I never streamed a video or recorded a podcast, and I don't know anything about sound recording and signal processing. So this post is purely about how I experienced the situations and not grounded in scientific or advanced testing.
In 2021, I bought the Bose QC45 headphones and I started using them at work. It's quite comfortable to isolate myself from my colleagues' voices in the open office setting. The sound quality is great for both calls and listening to music. The microphone is okay but I wanted to have a crisper, better sounding output.
I had been using a Samson Meteor condenser mic in the past. But in every call, I was told people could hear colleagues sitting next to me or across me way too much, to the point it was hard to understand me.
So I started looking for an alternative under 100€ that could pick less noise while still deliver a crisp output. I was thinking of going with another condenser mic that could pick my voice and reject the rest of the noise. Checking on the Internet a bit, I found the Samson C01U Pro, which features a super-cardioid pickup pattern: it is supposed to pick the sound coming from the front and cut the ones from the side and the back.
Trying it for a while, I figured that despite having a crisp sound, it was still picking way too much the voices of my colleagues over mine in calls. I asked on Reddit and it seems it was because the mic was standing on my desk, somewhere between 50 and 70 cm from my mouth. I should have had it way closer to my mouth (10 to 15 cm) with a lower gain so it would have only picked my voice.
Being in an office where we share desks, I couldn't just get a boom arm setup, as I'd need to mount and unmount on a daily basis. And this would also had been quite a expense for my low budget.
I ended up using the Blue Yeti (first generation) I bought 10 years ago for online gaming. I actually barely ever used it and it was delivered with quite a tall stand. Thanks to this, I can put on the edge of my desk so it's really close to my mouth for calls. With the knobs to adjust the pattern and the gain directly on the mic, I managed to have a crisp enough sound while barely picking any other voice than mine.