zavertnik

(no spoiler) From a storytelling perspective, Half Life 2 Episode 2 (the last mainline Half-Life game which Half-Life 3 was expected to be the sequel for) sets up Half-Life 3 as a proper ending to the game's story and confirmed by Mark Laidlaw's release of Epistle 3. I can see why they wouldn't want to rush quickly to the exit but at the same time, HL2E2 left them without much wiggle room. This issue alone is addressed in Half-Life Alyx.

From Valve's perspective, the Half-Life property is pretty valuable and its served them well as a means of demonstrating the newest source engine updates with a ton of hype bootstrapped because it is Half-Life.

Despite it's popularity, Half-Life is a delicate property in the sense that the story hinges on a lot of unknowns that are not revealed straight forwardly, similarly to LOST, the lack of detail stokes that mystery to the audience's delight, because they know there will be a big payoff that explains these mysteries at the end of the tunnel. That's a lot of pressure for the writers room @ Valve, especially so considering the development process of the game engine/assets itself.

It works for Valve. It sucks for fans, but the alternative is ending the property early or extending the property far beyond its intended shelf life. As long as I live to see the end of the narrative, I'll be happy. If not, then valve pls fix.

And here I was in bliss with the 32k context increase 3 days ago. 128k context? Absolutely insane. It feels like now the bottle neck in GPT workflows is no longer GPT, but instead its the wallet!

Such an amazing time to be alive.

I have been waiting for this, and now that it is finally here, it genuinely feels like Christmas.

AI bridged the gap between my wildest ideas and my present capabilities. It took some time to figure out how to use it efficiently with the 8K token limit, but once I did, I was able to break down any problem into small enough parts for GPT.

The quadrupled context window changes everything. I cannot wait to continue building. I am vibrating with excitement.

To be fair, the iPhone is not a camera, its a smart phone with a camera. There is no question that a DSLR with an interchangeable lens will beat out iPhone.

> The source camera and lenses were good enough - and the look was amazing compared to camcorders - to achieve this.

I would bet that the camera and lenses used didn't do the heavy lifting there, the selection of the location, the time of day, and the direction from the DP did. An expensive lighting setup isn't a requirement for a good image but good lighting is. Whether its natural or artificial is up to the DP.

The minimum viable quality you need for a project depends on the needs of the shoot.

> Forgetting "shoots" and professional lighting, the iPhone isn't going to have the massive range of other equipment, and when you're spending literally thousands of pounds on studio time, lighting hire, operator costs, etc etc, are you really going to pick up and shoot on an iPhone when you've got a Black Magic or Sony or Canon at hand? Unless you're being paid by Apple?

Of course not! I don't think anyone thinks the iPhone 15 is replacing the main camera on a shoot with a proper budget. But could the iPhone 15 be an additional camera for specialty shots or experimentation? Absolutely. For instance, if I'm shooting an unscripted scene with two people talking alone and I can mount my phone for a passable wide shot that can roll the whole time, giving me flexibility to shoot both individuals without losing coverage, then that's a huge value add.

I think in cases like that and in cases where people want to experiment without buying into a camera system, the iPhone becomes elevated beyond being just a mobile camera.

> I get that it's viable but there really is a lot of road between viable and superlative.

Totally agree, however I don't think I argued that. From my perspective, a viable camera is a viable camera for a job. The iPhone 15 seems viable from my viewing and it lives in my pocket while every other viable camera does not. That to me is the impressive feat. iPhone not living up to a proper DSLR is expected.

A lot of people here are bringing up all of the expensive gear surrounding the iPhone that helped give it the professional look. This isn't unique to iPhone as a sensor.

I work in TV and have spent a great deal of time on set shooting. The only time I've ever relied entirely on the camera's sensor and lens for a high quality image is shooting outside, and even then that requires adjustments, such as facing away from the sun, moving away from contrasty shadows, ect.

Outside of documentaries, every other shoot will have a great deal of time, effort, and money spent on lighting and set design to elevate what is being shot. For scripted projects/films, an even smaller % of shots will be shot with the raw, available light/environment.

What Apple did with the iPhone 15 proved that the iPhone can be used in a professional setting without being the on set bottle neck. For example, a short film shoot which had it's budget blown entirely on renting an Alexa will be bottle necked by the lack of lighting for the scene. Similarly, a short film which had its budget blown entirely on renting lights will be bottle necked if its shot on an iPhone 4.

The goal is balance and for smaller productions, that balance is found in budgeting. If anyone on set has an iPhone 15 Pro in their pocket, the shoot suddenly has a viable second camera-- maybe its not good enough for the entire shoot, but its surely going to be good enough as a B-Cam or even as an A-Cam in certain scenarios where a smaller form factor is required to get the shot.

I don't think Apple is sugar coating their demonstration here with all the expensive toys being used in parallel with the iPhone. The use of these tools in parallel with the iPhone IS the demonstration.

Like any good video, if its shot correctly and edited correctly, you won't have an easy time visually identifying what sensor is being used.

