Last night I received an email from an old friend, who asked me if there is so much more to choosing a rig, why isn't that mentioned on "the list"...(apparently he was reading this thread, but didn't want to stir things up?)
My answer was....well, it was long....but, in brief I wrote that if you read the rest of what Rob writes and read thru his equipment reports carefully, you'll see these recommendations, or at least some comments on troublesome issues....and, I also mentioned that (in my opinion) the "list" does serve a purpose, as long as don't use it to decide on what rig to buy! (certainly not as the only criteria!)
In addition to Rob's (and many others) recommendations to
not concentrate on one "number" (close-in 2-tone IMD3), and actually use other / many criteria when choosing a new rig....when I read VK5ISO quoting Rob Sherwood, it reminded me of the oft misunderstood fact that even if someone does actually wish to use Rob's list to better evaluate various rigs, they really do need to look at more than one "number"...and understand that sometimes even some of those numbers belie the actual real-world results (see details below)...
Then I thought....why not just cut-n-paste my reply here? So, here goes....some of the below are direct quotes from Rob Sherwood's papers, and some are my opinion, based on my readings and experience in the last 45 years...

If you're thinking the "king of the hill" (the rig at the top of the list
http://www.sherweng.com/table.html ), is always a great choice,
you really should read some more about whatever rig it may be...and, while reading what Rob himself writes about them is very good, you should also read the ARRL Product Reviews...
http://www.arrl.org/reviews-listed-by-manufacturer (I know, some here see a mention of the ARRL and their blood begins to stir....but, please let's not drift off into ham radio hysteria....can we just all agree that getting more info / education is good, no matter where it comes from?)
Regarding "numbers" and position on the "list", here are some cases in point:
---- 10 years ago, let's not forget the truly horrible audio out of an
Elecraft K3, and its surprisingly rather noisy synthesizer which made its DR noise-limited, i.e. poor RMDR....And, of course, it's poor transmit IMD! (now, except for its audio amp / path, no question it did had a good receiver, but was really only designed to be a CW rig) And, yes, improvements have been made over the years...
https://www.dj0ip.de/app/download/5794190002/Elecraft+K3+rev+c.pdf(Oh, and don't forget that the highly touted K2 was easily trounced by rigs many years older than it!)
---- And, the current top of the list, the
Yaesu FTdx-101D....well, just read Rob's report! Transmit ALC issues, ALC-overshoot, etc., and overall power spikes....rather harsh keying waveform, etc., and even QSK issues...(the ALC issues and keying waveform continue a poor Yaesu tradition, I'm afraid), but also has a bandscope anomaly, and like most modern rigs (sans the Apache ANAN's and the K3) has serious receiver AGC issues on noise pulses...not to mention, like many "modern" rigs, has rather average-to-poor transmit IMD, and unlike the FTdx-5000 (and 9000, and even the old FT-1000), no provision for "Class A" transmit operation (although with their ALC issues, Class A operations from these other Yaesu's needed to be done with precision...little-to-no processing and NO ALC action...in order for any significant improvement in transmit purity)
But, heck....it's on the top of the list, it must be great, right? Shaking my head...
https://www.dj0ip.de/sherwood-forest/sherwood-hf-xcvr-tests/yaesu-ftdx-101d/---- Of course, the recent "hot girl" that every ham wants to get their hands on, and test drive....the
Icom IC-7300....no dispute that, for the price, it's got a lot going for it.....but, it ain't perfect...(of course, no rig is perfect!) And, again, some improvements have come and sometimes errors are made in the tests / reports (see the two-samples report)...
In some hams' opinions, the IC-7300 is a better radio than the current king-of-the-hill, the FTdx-101D...even with its close-in 2-tone IMD3 being 13db to 16db worse than that of the FTdx-101D, many hams think the 7300 to be a better radio...oh, the horror!

