Commercial Release of Ethereal 13.00 (NNUE) for AVX2 Systems

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
AndrewGrant
Posts: 1392
Joined: Tue Apr 19, 2016 4:08 am
Location: U.S.A
Full name: Andrew Grant
Contact:

Commercial Release of Ethereal 13.00 (NNUE) for AVX2 Systems

Post by AndrewGrant » Fri Jun 04, 2021 12:21 pm

Please see: http://chess.grantnet.us/Ethereal/

Ethereal 13 has been released, with both a free (Standard) version and a commercial NNUE (Efficiently-Updated Neural Network) version which requires an AVX2 compatible system. The free version is available on Github, and offers minor elo gains over the previous version, with increased functionality for analysis. Ethereal continues its legacy of stability and reliability for analysis and engine play.

Ethereal 13 (NNUE) includes two NNUE Networks for evaluation. These Networks are the second of their kind, boasting themselves as the only other high level NNUEs not derived from, trained on, nor duplicated from the works of the Stockfish team. With it, Ethereal adds new ideas and techniques to extend existing machine learning algorithms, and adds original practices into the implementation. These new approaches in software design allow for older AVX2 processors to outperform NNUEs found in other engines, while still harnessing the full power of the AVX2 instruction set.

Additionally, Ethereal is the first engine to offer a secondary Network trained exclusively for Fischer Random Chess (FRC). This Network outperforms the standard Network by significant margins in FRC play, showcasing a creative style by Ethereal, not reliant on Opening Theory. These Neural Networks offer up to +120 elo over Ethereal 12.75 in self-play for Standard and Fischer Random Chess, the culmination of over 238,000,000 games played between various versions of Ethereal.

The Standard version will run on any 64-bit machine, with source provided for Linux users, and binaries provided for Windows users. The NNUE version requires a CPU with the AVX2 instruction set, and will not run on systems lacking these instructions. AVX2 has become ubiquitous in modern processor design, and is necessary to capitalize on the full strength of Neural Network technology. If you are unsure if your system is supported, please reach out to me via email at [email protected] for assistance.

Stripe, an online payment processor, will verify your payment of $40 USD. Afterwards, you will receive an email from [email protected] within 24 hours containing a download link for a Standard NNUE Network, a Fischer Random NNUE Network, as well as an assortment of binary or executable files. You should select the binary matching your Operating System. Users with Intel CPUs containing BMI2 support (Also known as PEXT), should use the pext binary for optimal performance. AMD CPU users should use the popcnt binary.

-----

Code: Select all

Ethereal's Current Best FRC Network:
ELO   | 97.00 +- 3.47 (95%)
CONF  | 10.0+0.1s Threads=1 Hash=8MB
Games | N: 27035 W: 12816 L: 5458 D: 8761
http://chess.grantnet.us/test/10243/

Ethereal's Current Best Standard Network:
ELO   | 121.51 +- 5.35 (95%)
CONF  | 10.0+0.1s Threads=1 Hash=8MB
Games | N: 10436 W: 4966 L: 1458 D: 4012
http://chess.grantnet.us/test/10215/
-----

While Ethereal without NNUE still exists, and can be compiled directly from the source with ease, I have not provided binary files. This is because the strength of NNUE makes the standard version obsolete for high tier game-play, like rating lists, tournaments, and correspondence analysis. In the future I may add support for machines lacking AVX2 support, but have not done so at this time. All new CPUs being released contain the instruction set for many years now. As Neural Network technology takes over computer chess, those lacking modern vector support will have difficulty seeing the same gains as those with advanced vector support.

mar
Posts: 2494
Joined: Fri Nov 26, 2010 1:00 pm
Location: Czech Republic
Full name: Martin Sedlak

Re: Commercial Release of Ethereal 13.00 (NNUE) for AVX2 Systems

Post by mar » Fri Jun 04, 2021 1:12 pm

I wish you good luck, hopefully the commercial version will bring additional motivation and at least some financial compensation for your effort.

