The 100:1 Time Odds Massacre

[goni-K26]

The 100:1 Time Odds Massacre | Is the "Magic" of Houdini 6 Officially Dead?

Tournament Setup:

• Engine 1: Houdini 6.03 x64 (BMI2) | TC: 15m + 10s (Rapid) | Hash: 256 MB
• Engine 2: Stockfish 18 x64 (AVX2) | TC: 9s + 0.1s (Ultra-Bullet) | Hash: 16 MB
• Hardware: Intel(R) Core(TM) i5-4570 @ 3.20GHz | 1 Thread
• Ratio: Exactly 100:1 Time Odds

The Final Score (100 Games):

Code: Select all

Stockfish 18:  65.5  (41 Wins, 49 Draws, 10 Losses)
Houdini 6.03:  34.5  (10 Wins, 49 Draws, 41 Losses)

The Analysis:

I decided to conduct a brutal experiment to see if a massive 100-to-1 time advantage
could save a classical legend from the modern NNUE slaughterhouse.
Most people say time is the great equalizer in chess. Well, after 100 games, I can tell you: Time is dead. Logic is dead.
There is only Stockfish.

1. 9 Seconds vs 15 Minutes:
Stockfish 18 had exactly 9 seconds for the entire game. Houdini 6.03 had 15 minutes. Let that sink in.
Stockfish was playing at the speed of light, with a tiny 16 MB Hash (basically thinking with its eyes closed), and it still treated a former
World Champion like a club player.

2. The Evaluation Gap:
Even with 100x more thinking time, Houdini’s hand-coded evaluation was blind to the deep positional traps of the NNUE era.
It’s not just a strength gap; it’s a biological evolution. Houdini looks like a calculator, while SF 18 looks like a God.

3. A 13.5% Disaster:
In my wider 'Elite-25' tournament, Houdini finished dead last with a pathetic 13.5% score.
Even with 5x time odds against the whole field, it couldn't survive. But this 100:1 match against SF 18 is the final nail in the coffin.

The Provocation:

Is Robert Houdart’s masterpiece now officially a "legacy engine" that belongs in a museum? Or is Stockfish 18 so efficiently optimized that time odds have become a joke?

Warning: If you are a fan of classical 'Human-like' evaluation,
the PGNs of this match will break your heart. Houdini wasn't out-calculated; it was out-classed in every single phase of the game.

What do you think? Is there any amount of time that can make Houdini competitive again, or has the "Magic" finally run out?

[mar]

what are you talking about? Houdini 6 was an illegal Stockfish derivative (H6 source was leaked some time ago)

[goni-K26]

what are you talking about? Houdini 6 was an illegal Stockfish derivative (H6 source was leaked some time ago)

Listen Martin, I’m well aware of the history behind Houdini 6 and the source code controversy. That’s old news for anyone who follows the scene. However, you're missing the point of my experiment.

The focus here isn't on the "legality" or the "origin" of the code, but on the sheer technological leap we've seen since that era. Back then, whether the code was "borrowed" or not, Robert Houdart managed to optimize it in a way that gave it a significant Elo boost and a very specific positional "feel" that dominated the field for years.

The 100:1 Time Massacre was a test of evolution, not a legal audit. I gave Houdini 15 minutes against 9 seconds of modern Stockfish 18 on a single thread (Intel i5-4570). The fact that a 100x time advantage can’t bridge the gap between a classical legend and a modern NNUE engine is what’s fascinating here.

I’m a fan of the performance and the history, not the license agreement. It’s about the end of an era. The "Magic" is officially dead not because of the code’s history, but because modern engines have reached a level where time odds simply don't matter anymore.

