Again, while I think the whole machine-learning thing sounds really fucking cool, I'm still really doubtful about the whole thing, and I want to raise certain key areas which others may not feel like expressing due to the overwhelming number of survey participants, but lack of forum replies. And just to add this disclaimer: there is no hate or spite here from me, more just confusion over anything else and seeking answers.
mastercomsHello everyone! After having a great discussion with the community about mastercomfig+
Where has this discussion taken place, or was that sarcasm? Only myself and Zetos have replied to this thread, both expressing uncertainty in the idea, plus your survey of 106 responses at the time of this post are negative. 68.9% of people said they wouldn't be open to subscribing to mastercomfig+, not to mention the16% maybe and 2.8% don't know. That's 87.7% of 106 people who are against/uncertain to the idea.
Another issue I'd like to raise is that dummies like myself still don't really understand the whole idea, so along with the survey results, I don't think you've really sold the community on what you're trying to achieve here. What is machine learning? How does it work? How exactly will that be used here? What sort of performance impact will it have on my game, if it is gathering data specific to my system? What sort of performance boost would I be expecting to obtain from all of this? Would this all be included in the .exe file, or will that be a separate download? What sort of data is being collected? Have I misunderstood the idea completely? I still have many more questions, and I only have one or two very short sentences by yourself to go by. And as you're the salesperson, I can't say I'm sold yet.
What exactly will the difference be between the paid version and the free version? If we set aside cvars which are mostly personal preference (decals, props, gibs, ragdolls, ropes, effects, sound, lighting, shadows, post processing, character visuals etc etc), what exactly will mastercomfig+ be suggesting to improve performance that we don't already know thanks to your efforts so far?
If the number of people subscribing is lower than you originally thought, how long will you maintain the project for?
Finally I'll be blunt - I have no intention of using mastercomfig+, but I appreciate your efforts and enthusiasm for this 10 year old game. Nevertheless, will I be able to donate a single lump sum of money your way, instead of seeing -$2.70's on my bank statement each month?