If the gains hold against other engines, we may see a new #3 (or even #2 if we don't count FF2), looking forward to see the results.
Martin Sedlak

User avatar
pohl4711
Posts: 1690
Joined: Sat Sep 03, 2011 5:25 am
Location: Berlin, Germany
Full name: Stefan Pohl
Contact:

Re: Commercial Release of Ethereal 13.00 (NNUE) for AVX2 Systems

Post by pohl4711 » Fri Jun 04, 2021 1:21 pm

I do not have a credit card (and dont want to get one). So, without the possibility to pay with PayPal or GooglePay, I cannot buy Ethereal 13 and cannot test it.

schack
Posts: 160
Joined: Thu May 27, 2010 1:32 am
Contact:

Re: Commercial Release of Ethereal 13.00 (NNUE) for AVX2 Systems

Post by schack » Fri Jun 04, 2021 2:35 pm

Zen 3 processors have native BMI instructions - you may want to update your advice here.

Do you have any test results against other engines?

toliveanddie76
Posts: 13
Joined: Thu Mar 26, 2020 10:35 pm
Full name: Norman Cantoral

Re: Commercial Release of Ethereal 13.00 (NNUE) for AVX2 Systems

Post by toliveanddie76 » Fri Jun 04, 2021 2:49 pm

chess engine: dime a dozen

finding a chess engine over 3000 ELO: simple google search

but this...
Additionally, Ethereal is the first engine to offer a secondary Network trained exclusively for Fischer Random Chess (FRC)
:idea:
is Priceless?

Damir
Posts: 2596
Joined: Mon Feb 11, 2008 2:53 pm
Location: Denmark
Full name: Damir Desevac

Re: Commercial Release of Ethereal 13.00 (NNUE) for AVX2 Systems

Post by Damir » Fri Jun 04, 2021 2:55 pm

AndrewGrant wrote:
Fri Jun 04, 2021 12:21 pm
Please see: http://chess.grantnet.us/Ethereal/

Ethereal 13 has been released, with both a free (Standard) version and a commercial NNUE (Efficiently-Updated Neural Network) version which requires an AVX2 compatible system. The free version is available on Github, and offers minor elo gains over the previous version, with increased functionality for analysis. Ethereal continues its legacy of stability and reliability for analysis and engine play.

Ethereal 13 (NNUE) includes two NNUE Networks for evaluation. These Networks are the second of their kind, boasting themselves as the only other high level NNUEs not derived from, trained on, nor duplicated from the works of the Stockfish team. With it, Ethereal adds new ideas and techniques to extend existing machine learning algorithms, and adds original practices into the implementation. These new approaches in software design allow for older AVX2 processors to outperform NNUEs found in other engines, while still harnessing the full power of the AVX2 instruction set.

Additionally, Ethereal is the first engine to offer a secondary Network trained exclusively for Fischer Random Chess (FRC). This Network outperforms the standard Network by significant margins in FRC play, showcasing a creative style by Ethereal, not reliant on Opening Theory. These Neural Networks offer up to +120 elo over Ethereal 12.75 in self-play for Standard and Fischer Random Chess, the culmination of over 238,000,000 games played between various versions of Ethereal.

The Standard version will run on any 64-bit machine, with source provided for Linux users, and binaries provided for Windows users. The NNUE version requires a CPU with the AVX2 instruction set, and will not run on systems lacking these instructions. AVX2 has become ubiquitous in modern processor design, and is necessary to capitalize on the full strength of Neural Network technology. If you are unsure if your system is supported, please reach out to me via email at [email protected] for assistance.

Stripe, an online payment processor, will verify your payment of $40 USD. Afterwards, you will receive an email from [email protected] within 24 hours containing a download link for a Standard NNUE Network, a Fischer Random NNUE Network, as well as an assortment of binary or executable files. You should select the binary matching your Operating System. Users with Intel CPUs containing BMI2 support (Also known as PEXT), should use the pext binary for optimal performance. AMD CPU users should use the popcnt binary.

