Hashcat benchmark comparator

[whoisroot]

Ok ok, I ran the short benchmarks, 5 runs of CUDA vs OpenCL, but I had a few slow-downs due to memory usage because I was running them while GNOME was active (no other programs open). These slow-downs affected both the CUDA and OpenCL benchmarks, but this test was not very "scientific" and not done in a "controlled environment" at all, so take them with a grain of salt...

Here are the most significant (>25%) differences:

Code:

{
  "CUDA": {
    "3200": {
      "name": "bcrypt $2*$, Blowfish (Unix)",
      "winner": "hashcat-7.1.2_CUDA_short/hashcat-7.1.2_CUDA_short.json",
      "diff": 0.5147074047235245,
      "raw_diff": 6960.799999999999,
      "pretty_winner": "20.48 kH/s",
      "pretty_diff": "6.96 kH/s",
      "percent": "51.47%"
    }
  },
  "OpenCL": {
    "1500": {
      "name": "descrypt, DES (Unix), Traditional DES",
      "winner": "hashcat-7.1.2_OpenCL_short/hashcat-7.1.2_OpenCL_short.json",
      "diff": 0.41465526166638833,
      "raw_diff": 212014724.40000004,
      "pretty_winner": "723.32 MH/s",
      "pretty_diff": "212.01 MH/s",
      "percent": "41.47%"
    },
    "1800": {
      "name": "sha512crypt $6$, SHA512 (Unix)",
      "winner": "hashcat-7.1.2_OpenCL_short/hashcat-7.1.2_OpenCL_short.json",
      "diff": 0.29825885050210843,
      "raw_diff": 20398.399999999994,
      "pretty_winner": "88.79 kH/s",
      "pretty_diff": "20.40 kH/s",
      "percent": "29.83%"
    },
    "17200": {
      "name": "PKZIP (Compressed)",
      "winner": "hashcat-7.1.2_OpenCL_short/hashcat-7.1.2_OpenCL_short.json",
      "diff": 0.49852851918197083,
      "raw_diff": 563290431.0,
      "pretty_winner": "1.69 GH/s",
      "pretty_diff": "563.29 MH/s",
      "percent": "49.85%"
    },
    "15700": {
      "name": "Ethereum Wallet, SCRYPT",
      "winner": "hashcat-7.1.2_OpenCL_short/hashcat-7.1.2_OpenCL_short.json",
      "diff": 0.2857142857142858,
      "raw_diff": 0.40000000000000013,
      "pretty_winner": "1.80 H/s",
      "pretty_diff": "0.40 H/s",
      "percent": "28.57%"
    }
  }
}

From the 107 hash modes ran, only these 5 were "significant" enough.

I hope someone with a better setup run the full benchmark later. I'll leave that to the community, hahaha.

The full results of the short benchmark are available here:
https://gist.github.com/whoisroot/c5db3d...0628d78e78