News:

Willkommen im Notebookcheck.com Forum! Hier können sie über alle unsere Artikel und allgemein über Notebook relevante Dinge disuktieren. Viel Spass!

Main Menu

Post reply

Other options
Verification:
Please leave this box empty:

Shortcuts: ALT+S post or ALT+P preview

Topic summary

Posted by RobertJasiek
 - December 31, 2023, 15:27:08
Quote from: A on December 31, 2023, 14:43:42
Quote from: RobertJasiek on December 31, 2023, 14:41:48iPad mini 4 16GB Wifi
And iPadOS version?

15.6

(I know, 15.8 is available. Due to the forced update via LAN, which takes ca. 15 ~ 30 minutes for an update and 5 minutes for security measures and dialog answers, and the old battery, which needs some "warming up" at lowest brightness before I can be sure that the iPad will not shut down during the following minutes, and the maximum brightness during every restart or update process, I need much time and caution to update so I have switched to infrequent minor updating after the iPadOS 15 major update. Do not tell me 15.8 would solve the issues countless earlier updates did not solve.)
Posted by A
 - December 31, 2023, 14:43:42
Quote from: RobertJasiek on December 31, 2023, 14:41:48iPad mini 4 16GB Wifi
And iPadOS version?
Posted by RobertJasiek
 - December 31, 2023, 14:41:48
Quote from: A on December 31, 2023, 12:26:33
Quote from: RobertJasiek on December 29, 2023, 09:05:51by disabling iPadOS updates via WLAN
What's your iPad model again?

iPad mini 4 16GB Wifi
Posted by A
 - December 31, 2023, 12:26:33
Quote from: RobertJasiek on December 29, 2023, 09:05:51by disabling iPadOS updates via WLAN
What's your iPad model again?
Posted by RobertJasiek
 - December 30, 2023, 20:15:43
inferencing
Posted by A
 - December 30, 2023, 19:39:05
Quote from: RobertJasiek on December 30, 2023, 18:47:14For KataGo and RTX 4070, an 8C/16T CPU usually has 16% load (GPU 92 ~ 100%), every couple of minutes for several seconds 20 ~ 30% load (GPU ~90%) and every ca. half an hour for several seconds 30% ~ 100% load (GPU 50 ~ 85%).
Is it training or inferencing.
Quote from: RobertJasiek on December 30, 2023, 18:47:14Go is a GPU bottleneck so several GPUs can mitigate this
The only metric to add or remove GPUs is if speed is enough, not GPU load - it will always be 100%.
Posted by RobertJasiek
 - December 30, 2023, 18:47:14
Quote from: A on December 30, 2023, 16:40:04Whatever it's doing on CPU will be the bottleneck, if it's like Leela Zero.

Hehe, Leela Zero is not the standard.

For KataGo and RTX 4070, an 8C/16T CPU usually has 16% load (GPU 92 ~ 100%), every couple of minutes for several seconds 20 ~ 30% load (GPU ~90%) and every ca. half an hour for several seconds 30% ~ 100% load (GPU 50 ~ 85%). Only the latter qualifies as significant CPU bottleneck. More CPU cores and faster single thread speed would slightly reduce, but not eliminate the CPU bottleneck. That's because of the cause of reaching same positions by  different tree paths - infrequent but when they occur they involve complex decision-making. Except for that, Go is a GPU bottleneck so several GPUs can mitigate this. Of course, with many GPUs in a server, one needs Threadripper or better but also for the needed lanes addressing them.
Posted by A
 - December 30, 2023, 16:40:04
Quote from: RobertJasiek on December 30, 2023, 15:57:55250 playouts (related to visits) per second of M2 is between iPhone 13 pro and iPad Pro M1, not fast enough to reliably read ladders, so occassionally weak amateur moves occur, and 1/26 the speed of RTX 4070 Desktop.
Whatever, I just wanted to make sure the AI you are constantly referring to can be ran on phones today.

Quote from: RobertJasiek on December 30, 2023, 15:57:55Since you dismiss AlphaZero theory as outdated, you do not understand the nature of Go AI theory and complexity
There are a lot of outdated theories that are still computationally expensive. Go is expensive because it has way more move combinations than chess. But NN itself isn't, because NN isn't _calculating_ anything. It's _predicting_ the outcome making a higher number or lower-complexity calculations with better parallelization than code that's actually trying to _calculate_ it. So in the end NN complexity is LOWER than if you try to really math out or bruteforce the result (almost infinite complexity).

