All Activity

"I'm getting too old for this shit. I'm only two weeks away from my retirement." I'm gonna be one of those craft people that refuses to use AI. Damn the consequences. This all sucks. As a documentarian, even something supposedly innocuous like audio transcribing is causing more inadvertent issues than it's solving. It's really getting in my way holistically -- although it "feels" like it's helping in the moment. It divorces me from the nuances and intimacy of the material. How can I be expected to make anything meaningful to myself if I allow an algorithm to make crafting determinations? That's a recipe to being superficial. I don't think I want to be superficial. Unless I'm doing corporate bull shit for the pay day. But then, the corporations using AI can now (or soon) cut me out of that calculus anyway. Get off my lawn you damn machines.

Man. Think hard about that.

That's the specific reason, though, that they are that different. If BMD charge me $60 to upgrade to the latest major version every year and I am not interested in the new features that they added, I pay them $0. If they go to a subscription model and charge me $5/month, I pay them $60 every year and if they add new features that I don't care about, I still have to pay them $60 for it or I can't use the existing features that I need. Basically, a subscription model removes the impetus for innovation and for developing features that customers are interested in. Like if the new feature is "we added a stock footage catalog that you can also pay to access," I wouldn't want or care about that new feature. I haven't used stock footage before and don't expect to use it any time in the near future. So yeah, there's a world of difference between a subscription and an optional yearly cost for extra features (and this is more or less how Luminar have been doing things and I'm here for it)

I think this was to be expected, regarding rolling shutter. It's similar to the Ursa Mini Pro G2 clock speeds == RS improvement situation over the G1, but this time in reverse.

Anything is possible, but based on Grant's past remarks about subscription models as well as how things were phrased here, I'd bet the outcome will be in favor up license upgrades (which may end up being some tiered annual payment) vs monthly subscription. That might not be all that different but the key is that with subscription, they cut off your service if you don't pay. Upgrade models, you stay at what version you want and aren't punished for not paying for the service at any particular point. Either way it may hurt the brand; A lot of what makes Resolve attractive is that is has been both freely accessible and stable for the most part. Even the more casual crowd that uses software like CapCut for social edits have found Resolve attractive after they moved to $10/mo subscription. Yeah, perhaps it may be a la carte for the AI tools. We'll see.

If it's subscription, I think there will be rioting in the streets and a lot of users migrating to Final Cut... including me. On the other hand, given that I paid like $200 for my license like 5-8 years ago and have been receiving free upgrades ever since then - and now I have a second license that came with my camera, if they were to ask for $50/year for license upgrades, I wouldn't be upset. After all, if Resolve either makes no money or loses money, they have no incentive as a business to keep supporting it. I'm not upset about that. If, however, they want me to hand them $20/month to use it, I'm done.

I remember Grant saying the same thing in these presentations in years previous. And I remember looking for anything that explicitly stated "free upgrades for life included" and could never find it. People just assume this because that's how it's been so far. I would hate a subscription model. Paying per major version would be fine, I would most likely skip a generation or two before feeling a need to upgrade. Same as I did with Adobe before their switch to subscription. Assuming the AI stuff is the major cost driver that would cause a switch to a different model, it would seem fairer to attach the cost more to the AI stuff- you get basic models/capabilities with the regular Studio purchase, then can pay to add-on better models or faster processing etc.

These are the numbers provided by the Blackmagic employee in the forum thread that I linked above. 8K 3:2 Open Gate 13.51 8K 16:9 11.61 8K 17:9 10.89 8K 2.4:1 8.59 8K 6:5 13.51

How to make someone addicted to drugs: first I will gift you drugs and then when you can't do without them I will ask you for money. Another clear example of enshittification (https://en.wikipedia.org/wiki/Enshittification)

Yeah If it becomes subscription based, and it is in the Adobe price range. I might jump ship, as photoshop and Lightroom would be nice to have. We will see. I really hate the adobe subscription model.

02:37:32 "We'll probably eventually charge some kind of upgrade for this." Looks like we can expect to pay for major version upgrades in the future. Still better than a subscription, thankfully.

12ms is pretty acceptable. I sort of suspected it would be a bit slower.

