Thanks a lot!
| Life In 19x19 http://www.lifein19x19.com/ |
|
| New android app "BadukAI" http://www.lifein19x19.com/viewtopic.php?f=18&t=17784 |
Page 22 of 27 |
| Author: | lovego [ Thu Mar 17, 2022 6:19 pm ] |
| Post subject: | Re: New android app "BadukAI" |
Thanks a lot!
|
|
| Author: | lovego [ Mon Mar 21, 2022 2:14 am ] |
| Post subject: | Re: New android app "BadukAI" |
Haha, Katago 1.11.0 is released! https://github.com/lightvector/KataGo/r ... ag/v1.11.0 Can you continue updating? 
|
|
| Author: | akigo [ Mon Mar 21, 2022 2:57 pm ] |
| Post subject: | Re: New android app "BadukAI" |
Fortunately, this is not necessary. There are only a few commits between release 1.11 and the revision that I built for BadukAI 1.14. And these commits are mainly concerned with the code for training and with documentation, which don't matter for BadukAI. So the KataGo code in BadukAI 1.14 is practically equivalent to release 1.11. |
|
| Author: | And [ Tue Mar 22, 2022 9:01 am ] |
| Post subject: | Re: New android app "BadukAI" |
in the worst case, depending on the hardware, what is the speed gain of an optimized network? and at the best? |
|
| Author: | akigo [ Tue Mar 22, 2022 1:28 pm ] |
| Post subject: | Re: New android app "BadukAI" |
If you compare the speeds of optimized and original network, you get: a) If you choose the optimal number of search threads (in terms of speed) for each network: In the worst (HW dependent) case the speed gain is factor 2 (e.g. 20 playouts/second instead of 10). In the best (HW dependent) case the speed gain is factor 24 (e.g. 240 playouts/second instead of 10). b) If you choose one search thread for both networks: In the worst (HW dependent) case the speed gain is factor 2. In the best (HW dependent) case the speed gain is factor 40. |
|
| Author: | lovego [ Tue Mar 22, 2022 6:27 pm ] | ||
| Post subject: | Re: New android app "BadukAI" | ||
I find a compatibility issue! I have two phones(Dimensity 720 and Snapdragon 662) When I use 40B weight, the output of winning rate is wrong.
|
|||
| Author: | lovego [ Wed Mar 23, 2022 12:27 am ] | ||
| Post subject: | Re: New android app "BadukAI" | ||
In my vivo y52s (Dimensity 720) I have test all weights Kata 10b -> ok kata 20b -> bad lz 15b -> ok lz 40b ->ok
|
|||
| Author: | akigo [ Wed Mar 23, 2022 2:29 pm ] |
| Post subject: | Re: New android app "BadukAI" |
Since version 1.14 BadukAI runs the optimized KG networks (20b and 40b) on NPU instead of CPU on Mediatek processors. This works well (and fast) on Dimensity 1100 (according to user reports). But your tests show that on the (slightly older) Dimensity 720 the NPU can't handle the KG network correctly and produces garbage output. What a pity. Since I don't know on which Mediatek processors it works and on which it doesn't, I will add a switch to KataGo's config settings in the next release, so that every user can turn this on or off depending on whether it works on his/her device. |
|
| Author: | lovego [ Wed Mar 23, 2022 5:14 pm ] |
| Post subject: | Re: New android app "BadukAI" |
OKļ¼thanks
|
|
| Author: | And [ Thu Mar 24, 2022 9:46 am ] |
| Post subject: | Re: New android app "BadukAI" |
akigo wrote: If you compare the speeds of optimized and original network, you get: a) If you choose the optimal number of search threads (in terms of speed) for each network: In the worst (HW dependent) case the speed gain is factor 2 (e.g. 20 playouts/second instead of 10). In the best (HW dependent) case the speed gain is factor 24 (e.g. 240 playouts/second instead of 10). b) If you choose one search thread for both networks: In the worst (HW dependent) case the speed gain is factor 2. In the best (HW dependent) case the speed gain is factor 40. stunningly!!! |
|
| Author: | lovego [ Fri Mar 25, 2022 7:50 am ] |
| Post subject: | Re: New android app "BadukAI" |
My friend bought a new phone (Redmi K30 Pro with Dimensity 9000 chip), but the speed of kata 20b is very slow, it seems that there is no acceleration, only 20n/s. The performance of Dimensity 9000 is stronger than that of Dimensity 1100, it should exceed 140n/s. Is it possible to provide a new version that allows me to test different acceleration schemes.
