There's going to be a long tail of domain-specific tasks that aren't well served by current models for the foreseeable future, but there's also no question the complexity horizon of the SotA models is increasing over time. I've had decent results recently with non-trivial Cuda/MPS code. Is it great code/finely tuned? Probably not but it delivered on the spec and runs fast enough.
The problem with this is that people are particularly bad at judging their own 'independence' of thought, regardless of their political views.
I would say the opposite of partisan would be someone who actively seeks to understand and relate to the views of those who they disagree with, or who are from their out-group. This would also imply independence of thought.
I used to worry about recall but eventually realised I made more and more interesting music when I treated my gear like a regular musical instrument and just recorded myself playing to audio. Perfect recall put me in a brain loop of endless tweaking that didn’t actually benefit the music at all, it would all just end up sounding overthought. Plus I had more fun doing it. This was a bit of a revelation for me. Obviously, whatever works for you works for you, but just a counterpoint.
Of course, that's what really matters most. I do appreciate that the infinite possibilities, permutations and even extreme convenience all-in-one digital integration provides can become a downside. It really depends on personal style, preferences and goals.
Confronted by all that boundless possibility, I have sometimes found myself freezing up with 'possibility agoraphobia' or just rat-holing into the tweak-cycling you describe. Ultimately, I figured out I have to enforce some discipline on myself - which felt a bit odd since I'm strictly doing this for fun. My realization was that the needed discipline can either be embedded in the tools or style I choose or I can choose to enforce it on myself - which is its own burden. Every approach has its benefits and costs. No free lunch :-).
In a broad sense, I suspect any of the different approaches (all acoustic, analog, digital or some hybrid) represent sweet-spots that balance latitude and constraints in different ways. All that truly matters is finding one that meets us where we are in the moment and feeds our soul.
Yeah, I think a bit of difference gives it a bit more character, fully agree. Recall is more about being in the transition between two different songs when performing, and needing to get to a start point that works with what you're transitioning to. For jamming, it's fun to spend 2-3 minutes finding a sound that works, but for performance or recording, it gets really tiresome to manually patch 20 cables on the fly.
But as you say, no right or wrong, we all do things under different circumstances and contexts, and what works for someone is wrong for another, and all that :)
I think this is a reasonable article and I do understand the perspective. It's a normal sentiment from those witnessing a craft that they have invested time in mastering become partially automated. However obviously the big picture (as the article alludes to) is that none of us are paid to write software, we're paid to solve problems as efficiently as possible and these tools, used wisely, can massively help with that.
> we're paid to solve problems as efficiently as possible and these tools, used wisely, can massively help with that.
They can also help reduce the number of devs a company needs to operate and arguably reduce the skill level needed to generate software changes that provide business value. While it is true that these tools could lead to companies keeping the staff count and increasing productivity, the reality is that the productivity increase from these tools isn't big enough to offset the big dev team expense while the tools are argubly increasing productivity high enough that you can do more with less.
And yes, companies laying off folks because AI might suffer in the next decade or two. But that won't save _you_ from being unemployed _now_ and struggling to find a role until the tide shifts. The market can afford to remain irrational far longer than your average dev can afford to remain unemployed.
Contrast is worse, it's a trade off. For me in most environments there will be bright reflections on the glossy screens, even indoors - your brain actually does a good job of ignoring them to the point you're not even conciously aware most of the time, but when you smear them out with the nano texture display it's just way more comfortable to look at, for me at least.
If I was focused on watching movies, or grading photos in a dark room then glossy would be the way to go.
Depends a lot on the screen and environment in my experience.
A glossy Apple Studio Display or iMac (both of which have a decent antiglare treatment despite being glossy) in an office setting for example isn’t too bad.
On the other hand, your average touchscreen laptop (which are always almost-mirrors with no hint of treatment, for some reason) with screen angled up slightly and overhead fluorescent tube lighting or a skylight on the other hand? Borderline useless if the screen isn’t bright enough to outshine the strong glare covering 40-60%+ of the screen.
I really like the nano texture display. But I occasionally like to dabble with my camera, sometimes pushing things around in editing apps and after seeing how an unedited photo rendered on a coworkers macbook with the nano texture, I sadly went for the gloss on my own purchase. It's not a huge difference and probably not even a dealbreaker to most, but it's already hard enough to find a best fit for varied color spaces out there who I share photos with and I didn't want to add another variable to the equation. When you're trying to deal with shadows it does get annoying when you're nerfed even just a little bit trying to determine how much detail is in the shadows.
It's admittedly a very obvious tradeoff, but part of me was hoping some new magic existed with the nano texture since everyone was ranting and raving about it on its release. Figured maybe just maybe it could've given me the best of both worlds. Nope. As it turns out people who make "review videos" often times don't know what the hell they're talking about. I've learned to tune those out and just source my purchasing info from people who do things im also interested in doing. In practical application it was an easy decision to make. For my use case glossy display just made more sense, I dork with photos/videos many times a year whereas I sit outside in the sun with my laptop maybe once a year if I'm traveling. And even in those scenarios the glossy display is fine.
The video covers their DSP56300 work briefly in the introduction, but the main topic is that of reverse engineering the undocumented Toshiba DSP used in Roland JP-8000.
Yeah, while I mostly agree with the sentiment, I don't actually recognise any of the behaviours described in this article. It does sound like the behavioural traits of a certain subsection of certain generations, who's expectations and norms have been warped by overuse of social media. It all sounds incredibly exhausting and I genuinely feel sorry for those growing up in this climate.
reply