[goni-K26]

The 100:1 Evolution Chronicles | Houdini vs. The NNUE Era

Tournament Setup:

Code: Select all

-----------------------|--------------------------------------------|
 Feature               | Tournament Details                         |
-----------------------|--------------------------------------------|
Tournament Name:       | The 100:1 Time Odds Massacre               |
Engine Threads:        | Single Thread(1 CPU)                       |
Tournament Type:       | Gauntlet                                   |
Hardware:              | Intel(R) Core(TM) i5-4570 3.20GHz 8GB RAM  |
GUI:                   | Arena Chess GUI                            |
Operating System:      | Windows 11 Pro x64                         |
-----------------------|--------------------------------------------|
Time Control(Houdini): | 15m + 10s (Rapid)                          |
Hash(Houdini):         | 512 MB                                     |
-----------------------|--------------------------------------------|
Time Control(Oppon-20):| 9s + 0.1s (Ultra-Bullet)                   |
Hash(Opponents-20):    | 16 MB                                      |
-----------------------|--------------------------------------------|
Rounds:                | 50                                         |
Tablebases:            | 5 men Syzygy                               |
-----------------------|--------------------------------------------|

Openings
________________________________

Code: Select all

|---------|---------|---------|---------|---------|---------|---------|---------|---------|---------|
| Round   | Opening | Round   | Opening | Round   | Opening | Round   | Opening | Round   | Opening |
|---------|---------|---------|---------|---------|---------|---------|---------|---------|---------|
|  1 -  2 |   -     |  3 -  4 | 1. e4   |  5 -  6 | 1. d4   |  7 -  8 | 1. c4   |  9 - 10 | 1. Nf3  |
| 11 - 12 | 1. g3   | 13 - 14 | 1. f4   | 15 - 16 | 1. b3   | 17 - 18 | 1. Nc3  | 19 - 20 | 1. b4   |
| 21 - 22 | 1. e3   | 23 - 24 | 1. d3   | 25 - 26 | 1. a3   | 27 - 28 | 1. g4   | 29 - 30 | 1. c3   |
|---------|---------|---------|---------|---------|---------|---------|---------|---------|---------|

After 120 games, the Houdini is proving that it’s not ready for retirement just yet.
Despite the massive 100:1 time disadvantage against a field of modern NNUE contenders,
Houdini 6.03 is maintaining a solid 53.3% score.

The Performance Analysis:
Houdini’s resilience is most evident in its defensive stability.
While the top-tier engines like Stockfish 17 and PlentyChess 7.0
are constantly probing for weaknesses, they are finding it remarkably difficult to break
Houdini's classical structure when it has 15 minutes of thinking time. Stockfish 17,
even with its superior NNUE evaluation, has only managed a single win in 6 rounds, proving that
depth and time can still act as a shield against raw tactical speed.

The Rising Threat:
The standout performer so far is PlentyChess 7.0.0, which has become Houdini's primary antagonist.
With a 66.6% individual score and two wins against the lead engine, it seems to handle the ultra-bullet time control with higher efficiency than even Stockfish.

The Bottom Field:
Engines like Motor 0.9.0 and Horsie 1.1 are struggling significantly.
They are unable to compensate for the lack of time, providing Houdini with the necessary points to stay in the positive territory.

We are moving forward with the marathon.
The goal remains 1,000 games (50 rounds) to see if this 53% stability holds or if the modern engines will eventually grind the veteran down.
________________________________________________________________________________

Tournament Standings after 120 Games (6 Rounds Completed)[/b]
Crosstable:

