HP Enterprise unveils The Machine, with 160 TB of memory

[duncan]

8-man is 25% of the whole set.

Is it accurate to say Chess is solved by 1/ 4 ?

Each set is exponentially larger.

At the very most one part in 72 quadrillion solved, but probably a lot smaller portion than that.

Plus the fact that there are no 8 man file sets available.

is there a way to use finalgen to get automated results for 8 men with one piece for each side ?
http://finalgenchess.ovh/home_ing.php

[Dirt]

is there a way to use finalgen to get automated results for 8 men with one piece for each side ?
http://finalgenchess.ovh/home_ing.php

Pawns can promote. A complete solution is almost as hard as the full 8 man tablebase.

(Usually, I count pawns as pieces when talking about tablebases, so this took some thinking. Also, aren't kings pieces?)

[duncan]

40 nodes.
I am guessing this thing costs some serious money

8-man lomonosov tablebases not considered earlier than 2018.

http://rybkaforum.net/cgi-bin/rybkaforu ... pid=562705

maybe this year if something will start if we are lucky

[Milos]

Production systems will ship with hundreds of petabytes installed, out-ot-the-box using memrister technology for the byte-addressable memory fabric (or similar, non-volatile, low/zero power, memory technology).

Please stop that non-sense. Do you even have a clue what memristor is?
It's a generic term for non-existent research dead-end type of devices.
The only even remotely viable and currently (2018) existing novel NVM device is 3DXP and it is there exclusively as fast flash replacement. We might see it in 2019 as far-memory, but that's not even remotely close to what you are writing up there.

[jdart]

This kind of thing is already kind of existing, because there are in-memory big data systems that run on distributed clusters of nodes. See for example Apache Ignite (https://ignite.apache.org/use-cases/spa ... layer.html). HP may have a bigger higher performance version of that but it has been done before.

--Jon