Hi,
I got new xiaomi 13T pro with Mali-G715-Immortalis MC11 r1p2 and méditek 9200+.
And the speed of my program is 20% less than on my old hauwei honnor play with mali G72 MC24.
I added
#pragma OPENCL EXTENSION cl_khr_priority_hints : enable
And this improved the speed from 140ms to 110 ms but still superior to the 80ms of the hauwei.Does using CL_Buffer with Mali-G715 could drop the speed.
OpenCL support definitely varies from OEM to OEM, and over time as well, which can be frustrating.
Unfortunately we can't provide a driver, we deliver to our partners who adapt the driver code to put it on their phones, so it has to come from them.
Xiaomi forums are in theory at https://c.mi.com/global/ but I can't get to it (from UK). However, there's an unofficial EU forum that apparently works with Xiaomi at https://xiaomi.eu/community/ - I've not used it to see what the quality of support is, but it seems active at any rate.
That said - looks like you've done great work yourself to get better performance!
Thanks for the information but i already try to contact xiaomi by mail concerning opencl. But after that, my e-mail was refused to create account. But they said that they do not have information about it. Xiaomi is just an integrator not a chip producer like hauwei. They may not even know that opoencl exist ;))
Yes i spend nearly 2 months on the xiaomi problem, not full time of course, but thinking about the possible problem until i found it. ;))
But i think that i anderstoud why OpenCL is a problem for OEM. It use a lot of batterieand produce a lot of heat when running it with CPU multithreading. When openGL is not so consumer of batterie and heat.
Yes i it quite frustrating that OpenCL is so different from one OEM to another and we cannot find so much help from any OEM.
But i noticed that OpenCL performance is related with the number of core on the GPU. So G72 with 24 core should be faster than G715 with 11 core. I may be wrong but that what i think.
The only imprvment i could see for openCL would be to be able to compute group in order not in aléatoir. that would be great we could do processing that need to be done with the CPU. But may be in the futur ;)).
PS. concerning the xiaomi. It look like they use 64bit like two 32bit for GPU and CPU driver command. That is may be why 64bit use all the core and 32bit compilation only 4 core. On the hauweil i found out that 64bit or 32 bit give the same result, and that 32bit is a little bit faster than 64bit. I did not anderstoud why 64bit use all the core and that 32bit only 4 core. MAys be someone could explain it.
regards.
i want to correct my previous remark concerning xiaomi.
xiaomi is a all more faster when the volume of information to be computed increase. And the CPU multithreading is a lot faster than the hauwei honnor play.
so i apologize for the previous remark.
hi,
Last test with all optimization. Xiaomi 13T PRO is 3 to 5 time faster than hauwei.
remove all the .wait and .flush at the end of the JNI function where you are doing your OpenCL kernel and CPU processing.
hi, Last test after using 64bit on hauwei mali G72 mp12. Using 64bit increase the performance and i can get 25% more than xiaomi 13T pro mp11. Anaway booth of them start doing wrong over 100 ms for image processing. I just test it today i wanted to be sure about what i said before. So using 64bit in OpenCL is a lot much better. ;))
But xiaomi is slower but stable in time. Hauwei can goes from 100% to 400% time consuming during some time and then going back to 100%. I think it is comming from the focus of the camera, or memory GPU cleaning. I am not sure yet. I will let you know. ;))
In fact the diferrence come from the camera. I got more data with the Xiaomi, more détail. So more data to process and more time consuming. So let said xiaomi is more accurate.