Testing And Certification Of Online Slots

Have you ever asked yourself how online slots are tested prior to release? We’ve had a look at 2 certification reports issued by an acclaimed testing agency

Before being launched and made available to the public, all online slots must be tested by independent labs to ensure that the compliance standards and regulatory requirements are upheld. This means that every slot needs to be individually certified for each and every regulator (i.e. MGA, UKGC, Gibraltar, Latvia and Spillemyndigheden in Denmark). There are several testing agencies available for this purpose, with the most well-known being eCOGRA, iTech Labs, TST, GLI, and NMi.

Curacao, great for vacations, not so great for online slots enthusiasts.

Unfortunately, many online casinos and software developers take advantage of off-shore regulators as a way around the high requirements set by the European regulators and in an apparent attempt to increase revenue. The most common of these is the Curacao eGaming License. It is worthless, holds no value in terms of player safety, and is nothing more than a dirty alliance between the country of Curacao and the European iGaming industry.

If you play online slots, it’s very likely you’ve come across the “You are now leaving Maltese Jurisdiction” disclaimer. In theory, this means that the RTP could be anything from ZERO and upwards and they don’t even have to tell you which one it is. Our advice – stay away from sites promoting Curacao casinos. They have no interest in your safety. On Bigwinboard.com we only promote casinos with real European licenses.

Since these reports are confidential, we’re not going to publish the titles of the online slots tested, nor the name of the testing agency. It’s not of significant interest either way. What’s important is how these tests are conducted and what the testing agency is looking for when carrying out their tests.

The Test Report

Testing of a physical land based slot machine

In most reports we’ve looked at, two types of files have been tested – a config file with an .xml extension and an RNG file with a .jar extension. Along with the file name, there’s also notes about version number and MD5 Hash Sum. Hash sums are unique identifiers and is used to ensure that the original files are the same as those used during testing. The config file contains all the game settings, while the RNG file contains the very heart of the game – the Random Number Generator (RNG). The RNG is the mathematical formula that randomly decides the outcome of a game.

In one of the reports that we got our hands on, statistical and mathematical tests were performed on a 1 000 000 rows of sample data (provided by the software developer) of the game output to verify the RNG operates in a random and fair manner. This included tests for probability (to ensure the expected number of occurrences), randomness (so that the following occurrence cannot be predicted with any degree of certainty) and uniformity (to determine that each possible outcome is equally likely over the long-term). Looking at the report, it was interesting to see the RTP having been recalculated from 97.1% as reported by the software developer, to 96.7% as calculated by the testing agency.

The evaluation is performed to ensure the following requirements are met:

  • The game output must be randomly generated
  • The game output must be unpredictable
  • The series cannot be reproduced

The test suite used to perform the evaluation consists of the following:

What They Look For In Detail

Some sections of the test may be skipped depending on the “physics” of a particular game. For example, if physical devices are not stimulated in the game, or if there is no perception created that the game outcome is influenced by either speed or skill, there’s no need to test for such factors.

These are some of the technical and operating standards tested for compliance by the compliance agency:

  • Games should be implemented and operate fairly and strictly in accordance with the published rules and prevailing RTP where applicable.
  • Games should not falsely display near-miss results, that is, the event may not substitute one random losing outcome with a different losing outcome.
  • If artwork contains game instructions specifying a maximum win, then it should be possible to win this amount from a single game (including features or other game options).
  • Casino games should operate at a speed that allows the player to establish the result of each cycle of the game. Players may be permitted to accelerate the speed of a game where they are still able to establish the result of each cycle.
  • Games should not be adaptive or compensated, i.e. the probability of any particular outcome occurring should be the same every time the game is played, except as provided for in the rules of the game.
  • The customer should not be forced to play a game simply by selecting it.
  • A mechanism should be implemented to prevent repeated gamble instructions, (for example, where a customer repeatedly presses play while waiting for a game result to be executed.
  • The functions of all buttons represented on the website and games should be clearly indicated.
  • Edges of the “hot” area of buttons should be clearly defined in the artwork to prevent clicking near buttons creating a gamble.
  • The output obtained through the use of the RNG shall be proven to: (a) be statistically independent. (b) be uniformly distributed over their range. (c) pass various recognized statistical tests intended to demonstrate (a) and (b) above and the absence of patterns. (d) be unpredictable without knowledge of the algorithm, its implementation, and the current seed value.

DMCA.com Protection Status © Bigwinboard.com, 2019. Unauthorized use and/or duplication of this material without express and written permission from this site’s author and/or owner is strictly prohibited and may result in a Google DMCA (Digital Millennium Copyright Act) takedown and/or legal action.

Leave a Reply

1 Comment threads
1 Thread replies
Most reacted comment
Hottest comment thread
2 Comment authors
AdminCasinomon Recent comment authors
newest oldest most voted
Notify of

Great article! Always wanted to take a closer look at this myself. Really surprised about the following though: “statistical and mathematical tests were performed on a 1 000 000 rows of sample data (provided by the software developer)” ; provided by the SOFTWARE DEVELOPER?

I feel that for many games, a serie of 1.000.000 spins can result in varying outcome; what prevents the software developer from simply providing the 1.000.000 series of spins with highest RTP? Tests should IMO be conducted over 10.000.000 spins as a rule of thumb