|
|
| Author: | akigo [ Fri Mar 25, 2022 1:49 pm ] |
| Post subject: | Re: New android app "BadukAI" |
Can you share the android system log for the time period when KataGo is starting ? Then I could check for error messages which give a hint what's going wrong. Otherwise I wouldn't know what to change ... |
|
| Author: | akigo [ Sat Mar 26, 2022 3:09 pm ] |
| Post subject: | Re: New android app "BadukAI" |
I made a new version (1.15) available on play store https://play.google.com/store/apps/details?id=net.kir.baduk_ai and github https://github.com/aki65/aki65.github.io/releases/tag/v1.15.0 with the change I announced in post #428. |
|
| Author: | akigo [ Sun Mar 27, 2022 9:42 am ] |
| Post subject: | Re: New android app "BadukAI" |
lovego wrote: My friend bought a new phone (Redmi K30 Pro with Dimensity 9000 chip), but the speed of kata 20b is very slow, it seems that there is no acceleration, only 20n/s. The performance of Dimensity 9000 is stronger than that of Dimensity 1100, it should exceed 140n/s. Is it possible to provide a new version that allows me to test different acceleration schemes. You can do this with the version from https://github.com/aki65/aki65.github.io/releases/tag/v1.15.1 (see instructions there). |
|
| Author: | lovego [ Sun Mar 27, 2022 5:35 pm ] |
| Post subject: | Re: New android app "BadukAI" |
Haha, my Dimensity 720 works well when useSpecificNpuAccelerator is 3/4/5 
|
|
| Author: | lovego [ Mon Mar 28, 2022 2:37 am ] |
| Post subject: | Re: New android app "BadukAI" |
Dimensity 9000 is about 300n/s (KG 20b) when useSpecificNpuAccelerator is 3 The fastest chip at present! 
|
|
| Author: | lovego [ Mon Mar 28, 2022 3:55 am ] |
| Post subject: | Re: New android app "BadukAI" |
Oh, I get 400+n/s when numSearchThread is 3 and paralletEvalution is on |
|
| Author: | akigo [ Mon Mar 28, 2022 1:36 pm ] |
| Post subject: | Re: New android app "BadukAI" |
It's great to see that. Thank you very much for your testing. I would like to add an entry for Dimensity 9000 in the performance table on BadukAI's homepage. Since all the visit counts in this table were measured in this position (black to play) it would be nice if you could check the speed of Dimensity 9000 in this position. It's important to compare speeds only in the same position, since the type of position affects the visit count: In a highly symmetrical position (for example an empty board), the search tree will include a lot of positions which are mirrors of already visited positions. KataGo will handle a visit of such a position by cache lookup instead of computation, so the visit count will be higher. By the way, correct move is E3 (not too difficult, I am told, but already beyond my level), so also a good way to check if computations are correct. |
|
| Author: | lovego [ Mon Mar 28, 2022 5:42 pm ] |
| Post subject: | Re: New android app "BadukAI" |
OK, I'll let my friend try. Another thing, in version 1.16, the LZ engine crashes when the analysis is paused, and the speed is very slow on Dimensity 9000, can LZ choose the NPU accelerator to use? |
|
| Author: | lovego [ Mon Mar 28, 2022 7:02 pm ] | ||
| Post subject: | Re: New android app "BadukAI" | ||
Regarding board recognition, I found a good solution! Use yolov3 to identify black stone, white stone, corners (no stone), sidelines (no stone) and empty position. https://blog.csdn.net/wyzegg/article/details/108029789 https://github.com/zhuoyiyao97/YOLO-GO
|
|||
| Page 22 of 27 | All times are UTC - 8 hours [ DST ] |
| Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |
|