Who Controls the User Experience? AMD’s Carrizo Thoroughly Tested
by Ian Cutress on February 4, 2016 8:00 AM ESTBenchmark Results: Web and Synthetic
Here are our results from our web and synthetic tests. A reminder of our systems:
| System Overview | |||||
| µArch | APU | Base / Turbo MHz |
Memory | Channel | |
| HP Elitebook 745 G2 | Kaveri | A10 PRO-7350B (19W) | 2100 / 3300 | 8 GB | Dual |
| HP Elitebook 745 G3 | Carrizo | PRO A12-8800B (15W) | 2100 / 3400 | 4 GB | Single |
| Toshiba Satellite E45DW-C4210 |
Carrizo | FX-8800P (15W) | 2100 / 3400 | 8 GB | Single |
| HP Pavilion 17z-g100 |
Carrizo | A10-8700P (15W) | 1800 / 3200 | 8 GB | Single |
| Lenovo Y700 | Carrizo | FX-8800P (15W) | 2100 / 3400 | 16 GB | Single |
Google Octane 2.0
Lots of factors go into web development, including the tools used and the browser those tools play in. One of the common and widely used benchmarks to judge performance is Google Octane, now in version 2.0. To quote: 'The updated Octane 2.0 benchmark includes four new tests to measure new aspects of JavaScript performance, including garbage collection / compiler latency and asm.js-style JavaScript performance.' We run the test six times and take an average of the scores.
Octane splits hairs between the Kaveri and A10-8700P, but the Toshiba has the higher skin temperature and can turbo for longer than the Elitebook G3.
Mozilla Kraken 1.1
Kraken is a similar tool to Google, focusing on web tools and processing power. Kraken's tools include searching algorithms, audio processing, image filtering, flexible database parsing, and cryptographic routines.
Kraken mirrors Octane, except this time the A10-8700P gets a jump on the Kaveri.
WebXPRT 2013/2015
WebXPRT aims to be a souped up version of Octane and Kraken, using these tools in real time to display data in photograph enhancement, sorting, stock options, local storage manipulation, graphical enterfaces and even filtering algorithms on scientific datasets. We run the 2013 and 2015 versions of the benchmark.
In both versions of the benchmark, the Kaveri system beats all the 15W Carrizo platforms. It was inevitable that at some point during the benchmarking that those extra four watts of thermal headroom in the chip might allow the CPU to turbo for longer – as WebXPRT is by nature a bursty workload, if it can use this to its advantage then we’ll surely see a regression.
I want to pull out some power numbers a little early here to show what I mean. Here are the two Elitebooks in WebXPRT 2013, whose scores differ by 6%:
These power numbers were taken under the ‘all else equal rule’, so each screen was at the same brightness and almost zero applications requesting run time in the background. Here we see that the Carrizo system is drawing less power on average in idle and load (a common theme), but suffers from higher peak power draw and a much larger average-to-idle change in power (which can be overshadowed by onboard components coming out of sleep). It means we get the very uneasy metric of 1208.7 J of energy consumed for the Kaveri over idle and 1932.8 J of energy consumed for Carrizo, though it does depend on how much idle is truly idle across the whole SoC and platform.
This might be where the performance deficit lies though – in a Carrizo system that boasts lower power at idle and lower power draw on average, in a bursty workload environment it is actually wasting time and power to switch things on and off constantly.
Cinebench 15/11.5
Cinebench is a widely known benchmarking tool for measuring performance relative to MAXON's animation software Cinema 4D. Cinebench has been optimized over a decade and focuses on purely CPU horsepower, meaning if there is a discrepancy in pure throughput characteristics, Cinebench is likely to show that discrepancy. Arguably other software doesn't make use of all the tools available, so the real world relevance might purely be academic, but given our large database of data for Cinebench it seems difficult to ignore a small five minute test. We run the modern version 15 in this test, as well as the older 11.5 due to our back data.
Cinebench shows the spread of performance relating to the microarchitecture advantages of Carrizo compared to Kaveri, as well as the benefits that a 35W part can give over a 15W part. That being said, this spread of results, while perhaps an academic answer to ‘which is the fastest’ is not often seen in the real world.
x264 HD 3.0
The x264 HD 3.0 package we use here is also kept for historic regressional data. The latest version is 5.0.1, and encodes a 1080p video clip into a high quality x264 file. Version 3.0 only performs the same test on a 720p file, and in most circumstances hits its limit on high end processors, but still works well for mainstream and low-end. Also, this version only takes a few minutes, whereas the latest can take over 90 minutes to run.
As with Cinebench, we get an ideal academic spread of data.
Facebook
Twitter
RSS
175 Comments
View All Comments
alexruiz - Friday, February 5, 2016 - link
Ian,What specific model of the Elitebook 745 G3 did you test?
You listed the QHD screen (2560 x 1440 IPS) but you also wrote that it came in at just under $700. Is this correct? The one that you received for $700 included the QHD screen?
