I’ve been looking at this issue myself and there’s one important wrinkle to keep in mind. While you can take your total RMDs from any or all of your affected accounts, the accounts must all be of the same type. For example, if you have both IRAs and 401ks, you can’t take the IRA RMD from the 401k and vice versa.

RMDs from each 401K must be taken separately. They cannot be aggregated even though they are of the same type. Aggregation is possible with 403(b)s.
https://www.irahelp.com/system/files/ira-focus/25060/june-elite-analysis-rmd-aggregation-rules.pdf
RMDs for inherited IRAs must generally be taken separately from each IRA. An exception is if they IRAs are of the same type and they are inherited from the same person.

since they started 'adjusting' them in the 80s.
CPI figures have been adjusted since day one (1919). In 1921 the government cobbled together a national figure from an unweighted average of 32 city figures.
https://www.bls.gov/cpi/additional-resources/historical-changes.htm
Would it be preferable to keep counting buggy whips, or whatever was in those 1919 averages rather than 'adjust' the inflation components over time?
Ever hear of Hedonic adjustments? That's when a product is 'new and improved' they can charge more and the increase is no included.
Sure, I've heard of hedonic quality adjustments. They can go up or down, based on the product change and the value of that change.
https://www.bls.gov/cpi/quality-adjustment/questions-and-answers.htm
Certainly $300 buys more computer today (Best Buy's sub-$300 computers) than it did with the Commodore VIC-20 in 1981. Should we 'adjust' the CPI for this increase in value (i.e. recognize that computers are cheaper today)? Or do we stand firm and insist on making no adjustments based on product quality?
We could actually continue to include the Commodore and its ilk in the CPI. It looks like they're still available at around 2/3 the original selling price.

@msf said,
"- Inherited Roth IRAs have RMDs."
There is no Require Minimum Distribution for Inheirted IRAs, but instead, a Required Full Withdrawal following the 5 or 10 year rule. One could wait 10 years before making that one full required withdrawal providing an additional 10 years of tax free growth from the date of inheritance.
This article does a good job of explaining Inherited (Roth) IRAs:
https://fool.com/retirement/plans/inherited-iras/
1. A spouse (as a beneficiary) can rollover an Inherited Roth IRA (from a deceased spouse) and continue to enjoy no RMDs.
2. Withdraw the funds as a lump sum. You may withdraw all of the money from the original owner's IRA as a single lump sum. Doing so gives you a lot of money now, but also results in a high tax bill for the current year, unless you're withdrawing the funds from a Roth IRA that the original owner held for at least five years. In that case, you won’t owe any taxes on these withdrawals. However, if the owner didn’t have the account for at least five years, then you could owe income taxes on the Roth IRA earnings.
3. Use the five- or 10-year withdrawal method. The five- or 10-year withdrawal method enables you to withdraw money as often as you'd like and in whatever increments you choose, as long as the money is completely withdrawn within five or 10 years. If you fail to withdraw all the funds in time, then you'll pay a 50% penalty on whatever remains in the account.
You have five years to withdraw all the money from an inherited IRA if the account owner died in 2019 or earlier, and 10 years if they died in 2020 or later.
For all of us, this can be very confusing. If you have a specific scenario (question). I would suggest reader's ask their questions on the Ed Slott (Discussion Forum). It is a great IRA resource.
https://irahelp.com/phpBB

Howdy folks,
The government statistics on inflation are rubbish and have been since they started 'adjusting' them in the 80s. Too many COLAs and other ties to the CPI. feh. It's called screwing the public and hiding the facts about the dollar.
@Hank you mentioned that housing was eliminated in the early 1980s. Other things also. Here is Shadow Stats showing CPI like they calculated it in the 80s' and later the 90's. Ever hear of Hedonic adjustments? That's when a product is 'new and improved' they can charge more and the increase is no included. That's fine if you can still buy the old and unimproved item. Right.
Notice that he has the earlier version of inflation running about 12-13%
http://www.shadowstats.com/alternate_data/inflation-charts
They're still dealing with the Unfunded Liabilities overhang that we've talked about for years now. What some $100 T. Their choices were to break as many promises as politically possible and deflate the currency by about 50%. Please note that it's down 96% since '13 so it only has to go to 2%. Easy. And let's not talk about the strong US dollar. Er, it's the cleanest pair of dirty socks in the hamper.
and so it goes,
peace and wear the damn mask,
rono

