Not in this case. Pretty much all non-reference designs have flaws, regardless of whether it makes sense or not.
I think 30-series is the worst, most flawed, and most prone to failure out of all GPU generations that ever existed(including infamous AMD 200/300 series).
They are still awesome and powerful cards, but once again ODMs f#$%ed it up. I'm afraid to even think what's going to happen with upcoming GPUs that have a TDP of my water heater.
Just to put it into perspective:
* lots of 3060 Ti are prone to VRAM failures. Vik-On(notorious GPU repair youtuber) already started to fill up his "jar of shame", which he wants to send to SKHynix
* 3080/ti overheating is a widespread problem. That 300-400W TDP and huge cooling systems are the ones causing problems (factory defects, misalignment and bending, especially overtime).
Out of all these cards that I've seen personally there's only one Palit 3090 that my friend owns which still holds up like a champ, though the previous owner did repaste/re-pad it last December.
All the other ones already went through my office with one of the 2 symptoms: 1) VRAM hitting 110C(that's when card starts to severely throttle); 2) Heatsink skews a bit and causes GPU to spike above 90C due to uneven pressure.
Some cards have ridiculous backplates, like MSI 3080 that I modded recently. It has plastic backplate with something that looks like aluminum composite embedded just around the GPU bracket. It kinda works, but before pad replacement it was hitting over 1000C with stock clocks, while after replacing pads with copper shims it became a comfortable 75C.
Gigabyte is the worst. I think it's due to their choice of cooling system: it is huge and heavy, but has a giant 1.5-2mm gap for VRAM pads, and unlike MSI - it has no metal top plate that acts like a VRAM heatspreader. Basically the heat exchange on VRAM is bad on the front, and severely limited on the back. VRAM may hit 110C easily, but the backplate stays comfortably warm to the touch.
