I wonder if it would make any sense, assuming regular oil changes, to drop the oilpan after X miles to clean it (does sludge build up in it?) and the oil pickup.
I’ve seen discussions regarding BMW motorcycles from the late '90s and early '2000s, which have a 6000-mile oil change interval specified in the manual. Several of the participants who had had samples of their used motorcycle oil tested by Blackstone reported that the oil could have gone 12,000-15,000 miles before being problematically contaminated and/or depleted of its additives. I would guess manufacturers of cars, trucks and other vehicles are similarly conservative with their oil change intervals.
Totally agree.
I’d also assume (and maybe this is part/all of what you meant) that the manufacturers’ experience tells them that:
RECOMMENDATION: 3 months/3,000 miles
TRANSLATES TO: 6-9 months/6-9,000 miles (or worse)
[In much the same way that a patient who tells the doctor they drink three drinks a day probably drinks 6+ drinks a day and every dentist’s patient flosses at least once every single day ;-)]
If I didn’t vary my driving too much over the course of a year or so – particularly with a lot of short trips and the car sitting a great deal, unused – I’d probably pay for the oil analysis once, and then assume I’m good until either my driving habits changed or I’d racked up maybe … 50k more miles on the car … at which point, I’d pay for the analysis again.
Some also suggest seeing if there’s a “severe service” maintenance schedule for your car and following that schedule, regardless.
But I’ll reiterate: everything other than oil analysis is basically just a guess 
Even with a test, though, wouldn’t you need several samples to understand how it wears, like a fresh one vs one at 3k, 6k, 12k miles, etc.? Or at least multiple data points from other people’s similar cars?
Or is there some generalized oil wear formula that they can apply to your sample at any mileage, and extrapolate from?
More data is always going to be better, but it’s a continuum – like so many other things. And finding the sweet spot really depends on the number of variables relevant to your vehicle, its ownership, and its use.
For more info on the subject than most of us would ever care to have …
ETA:
Given not a whole lot of variability in one’s driving style, mileage, the operating conditions, etc., these labs have the data on basically every kind of commercially available motor oil, so they can see the degree to which every variable has changed or degraded from new.
From there – given limited changes in operation of the vehicle – they can pretty accurately extrapolate out to the near future.
What is the mileage recommendation from the manufacturer? Because that is what I would follow.
As @Crafter_Man indicated above, the mileage interval is the more important factor than the time interval.
Also note that synthetic motor oil lasts considerably longer than conventional motor oil, typically twice as long between oil changes.
My son has a 2019 Subaru Forester that uses synthetic oil. The recommended oil change interval is 6,000 miles, so that’s what he follows.
I actually think that is a bit too frequent for synthetic oil because my 2015 Toyota 4Runner (which also has synthetic oil) has a recommended oil change frequency of 10,000 miles.
Nevertheless I think your best bet is to follow the manufacturer’s recommendation for mileage, and that is especially true if you are still in the warranty period. You don’t want them to deny a warranty claim by not following the recommended maintenance interval. It’s possible they could ding you for going beyond the recommended time interval, but I think that is unlikely. Assuming your car is being regularly driven, mileage is much more important.
ETA: I just saw this: are you saying that the recommended oil change interval is 3 yrs/30k miles? Because there is no way that is true. Per the manufacturer, most Subarus on the road have a recommended oil change frequency of 6 months/6,000 miles.
Sorry for the confusion/my lazy post.
They generally recommend maintenance - including an oil change, every year/10k miles. That is when I regularly realize my mileage is so short.
The oil change rec is 6 mos/6k. The time on the on the windshield sticker they apply alwasy comes well before the mileage.
I’m thinking I’ll go to every 12 mos - which should be approx 6k.
That’s what I would do, too.
Then the question is what does the curve of oil protection vs oil life left. For example an oil at 50% of its life has 50% of the protection. I know it is not that straightforward but using it as an example. ISTM that a lot of talk about oil changes assume (maybe unconsciously) that you get 100% protection for 10000 miles and after that no protection.
A plot of oil protection vs. oil life is almost certainly not linear though. My understanding is that the detergents and additives in motor oil are intended to suspend contaminants and wear particles so that they can filtered out by the oil filter. Oil life is largely dependent on how long these additives work, along with the buildup of any contaminants that do not get filtered. Modern motor oil is designed to perform at 100% (or very nearly so) of its intended function until the end of the recommended maintenance interval.
The oil analysis tests refer to upthread allude to the fact that motor oil will actually protect even longer than this, but is heavily dependent on driving habits and individual variation. So that is why they recommend changing the oil well before the theoretical maximum life.
A Subaru flat four, and a Toyota V-6 are very different engines, so it is possible the recommendations reflect real differences in the expected life of the oil. Sometimes it does happen that differences in recommendations are almost certainly fake. I think the Volkswagen 1.8T engine in a VW oil change interval was 12/5000, but the same engine in an Audi was 12/10000.
… or different oil-change intervals for the same car/model in differnt countries.
Let me add, that possibly for Canada the car makers are extremely conservative, b/c they can never know if a car is being used basically at the arctic circle or under other extreme conditions.
I do mine at 5K with synthoil because I think (I am not a tribologist) that it is better to change oil before it starts breaking down rather than when it starts breaking down.
Like I said, IANAT
Can we start talking viscosity? Would a 15W 40 vs 5W 30 vs 0W 8 make a difference in the real oil change intervals?
What outside temps are you operating at?
Viscosity is not about what works when the engine & oil is fully up to operating temp. It’s about getting from shutdown to running steady-state wo trashing the engine.