bugs included). Intel will list the requirements for the motherboard's power supply, and don't expect them to draw more than that (even then they may include a certain capacitance and frequency limits for the power supply. Intel is still free to draw however much they wish as long as the average within one period of the switching frequency is within the limits and can be supplied by the capacitance).
Breaking the specs is huge sin in this industry. Breaking what consumers think the specs are is irrelevant.
name99 - Friday, June 28, 2013 - link
This reminds me of the whining that occurred when Intel first added turbo support to its CPUs --- apparently it's "unfair" to use the laws of thermal physics to improve the performance of devices.Look, Intel has done an INCREDIBLE job of allowing its devices to run at short high speed bursts, for responsiveness, while generally using extremely low power. This is a tradeoff that meets most people's needs very well, even if it's not an appropriate tradeoff for a server chip that's going to be running at 95% utilization 24/7.
Others should be emulating Intel, not complaining that what they are doing is "unfair".
MrSpadge - Friday, June 28, 2013 - link
Excellent posts here by you, wumpus, darkstone and others!