@crash - it's not for everyone but
IOFIX - 1yr.: +18.29% although after last year there really wasn't much place to go but up.
YTD: +13.7%
Yield: 3.98%

IOFIX generated excellent category returns from inception (05/28/2015) through 2019.
IIRC correctly, volatility was low and the Sharpe ratio was high during this period.
The fund then delivered an unpleasant surprise when it returned -36.18% during Q1 2020.
IOFIX seemed like a safe fund for years...

Two petty technical points:
- You're supposed to calculate the RMD for each IRA and then add them together. Usually that comes out the same as adding the values together and then dividing, since:
$A / N years + $B / N years = $(A+B) / N years.
But in rare cases you could have a different N for two IRAs.
If on one IRA the sole beneficiary is your spouse, who is more than 10 years your junior, you use Table II (Joint Life and Last Survivor Expectancy) for the divisor. If the situation is different on another IRA (e.g. beneficiary is sibling), then you use the customary Table III (Uniform Lifetime) to find the divisor.
- Inherited Roth IRAs have RMDs.

@msf
Grateful for all that info. It's an eye-opener. And yet, "what's a mother to do?" Short-term funds, to me, are "return-free risk." Not like STOCKS, of course. Or rather, the risk is in not getting much of anything back on your investment. IG-rated stuff might offer you (me, that is) a monthly dividend which MIGHT cover the cost of an Uber ride somewhere.
PTIAX was paying a 14 cent div. during the mortgage boom. Now it varies, but still better, per share, than my other two. And nowhere near 14 cents anymore. They are all actively managed, and I like that.
We don't normally talk in terms of hard-dollar figures here, about the size of our portfolios. I'm still below a quarter million. 56% in bonds, trying to grow the bond-stake and reduce the proportion in stocks. Last week was a good week for stocks. I'll take it.

The Barron's quote of Carson which elicited your comment was about how including home costs in the CPI-U would make that number much higher. It concerned the CPI-U, or what Carson calls "the official price report".
You posited (in)frequency of transactions as a rationale for excluding home prices from the CPI-U. Not from some sort of core CPI, not from some "embedded" inflation calculation, but from the CPI-U. I simply pointed out that if that were the rationale, then vehicle prices would also be excluded.
ISTM that given this exclusion rationale there are three alternatives:

1. vehicle prices should not be included in the CPI-U. They are purchased at about the same frequency as homes and can be just as volatile: used vehicles up 24% Y/Y, homes up 19.5%.
2. there is some other continuum (what?), aside periodicity or potential volatility, that distinguishes the housing component from the used vehicle component
3. the periodicity rationale given is at best incomplete (kindly complete)

The bottom line is simple: CPI-U is a measure of inflation as seen by the "average" consumer. The average consumer sees wildly fluctuating prices, including gasoline (up 42% Y/Y) and medical supplies (down 1.6%). Welcome to the real world.
Whether one should worry about this or how one should smooth out the volatility is a different question, and not the point of the Barron's piece.

Age is one substantial factor in my sense of risk tolerance. Current expenses are another. I'm not getting younger, and my expenses are actually rising. Safety, rather than growth is my primary concern by now. Yes, bonds are paying next to nothing, and the safest of bonds barely pay anything at all. Looking at the big picture, I'll take the step into riskier bonds, which after all are not so VERY risky--- in order to produce SOME kind of measurable profit. PRSNX is barely holding its own. Weaker dollar. I suppose it's due to the expected reaction to the Demublicans' (sic) big spending plans? The stock market at these current price-levels is prohibitive. The monthly divs I'm still re-investing from my bond funds will be a lifesaver, if I ever DO need them to pay monthly bills. Inflation will NOT be so very "transitory." (Yellen, Powell.)