I think the best comparison for the iPhone is looking at DSLRs that do not shoot RAW but do shoot in Log. The workflow out of iPhone compared to say, an a7s, should be relatively similar, aside from the differences in sensor capabilities such as color depth, dynamic range, ect.

I can see a lot of potential use of the iPhone 15 as on hand as a "deploy anytime, anywhere" B-Cam. For the editor, having that B-Cam to cut away to is invaluable and for small shoots or tight schedules, slapping the iPhone 15 on a cheap tripod is at the right level of convenience to give that a shot.

I've had similar luck with peanut oil too!

I love movie theater popcorn so I often try to get that exact experience at home. The biggest leap in quality for me is using a tall pot and a colander on top, it achieves the same goal as your paper towel between the lid-- getting that excess moisture out of the pot really makes all the difference!

> The question that I think tells the kind of potential that ML has for music is, how good is written language in describing music?

If you took a DAW project file of a song.wav that was completely written and produced digitally using virtual instruments and compiled all of the parameters a user had to set to achieve their output.wav into a .csv file, you may be surprised to see (1) how few parameters were used (2) how often those parameters are unchanged from their defaults and (3) the amount of those parameters that would be expected in any other project file.

When you break it down, you really only have 6 layers to parse, all of which are dynamic but within a relatively small and consistent sandbox, at least relative to image generation.

1. composition layer - the midi or notes of the song. 2. arrangement layer - the selection of instruments used in the song and the division of the song's midi to the song's respective instruments. 3. instrument layer - the parameters of each instrument, such as a synth path or a virtual piano's room setting. 4. post processing layer - the effects placed on the output of each instrument, such as reverb, compression, delay, ect. 5. mixing layer - the volume of each instrument + post processing channel 6. mastering layer - processing on the master track

All of these things are more or less standardized. Developers always add their own flair (read: custom parameters) for their plugins, but they can be decompiled to be a composition of each of these layer's fundamental parameters. All these parameters + the midi of a song would be a few kb

I feel like a LLM trained on the parameter sets which interacts with the software used to manipulate these layers could really produce amazing tools and open the door to writing high quality songs to everyone, just as other AI products have opened so many similar doors.

The DALL-E app for music, in my mind, probably wont be a text based description -> .wav output. Instead, it would be the generation of the elements of each layer with options that can be listened to in real time using whatever VSTs were used in training. When you ask ChatGPT to write a complex python script, it starts with an outline of all the methods in the script as placeholders and then takes you step by step until you're done, then you troubleshoot it or flesh it out. The best part of a generative music like this is that it leaves the user really only with having to decide if something sounds good or not.

As a mostly musically illiterate producer myself, I've produced hundreds of songs and a few albums without ever really learning how to do anything other than manipulate the parameters. When I started learning to produce music I was 15 years old and knew nothing about music production. But, I was really good at was using computers and software so I learned to play the DAW, the plugins, and the sample packs. The only layer that I couldn't learn through learning software was the composition, the writing of the midi. Fortunately, the midi of a song becomes very easy to brute force over time, so I learned to brute force midi. Once I became efficient with my workflow, producing music became a task of "make this idea sound good." And without ever really feeling like I was a musician or composer, this became an enormous passion and outlet for me that I did every day for a decade.

I was able to do this because at its core, all the mechanical parts of a song are simple machines and a song's quality is the way those machines are used together. As an outside, this feels like a workflow that would be very machine learning friendly. But I could be wrong!

I totally agree, even if I hate it. There are a lot of internet users out there who have this expectation that they are owed a certain level of access for free to these services. Those expectations aren't unfounded: most content on the internet has been, more or less, free to access. It was a good time, but surely we all knew that it couldn't last forever?

The era of ultra monetization we're being ushered into is new and obviously its going to create some discomfort.

But that doesn't change the reality that these expectations are totally irrational and unreasonable. We can argue that the information should be free, that content should be accessible, that the sharing and proliferation of information without barriers of entry benefit all of us in the short and long term. But those esoteric philosophies can't win a debate against a corporations desire or fiduciary responsibility to maximize profit.

Money is tight for everyone across the board, borrowed or not (except maybe for apple). Why wouldn't they amputate the minority of users who are only costing Youtube money? I can't think of a good argument against it, other than "I hate it".

People are enraged by Netflix cracking down on account sharing, people were enraged by article paywalls, and people are and will be enraged by Youtube blocking ad blockers.

Unlike Netflix and article paywalls, Youtube has no real viable competitor. I get why the NYT allows me to bypass their paywall with a browser extension: if they truly make it impossible for me to bypass the article, I'll be able to find some other website that has reposted or reworded it. I get why Netflix allows me to bypass their account sharing: if they make it impossible for me to view the content without paying, I can just torrent it.

But Youtube is uniquely different. The product there is not just the body of content but their ecosystem which contains essentially every video on the internet and all the connective tissue to traverse billions of minutes of video. I use an adblocker on Youtube, but if they lock me out of that option, I'll still use YouTube, because where else would I go? No where.

Maybe I won't go to Youtube as often, but they'll have effectively converted me from being an extremely active viewer getting a free ride to an infrequent semi-active viewer that generates revenue.