-----

Code: Select all

Ethereal's Current Best FRC Network:
ELO   | 97.00 +- 3.47 (95%)
CONF  | 10.0+0.1s Threads=1 Hash=8MB
Games | N: 27035 W: 12816 L: 5458 D: 8761
http://chess.grantnet.us/test/10243/

Ethereal's Current Best Standard Network:
ELO   | 121.51 +- 5.35 (95%)
CONF  | 10.0+0.1s Threads=1 Hash=8MB
Games | N: 10436 W: 4966 L: 1458 D: 4012
http://chess.grantnet.us/test/10215/
-----

While Ethereal without NNUE still exists, and can be compiled directly from the source with ease, I have not provided binary files. This is because the strength of NNUE makes the standard version obsolete for high tier game-play, like rating lists, tournaments, and correspondence analysis. In the future I may add support for machines lacking AVX2 support, but have not done so at this time. All new CPUs being released contain the instruction set for many years now. As Neural Network technology takes over computer chess, those lacking modern vector support will have difficulty seeing the same gains as those with advanced vector support.
Hi Andrew,

It would be nice if you can put more buying options like paypal on your list...

When only Credit Card, it would seem you are trying to avoid paying taxes.... :wink: :wink:

Modern Times
Posts: 2915
Joined: Thu Jun 07, 2012 9:02 pm

Re: Commercial Release of Ethereal 13.00 (NNUE) for AVX2 Systems

Post by Modern Times » Fri Jun 04, 2021 2:56 pm

Debit cards like debit mastercard could work ?
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

peter
Posts: 2471
Joined: Sat Feb 16, 2008 6:38 am
Full name: Peter Martan

Re: Commercial Release of Ethereal 13.00 (NNUE) for AVX2 Systems

Post by peter » Fri Jun 04, 2021 3:02 pm

AndrewGrant wrote:
Fri Jun 04, 2021 12:21 pm
Please see: http://chess.grantnet.us/Ethereal/
Great news, Andrew!
Will Ethereal 13.00 NNUE compete at TCEC?
Good luck regards
Peter.

AndrewGrant
Posts: 1392
Joined: Tue Apr 19, 2016 4:08 am
Location: U.S.A
Full name: Andrew Grant
Contact:

Re: Commercial Release of Ethereal 13.00 (NNUE) for AVX2 Systems

Post by AndrewGrant » Fri Jun 04, 2021 3:03 pm

Damir wrote:
Fri Jun 04, 2021 2:55 pm
Hi Andrew,
It would be nice if you can put more buying options like paypal on your list...
When only Credit Card, it would seem you are trying to avoid paying taxes.... :wink: :wink:
pohl4711 wrote:
Fri Jun 04, 2021 1:21 pm
I do not have a credit card (and dont want to get one). So, without the possibility to pay with PayPal or GooglePay, I cannot buy Ethereal 13 and cannot test it.
I am in the process of getting Paypal to work. Originally I was planning to use Paypal, as it has greater name recognition, and charges lower fees on my end, where as Stripe takes a fair chunk for non US transactions. Paypal integration seemed like it was only meant for sending physical products, not making digital purchases. As a last resort, I might end up adding a Paypal link which will not be a "sale" page, but simply a money transfer page. Downside is that I will not be able to automate the process as well.

AndrewGrant
Posts: 1392
Joined: Tue Apr 19, 2016 4:08 am
Location: U.S.A
Full name: Andrew Grant
Contact:

Re: Commercial Release of Ethereal 13.00 (NNUE) for AVX2 Systems

Post by AndrewGrant » Fri Jun 04, 2021 3:04 pm

peter wrote:
Fri Jun 04, 2021 3:02 pm
AndrewGrant wrote:
Fri Jun 04, 2021 12:21 pm
Please see: http://chess.grantnet.us/Ethereal/
Great news, Andrew!
Will Ethereal 13.00 NNUE compete at TCEC?
Good luck regards
A slightly older version is playing at TCEC, but they should be fairly equal in strength. I would say I expect another Div P finish, but NNUE has equalized the gap that Ethereal had over the rest of the alpha-beta field. We will have to see.

Post Reply