When I was part of a DIR program, we were required to use standard gases for all dives. We did that diving in a place with no ability to get helium or oxygen nearby, so we had to bring our own, with no booster. We had to mix by partial pressure, topping off with air from the only local compressor. To cut to the end of the story, we vented a whole lot of helium into the atmosphere and then to outer space so we could get our tank pressures low enough to make standard mixes in our partially filled tanks. Today I don't have that constraint, and I do have a booster. Our group has undoubtedly saved many thousands of dollars and many thousands of cubic feet of helium this way.
The idea of standard gases was originally a good one for several reasons, and where it was invented, it saved money. Before there were apps for blending, you had to go through a mathematical process to calculate the three step process of making trimix. Standard gases were ever so much easier. You just went to the local fill station, had them put in the easy-to-calculate amount of helium, and then top it off from their banked 32% nitrox. It especially made switching from one standard gas to another for a dive at a different depth a breeze. You just topped off the helium by adding the correct amount to what you had in there already and finished off with banked EANx 32.
So, if you happened to be diving a few miles from a full-service fill station, standard gases were the way to go. For expedition filling, not so much.
It was also a time when helium was cheap. Very cheap. It was also a time when computers were awful, hence bottom timers and bio-computers following I-got-rithms. Kit configuration and technical techniques were still being developed, synchronised team diving cut back on deaths.
Nowadays with helium reaching ever higher prices and with little chance of a miracle discovery of massive reserves, it's got to be conserved which conflicts with standard gasses. Rebreathers are ever more common on technical dives (a simple 45m/150' dive yesterday had 12 on the boat with only one bubble blower on air, everyone else on CCR. The skipper having problems with his compressor to fill that person's twinset, whereas all the rebreather divers had plenty for the following day's dive).
Computers are now reliable enough to do what they're good at, computing. All hail Shearwater. Hell, take two as you're reliant upon it. I take three.
Diving in teams is still fine. Deco obligations can be balanced out as you go; just use the same gradient factors.
The DIR-wars have run their course. George Irving and his ilk would still whine like a 747 at LAX. Technical kit configuration and training is pretty standardised now, sure some agencies still plough their own furrows, but most are similar (think the entry level ANDP vs ART - basically the same now, similarly with normoxic trimix, "MOD"1 and 2 for CCR).
Back to mixing... Standard gasses were developed for, amongst other reasons, convenience of mixing. Who cared when helium was cheap that your mix was too helium rich. Now you have to be rich to afford it.
Optimise your gas for the dive you're doing. If someone's qualified or only interested in max 45m/150' diving, there's little point in getting an expensive standard mix (21/35) which is good for 57m/180', when 25/25 is a perfectly good gas. Or even use lower helium levels if you're happy with it (when I dived OC I'd be happy with double or triple dregs, even 10% helium was enough to wind the narcosis down, YMMV).
I've 4 deeper dives this week around the 60m/200' level (weather permitting). Using 18/45 on OC would be excruciatingly expensive in terms of cost, but mostly with logistics with multiple twinsets and goodness knows how many ali80s. This renders dives like this as out of the question on OC with a gas bill in excess of $1,000. On CCR this is just not an issue as I can mix it in my garage -- or even the back of my car -- without a booster/compressor and using the old twinsets for banking. Even the dive planning's trivial: set your overall runtime (2h30) then start the ascent when the runtime + TTS is 2h15 (add a bit of contingency) or when it reaches your max bailout plan. Then let the computers work out the deco whilst chilling out and dodging jellyfish.