If you're in LA and need a computer to use or an environment with wifi and power and what not to trouble shoot this, you're welcome to use my apartment. I have a spare ATT phone if you need something to act as a intermediate device (if your son wants their phone back :p).

If I worked in tech I'd offer you access to a solution but this is all I've got. I hope this problem doesn't last long for you.

> I now subscribe to the WSJ. That way I can get the business and financial news I'm interested in by people who are educated in business and finance, rather than activism.

I don't think that's a fair characterization of NYT. I'm curious if you're putting NYT's reporting and NYT's op-eds into the same bucket.

> I let my NYT subscription lapse because their articles were clearly pushing an agenda rather than reporting the news.

There aren't many large publications that publish w/o an angle, whether that effort is in the name of truthiness or personal gain is difficult to discern.

In my opinion, the only solution to the 'getting unbiased news consistently' problem is solved with news aggregations. Looking at the diff between the major publications on any specific story may not reveal which angle/agenda is the most true, but it will reveal the variables that are being used to steer the ship.

> Once a creator becomes popular, they often become sponsored, and a lot of the gear they use (which also sets them apart from regular creators) is given to them through sponsorships.

I'd wager in terms of 'gaming gear', this could be true. But I don't necessarily believe that these bigger streamers are going w/o an equipment bill at some stage of their career.

> This kind of ecosystem completely destroys the value of independent artists and favors big industry every time.

The problem I have with this is that this is just a criticism of capitalism. Independent artists all share a goal: being recognized for their skills, and if they choose to, to make money off of it.

> These platforms don't have separate categories for content creators based on their funding and content to fairly display everyone as equals,

I'd argue that it's completely okay to say that not all creators create equally good content.

> or among people operating on each level of ability/funding, why?

I think we should divorce this idea that good content comes from good funding. Many of these popular streamers are still using shitty webcams in their totally normal bedroom, being lit entirely by their monitor, using a headset microphone.

When the cost of getting the "premium" look boils down to any a7 mirrorless camera with a lens that has a minimum f/ of 2.8, we're not talking big bucks here.

> because platforms favor proven earners and big industry -- Thy also don't want unknown creators to potentially replace chosen creators as well in a lot of cases.

I think this is backwards. The platform favors popular streamers, who by virtue of being popular, make money.

> Money/Profit drives platforms, not content, and not a concern for discovering new talent, but platforms want independent content contributed without providing opportunity and without paying independent creators for it (of course). Platforms create the dream of discovery and work hard to lure independent creators with the ideal that they can grow on them purely as fish bait to cover the fact that most already engineer their featured content heavily to be pre-determined and fixed.

But this dream of discovery isn't dead. It's just not the first thing you see when you go to Twitch. I think that makes sense, no?

> Don't get me wrong, occasionally new talent slips through, but it's very rare in frequency.

This is true, and I totally agree. There are a lot of people who will not get the spotlight. There are a lot of people who are extremely talented who will not get the spotlight.

This is because it takes more than just a good visual aesthetic/quality to get popular on these services. You have to be a good marketer, you have to mesh within a community, and you have to know how to build an audience vs. knowing how to entertain an audience.

This dilemma is no different than all the people who want to be actors, singers, writers, ect.

> If you look at black hat forums, there are hundreds of people paying and faking their way in order to become popular and successful, and even some scammers are making more money off of fake music and streaming than really talented independent artists in music and video production. The current state of social media algorithms is not based on discovering talent, it is based on making companies money, and in covertly limiting user options for choices in featured content to make platform revenue stable.

And we see this fail time and time again, IMO. If you buy 100k followers, you're gonna have to buy a ton of likes every time you tweet/post to sustain the image of success. I don't think people who do this often succeed, but more power to the ones who do because I think this is a totally practical and fair way to get into an industry.

> A better way to look at the process of creating things as an independent artist is to focus on building locally, and in not trying to emulate nor compete with big industry, because the burnout is real, and the chances of growing, lasting, and thriving in massive creator-based ecosystem now are very very slim. Create your own web site, and own your own platform, because if you're an unknown or independent artist, major platforms aren't promoting you as well as a more dedicated online presence can.

I hate to say it, but I think deep down, people do not want to explore unknown creators. That's why your chances are slim. I don't want to watch 10 streams of someone who has not learned how to entertain an audience (which is the reason why I'd be watching) to maybe come across 1 stream that does entertain me, only to find out that this person streams every "once in a while".

I think you brought up a lot of valuable thoughts about this streaming industry, but I hesitate to agree with a lot of what you said simply on the premise that any given job in which you have more people wishing to hold a position than positions available will inevitably lead to a lot of people being left behind, whether it be due to poor luck, underdeveloped abilities, ect.

What was censored? The mods here on HN are good people, if you had a comment unfairly removed, I'm sure it was in error rather than malice.

The torrent is on her website, a few people tweeted that they finally had seeders, but I'm unsure if they got the entire archive.

I'm currently trying to download it now, but the torrent file is so large that it's crashing most torrent software (pico, deluge, webtorrent) I throw at it, on 2 machines!