The V5.0 depth evaluation of the millet tablet

The Yesterday, the Rabbit Rabbit V5.0 beta version of the official launch. On the Rabbit V5.0 version, the industry has also had a joke, is the glory of 6 released when said to rewrite the list of the rabbit, and then An V5.0 just exposure, so netizens joked that the rabbit in order not to let the Glory 6 quickly launched a new version. Is that true? Today we will solve the mystery, and the Rabbit launched the V5.0 of the cause and effect of a more comprehensive analysis. Let's take a look at the major changes in the V5.0 of the Rabbit, V5.0 new single core performance and HTML5 test, with a new 3D image test engine, and adjusted the weight of each test item. As everyone commonly used in the mobile phone scoring software, its every move will affect the specific score of each cell phone, then the rabbit V5.0 What is the impact of running points? Will the seating of the flagship mobile processor 骁丽 801, Tegra K1, Kirin 920 change? Let's experience this: the evaluation software is actually a special application, but it can mark the pleasure of running on a mobile phone. It's like a car running on a highway, and the evaluation software is a car, and a different road is like a mobile phone. The better the road, I run on the happy, other cars (microblogging, QQ, games and other conventional applications) run naturally also carefree, and vice versa. And the car will be on the road surface formation, the signs are correct, Rab Rao Road and so on detailed records, so it provides a good advice to all kinds of vehicles, the decision makers choose to go that road (player machine) also has a great reference. But the road is more and more wide (mobile phone performance is more and more good), and the test car itself is there, fire can not reach the maximum speed of the new road allowed, that is, can not detect how good the road. Looking at the car around the Roar, had to face the shouts: I should change the engine! This is a simple understanding of why the software needs to be upgraded. In official words, the original test items have come to the bottleneck, and will not be able to fully reflect the next generation of chip performance differences, this is particularly prominent in 3D graphics. In our general understanding, for An run points, the current top mobile phone in 37000 points, the middle end is generally maintained at 25000 points, and the entry level most of the 15000 points. These scores have basically formed a fixed thinking, if hastily introduced a new scoring mechanism, will no doubt the user's comparative understanding caused trouble. In this regard, the rabbit still uses 3.0 liters 4.0 of the train of thought. That is, to increase and upgrade the test project, as appropriate, adjust the proportion of a single item, and ultimately stabilize in a close to the previous version of the range, reduce the user's understanding of the score difficulty. So the Rabbit rabbit V5.0 in the end what new test items, upgrade what, the purpose of the proportion is how to adjust it? We went on to look down. Rabbit Rabbit Evaluation V5.0 Increase (upgrade) project has three main: 1, increase CPU single-threaded test; everyoneIt is often heard that day-to-day applications do not optimize multi-core, and it is true that many games run on four or eight-core handsets to only open one or two cores. And some power-saving mode will deliberately let some of the core sleep, at this time, the single core performance of the system to improve the flow of the speed, the start of a greater impact. The following picture shows an official comparison of the individual nuclear performance of the rabbit (the specific score is related to the frequency).