VG Conformance Test, the kill of khronos!

...... What is the draft I wrote yesterday? Where did I go ?? Forget it. If the disaster happens, blame yourself. If you lose it, lose it.

The book was written last night. It seems that I am too lazy to write something, and I plan to look down on the world where I write a book on my own. Then I save the draft, lost the draft-today's work is really full, because the new task is to deal with the Conformance Test of vg11.

Well, it's been a year since VG was officially started. From the first time I saw the excitement of my tiger head, and XX, which is constantly entangled with the Conformance Test of khronos (I really don't know what words to use). It seems that VG has become a part of my life, of course, there are also es to adjust the swap. I am quite dissatisfied with khronos. I have mentioned some in my previous blog. Compared with gles, VG still has a long way to go, not to mention how to use HW to provide these features-of course, people do not require you to use hardware acceleration, as long as you run flash-like applications fast on your mobile phone-1.0 and 1.1 alone, the fight between each other is amazing. There are two results in one case. I really don't know how to choose the version we should follow ??

Another dissatisfaction is the irrationality of the Conformance Test test itself. A feature is tested, but a large number of other relevant feature code is introduced, and the evaluation results are still greatly affected. In Version 1.1, it seems that it should have changed a little, so it improved the reference to make it more "accurate", and at the same time it scored off the fault tolerance score of some cases, what is the impact on our driver developer? I think it is self-evident that everyone should understand it. So tomorrow I will continue to make a good score to pass the test.

It's all about khronos. I hope it won't see my article.

-- Safari is used today. It has been useless for a long time. It feels a little warm,