Code: Select all

----|------------------------------|------|-----|--------|-------|-------|
Rank|             Name             |Points|Games|  Elo   |  +/-  |Score% |
----|------------------------------|------|-----|--------|-------|-------|
  1 | Houdini 6.03 x64(BMI2-Rapid) | 64,0 | 120 | 2866,0 |  -4,0 | 53,3% |
  2 | PlentyChess 7.0.0 x64(BMI2)  |  4,0 |   6 | 2871,3 | +10,4 | 66,6% |
  3 | Obsidian 16.0 x64(BMI2)      |  3,5 |   6 | 2825,3 |  +9,3 | 58,3% |
  4 | RubiChess 20240817 x64(BMI2) |  3,5 |   6 | 2816,1 | +10,1 | 58,3% |
  5 | Stockfish 17 x64(AVX2)       |  3,5 |   6 | 2872,1 |  +5,1 | 58,3% |
  6 | Viridithas 19.0.1 x64(AVX2)  |  3,5 |   6 | 2789,3 | +12,4 | 58,3% |
  7 | Alexandria 9.0.0 x64(BMI2)   |  3,0 |   6 | 2846,1 |  +2,2 | 50,0% |
  8 | Berserk 13.0 x64(BMI2)       |  3,0 |   6 | 2801,1 |  +6,1 | 50,0% |
  9 | Cinder 0.3.1 x64(AVX2)       |  3,0 |   6 | 2766,1 |  +9,1 | 50,0% |
 10 | Hobbes 1.0.0 x64(AVX2)       |  3,0 |   6 | 2759,6 |  +9,6 | 50,0% |
 11 | Komodo Dragon 3.3 x64(AVX2)  |  3,0 |   6 | 2800,1 |  +6,2 | 50,0% |
 12 | Pawnocchio 1.9.2 x64(AVX2)   |  3,0 |   6 | 2801,1 |  +6,1 | 50,0% |
 13 | Reckless 0.9.0 x64(AVX2)     |  3,0 |   6 | 2879,1 |  -0,8 | 50,0% |
 14 | Starzix 6.0 x64(AVX2)        |  3,0 |   6 | 2785,5 |  +7,5 | 50,0% |
 15 | Caissa 1.25 x64(BMI2)        |  2,5 |   6 | 2794,5 |  +1,5 | 41,6% |
 16 | Clover 9.1 x64(AVX2)         |  2,5 |   6 | 2794,5 |  +1,5 | 41,6% |
 17 | Stormphrax 7.0.0 x64(BMI2)   |  2,5 |   6 | 2765,0 |  +4,0 | 41,6% |
 18 | Integral 7.0 x64(BMI2)       |  2,0 |   6 | 2756,3 |  -0,6 | 33,3% |
 19 | Tarnished 5.0 x64(AVX2)      |  2,0 |   6 | 2762,8 |  -1,1 | 33,3% |
 20 | Horsie 1.1 x64(AVX2)         |  1,5 |   6 | 2760,8 |  -6,1 | 25,0% |
 21 | Motor 0.9.0 x64(AVX2)        |  1,0 |   6 | 2763,3 | -11,6 | 16,6% |
----|------------------------------|------|-----|--------|-------|-------|

HEAD-TO-HEAD MATRIX
________________________________________________________________

Code: Select all

----|-------------------------------|-----|------|-----|-----|--------|
No: |              Opponent         | Wins|Losses|Draws|Games| Score% |
----|-------------------------------|-----|------|-----|-----|--------|
  1 | Reckless 0.9.0 x64(AVX2)      |   1 |   1  |   4 |   6 | 50,00% |
  2 | Stockfish 17 x64(AVX2)        |   0 |   1  |   5 |   6 | 41,67% |
  3 | PlentyChess 7.0.0 x64(BMI2)   |   0 |   2  |   4 |   6 | 33,33% |
  4 | Alexandria 9.0.0 x64(BMI2)    |   0 |   0  |   6 |   6 | 50,00% |
  5 | Obsidian 16.0 x64(BMI2)       |   0 |   1  |   5 |   6 | 41,67% |
  6 | RubiChess 20240817 x64(BMI2)  |   0 |   1  |   5 |   6 | 41,67% |
  7 | Berserk 13.0 x64(BMI2)        |   0 |   0  |   6 |   6 | 50,00% |
  8 | Pawnocchio 1.9.2 x64(AVX2)    |   1 |   1  |   4 |   6 | 50,00% |
  9 | Komodo Dragon 3.3 x64(AVX2)   |   1 |   1  |   4 |   6 | 50,00% |
 10 | Caissa 1.25 x64(BMI2)         |   1 |   0  |   5 |   6 | 58,33% |
 11 | Clover 9.1 x64(AVX2)          |   2 |   1  |   3 |   6 | 58,33% |
 12 | Starzix 6.0 x64(AVX2)         |   0 |   0  |   6 |   6 | 50,00% |
 13 | Viridithas 19.0.1 x64(AVX2)   |   0 |   1  |   5 |   6 | 41,67% |
 14 | Motor 0.9.0 x64(AVX2)         |   4 |   0  |   2 |   6 | 83,33% |
 15 | Horsie 1.1 x64(AVX2)          |   3 |   0  |   3 |   6 | 75,00% |
 16 | Tarnished 5.0 x64(AVX2)       |   2 |   0  |   4 |   6 | 66,67% |
 17 | Stormphrax 7.0.0 x64(BMI2)    |   1 |   0  |   5 |   6 | 58,33% |
 18 | Cinder 0.3.1 x64(AVX2)        |   0 |   0  |   6 |   6 | 50,00% |
 19 | Integral 7.0 x64(BMI2)        |   2 |   0  |   4 |   6 | 66,67% |
 20 | Hobbes 1.0.0 x64(AVX2)        |   1 |   1  |   4 |   6 | 50,00% |
----|-------------------------------|-----|------|-----|-----|--------|