LOL
https://www.dj0ip.de/app/download/5806181195/Icom+IC-7300+A.pdfhttps://www.dj0ip.de/app/download/5811576452/Icom+7300+Two+Samples+D.pdf{BTW, personally I'm still not a fan of "computer radios".....but, the Apache ANAN 7000 / 8000 have great transmit purity / IMD (with their pre-distortion)...and, while I'm still not a convert to direct-sampling SDR's (not even ones with "knobs"), the Icom IC-7610 does look nice, even in my eye.....and while I'm
not going to buy an IC-7300, to be clear I don't think it is evil, nor one of the seven signs of the apocalypse!}
To get a better understanding of the "list" and the numbers there....have a look at what Rob writes:
"Blocking:
Blocking occurs when the radio is just beginning to overload from a signal outside the passband. It is usually about 30 dB above the Dynamic Range of the radio (to be described below). If a radio has a good dynamic range, then it will have a good blocking number. 130 dB is a good number. With direct sampling radios, blocking is technically not the correct term. An A to D converter has an absolute overload point, unlike a 1 or 3 dB gain compression point. Note: Instantaneous overload from many strong signals may cause the overload indicator flicker, but may not have an audible side effect."
"Phase Noise:
Old radios (Collins, Drake, Hammarlund, National) used a VFO or PTO and crystal oscillators to tune the bands. Any noise in the local oscillator (LO) chain was minimal. When synthesized radios came along in the 70s, the LO had noise on it. It is caused by phase jitter in the circuit, and puts significant noise sidebands on the LO. This can mix with a strong signal outside the passband of the radio and put noise on top of the weak signal you are trying to copy.
This is a significant problem in some cases: You have a neighboring ham close by, during Field Day when there are multiple transmitters at the same site, and certainly in a multi-multi contest station. You would like the number to be better that 130 dBc / Hz at 10 kHz. [corresponding to a RMDR of about 103db] A non-synthesized radio, such as a Drake or Collins, has so little local oscillator noise the measurements were made closer-in between 2 and 5 kHz.
Note: Very few legacy superhet radios have low phase noise, though most direct sampling radios have low phase noise. The ARRL has clearly emphasized low phase noise (RMDR) since 2013. (RMDR = Reciprocal Mixing Dynamic Range) To convert my LO Noise (dBc/Hz) column data to RMDR subtract 27 dB for a 500-Hz bandwidth."
"Dynamic Range:
Now we get to the nitty gritty. I started testing radios in 1976 because the ARRL rated the Drake R-4C very good, but in a CW contest it was terrible. The radio overloaded in a CW pile-up, so I decided to figure out what was wrong with their testing. In 1975 the League [the ARRL] had started testing for noise floor and dynamic range, new terms for most amateurs. Spurious Free Dyanmic Range measures how the radio can handle strong undesired signals at the same time as a weak desired signal, without overload. When a radio overloads, it starts generating spurious signals on its own.
Dynamic range is defined as the level in dB when two strong test signals make distortion in the radio equal to the noise floor. The radio thus can handle that range of signals before the strong signals just start to overload the radio.
The League originally only tested the dynamic range at 20-kHz test spacing, which was reasonable at the time. But as multi-conversion radios became the norm, this test was inadequate. The Drake example was a case in point. When the two test signals are 20 kHz apart, the overload distortion products are 20 kHz each side of the pair of test signals. In other words, the League was testing as if the QRM was always going to be 20 and 40 kHz away! In reality the QRM is likely going to be close by.
In 1977 I published an article in “ham radio magazine” [Dec 1977, Ham Radio, page 10 - 18]discussing this subject. I tested the offending R-4C at 2 kHz in addition to 20 kHz. In that case the 20-kHz dynamic range was over 80 dB, but the 2-kHz dynamic range was less than 60 dB.
The roofing filter of the R-4C is 8-kHz wide, and in a CW contest, there were many signals inside that 8-kHz filter, overloading the radio. I installed a 600 Hz roofing filter in the R-4C, and the problem went away. When testing the Sherwood modified R-4C at 2 kHz, the dynamic range was over 80 dB, just like it was with the 20-kHz test.
