I'd like to point out that MGSV does have an enemy attention indicator when an enemy has seen you but not clearly enough to cause an alert. It's similar to the radial meter in the recent Splinter Cell and Far Cry games, but it only appears once an enemy's attention has been drawn, instead of warning you before attention is drawn like those games do. Changing this to be more like those games would require that AI become less reactive (so that there's time for the player to correct themselves before the AI reacts at all), which is something the designers seem staunchly opposed to.
This doesn't address your main point about the need for a preventative solution though. Personally, I think that an advanced version of MGS4's Threat Ring would be a great solution. However, the Threat Ring still doesn't answer the specific question of "can this enemy perceive me?" I'm not sure if having that question be very clear cut, like it was in MGS before 3, is ideal. MGSV is a game of estimation to some extent, and messing with that in any way would change it pretty drastically. Every stealth game where enemy perception fields aren't represented exactly has some amount of this estimation, and I think the estimation itself is a powerful gameplay mechanic.
To address your specific desire for something like Splinter Cell's visibility indicator, visibility in MGSV isn't binary, so such an indicator would be imperfect most of the time just like it is in Splinter Cell, but a "perfectly hidden" state while in certain cover (shadows, foliage) might work with some Splinter Cell light flare-esque indicator. I think if you know the enemy's state, comparing it with your own pretty much always gives you enough information to get a good estimate of visibility. That's why I think a revised Threat Ring could be all MGSV needs.
MGSV actually provides an indirect way of knowing the limitations of enemy perception fields as well, though only hardcore players will make use of it since it involves pure trial and error: the distance meter that appears when an enemy is marked can be used to figure out at what distance an enemy percieves you under various conditions. For example, through experimentation I found that the distance enemies in Ground Zeroes can hear you sprinting from is 15 meters. One can then use this information to more precisely estimate whether they're skirting enemy perception fields correctly. Still, in the end, I agree that MGSV doesn't have enough, and I'm sad that it appears we'll have to make due with trial and error to a certain extent ultimately.
On regenerative versus non-regenerative health, I think the Far Cry games found a really good balance that MGSV could learn from, if not just copy. The health meter is segmented, and only segments that are not empty when regeneration activates will regenerate. Otherwise consumables must be used. If you have no consumables you must manually heal, which leaves you vulnerable. MGSV's wounded state is similar to that last bit, but it feels extremely arbitrary. There's feedback that I know of to let you know you're at risk of a serious injury, and I've had times where they've seemingly come out of nowhere. It exists to force you to go defensive, but it feels random. It's definitely not ideal, and it's a band-aid on a system that isn't very functional in general.
The issue of the transformation into more of a shooter as a result of more lenient options is extremely complex. We don't know how harshly TPP will punish failed stealth, but we probably agree that it likely won't be too harsh for most of the game. Finding the proper balance is a very serious game design task.
How about this. I'm going to try to express my personal perspective here. I think that Ground Zeroes functions very well as a pure stealth game with some exceptions (some of them being really significant oversights in my view, but they're things that much of the userbase will probably never take note of), and I actively want to play it as one, so the loosening of the general structure doesn't really apply to me. It literally doesn't matter to me that non-stealth options have expanded because I've fully bought into the idea of this being a stealth game and am willing to self-limit to give myself the best experience I think I can get. I understand that this can be a tough mindset to gel with, but if you're willing and able to discuss mechanics at this level adopting it might help you. Don't think of the experience the total userbase gets from the game, focus on making your own experience the best it can be by playing to what you see as the strengths of the game mechanically. I do this very often; it's rare that a game has accessible mechanics, enough depth to be satisfying in the long term, and no ways to be broken all at the same time when it has a certain base level of complexity. Stealth games have always struggled mightily with the first of those three things, so compromises have been made. MGSV is, I think we agree, boring as a shooter. Luckily, I think it's quite interesting as a stealth game, so I'm committed to playing it that way. Kojima Productions isn't trying to make the platonic ideal of a stealth game in the end, and that's something we have to accept, but what they are making is very good in general and incredible with the right mindset despite some flaws.
The natural argument against this philosophy is that allowing non-stealth options inherently weakens MGSV as a stealth game. I feel that this is not the case, but if anyone wants to argue that it is I'd love to see a detailed argument. This isn't Splinter Cell where level design (Blacklist) and pretty much an entire game (Conviction) were sacrificed to create shooting oriented gameplay, it seems safe to assume that outside of the obligatory rail shooting sequences TPP will be fully functional as a stealth game. I also think its core mechanics of movement, enemy AI and stealth oriented player-AI interactions stand well above those games, but that may or may not be evidence that designing a more shooting oriented game isn't necessarily harmful depending on how those mechanics were conceived. The real truth can only be known by the actual designers, and even then, none of them actually know absolutely everything about all of the game, so the debate is completely open.
Oh, and I absolutely agree about instant large scale alerts. They suck. Lots. In an absolute purist playstyle they don't matter, but come on.
Fuck, I wrote way too much. I hope this is coherent.
), I just don't think the basic, nitty-gritty stealth mechanics are up to snuff. And, honestly, it wouldn't take huge, sweeping changes to turn MGSV into the kind of stealth game I'd like. A simple indicator of when you're hidden would be easy to implement. It doesn't have to be an obtrusive HUD element like the camo index; think of Sam Fisher's NVGs lighting up to show that he can't be seen. Enemies having to physically call in alerts on their radios is an obvious one, same with the return of Caution and Clearing phases. Even Reflex Mode isn't beyond fixing. I think that could be a totally respectable mechanic if, say, it only worked for enemies you'd already tagged, and it simply gave you a rough indication of where the dude was rather than snapping your crosshairs onto him.