Note: Scores are given from Houdini’s perspective against each opponent.
________________________________________________________________________________
Status: RUNNING: 120/1.000 games played.
White Wins 15(12,50%) | Black Wins 15(12,50%) | Draws 90(75,00%)
Download Pgn Games(Zip)
________________________________________________________________________________

Why Houdini 6.03? The Philosophical Benchmark

I continue to use Houdini as my primary benchmark, especially given the historical controversies surrounding its source code.
My reasoning is not based on sentiment, but on objective historical context.

1. The Golden Standard of the Pre-NNUE Era:
For years, Houdini represented the pinnacle of "magical" chess evaluation.
Even when Stockfish existed as a rival, Houdini’s tactical sharpness and unique playing style captivated the chess world.
It was the yardstick by which all progress was measured before the neural network revolution.

2. Measuring the Evolutionary Leap:
By pitting Houdini against modern NNUE giants with a 100:1 time advantage, I am not just running a tournament;
I am documenting the giant leap in chess evolution. Using Houdini as a fixed reference point allows us to visualize
exactly how much the "Hand-Coded" logic has been surpassed by the "Machine Learning" era.

3. The Ultimate Stress Test:
If a modern engine can dismantle a former "World Champion" logic like Houdini’s while having only 9 seconds on the clock, it
provides a brutal and undeniable proof of the efficiency gains in modern search and evaluation.

I use Houdini because to understand how far we have come, we must measure ourselves against the very best of where we used to be.
The Boss (as I like to call this veteran) serves as the perfect bridge between the classical past and the neural future.
________________________________________________________________________________

[goni-K26]

________________________________________________________________________________

Tournament Standings after 240 Games (12 Rounds Completed)

