Michael Paycer — SQL Server DBA and lifelong computer-chess tinkerer
Computer Chess · Disputes

Poke the Bear

Every other page in this cluster states things crisply and links here for the asterisk. This is where the hedging lives — the claims that aren't settled, the numbers that don't mean what people think, and the "facts" the whole field takes on faith. It's the most contrarian page in the set, and the most fun. If you've ever watched a vendor benchmark collapse under a second look, you already have the instinct this page runs on: trust, but verify.

Computer Chess Cluster
This is the Disputes page — part of Computer Chess, from a DBA's Chair. It's the caveat-sink for the whole cluster: History, Clone Wars, and Architecture all point here for their footnotes. Terms: Glossary.
Why This Page Exists

A pressure-release valve for the whole cluster

Here's the architectural honesty. Every narrative page wants to say something clean — "Deep Blue was around 2850," "AlphaZero beat Stockfish," "Rybka copied Fruit." But each of those carries an asterisk, and if I hedged inline on every page the prose would turn to mush. So the other pages state things crisply and link here for the caveat. This page absorbs all the "well, actually," which is exactly what makes it the most citable, most expertise-signaling page in the set. Each section below is a clean split: the popular claim, then the messier truth.

The Claims

Popular claim vs. messy truth

The messy truthDeep Blue never had an official rating. It played almost no rated games under standard conditions, so any number is a band-width guess. Worse, it's permanently unretestable: IBM dismantled the machine after the 1997 match, so it can never be measured. Add Kasparov's accusation of human intervention (the strangely un-machine-like moves — most famously in Game 2 — that he never got explained during the match; IBM denied intervention and only published logs later, and the odd Game 1 move was eventually attributed to a bug), and Deep Blue's true strength isn't just unknown — the opacity is permanent by design.
The messy truthThe 2017 match ran under conditions many developers called unfair to Stockfish: an old Stockfish version, only 1 GB of hash, a fixed time per move, no opening book, no tablebases, and a TPU-vs-CPU mismatch. Tord Romstad (a Stockfish author) objected publicly. A 2018 rematch under fairer conditions — proper hash and tablebases, 1,000 games — was more legitimate, and AlphaZero still won convincingly (+155 −6 =839). But AlphaZero was never released, so the result is unreproducible. Impressive, real, and permanently un-rerunnable.
The messy truthThe ICGA did rule that way, and banned Rajlich. But even many who accept the verdict think the process was procedurally messy, and the distinction between "inspiration" and "line-by-line copying" was never cleanly drawn — by 2011 a lot of the cited ideas had diffused across the whole field. Officially ruled against; never fully resolved in the community's mind. (Full story: The Clone Wars.)
The messy truthProbably — Rajlich himself called the derivation "obvious." But Rybka was itself ruled to have derived from Fruit. So Strelka is a clone of a clone, and the honest question the whole saga raises is: where does derivation actually begin? At some point "borrowed" and "learned from" blur, and the field has never drawn that line to everyone's satisfaction.
The messy truthNot the same engine at all. The "Fritz" brand rode different engines over the years — Frans Morsch's engine, then Pandix, then a Rybka-based version, then later bases. So "Fritz beat Kramnik" and "Fritz 21, 2026" share a label, not a lineage. Naming is not lineage, and brand continuity quietly hides engine discontinuity.
The messy truth (the thesis of this page)This comparison is close to meaningless. Ratings moved with hardware, time controls, tablebases, opening books, and decades of list inflation — the yardstick itself changed length. Comparing a 2026 CCRL number to a 1997 guess isn't measuring the same thing twice; it's comparing two different rulers. Cross-era engine Elo is barely real as a strength claim, and the marquee example below proves it with receipts.
The messy truthThe Ippolit/RobboLito family Houdini drew on was anonymous, and the belief that it was Rybka-derived is widespread but never legally proven. Houdini's author acknowledged influence, not copying. Widely believed is not the same as established — a distinction this whole page insists on.
The Marquee Example

One engine, one rating body, a 130-point swing

If you want the single cleanest proof that engine Elo is soft, it's this — and it's fully receipted. The same engine, on the same rating body (CCRL), takes four different "current" values depending only on which build, index, and snapshot you cite:

Rybka's "current" CCRL rating — same body, different knobs
Engine / buildCCRL list & dateRating
Rybka 4, 64-bit 4CPU40/15 "best versions only" · Jan 8 20263186
Rybka 4, 32-bit40/15 complete list · Feb 28 20263057
Rybka 3, 64-bit 4CPU40/15 complete list · Feb 28 20263152
Rybka 3 Human, 64-bit 4CPU40/15 complete list · Feb 28 20263142

Roughly 130 points of spread on "the Rybka 4 number" alone — from build, index, and measurement date, with no change to the engine's actual play. That's not a rounding error; it's the whole point.

CCRL is recomputed continuously, so every figure above is a dated snapshot, not a fixed fact — by mid-2026 the same builds had already drifted again (Rybka 4 64-bit 4CPU reading around 3139). That the "current" number won't hold still is the argument.

The DBA bridge · the metric everyone quotes isn't what they think

Same query, different server, build, cache state, or measurement day → a different "the number." Nobody sane trusts a benchmark without its configuration, and nobody should trust an engine Elo without its (build, list/index, snapshot date). "Rybka 4 = 3186" is a benchmark ripped from its config — the kind of claim every DBA has watched fall apart the moment someone asks "measured how, on what, when?" The publishing rule for this whole cluster follows from it: no engine rating ships without its receipt.

Frequently Asked Questions

The unsettled questions — FAQ

Did AlphaZero really beat Stockfish fairly?

The 2017 result came under conditions widely called unfair to Stockfish (old version, 1 GB hash, fixed move time, no book, no tablebases, TPU vs CPU). The fairer 2018 rematch — proper hash and tablebases, 1,000 games — AlphaZero still won (+155 −6 =839). But AlphaZero was never released, so it's unreproducible, which is why it stays partly contested.

Was Rybka really a clone?

The ICGA ruled in 2011 that it derived substantially from Fruit and banned its author. Many accept the finding but consider the process messy, and "inspiration vs. copying" was never cleanly resolved. Officially ruled against; never fully settled in the community.

Why do the same engine's ratings differ between lists?

A rating is a measurement, not a property. It depends on the build, the list and index, the hardware and time control, and the snapshot date. Rybka 4 shows about 3186 on one CCRL index and about 3057 on another — the same engine, ~130 points apart, from configuration alone.

How strong was Deep Blue, really?

Unknown, and unknowable. It had no official rating, played almost no standard rated games, and was dismantled after 1997, so it can never be retested. Any specific number is a guess.

Is comparing engines across eras meaningful at all?

Only loosely. Within a single rating list and era, relative numbers are informative. Across eras — different hardware, time controls, tablebases, books, and list inflation — the comparison is closer to storytelling than measurement.

Source Discipline

Every claim here traces to a real primary source

This is a page about not taking claims on faith, so it can't take shortcuts itself. The material here traces to primary sources — the ICGA report, DeepMind's AlphaZero paper, Stockfish and ChessBase's own legal posts, the CCRL lists, and the Chessprogramming Wiki — not to "as everyone knows." Where a claim is genuinely first-hand or genuinely disputed, it's labeled as such. That's the trust-but-verify loop this whole cluster is built on, running on itself.

  • Deep Blue — IBM Research history; contemporaneous 1997 match coverage; Kasparov's public statements.
  • AlphaZero vs Stockfish — DeepMind (Science, 2018); Stockfish developer commentary (Tord Romstad); the 2018 rematch results.
  • Rybka / ICGA — the ICGA panel report; Chessprogramming Wiki, "Rybka Controversy."
  • Rybka rating swing — CCRL 40/15 "best versions only" (Jan 8 2026) and complete list (Feb 28 2026), computerchess.org.uk; figures cross-checked by the author.
  • "Fritz" brand lineage — Chessprogramming Wiki (Fritz) and ChessBase release history.
Keep Exploring the Cluster

Back to the stories these asterisks belong to

Now that the caveats live here, the narrative pages get to state things cleanly. Read the ones that point here.

The Clone Wars →  ·  History →  ·  Back to the Hub →