@Crash. I have quite a bit in bond funds myself. It’s definitely not a comfortable feeling. :) My model is about 35% growth, 30+% income and 30+% alternatives. I use DODLX and DODIX quite heavily on the income end. Also PRIHX which has been a steady-Eddy. Who knows? You need to allocate somewhere and equities don’t strike me as a bargain right now, nor a place where a 75 year old should be tying up all his money.

There has been plenty of discussion about whether recent inflation is transitory or not.
The "experts" disagree on this topic.
Here's a look at S&P 500 (and its predecessor) calendar year returns along with annual inflation.
Link

Why should rare years-apart purchases be included in widely impacting run-of-the-mill inflation calcs? ...

Do you disagree with the BLS for including the prices of new and used motor vehicles in its CPI calculations?

It's a continuum, arguable, debated, ...

The only antecedent for "it" I can see is periodicity of purchases, so I'll infer that this continuum is the length of time between purchases.
Until the past decade or so, average time of home ownership was about four years. For example:

Source page: https://ipropertymanagement.com/research/average-length-of-homeownership
New car ownership in 2017 was almost seven years, or just a shade less than home ownership since the GFC.
https://www.cnbc.com/2017/05/28/car-owners-are-holding-their-vehicles-for-longer-which-is-both-good-and-bad.html
Is this minuscule difference in holding periods really what you want to pin your continuum premise on?
I'm sure you're aware that when housing costs (including investment attributes) had been used in calculating the CPI, those costs included mortgage interest, property taxes, insurance, and maintenance expenses. That's in the paper I cited.
Paraphrasing your original question, why should frequent, periodic cash outflows (interest, taxes, insurance, and maintenance costs) be excluded from widely impacting run-of-the-mill inflation calcs? Even if those costs fluctuate wildly from month to month as construction (maintenance) costs have done this year.

Rather than spend a fair amount of time searching for an ratings agency (NRSRO) default report, I'll just refer you to M*'s figures. In Exhibit 2 on p. 4 of this M* paper is a table that includes "default score" by credit rating. Those aren't default rates, but represent relative rates of default. BBB has a score of 5.0, BB has a score of 17.78, meaning that BB bonds default at roughly 3½ times the rate of BBB defaults.
There really is a big difference, which is why M* doesn't simply score A's as 1, B's as 2, C's as 3 and so on when calculating a portfolio's average quality.
https://www.morningstar.com/articles/354597/credit-quality-demystified
PTIAX may be a mere poseur. Between 2012 and 2020 it was classified as a multisector bond fund, typically meaning that it had even more junk than a core plus bond fund. In 2021 it was classified as an intermediate core plus fund, and in 2011 it had been classified as an intermediate core fund (before core plus funds were given their own category).
As the linked article (about the new core plus category) states, the median amount of junk in a core plus fund is (or was, at the time) about 8%. DODIX has 11%, all BB (including NR bonds). PTIAX has more than that (12½%) in bonds rated lower than that (or NR). Plus another 5% rated BB. Then there's BCOIX, with less than 4% junk (including NR), nearly all at BB.
I suspect you'll find a fair degree of correlation between funds' YTD performance and the amount of junk in their portfolios. PTIAX > DODIX > BCOIX.
Note:edited to fix typo, per @BaluBalu's suggestion.

Thanks @msf for the pricing of a new or used vehicle question.
One person told me he had been ask to cough up an extra 10 k on a new drive at a dealer , 5k at another place & finally coughed up an extra $750 for his new ride.
EXTORTION NOT ? , Derf

Yellen explaining a few weeks ago;
https://www.cnbc.com/amp/2021/10/05/yellen-sees-inflation-staying-higher-for-the-next-several-months.html
And from this morning:
https://www.cnn.com/videos/politics/2021/10/24/full-interview-with-secretary-janet-yellen.cnn
We’ll see.