Code: Select all

----|------------------------------|-------|-----|--------|-------|-------|
Rank|             Name             |Points |Games|  Elo   |  +/-  |Score% |
----|------------------------------|-------|-----|--------|-------|-------|
  1 | Houdini 6.03 x64(BMI2-Rapid) | 126,0 | 240 | 2861,3 |  -8,7 | 52,5% |
  2 | Obsidian 16.0 x64(BMI2)      |   7,0 |  12 | 2833,8 | +17,9 | 58,3% |
  3 | Pawnocchio 1.9.2 x64(AVX2)   |   6,5 |  12 | 2811,5 | +16,5 | 54,1% |
  4 | PlentyChess 7.0.0 x64(BMI2)  |   6,5 |  12 | 2867,1 |  +6,1 | 54,1% |
  5 | Reckless 0.9.0 x64(AVX2)     |   6,5 |  12 | 2883,0 |  +3,0 | 54,1% |
  6 | Starzix 6.0 x64(AVX2)        |   6,5 |  12 | 2797,1 | +19,1 | 54,1% |
  7 | Stockfish 17 x64(AVX2)       |   6,5 |  12 | 2872,1 |  +5,1 | 54,1% |
  8 | Viridithas 19.0.1 x64(AVX2)  |   6,5 |  12 | 2796,1 | +19,2 | 54,1% |
  9 | Alexandria 9.0.0 x64(BMI2)   |   6,0 |  12 | 2848,1 |  +4,2 | 50,0% |
 10 | Berserk 13.0 x64(BMI2)       |   6,0 |  12 | 2806,8 | +11,9 | 50,0% |
 11 | Komodo Dragon 3.3 x64(AVX2)  |   6,0 |  12 | 2806,0 | +12,0 | 50,0% |
 12 | RubiChess 20240817 x64(BMI2) |   6,0 |  12 | 2816,1 | +10,2 | 50,0% |
 13 | Stormphrax 7.0.0 x64(BMI2)   |   6,0 |  12 | 2778,0 | +17,0 | 50,0% |
 14 | Cinder 0.3.1 x64(AVX2)       |   5,5 |  12 | 2769,8 | +12,9 | 45,8% |
 15 | Clover 9.1 x64(AVX2)         |   5,5 |  12 | 2800,5 |  +7,5 | 45,8% |
 16 | Hobbes 1.0.0 x64(AVX2)       |   5,5 |  12 | 2763,8 | +13,9 | 45,8% |
 17 | Tarnished 5.0 x64(AVX2)      |   5,5 |  12 | 2775,8 | +11,9 | 45,8% |
 18 | Caissa 1.25 x64(BMI2)        |   5,0 |  12 | 2795,8 |  +2,9 | 41,6% |
 19 | Horsie 1.1 x64(AVX2)         |   4,0 |  12 | 2764,5 |  -2,5 | 33,3% |
 20 | Integral 7.0 x64(BMI2)       |   4,0 |  12 | 2755,8 |  -1,1 | 33,3% |
 21 | Motor 0.9.0 x64(AVX2)        |   3,0 |  12 | 2762,0 | -13,0 | 25,0% |
----|------------------------------|-------|-----|--------|-------|-------|

HEAD-TO-HEAD MATRIX
________________________________________________________________

Code: Select all

----|-------------------------------|-----|------|-----|-----|---------|
No: |              Opponent         | Wins|Losses|Draws|Games| Score%  |
----|-------------------------------|-----|------|-----|-----|---------|
  1 | Reckless 0.9.0 x64(AVX2)      |   1 |   2  |   9 |  12 |  45,83% |
  2 | Stockfish 17 x64(AVX2)        |   1 |   2  |   9 |  12 |  45,83% |
  3 | PlentyChess 7.0.0 x64(BMI2)   |   1 |   2  |   9 |  12 |  45,83% |
  4 | Alexandria 9.0.0 x64(BMI2)    |   0 |   0  |  12 |  12 |  50,00% |
  5 | Obsidian 16.0 x64(BMI2)       |   0 |   2  |  10 |  12 |  41,67% |
  6 | RubiChess 20240817 x64(BMI2)  |   2 |   2  |   8 |  12 |  50,00% |
  7 | Berserk 13.0 x64(BMI2)        |   0 |   0  |  12 |  12 |  50,00% |
  8 | Pawnocchio 1.9.2 x64(AVX2)    |   2 |   3  |   7 |  12 |  45,83% |
  9 | Komodo Dragon 3.3 x64(AVX2)   |   2 |   2  |   8 |  12 |  50,00% |
 10 | Caissa 1.25 x64(BMI2)         |   2 |   0  |  10 |  12 |  58,33% |
 11 | Clover 9.1 x64(AVX2)          |   3 |   2  |   7 |  12 |  54,17% |
 12 | Starzix 6.0 x64(AVX2)         |   1 |   2  |   9 |  12 |  45,83% |
 13 | Viridithas 19.0.1 x64(AVX2)   |   1 |   2  |   9 |  12 |  45,83% |
 14 | Motor 0.9.0 x64(AVX2)         |   6 |   0  |   6 |  12 |  75,00% |
 15 | Horsie 1.1 x64(AVX2)          |   4 |   0  |   8 |  12 |  66,67% |
 16 | Tarnished 5.0 x64(AVX2)       |   2 |   1  |   9 |  12 |  54,17% |
 17 | Stormphrax 7.0.0 x64(BMI2)    |   1 |   1  |  10 |  12 |  50,00% |
 18 | Cinder 0.3.1 x64(AVX2)        |   1 |   0  |  11 |  12 |  54,17% |
 19 | Integral 7.0 x64(BMI2)        |   4 |   0  |   8 |  12 |  66,67% |
 20 | Hobbes 1.0.0 x64(AVX2)        |   3 |   2  |   7 |  12 |  54,17% |
----|-------------------------------|-----|------|-----|-----|---------|

