Do all modern DMMs have TRUE RMS, even if they don't specify it?

[BillyO]

Does this have any advantages, or is it just a way to sell the “better” models at a higher price?

From my perspective, lack of True RMS today is strictly a cost cutting measure and/or a way of getting more life out of outdated technology.  When Zoyi, Aneng or FNIRSI can bring out very accurate True RMS meters for under $40, there really is no excuse.

Is this also the case with these non-TRMS DMMs?

Bandwidth is a different consideration.  Analog meters will have a better inherent BW than digital.  They are real time and don't have to deal with sampling waveforms.  Higher speed ADCs cost more money and are power hungry.  So, again, low BW DMMs are a sign of keeping costs down, but it has very little to do with having, or not having True RMS

[Aldo22]

@BillyO:
Thanks for the explanation.
That makes sense.
So non-TRMS DMMs don't have any advantages.

[Kleinstein]

In most cases True RMS response is better than the simpler rectifier value response.

The main reason to still have the simpler AC version is costs. The old style analog RMS chips are still relatively expensive. It is the relatively new digital RMS as part of the chip set that makes it affordable also at the low end. If the chip set uses an SD ADC chip (the more modern alternative to old style dual slope), the extra effort for the chip-set is rather small, especially for only limited bandwidth.
So far I don't know a DMM chip set with fast digital RMS included, but this is likely just a question time and maybe power consumption.

Another odd case are (old) service manuals for aviation and military that sometime still call for a meter that shows the average of the rectified value. So these explicitely don't want true RMS. AFAIK Fluke has a militry version of a meter that intentionally changed from true RMS to rectifier response, using the same RMS chips, but dropping a capacitor and a resistor change.

The old style non RMS meters can also fail at the high frequencies. One point is the input divider that needs matching parallel capacitors to get accurate higher freuency (like > 1 kHz) readings. With the higher frequencies the divider is set by the capacitors and not the resistors.

Even old analog meters with just a diode rectifier may fail with higher frequency AC: my old analog meter fails big at some 50 kHz showing near full scale on most ranges, in cluding 900 V  .

A cheap meter without trimmed capacitors may as well also use a slow, lower power active rectifier and amplifier, as the higher frequencies would not be accurate (like +-10%) anyway. A reduced BW also comes with less noise.

[TimFox]

For high frequency measurements, the old-school peak or peak-to-peak voltmeters (6AL5 voltage doubler in RCA VTVMs or the Eimac 2-01C “hot diode” rectifier with careful probe design in the -hp- 410A and 410B VTVMs) could perform well at high frequencies.  The 410B was spec’ed to 1 dB up to 700 MHz, useful up to a few GHz.

[Aldo22]

Even old analog meters with just a diode rectifier may fail with higher frequency AC: my old analog meter fails big at some 50 kHz showing near full scale on most ranges, in cluding 900 V  .

My old analog meter is actually still quite usable, even at 100kHz.
1.8V vs. 1.71V on the little oscilloscope.
That's not accurate, but still close.
At 200, 300kHz, it shows values that are too high, but not full scale.

[gamalot]

I found a video by Dave about how to convert a Fluke 77 IV average responding multimeter into a True RMS model 177.

I actually have a practically brand new Fluke 83V myself, which is also an average-responding meter. I just snagged another used 83V on eBay, and as soon as I get it, I'm going to convert that one to True RMS 87V.