DxO stated: "Comparing the EOS R”s ISO 2742 against the EOS 5 D Mk IV’s ISO 2995, there’s less than 0.13 EV difference in low-light ISO (Sports), just a 0.3-bit difference in color depth (Portrait) at base, and a difference of less than 0.1 EV at the base ISO of 100 in dynamic range (in other words, both cameras have effectively the same dynamic range)."
Linking TN videos is not helping your case, he has the technical comprehension of a bowling ball.
All of those differences are essentially meaningless, the sensors have effectively identical performance. If you're seeing a real-world difference, there's something else going on, e.g. your R was defective.
DXO shows the sensors are not identical in performance (regardless of the amount) which points to something different going on with the signal processing or something else. There should be no variation like this if they had identical sensors, identical signal processing and identical output. But again, a chart will not show things like banding and pattern noise. Real world is worse as I said I've shot thousands of images with both cameras and the R is fine in normal light, when you are not pushing/pulling files. But in lower light or when you need to recover shadows the R is significantly worse than the 5DIV. I returned the R and tried 2 more copies before I decided to keep one. They were all the same.
I normally agree that Tony is normally not somebody I admire for technical info, but what he showed in his videos mirrored my experience EXACTLY. You don't have to be a technical genius to see the photos/comparisons for yourself. At least Tony did some investigating to compare the IQ of the cameras rather than just say "it has the same sensor as the 5DIV" and call it a day like 99.9% of "influencers".
The fact is, most people do not shoot at ISO 6400 and up a lot or pull up shadows. I do. That's where I do most of my work and I pull up shadows a fair amount. I've done this for 15 years and various cams. The R has the worst "banding/pattern noise/colour shift" of any camera I used since the 1D2N. 5DIV has none of it. At all. It's an impressive camera for IQ. The R was better for AF and more fun to use but the worse IQ (vs the 5D IV) was the reason I finally sold it (and am now waiting on the R6 II). I could shoot 5DIV however I wanted and never have issue (aside from unreliable AF). But with the R I had to be VERY careful to either nail exposure or overexpose and try to recover highlights instead (not ideal for a digital sensor). The R made me very hesitant to shoot things I would never give a second thought to with the 5DIV.
I attached an image shot at ISO 100 on the EOS R, shadows slider brought up in ACR. There horizontal pattern noise is pretty bad (but it's worse in context with the full image). I have ISO12,800 shots taken with the 5DIV with shadows pulled up and there is nothing even remotely close to this going on.
I hate when people with 0 experience try to tell me that my experience (as a 18 year pro though that is neither here nor there) isn't real or valid. I shot both cams for quite a while, thousands of images, and 3 copies of the EOS R. Pattern noise in shadows existed on the EOS R where it didn't on the 5DIV. You can show line charts all you want but that doesn't change the facts. And the fact that DXO showed that that the IQ from the sensors was not identical (even if the differences were "small") proves that there is a difference. Apparently it's from the signal processors or on sensor autofocus systems from my research.
I find most mirrorless have more of an issue with shadow banding than their DSLR counterparts. I know the newer Nikons had similar issues vs the D750 and D850 sensors. But, most people dont push their image files more than a simple colour grade in LR so they don't notice or care. I think the R3 is better than most because of the BSI stacked sensor so I'm hoping the R6 II will get the same sensor (or similar). Though funny enough, my buddy did a comparison of the 5DIV and his R3 and they were similar with noise but the R3 had some weird circular banding we couldn't explain. It COULD have been the lens (since it was circular) or corrections applied (though I didn't think they affected Raw files). But in either case, that 5DIV file held up shockingly well against the new Canon flagship mirrorless. The R just wouldn't be able to compete.
Ps. the R had another notorious "banding" issue that was eventually solved with firmware. This banding issue was something different entirely.