Note: Scores are given from Houdini's perspective against each opponent.
________________________________________________________________________________
Status: RUNNING: 240/1.000 games played.
White Wins 30(12,50%) | Black Wins 32(13,33%) | Draws 178(74,17%)
Download Pgn Games(Zip)

________________________________________________________________________________
The Reality Check:
____________________________________
In Game 10, Stockfish 17 peaked at a +2.50 advantage, yet it completely lost its way in the tactical fog. This proves that while NNUE provides a superior 'positional intuition,' it can become a liability without the search depth to back it up. Houdini didn't just survive the pressure; it outcalculated the 'Boar' until the evaluation hit a staggering +148.87. It's a classic case of raw calculation speed (Houdini) vs. neural depth (Stockfish) in a time-starved environment.________________________________________________________________________________
[pgn]
[Event "Evolution Chronicles - Houdini vs. The NNUE Era"]
[Site "Intel(R) Core(TM) i5-4570 3.20GHz 8GB RAM"]
[Date "2026.05.03"]
[Round "10"]
[White "Stockfish 17 x64(AVX2-5Pct)"]
[Black "Houdini 6.03 x64(BMI2-Rapid)"]
[Result "0-1"]
[BlackElo "2870"]
[ECO "?"]
[Opening "?"]
[Time "00:09:18"]
[WhiteElo "2867"]
[TimeControl "900+10"]
[Termination "adjudication"]
[PlyCount "143"]