Most radios in the 70s and 80s had gone to up-conversion for two reasons. This got rid of the necessity of a preselector, and it allowed general coverage without a dead spot equal to the first IF frequency. In the up-conversion radio, the first IF was always above 10 meters, and often above 6 meters. All first IF filters were at least 15 kHz wide, and there was the problem. The Drake 8-kHz first IF was bad enough, and now almost all the radios for 20+ years had a first IF what was at least 15-kHz wide. Almost all of them had a close-in dynamic range around 70 dB. That was barely adequate for SSB and inadequate for CW.
For more than 40 years I have been testing radios, and I decided to sort the table on my website by close-in dynamic range at 2-kHz spacing. This was the “acid test” for CW contest / DX pile up operation.
In 2003 the Ten-Tec Orion came along, and it went back to a 9 MHz first IF (instead of 40 to 70 MHz), and offered a narrow CW roofing filter, like I had added to the Drake. It was the first commercial rig to be better than the Sherwood roofing filter modified R-4C. Later the Elecraft K3 came to market, and now Yaesu and Kenwood have what is now called “down-conversion” radios with a low frequency first IF.
What do you need in the way of close-in dynamic range? You want a number of at least 70 dB for SSB, and at least 80 dB for CW. A 10 dB safety factor would be nice, so that means you would prefer 80 dB for SSB and 90 dB for CW. Now there are approximately 20 radios that meet that specification."
Now, in addition to
reading the test reports and product reviews....even when you look at just some test result "numbers" of different rigs on the list, you will find that sometimes things are more complicated than just the "numbers" might show (similar to how things were 45 years ago, when Rob started his quest for a better receiver....when a stock Drake R4C "tested" good in the lab in Newington, but fell apart on-the-air in 160m CW contest....the numbers weren't the story...and, looking on the other side of the transceiver just like today's receiver "numbers" of the FTdx-101d, K3s, etc. etc. don't tell the story, nor do they show transmitter issues!!)
In addition to the modern radios mentioned above, here's an
old case-in-point here, regarding receiver numbers:
A Drake TR-7....a high-quality legacy HF rig, from the 1970's (this is a high-quality DBM "up-conversion" HF rig, with a circa 1975 design VCO/synthesizer, mixer, etc.) that I personally know very well...(but, hey...I'm no DeMaw, Hayward, Rohde, etc....these are the guys I learned from, and have probably forgotten more than most will ever grasp about receiver design!)
I own/operate two TR-7s, the first one I bought new in 1978, the second I bought used about 20 years ago.....I understand their design pretty well, as well as have a good deal of time operating them in various environs....used them at home and in the field on 160m contests, Field Days, 75m DX'ing and ragchewing, casual 40m and 20m operating....and also one on-board, offshore at sea, on ocean-going yachts...
A cool fact of life for the TR-7 is that, except for the comparatively-noisy VCO/synthesizer, it actually holds its own against many of the modern 21st Century HF ham rigs made in the past few years (that have 40 years newer / more advanced technology)....and, of course, the TR-7's transmitter is a lot nicer than just about any other ham HF rig made today, or in the past almost 40 years!
So, using the Drake TR-7 "numbers"....
a) It has a BlockingDR of 146db (only recently beaten by the $12,000 IC-7851, and the multi-thousand dollar rigs, the Elecraft K3s / Yaesu FTdx-101d).....which if you use Rob's general rules / calculations, would mean that the TR-7 has a "Dynamic Range" (2-tone IMD3 ) of 116db? But, of course, it doesn't....(it's actually 99db wide-spaced, and 75db close-in)
b) It's Reciprocal Mixing Dynamic Range (RMDR), which is a strong signal (one signal) mixing with the "noise" of the synthesizer/VCO is approx. 90db, which is adequate even for today...(10 - 20db worse than the modern hi-end 21st Century rigs near the top of the "list", but pretty good in comparison to many radios out there....and certainly state-of-the-art 45 years ago, when, in 1975, the TR-7 was designed)
c) The measured "Dynamic Range" (2-tone IMD3) is 99db wide-spaced, and 75db close-in...not surprising as the 1st IF filter (aka "roofing filter") is only 4-poles and 9khz wide...
