On a scale from 1 to 10, what would you say the chances are that except you and Bram, you are not the only one having such an un-attended distributed system, doing real-time searching?
If there are 100 people searching, I would say (hope) at least 50%. so from 1 - 10, a 5 LOL. But I think most will limit the search to a single machine (which is fine, use what you have); but even then, they should have (at least) some type of python script to check off/track ranges searched - to ensure no duplicate work is happening.
But if they have multi machines in their home, or can rent a few, I would definitely incorporate some type of client/server scripts, even if only in Python. Having the ability for all machines to work together is a HUGE bonus. Even if one merely starts by using their local LAN resources and then scaling/opening it up to non LAN machines.
You think most have an unattended distributing system?
Respectfully, I wrote the full software from scratch which to this day is still 15% faster than anything ever mentioned here while 99.9% of the people in this thread are vanity search script kiddies.
Lets also mention the system handling the distribution of the workload on thousands of GPUs, which I also wrote from scratch.
We can also talk about how I gathered the funding for the puzzles by convincing people on the internet to trust me with hundred of thousands of dollars.
I executed all this flawlessly. And broke 2 puzzles months appart as a result, while the previous ones took years.
Let me know when you spot a higher level result around here.
I agree 100%. You can knock him for whatever, but you can't knock Bram for the results, whether you agree with the method or not...results speak for themselves. And I imagine, the more he does it, the more he learns from the trials and errors and will keep tweaking his entire ecosystem, to become more efficient, in some form or fashion.