1. Nf3 e6 {-0.17/24 45} 2. e4 {+0.27/19 0} d5 {-0.08/23 15} 3. e5 {+0.31/16
0} c5 {-0.04/23 14} 4. c3 {+0.22/20 0} Nc6 {+0.01/23 33} 5. d4 {+0.33/17 0}
Nge7 {-0.13/25 48} 6. Na3 {+0.33/21 0} Bd7 {-0.07/22 25} 7. Nc2 {+0.47/17
0} Qc7 {-0.10/24 52} 8. Be2 {+0.51/17 0} Rc8 {-0.05/23 21} 9. Bd3 {+0.58/16
0} c4 {0.00/22 26} 10. Be2 {+0.49/21 0} Nf5 {-0.33/26 54} 11. g4 {+1.21/18
0} Nfe7 {-0.38/24 23} 12. Nh4 {+1.22/19 0} Ng6 {-0.42/25 110} 13. Ng2
{+1.20/16 0} Be7 {-0.36/24 20} 14. h4 {+1.48/17 0} h6 {-0.33/26 22} 15. f4
{+1.30/18 0} b5 {-0.32/22 36} 16. a3 {+1.22/17 0} Qd8 {-0.23/23 18} 17. g5
{+1.45/16 0} Na5 {-0.38/24 40} 18. Rb1 {+1.48/19 0} hxg5 {-0.36/24 29} 19.
fxg5 {+1.83/15 0} Nb3 {-0.28/26 28} 20. Be3 {+1.93/17 0} Bc6 {-0.48/26 42}
21. Kf2 {+1.89/17 0} Kd7 {-0.46/24 22} 22. Bh5 {+1.84/19 0} Qe8 {-0.69/25
49} 23. Kg3 {+1.99/19 0} Nf8 {-0.53/23 26} 24. Rf1 {+2.43/19 0} g6
{-0.44/25 16} 25. Bg4 {+2.24/20 0} Rh7 {-0.63/28 57} 26. Bf4 {+2.21/18 0}
Kc7 {-0.22/25 12} 27. Qe1 {+2.39/20 0} Nd7 {-0.71/26 126} 28. Nce3
{+2.40/18 0} Qd8 {-0.49/23 14} 29. Bd1 {+2.37/16 0} a5 {-0.71/25 39} 30.
Ng4 {+2.35/16 0} Kb8 {-0.65/23 18} 31. N2e3 {+2.42/18 0} Kb7 {-0.82/22 27}
32. Qe2 {+2.35/19 0} Qb6 {-0.70/20 9} 33. Nh6 {+2.50/17 0} b4 {-0.50/21 9}
34. axb4 {+2.44/17 0} axb4 {-0.94/22 33} 35. Neg4 {+2.48/18 0} Rg7
{-1.02/22 34} 36. Rf3 {+2.38/18 0} Qa5 {-1.11/22 14} 37. Be3 {+2.40/12 0}
Qa2 {-1.01/21 8} 38. Bc2 {+2.06/17 0} Nb6 {-1.20/24 27} 39. Nxf7 {+2.30/15
0} bxc3 {-0.93/19 5} 40. bxc3 {+1.64/14 0} Na4 {-0.37/22 6} 41. Rbf1
{+1.02/15 0} Nxc3 {-0.33/22 7} 42. Qf2 {+1.21/13 0} Nb5 {-0.87/23 26} 43.
Nd6+ {+1.12/12 0} Bxd6 {-0.54/21 4} 44. exd6 {+1.07/13 0} Nxd6 {-0.13/21 5}
45. Rf6 {+1.09/13 0} c3 {-0.57/21 21} 46. Rb1 {+1.31/14 0} Ka7 {-0.55/20 6}
47. Kg2 {+0.44/15 0} Nd2 {-0.40/18 5} 48. Bd3 {+0.46/11 0} N6c4 {-0.52/19
18} 49. Re1 {+0.68/12 0} Qb2 {-0.19/17 4} 50. Re2 {+0.83/16 0} Ba4
{+0.10/18 5} 51. Ne5 {+0.32/14 0} Bd1 {-0.18/23 19} 52. Bxd2 {-0.16/14 0}
Bxe2 {+0.68/20 10} 53. Bxc3 {-0.18/12 0} Qxc3 {+0.69/20 0} 54. Bxe2
{-0.13/13 0} Ne3+ {+0.95/20 5} 55. Kh3 {-0.19/15 0} Nf5+ {+1.26/19 5} 56.
Bd3 {-0.08/14 0} Rh7 {+3.07/20 6} 57. Rxe6 {-0.01/15 0} Kb7 {+3.08/19 0}
58. Rxg6 {-1.93/19 0} Rxh4+ {+3.26/22 7} 59. Kg2 {-1.88/12 0} Rch8
{+3.99/22 6} 60. Rh6 {-3.10/16 0} Nxh6 {+5.07/23 6} 61. Qxh4 {-2.37/13 0}
Nf7 {+5.35/22 7} 62. Qg4 {-3.67/20 0} Nxe5 {+148.87/33 7} 63. dxe5
{-4.06/16 0} Qxd3 {+148.88/33 2} 64. Qb4+ {-4.72/18 0} Kc6 {+148.89/38 14}
65. Qd6+ {-5.56/16 0} Kb5 {+148.90/38 3} 66. Qd7+ {-5.84/13 0} Kb4
{+148.91/40 11} 67. Qb7+ {-7.32/16 0} Kc3 {+148.94/40 11} 68. Qc7+
{-9.56/15 0} Kd2 {+148.95/40 4} 69. Qa5+ {-7.56/18 0} Ke3 {+148.96/44 12}
70. Qc5+ {-199.84/24 0} Kf4 {+M7/21 0} 71. Qf2+ {-199.88/20 0} Kg4 {+M6/20}
72. Qf6 {-199.90/18 0 Arena Adjudication} 0-1
[/pgn]

