Hi, I'm kind of a beginner home 'producer', but really I can't mix very well in terms of knowing how to do a specific thing. I'm interested in this one track, ( https://www.youtube.com/watch?v=3zbIBF97Sys ) especially as I'm trying to record vocals myself, I only have a home setup, at4040 + boom arm etc...

First off I'm not sure like how hard you actually have to go in the recording take to get vocals like this, (don't want to annoy the neighbours) vs the fx, distortion, ott, whatever carrying it after, because to me it sounds like a very shouty take. Second, should I double up the same take and pan it or any technique like that and is there anything specific to focus on when when trying to get that same overall sound, i've played around with Ableton amp, roar distortion, ott, reverb room/ir types etc, but can't quite get that same feel.

Thanks for any tips

1) your AT2020 will work perfectly since AT mics are mostly "linear" mics.
2) record 15-30 cm to the mic. don't forget the nylon pop filter. the performance are LOUD and very emotional - this is the key!
3) I would recommend to subscribe and watch Chris Liepe videos on how to warmup and perform vocals fry - there are very similar technique with your reference track, where Chris will "break" the voice to find that raspy fry tone.
3) the vocals are heavily distorted and you can spent a lot of time to find the "proper" distortion, so I will give you a couple of hints..
4) Process Audio SpiceRack, preset "star of the show" - instant sound you need!
5) Pulsar Audio Vocal Studio - you can add additional compression, reverb and delay using this plugin. It is my go-to plugin for fast vocals processing. You can also try it's distortion module (3 different characters), so you can possibly can get rid of SpiceRack.
6) Then you will definitely need the digital EQ to tame harshness and to increase loudness of the important parts. I would recommend "pro-q4 222 presets" - there are tons of lead/backing vocals presets, mix test, soothe emulation and more.
6) don't forget to add ambience reverb with UAD EMT 250 emulation or Wave Alchemy Dawn. It's mainly Early Reflections which put your voice "in the mix".
7) use Waves RVox in the end of the chain to add even more power
8) use Waves CLA Vocals to add wideness and other effects (to your liking)

There are questions like this all the time.

"How do I get vocals like these?" With a post of an Adelle recording or some other extraordinary vocalist. Hoping to find some sort of mic technique or specific mic to make it happen, forgetting that a part if the signal chain is the performance the artist gave...indeed it is the most important part of the signal chain.

Now, I know you are not looking to sound like this singer...However, the same idea applies. You do seem to understand that the performance is a part of it, which is good.

I don't know how loud you would have to be to produce this type of intensity or emotion...however you would categorise this style of singing. But, if you have to get loud to achieve it...then you have to get loud to achieve it. The performance IS the largest part of this.

There is certainly some EQ on the vocal to narrow the bandwidth. And of course some distortion. And those would be required for the end result...but ultimately... The success of those tools will come down to the performance.

Taking Judy Garland's performance of somewhere over the rainbow, and applying the same EQ and distortion FX will not sound the same.

However, knowing that, there is still the matter or processing g the vocal performance to complete the effect.

This sounds like what I would call a "megaphone effect".

Some people will get an actual megaphone and sing through it into a mic...to me this has a low cut higher than usual and perhaps a bit of a high cut...you'd have to play with that. And then some distortion.

If you don't get the results you are looking for with just that...what you could try is something like this.

Double the vocal recording. Process one more extremely. The high and low cut + distortion. Then process the other version of the track "normally"...regular EQ, compression...then blend the two versions together. I would not own them.

You can other sing the part twice rather than just copying one. Though the results may not be great...or they may be very good...I would try to record them as closely performances wise as possible.