Experiment Log 384
rating: 0+x

Experiment Log SCP-384

Approved by O5-█
Monitored by O5-█, O5-██, O5-█
Project Head: Dr. ████████

Note: This is NOT a comprehensive list, only select declassified tests are shown. Notably, the method and experimentation used for producing SCP 384 Usage Guide will remain classified under a need-to-know basis (Protocol NTN-1) for release only.

-O5-█


All researchers working with SCP-384 are encouraged to append their results to this experiment log in the following format:

Date: ██/██/████
Experiment Designation: SCP384-TX (where X is test number)
Experiment:

Procedure:

Results:

Notes:



Date: ██/██/████
Experiment Designation: SCP384-T1
Experiment:
To determine if SCP-384 can correctly determine integer sequences. Sequences used are from the OEIS website (http://www.research.att.com/~njas/sequences/Seis.html)

Procedure:
Following the Usage Guide, each integer was entered on a single line, 2 per screen. A maximum of 10 numbers were used per sequence. Calculation performed in "forward mode". Machine will be turned off and then on again to clear results between tests.

Results:

  • Test 1: Natural Numbers/Seq. A000027
    • Input: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10
    • Output: 11
    • Observations:
      • Calculation time - 5 sec.
      • Normal sounds produced by Object
  • Test 2: Even Natural Numbers/Seq. A005843
    • Input: 2, 4, 6, 8, 10, 12, 14, 16, 18, 20
    • Output: 22
    • Observations:
      • Calculation time - 5 sec.
      • Normal sounds produced by Object

Notes:
As expected, the Object tested with 100% accuracy. Object makes small clicking and clockwork like sounds while performing calculations. It seems to take longer than would be expected, but further investigation may conclude otherwise.
Similar tests to be performed with more complex sequences. Additionally, after sequence testing, "random" testing will proceed.

Date: ██/██/████
Experiment Designation: SCP384-T7
Experiment:
Further testing of OEIS sequences. Prime and computational science sequences will be used.

Procedure:
Following the Usage Guide, each integer was entered on a single line, 2 per screen. A maximum of 10 numbers were used per sequence. Calculation performed in "forward mode". Machine will be turned off and then on again to clear results between tests.

Results:

  • Test 1: Primes p such that 2^p - 1 is prime/Seq. A000043
    • Input: 2, 3, 5, 7, 13, 17, 19, 31, 61, 89
    • Output: 107
    • Observations:
      • Calculation time - 7 sec.
      • Normal sounds produced by Object
  • Test 2: Number of series-parallel networks with n unlabeled edges/Seq. A000084
    • Input: 1, 2, 4, 10, 24, 66, 180, 522, 1532, 4624
    • Output: 14136
    • Observations:
      • Calculation time - 12 sec.
      • Normal sounds produced by Object

Notes:
During Test 1, the "Determine" button was pressed accidentally a second time, which produced more sounds and apparent calculation by the Object. The result outputted was 127. With this apparent discovery, it can be theorized that repeated presses of the "Determine" button will produce the next result in the sequence until the machine is reset or powered down.

Date: ██/██/████
Experiment Designation: SCP384-T39
Experiment:
Initial experimentation of "random" variable predictive output.

Procedure:
A standard dice cup is used with a single 20-sided precision gaming dice as random input. Cup is shaken by experimenter, flipped upside-down onto workbench, and dice value is recorded by separate researcher. Only 5 rolls per test will be used. Once all 5 values are recorded, they will be entered following Usage Guide procedures, one value per line. Calculation will be performed in "forward mode". Once output is displayed, researcher will once again roll dice. Separate researcher will compare results. Machine will be turned off and then on again to clear results between tests.

Results:

  • Test 1
    • Input Dice Rolls: 20, 14, 11, 9, 15
    • Output: 11
    • Final Dice Roll: 11
    • Observations:
      • Calculation time - 45 sec.
      • Normal sounds produced by Object
  • Test 2
    • Input Dice Rolls: 11, 9, 3, 1, 6
    • Output: 5
    • Final Dice Roll: 5
    • Observations:
      • Calculation time - 76 sec.
      • Normal sounds produced by Object
  • Test 3
    • [DATA EXPUNGED]

Notes:
[DATA EXPUNGED]

The following email is declassified in interest of research methodology. Additional email will be forthcoming. - O5-██

Subject: Progress so far as of ██/██/████
From: Dr. ████████ (████████@████████)

It appears the tests are going as expected. I had my doubts about SCP-384's supposed acuity in making predictions, but so far the results have been astounding! Computation tests, mechanical randomness, radioactive decay, and even lottery numbers have come up as 100% accurate in all cases!

There are misgivings and annoyances to this Object, in my opinion. Sometimes the calculation takes a significantly long time to achieve output. I have waited at hours, or even days for the calculation to output a value. At times the output comes just seconds before the random event occurs. This might explain why Agent ██████ found the Object right next to [DATA EXPUNGED] during his investigation of [DATA EXPUNGED]

[DATA EXPUNGED]

[DATA EXPUNGED]

We need to continue in the spirit of this research and find more ways of throwing oddball values at this Object. I propose we setup a procedure to determine "non-standard" variables of prediction. We can easily encode arbitrary integers to categorize certain "systems" for prediction. One thing I would like to try is to predict what the next body-count of Keter level SCPs may be, to plan for additional staffing. I would propose that we enter values of the latest death toll, one per line, per Keter SCP, and press that shiny Determine button to see what we might be against. It could provide useful to budgetary constraints.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License