The inflation headline statistic is 5.4%. 5.9% is the adjustment for SS. That affects only SS receipients. It is based on CPI-W which is different from the marquee inflation metric CPI-U.
https://www.bls.gov/news.release/pdf/cpi.pdf
As I posted previously, CPI-E is even lower. According to the CPI-E excel spreadsheet data, this Y/Y figure as of Sept was about 5%.
There are also those who feel that items that are purchased infrequently shouldn't even be counted. On that basis, perhaps we should pull out the figures for vehicles like used cars and trucks. Their costs went up 24.7% over the past twelve months. As you wrote, some things we can control, and most people don't need to buy a vehicle in any given year.
Moving on to bonds, the three funds RPSIX, PRSNX, and PTIAX all have credit ratings of junk by M*. (This does not represent a tactical move by these funds; over the past several years they have been consistently rated junk.)
Junk bonds tend to have moderately high correlation with equities, and so share a similar (albeit muted) risk profiles. This has been a good year for equities and for junk bonds.
Here's a correlation matrix of the four funds. R² over the past year with respect to MWHYX (a junk bond fund) ranges from 49% for PTIAX to over 75% for RPSIX and PRSNX.

We will be the last Carbons standing. Actually, I have never met any of you.
Some of us carbons may already be silicon.
Here's META's holdings:

RPSIX. 1-year +7.95%
YTD. +2.42
yield 2.55%
my own performance since investing in the fund, according to TRP: +5.12%
PRSNX. 1-year: 3.77%
ytd: 0.71%
yield: 2.87%
account performance since inception: 4.71%
PTIAX. 1-year: 3.29
ytd: 0.95%
yield: 3.47%
I can't complain..... yet.
...And if someone can either increase income or decrease spending, inflation is not as big a monster... unless we're talking late '70s/early '80s-style inflation: outa control. Yes, the current inflation headline statistic is +5.9%. But the way the gov't measures inflation is.... bullshit. It's higher that 5.9%, for sure. SOME things we CAN control, to protect and assist ourselves. But I'm preaching to the choir: we who are regular contributors to this message-board are among the luckiest, most fortunate. The country is being hollowed-out from within. We are in a long, slow decline, the way all empires eventually go through.

The Economist: “House prices were included in America's CPI between 1953 and 1983 before being removed. This was partly because indexing benefits and pensions to inflation had become expensive and some politicians wanted to bring measured inflation down”
Source:

Here's a Monthly Labor Review (MLR) piece written contemporaneously (June 1982) with the announcement (late 1981) of the change in how housing costs would be included in the CPI.
https://www.bls.gov/opub/mlr/1982/06/art2full.pdf
Government expenditures, though not mentioned in the paper, could have been part of the reason for the change. As stated in The Economist, that would be strictly a data driven decision, i.e. based simply on the fact that housing prices were rising rapidly.
The MLR piece notes that there had been a recent statutory change in how government revenue would be affected by inflation. That change called for a more accurate CPI calculation:

In addition to problems of data adequacy, impetus to change the homeownership component stems from an important new use of the index. The Economic Recovery Tax Act of 1981 (Public Law 97-34) requires use of the CPI for All Urban Consumers (CPI-U) for escalation of income tax brackets and the personal exemption amount. The law requires announcement of the new tax brackets in December 1984 based on CPI-U data for the prior 2 years. This is a major new use of the index which will have a broad effect on total Federal Government revenues, and this new use underscores the importance of action to ensure that the CPI reflects consumption cost experience of consumers to the fullest extent possible.

I thought reading the article that Forsyth was saying (in roundabout fashion) that if housing was still included in the CPI today (as it was until 1983), than today’s inflation would be approaching the level of the 80s.
The Economist: “House prices were included in America's CPI between 1953 and 1983 before being removed. This was partly because indexing benefits and pensions to inflation had become expensive and some politicians wanted to bring measured inflation down”
Source:

Here’s a colorful (albeit slanted) blurb from this week’s Barron’s under the title “The Striking Price: A Smart Trade for Heady Times” - by Steven Sears
“Now, almost 100 years later, the market construct has seemingly changed. The shoeshine boy's figurative heirs have grown rich buying stocks, digital currencies, and other risk assets … The S&P 500 index just hit another record high, and activity has reached a fever pitch. The options market, which enables investors to magnify their stock bets at relatively low cost, has seen daily trading volumes rise to about 40 million contracts a day, up from maybe a million in 2000.”
Barron’s 10/25/2021
* Represents the views of one contributor. The publication contains both “bull” and “bear” case analyses.