Thanks
Anonymous Blowhard - Friday, February 5, 2016 - link
The link on the page goes to the T3L35UT#ABA, which is a 1080p screen. I'm guessing the spreadsheet and/or pricing is incorrect somehow.Ian Cutress - Friday, February 5, 2016 - link
I remember being told it was 700, though I may have misheard and perhaps they were speaking GBP. But AB is right, the larger screen is 1080p and more expensive.The HP website is surprising though - the UX on the website is crazy to find a product that isn't the latest and greatest. It is very difficult to find anything and you seem to pay a premium for speccing out a custom Elitebook. Trying to spec out the one I was given came to $1900 from a $1620 base, which clearly isn't right because the high end 'buy now' model was about $1120 with similar specifications.
http://goo.gl/5spQls
Anonymous Blowhard - Friday, February 5, 2016 - link
GBP certainly makes a lot more sense. The base USD$750 model on HP's site has an A8-8600B, 4GB of RAM, and a 500GB HDD. If you want to find a Carrizo system with an "awful user experience" there it is.Custom-building an HP business system from their site is also expensive for no reason other than "because we said so." They really want you to buy the mass-manufactured SmartBuy models, although you may have better luck with a custom unit (read: "reasonable pricing") if you buy from a 3rd-party reseller.
I would like to see some GPU benchmarks with a second stick of RAM added to the 745 G3 though. Single-channel memory gives integrated graphics the Tonya Harding treatment.
Intel999 - Sunday, February 7, 2016 - link
Another review of Carrizo in the HP 745 G3 showed a 30% improvement in gaming when the second stick of RAM was installed. It went from woefully behind Intel to just ahead of them after sufficient RAM was provided.It is remarkable that any OEM would damage their own reputation by putting out crap that is obviously intended to funnel sales to Intel.
Imagine an OEM putting out a model, let's call it HP100, and offering it with identical specs with the only difference being a Carrizo CPU vs. an I5 from Intel. Few if any, would notice a difference in performance in the real world, not talking benchmarks.
The OEM would have a $100 savings on the Carrizo version if not more. If they were to sell it for $50 less than the Intel version they would sell more and that extra $50 profit on the AMD machine would more than offset the "rebates" they are getting from Intel. So they could have a higher profit margin at a lower price point. Afterall, with all the Intel marketing you'd still have plenty of people opting for the Intel version with it's lower profit margin. Just not as many and overtime you could approach Intel and point out that you make more money off AMD machines and who knows, maybe for the first time in recent history the OEM would control how their company is run and Intel would say "I guess we can cut our price by $25" and then you would be making the same profit on both Intel and AMD machines since, of course, the OEM wouldn't pass that $25 to the customer.
They could still offer junk machines at lower price points. Fill em up with Celerons or whatever they wanted attached to HDDs, 4GB of RAM, and crap screens.
Surely, no OEM is still following the old Intel rebate plan that allowed Intel to determine the market share that the OEM is allowed to give AMD. Or, are they?
Lolimaster - Friday, February 5, 2016 - link
I hate HP store with all my might, back in the day I got better experience donwloading from japanese p2p's not knowing japanese and just using a simple guide.keeepcool - Friday, February 5, 2016 - link
Please, just dont let Toshiba make laptops, just DON'T.Sorry for the word, but Toshiba is a shit brand that deserves to die. All their laptops have cooling solutions that are badly designed from the start, and the they cut all the corners and go for paper thin 4mm heatpipes on top of deformed and pitted 1.5mm cooper plates that make poor contact with the dies, the fans are thin, the thermal paste is worse than toothpaste, just stop allowing them to disgrace AMD name.
HP is also guilty of this, after the disgrace that where the dv6 models almost no one in Portugal and a lot other countries ever want a laptop with AMD cpu or gpu, because DV6's are just know for being litteral toasters that crash and burned, yes part of that was due to lack of maintenance, but still, it left a very sour taste regard AMD/ATI equiped laptops.
Today only people with low monetary margins will go for an AMD laptop, because they are the cheapest ones, and even then HP and Toshibas just manage to make toasters out of 7 and 15Watts TDP, its like they actually spend time engineering them to became so hot with so little TDP's...
tipoo - Friday, February 5, 2016 - link
50.27 Wh, 3 cell Li-Po design, rated to 10.25 hours. Out of a 3.3GHz AMD APU.You know, HP, somehow, I don't believe you.
bluevaping - Friday, February 5, 2016 - link
Nice article. And I get the point of the article about OEM's and AMD. I mostly agree with Shadowmaster 625. But it would have been nice to see a dual channel memory test. The Elitebook supports it. Slap in the extra ram and test it. And try dual channel 1866 Ram too.tipoo - Friday, February 5, 2016 - link
Shame the Lenovo doesn't support dual graphics - any reason why, or will this be updated in drivers? Especially as the integrated one has the same number of SPs as the dedicated, it could add a lot to the power equation.