Just watched the first episode, it feels a little a bit gimmicky to me. Sometimes the lighting is a bit boring, as that is normal as you can't lock off shots like you do on static shots. I rather have something more stylistic. With cuts between ECU's, wides, medium shots, cutting between characters to build suspense etc. Might be a bit more classic approach, and maybe in my head that works better. Never the less they did an fantastic job nailing an entire episode in one shot. (The studio episode 2 is also done in similar style)

That 12k sensor does have some mojo going on. As good as Alexa? doubt. But the Alexa is soooo heavy + lowlight is better on the pyxis.

All I want to know is if the presentation has finished yet?

Haters of rolling shutter, weep with sadness. While the 8K readout speeds that the the Blackmagic employee posted in this thread are still very good for an 8K camera, they are about 2x the speeds from the Pyxis. For me, I wouldn't care much about ~12ms readout in 8K 16:9 mode and would barely see a difference from the ~6ms of the UC12K, but some people seem really upset by even a small change in RS. I suppose that means that the 12k readout times will be about 24ms which is still totally fine for most things, but I suppose if they send one to Gerald to review, he'll spend half the video whining about it. https://forum.blackmagicdesign.com/viewtopic.php?f=2&t=218593&start=50

If you can wait 2-3 months after it ships, you'll find it barely used for that much or even a bit less. That's the nice thing with YouTube-popular cameras these days.

Had to listen to the livestream while on the road for work. Man oh man does Grant love his broadcast stuff. I drove my full nearly two hour trip before he even hinted at cameras lol. Anyways, the Pyxis 12K is super exciting to me. It makes sense that its price is exactly halfway between the Pyxis 6K and Cine 12K…. But $4500 USD would have felt worlds better to me haha.

Well that's just great! Off to watch the video!

It is the S5ii as far as I’m aware. I’ve pulled this from a demo of the new functionality and shows switchable contrast and phase detect in the UI. Full video is here Yeah I think it will become clearer what BM’s intentions actually are for the 12K when they are doing demos and interviews at the show itself.

Even if the sensor has PDAF as an option, there's no guarantee that they bought a variant of it that has it enabled. Someone somewhere else mentioned that it's the same sensor as is used in the S1H which has autofocus, but not PDAF. Any BMD didn't say that they were giving their cameras PDAF that I saw - just continuous AF. It's entirely possible that they're rolling out contrast-based AF across the board. The fact that Petty said it would only work with L-mount lenses since they give distance information would suggest PDAF, though. Petty did strongly imply, however, that it would be coming to other cameras in time - he made some comment about how it was really important for large format sensors. Only time will tell, though - for now, Focus Pro gives me something like decent AF, though with all of the annoyances that come with doing focus based on a second camera that sits under the other camera.

Same as I said about the UC12k, its big and heavy, and eats batteries. Used Alexa classics are so cheap, I'd rephrase and say that I'd get a Pyxis over an Alexa because cost isn't an issue.

Mmm…In terms of AF, it was only actually stated to be coming for the Cine 6K and then by extension, when that firmware is out of beta, for the PYXIS 6K. I’m wondering if a misguided conclusion has been leapt to here by me that it will come to the PYXIS 12K. The other two share the same Sony sensor and it’s one we know to have PDAF capabilities because it’s used in a number of other cameras. The 12K sensor though is all their own work so I don’t know if it actually does?

Exactly, the worst part of the P4K is the body, one of the best parts of the OG is the body. With a different comparison, I saw a video today of the BMMCC6K held up next to the Pyxis and I couldn't believe how big the FF 6K is... it looks bigger than a 1DX and other than form factor, not much smaller than the Pyxis. The Pyxis seems like the obvious choice between the two. I sure do, that's one thing I do know. Ha, I know that even more. I feel like there are so many good cameras out now that you almost can't go wrong, but like always, the best camera is right around the corner and we're just in a betweener stage. Maybe you just need the Z mount Komodo X... problem solved. Canon should just buy BM, and then I can hate them both together.

There’s dumb L mount adapters for most legacy mounts so just use electrical tape over the PYXIS end contacts to be on the safe side.