recentpopularlog in

charlesarthur : benchmark   4

Fun with charts: The iPad bests the MacBook • Six Colors
Jason Snell takes a look at what the benchmarks say about the latest iPads v the latest MacBook (Pro or plain):
<p><img src="https://sixcolors.com/images/content/2018/multicore-value-redux3-6c.png" width="100%" />

You might say it’s not fair to compare these devices because the iPad Pro is a computer but not a PC. But even if you don’t buy the fact that the iPad Pro is perhaps the best value in terms of processor performance in Apple’s mobile product lineup today, you’ve got to admire the power of that eight-core A12X Bionic processor. The only MacBooks that can beat it right now in overall score are the two fastest 15-inch MacBook Pro models.

(For those who are concerned that only measuring multi-processor scores is unfair because so many software tasks aren’t multithreaded and have to push a single processor core to the limit—don’t sweat it. The iPad Pro still comes out on top in all the single-core versions of these measurements.)</p>


This is pretty dramatic: in price/performance terms, the iPads are miles ahead. That probably applies for non-Apple PCs too. Though as Snell says, there are other criteria too.
ipad  macbook  benchmark 
december 2018 by charlesarthur
Hardware Unboxed analyzes Intel's commissioned core i9-9900k benchmarks • HardOCP
:
<p>Hardware Unboxed did a short analysis of a few of the benchmarks as their team felt that the i7-8700K benchmarks and the AMD Ryzen 2700X numbers were incorrect. They found that Principled Technologies had allegedly gimped the AMD CPUs by using different coolers, incorrect ram timings, and possibly even disabled some of the cores on the AMD Ryzen 2700X. To put this into perspective, on the Ashes of the Singularity benchmark that Hardware Unboxed ran, the AMD Ryzen 2700X was 18% faster and the i7-8700K was 4% slower, than the commissioned testing that Intel has published. They even showed how over a suite of games that the i7-8700K was only 9% faster than the AMD Ryzen 2700X in previous pure gaming benchmarks conducted by Hardware Unboxed. Yet in Intel's commissioned benchmark results, the AMD Ryzen 2700X was far, far, behind the Intel i7-8700K in performance metrics. This is why we never trust a manufacturer's benchmarks. Always wait for the review before buying hardware.</p>

So Intel is "choosing" who benchmarks its processors for broader publication so that they will come out ahead of AMD. It feels weird to be living in a time when Intel cares again about AMD being competitive.

That said, unless you're building a PC from scratch, you don't have much choice about your processor, do you? (Thanks Stormyparis for the link.)
Intel  benchmark  cheating 
october 2018 by charlesarthur
The iPhone XS & XS Max review: unveiling the silicon secrets • Anandtech
Andrei Frumusanu:
<p>Apple promised a significant performance improvement in iOS12, thanks to the way their new scheduler is accounting for the loads from individual tasks. The operating system’s kernel scheduler tracks execution time of threads, and aggregates this into an utilisation metric which is then used by for example the DVFS mechanism. The algorithm which decides on how this load is accounted over time is generally simple a software decision – and it can be tweaked and engineered to whatever a vendor sees fit.

Because iOS’s kernel is closed source, we’re can’t really see what the changes are, however we can measure their effects. A relatively simple way to do this is to track frequency over time in a workload from idle, to full performance. I did this on a set of iPhones ranging from the 6 to the X (and XS), before and after the iOS12 system update.

<img src="https://images.anandtech.com/doci/13392/Ramp-A8.png" width="100%" />

Starting off with the iPhone 6 with the A8 chipset, I had some odd results on iOS11 as the scaling behaviour from idle to full performance was quite unusual. I repeated this a few times yet it still came up with the same results. The A8’s CPU’s idled at 400MHz, and remained here for 110ms until it jumped to 600MHz and then again 10ms later went on to the full 1400MHz of the cores.

iOS12 showcased a more step-wise behaviour, scaling up earlier and also reaching full performance after 90ms.

<img src="https://images.anandtech.com/doci/13392/Ramp-A9.png" width="100%" />

The iPhone 6S had a significantly different scaling behaviour on iOS11, and the A9 chip’s DVFS was insanely slow. Here it took a total of 435ms for the CPU to reach its maximum frequency. With the iOS12 update, this time has been massively slashed down to 80ms, giving a great boost to performance in shorter interactive workloads.</p>

Most of this multi-page review is just benchmark gobbledygook to me, but this page and those graphics really stand out because it shows iOS 12 getting performance improvements of as much as 50% on old hardware, through tweaks to the core OS.
benchmark  iphone  xs 
october 2018 by charlesarthur
Samsung TVs appear less energy efficient in real life than in tests » The Guardian
Arthur Nelsen:
<p>The lab studies found that Samsung’s ‘motion lighting’ feature reduced the TV sets’ brightness – and power consumption – under international electrotechnical commission (IEC) test conditions. These involve the playback of fast sequences of varied material, such as recorded TV shows, DVDs and live broadcasts.

But under real-world viewing conditions, no reductions in power consumption were registered, making the sets’ power consumption, fuel bills and carbon emissions correspondingly higher.

After tests in February, a ComplianTV report, which did not name Samsung, said: “The laboratories observed different TV behaviours during the measurements and this raised the possibility of the TV’s detecting a test procedure and adapting their power consumption accordingly. Such phenomenon was not proven within the ComplianTV tests, but some tested TVs gave the impression that they detected a test situation.”

“Samsung is meeting the letter of the law but not the spirit of the law,” Rudolf Heinz, the project manager of ComplianTV’s product lab, told the Guardian.</p>


Oh, come on, Samsung would never.. <a href="http://www.theguardian.com/technology/2013/jul/31/samsung-galaxy-s4-benchmarking-software">oh</a>.
samsung  benchmark  tv 
october 2015 by charlesarthur

Copy this bookmark:





to read