The reason I'm writing this here, is not convince anyone that my old favorite (Drake TR-7) is so much better than today's modern rigs, but rather just to show that the TR-7's excellent wide-spaced IMD3 of 99db and its outstanding146db BDR, are quite good.....and its RMDR is about 90db, and it's close-in IMD3 is 75db, are not that bad....I mean, we did actually make contacts back in the old days, you know!
Fyi, the TR-7's position on the list is
way above that of the Elecraft K2 and right next to the Icom IC-756ProII and ProIII, and near the IC-7600 / IC-7700 and the TS-590s (in up-convert bands)....and remember that the TR-7's design/manufacture is 30 - 35+ years older than those "modern, 21st Century radios"...yep a 45 year old TR-7 can trounce an Elecraft K2 in about category!!! And, is as good (better in some applications) than the '7600/'7700, and ProII's and III's....But, you never hear anyone say that, do you....nope....that's 'cuz it doesn't fit with the "accepted lore"....
but, it is true!!
Remember that while we've seen "100db radios" being advertised for decades, this wasn't really reality until the 2010's....things didn't really improve until about 10 years ago....and, what 'da know, somehow hams were making contacts (CW, SSB, RTTY, etc.) on the HF bands for many decades, in times of solar minimums and solar maximums....how was that possible without a "100db" radio? (maybe it was magic?...hmmm?)
And, let's please remember that except in serious CW contest pile-ups, these numbers (of the TR-7, and some others) are more than adequate for most other HF ham operations, even in today's crowded bands, and most contest environs....the TR-7 is actually to this day (along with the IC-781) is considered one of the best legacy HF rigs, especially in SSB operations!
And, while not 100% on topic, the TR-7's Noise Blanker (the $175 optional NB-7), is considered by most radio engineers and every ham that's ever used it, to be one of the best noise blankers ever made....it does not ever degrade receiver performance (even in crowded contest conditions) and it friggin' works! It remove impulse noise / most powerline noise, from S-9+ down to S-0...without disrupting the receiver....
{this is one reason that I decided to not try to narrow the first IF filter ("roofing filter") from its designed 9khz width (and only 4-poles), as I didn't want to change the input noise shape, thereby reducing the blanker's effectiveness, in order to just gain a few db of close-in rec IMD performance (heck, I'm mostly an SSB op, so no real improvement for me anyway)....
Although, I
have thought of changing the VCO's (maybe employing only one VCO, as KA2WEU, Dr. Ulrich L. Rohde, suggested?) and a new mixer....to improve the RMDR....but, as I still haven't found it lacking in that criteria, I haven't yet done more than just think about it!}
BTW, Rob just recently exchanged his longtime favorite IC-781 in operating position #1 with an IC-7610, moving the '781 to position #3....how's that for longevity....the '781 is a late-'80's rig, 'ya know!...
Now, I'm certainly not a CW contester, but the fact is my 45-yr old design TR-7's are plenty good enough for 80m Dx'ing, etc. and 75m rag-chewing down here in Central Florida (where summertime static crashes can be 20+db over S-9)! And, they along with many other radios, are certainly able to hold there own even with today's crowded bands!}
While the RMDR and close-in IMD3 don't place the TR-7 anywhere near the top of "the list", they ain't bad....and, its 99db wide-spaced IMD3 and outstanding BDR of 146db, go a long way to keeping the TR-7 very competitive these days!
Yes, I know, the commonly-accepted adage that a high BlockingDR is meaningless if you have poor RMDR (i.e. noisy VCO/LO/Synthesizer)....and on paper / in the lab, yes this can be "proven", but in the real-world (like Field Day), this does not always ring true!
Just shows-to-go-ya', that looking at one "number" in a spec sheet or where a particular rig places on a "list", doesn't always tell the whole story!

I hope this finds some of you smiling....
73,
John, KA4WJA