Quote from: RobertJasiek on December 30, 2023, 15:57:55It would make sense to execute inferenced KataGo on many A100s
You could, but makes zero sense. Whatever it's doing on CPU will be the bottleneck, if it's like Leela Zero.
Posted by RobertJasiek
 - December 30, 2023, 15:57:55
You are an optimist if you think that you do not need a manual. Compiling stuff might be the easy part. To use KataGo with a GUI, such as KaTrain, it needs to be told a command (or parameters) with which to call KataGo.

A b18 net is a good choice. There are older and new versions of nets.

250 playouts (related to visits) per second of M2 is between iPhone 13 pro and iPad Pro M1, not fast enough to reliably read ladders, so occassionally weak amateur moves occur, and 1/26 the speed of RTX 4070 Desktop.

Since you dismiss AlphaZero theory as outdated, you do not understand the nature of Go AI theory and complexity. The point is that application of nets scales well for quite some computation time: much more inference runtime gives much better results.

It would make sense to execute inferenced KataGo on many A100s. Just nobody wants to pay for such luxury.

Once more: not my AI. I use it but have not developed it.
Posted by A
 - December 30, 2023, 15:12:16
"With M2 processor of iPad Pro, the playouts are performed over 250 times per second with KataGo b18c384nbt"
There's your speed for iPad btw.
Posted by A
 - December 30, 2023, 14:30:28
Quote from: RobertJasiek on December 30, 2023, 13:44:56it relies on an AlphaGo Zero type engine with ELF OpenGo weight. This is probably stronger than humans (with sufficient runtime) but - I think but I am unsure - is not, and much weaker than, KataGo.
That forum post is from 2018. App description says
"It is a Go AI featuring a neural network of that strong, open source AI. Leela Zero, ELF OpenGo, PhoenixGo, Minigo, SAI, and optionally KataGo!"

So I presume your AI is seemingly something one can inference even on iPhone.

Quote from: RobertJasiek on December 30, 2023, 13:44:56If you want to run speed tests
Nah, not interested, all those Zero game AIs use outdated theory and are not computationally expensive by modern terms. I see they've added Apple Silicon support this May and seem to be still working on it, good for them.

Quote from: RobertJasiek on December 30, 2023, 13:44:56earn about KataGo and its tuning by starting from my webpage
I don't need instructions, I can build it from source.

Posted by RobertJasiek
 - December 30, 2023, 14:03:39
KataGo is developed by David Wu aka lightvector.

GUIs are developed by various people.

I have heard that KaTrain / Mac does not show visits/s in the GUI. KaTrain / Mac might cost something. It was said that it is the only local GUI for KataGo but this may have  changed without my knowledge.

I do not know if speed tests for KataGo are possible on a Mac command line. They are possible on Linux or Windows.
Posted by RobertJasiek
 - December 30, 2023, 13:46:58
KaTrain does rely on KataGo.
Posted by RobertJasiek
 - December 30, 2023, 13:44:56
According to

www.lifein19x19.com/viewtopic.php?p=238921#p238921

it relies on an AlphaGo Zero type engine with ELF OpenGo weight. This is probably stronger than humans (with sufficient runtime) but - I think but I am unsure - is not, and much weaker than, KataGo.

Unfrotunately, I do not know by heart which Apple device softwares use KataGo. Maybe this if its description is true:

katago goGame ai

apps.apple.com/us/app/katago-gogame-ai/id1509047602

KaTrain is also available for Mac but its GUI is very basic (might be good enough for your trials though) and I cannot know if it is as buggy as for Windows:

github.com/sanderland/katrain/releases

KaTrain runs local, I do not know whether the other softwares are local or just access to (slow) webservices.

AI KataGo Go

apps.apple.com/de/app/ai-katago-go/id1560986333

See also

senseis.xmp.net/?KataGo
senseis.xmp.net/?GoPlayingPrograms
senseis.xmp.net/?ComputerGo

For editing but not (much) for playing:

senseis.xmp.net/?GoEditingPrograms

If you want to run speed tests, learn about KataGo and its tuning by starting from my webpage and then following the various links:

home.snafu.de/jasiek/AI_Computer.html
Posted by A
 - December 30, 2023, 12:29:25
RobertJasiek, btw is this your KataGo and some other Go NNs running on iPhone/iPad?

apps.apple.com/us/app/a-master-of-go/id1442035374?platform=iphone