Haha yeah bro. My current PC build is from like 2012, built for audio editing, not video editing. Wanted to try DaVinci Resolve, but gfx card is too old to even boot it lol.
Since I’ve basically put all DJing on hold because of hearing damage, I’ve been looking into more graphics stuff again lately. Would be cool to fuck around with some imaginary/plausible Blender + Touch Designer combo. My DJ crew just got our own CDJs and looking into building sound system now, in combo with maybe try to expand to projector visuals with custom reactive graphics. Just smoked some medicinal, lost point a bit… but ye, wanna get into those things more. But gonna have to wait until I can build a new PC.
Yeah, my understanding is that this shit gets better exponentially. I know there’s been some recent papers that suggest massive improvements that are constantly coming out. Then take into account these systems would have been in training for the last 6 or so months, the next version theoretically would incorporate those improvements.
I don’t quite understand everything that’s going on but it seems like a lot of these technologies are going to get pretty wild pretty quick
I thought this was interesting to see someone try and manipulate generative ai in a commercial sense. I don’t really see ai bridging the gap with creative interpretation anytime soon and prompting ideas or concepts using keywords and phrases are always going to be a headache. As cliche as it sounds, it’s a vibe thing. Learning how to communicate with ai and using tools will be the key for creatives to stay relevant in the future, but for now, I can create much better content, although it’s much more time-consuming. None of my clients are capable of writing a proper brief and explaining their own thoughts so I’m not too concerned yet
you’re way too talented not to be working in a creative field. Have you thought about freelancing and seeing if you can get any traction on one of those soulless gun for hire platforms?
I’ve been tinkering with a couple things recently, inspired by Lauren Scarlett’s minimalist playlist series. Curating music and then making cover art for the “mixtape”. Mine is a bit less minimal coz I do love a good texture.
It seems to struggle with generating text, much like other AI image generators. Even with specific prompts, eg. the words “Crime Scene” printed on yellow tape - it’s all jumbled and archaic
I promoted some cool shades on a pic of me when I was missing my front tooth and it looked so bad. I also asked for a bird on my shoulder and it looked really incredible. Pretty gimmicky rn, but it will get much better
There’s lots of open source stuff you can set up yourself, if you have a machine for it, then just “teach” the algos to do whatever you want. Wizardhead has made some amazing stuff this way, I’ve linked it here (somewhere on the boards) before. I assume it requires a lot of technical knowledge, a beefy setup and lots of time.