[goni-K26]

Stockfish 17 outplayed a deeper-searching Houdini strategically, not just tactically
The attack (...Ng3, ...Kd7, ...Rh4) required positional judgment that raw speed alone shouldn't produce
Houdini had every time advantage and still couldn't unravel the position

The opening: 1. b3

[pgn][Event "Evolution Chronicles - Houdini vs. The NNUE Era"]
[Site "Intel(R) Core(TM) i5-4570 3.20GHz 8GB RAM"]
[Date "2026.05.05"]
[Round "15"]
[White "Houdini 6.03 x64(BMI2)"]
[Black "Stockfish 17 x64(AVX2)"]
[Result "0-1"]
[BlackElo "2867"]
[ECO "?"]
[Opening "?"]
[Time "16:52:10"]
[WhiteElo "2870"]
[TimeControl "900+10"]
[Termination "normal"]
[PlyCount "65"]

1. b3 e5 {+0.21/18 0} 2. Bb2 {+0.08/24 40} Nc6 {+0.18/17 0} 3. e3 {+0.05/22
4} d5 {+0.13/20 0} 4. Bb5 {-0.05/24 30} Bd6 {+0.22/20 0} 5. Nf3 {+0.14/23
14} f6 {+0.30/14 0} 6. O-O {+0.06/23 39} a6 {+0.31/21 0} 7. Bxc6+ {+0.25/22
18} bxc6 {+0.50/14 0} 8. c4 {+0.31/21 3} Be6 {+0.50/19 0} 9. c5 {+0.05/24
43} Be7 {+0.50/16 0} 10. Nc3 {-0.06/24 56} Qd7 {+0.42/14 0} 11. d4
{-0.07/25 39} e4 {+1.57/16 0} 12. Ne1 {-0.21/31 42} h5 {+1.59/15 0} 13. f3
{-0.29/25 57} f5 {+1.54/16 0} 14. f4 {-0.41/27 39} h4 {+1.59/17 0} 15. Qd2
{-0.40/30 26} Bf7 {+1.43/20 0} 16. Nd1 {-0.52/29 32} Bh5 {+1.57/19 0} 17.
Nf2 {-0.20/29 27} Nf6 {+1.56/15 0} 18. Nh3 {-0.20/29 26} Bg4 {+1.54/17 0}
19. Nf2 {-0.53/27 160} Nh5 {+1.74/21 0} 20. h3 {+0.72/25 12} Ng3 {+2.08/15
0} 21. hxg4 {+0.70/28 34} fxg4 {+1.91/17 0} 22. Nc2 {+0.61/28 30} g5
{+2.21/16 0} 23. Rfe1 {-0.13/24 39} Qf5 {+2.43/15 0} 24. Qd1 {+0.03/24 18}
h3 {+2.76/15 0} 25. gxh3 {-0.19/28 35} Rh4 {+2.77/16 0} 26. Bc1 {-2.48/25
54} Kd7 {+3.23/15 0} 27. Nb4 {-5.62/26 89} gxf4 {+3.79/19 0} 28. exf4
{-5.68/26 18} Rg8 {+4.70/20 0} 29. Nbd3 {-9.85/26 87} Bf6 {+5.09/16 0} 30.
Be3 {-12.62/24 47} exd3 {+6.38/19 0} 31. Qxd3 {-14.57/24 19} Qh5 {+6.50/17
0} 32. Bd2 {-19.18/20 16} gxh3 {+7.34/18 0} 33. Kh2 {-16.28/24 5 White
resigns} 0-1[/pgn]