Origami Cactus

You nailed it. I remember getting downvoted just earlier this week for saying that maybe Samsung knew what they were doing and should be at least given the benefit of the doubt and the chance to show us what they've been working on. I also saw that modem argument being thrown here and there, which doesn't make sense because the other Exynos like the one in the S23 Fe don't have any issue like that. For many, Exynos had to be bad no matter what.

Damn it... I ordered the S24U

My workplace just announced that they've signed a deal with Samsung. To stay on topic, it would make the S24U 230€ cheaper...
I'm... Perplexed...

The 5x camera on the S24 ultra is 50Mp, pixel size 0.7µm, F3.4, 115mm eq (real focal length 18.6mm), sensor size 1/2.52 (sony IMX854, 5.75 x 4.32 mm).
The 10x camera on the S23 ultra is 10mp, pixel size 1.12µm, F4.9, 230mm equiv (real focal length 27.2mm), sensor size is 1/3.52 (Sony IMX754, 4.3mm x 3.22mm)
 
The new S24 ultra at 10x will be 5.75 x 4.32 mm / 2 = 2.8mm x 2.16mm.
So the sensor area is about 2x smaller, 13.8cm2 vs 6cm2. But it will be 50/2/2 = 12.5mpix vs 10mpix on the old sensor.
 
The new lens is exactly 1 EV brighter, F3.4 vs F4.9, so it lets in exactly 2x more light, effectively counteracting the fact that the sensor collects about 2x less light due to size. So image quality at 10x should be about the same, if not a bit sharper due to more megapixels, and it can use a 2x higher shutterspeed in all situations, so it should give better results in lower light.
 
As far as higher zooms go, the extra few megapixels will help, but more difference should come from the new ISP and the better frame stacking methods the new phone can use.
 
All the other cameras are the same, so no need to do any calculations, they will just benefit from the new ISP and better AI processing, so images should look better, how much is a different question. I like the 4k 120fps option on the main lens too, and that you can use all lenses at 4k 60fps, instead of being locked into the single lens when starting to record.

I can put it in simpler terms or answer some questions about it lol. I mainly made the math in my own curiosity, but thought it can also be interesting to other photography nerds.

Simply put: Lens lets in 2x more light, but sensor area at 10x zoom is also about 2x smaller. Quality at 10x at higher should be a little better on the s24u thanks to 12.5mpix vs 10mpix, while the sensor area vs lens brightness cancel each other out basically.

This gives me hope:

You might not care, but other people might care if their phone locks up for several seconds.
It's not a good user experience.
 
Which features were available already that Samsung was "bragging" about?
Also, it is entirely possible that something exists in other phones but that are new to the Galaxy series. 
 
 
Yeah, bizarre arbitrary rules like "doesn't lock the phone up for 10 seconds because it results in a bad user experience" or "can't do it in real-time in a scenario where real-time is absolutely necessary, like during a phone call". Very arbitrary and bizarre...
I think we should wait and see which features get ported and which don't before complaining about how it's an arbitrary limitation. I mean, you do realize that you are complaining and attacking them for locking features away without even knowing which features are being restricted, right? You're complaining before even knowing what you are complaining about.
 
 
I find that VERY hard to believe.
 
Please note that the Xperia X Compact also had a significantly longer exposure time.

The S24+ and OnePlus 7 are pretty much the same size.
The S24+ is 0,8mm taller, 1,1mm wider, and 1,5mm thinner.
 
Volume wise the S24+ is 4,4% smaller.
 
https://www.phonearena.com/phones/size/OnePlus-7,Samsung-Galaxy-S24+/phones/11090,12114
 
 
I don't agree that the OnePlus 7 would be considered compact by today's standards. It's the same size as the + model of Samsung's phones.
If you think it's too big then the S23 (non-plus) is significantly smaller.
https://www.phonearena.com/phones/size/OnePlus-7,Samsung-Galaxy-S24/phones/11090,12113
 
I think it's best to remain open-minded about the Exynos 2400 until it is released. Chances are it will not be the best SoC in the world, but it shouldn't be bad either.
 
 
That's not the same thing either...
At this point I think you're deliberately misunderstanding so that you can keep having something to complain about.
And like I said, it doesn't really matter who did it first.

It's been used primarily for biometric authentication for at least a decade: https://en.wikipedia.org/wiki/Speaker_recognition
 
In phones for certain since about 2016, so if it worked on a 2016 phone I'd expect it would be no issue for a 2024 phone in computational terms. CNN inference isn't that computationally cheap either to be fair.
 
Interestingly enough it's been made obsolete as a biometric authentication method by AI voice synthesis.

Me trying to understand all this

The first few, maybe 3-4 generations of the pixel phones had the same exact cameras every year, but they also improved massively yearly too.
Image sensors really aren't the barrier in these age of platoed tech development, it is more about the image processing.
I guarantee the S24 pictures will come out better than the S22 ones in all situations, even if you aren't interested in AI, this is exactly where it helps.

The Galaxy S24+ is worth upgrading if you have an older model, since it brings back a 1440p display (not seen since the S20) and includes a larger battery on top of the other minor upgrades. But I will agree the regular model is a tough sell for many people.

Personally I'm very interested in the Exynos 2400. Rumors and older geekbench results indicate slightly worse single core performance and equal multicore performance against the 8G3, with decent power consumption and efficiency.
 
I know Exynos has been doing quite poorly for a while now, but Samsung has lately been putting a lot of work into their fabs and SoCs - and so far the results have been promising. I'm curious to see just how much their processes have stabilized & improved.
 
(edit) So far, looking at early benchmark data from hands-on reviews, this seems to be the general result:
Single-core: 5~10% behind 8G3
Multi-core: 5~10% behind 8G3
GPU: Better than 8G2/A17 Pro, worse than 8G3
 
Of course, we do need to wait for actual benchmark results from professional reviews to make a final judgement, but so far Exynos isn't doing too bad.

I think the hype around it it's largely over-promising.
 
The most post practical uses of AI is in assistive technologies (eg auto-captioning, text-to-speech, machine vision identification, early-warning systems in safety/weather/disaster-prevention systems)
 
But the AI we keep hearing about is generative AI. A type of AI that has no practical use, and only works by being trained on copyrighted materials, without any supervision or curation of the data ingress.
 
Generative AI is going nowhere. It's used for deepfakes, and there will eventually come a point that regulations come down the pipe and kill all viable commercial use of generative AI. This will either come as regulations on what content the AI can consume or liability for the content it generates, as well as the need to clearly identify materials that aren't AI generated from AI generated due to the erosion of trust caused by AI use in journalism.
 
Like the idea that a car will learn how to drive using cameras and AI alone is pretty foolish. What needs to come down the pipe for that is municipal traffic control systems (eg traffic lights, signs, and temporary diversions) being communicated to the car in real time, as the primary traffic control scheme, and using something similar to CBTC in train signaling to actually control the car spacing. This leaves the car and the driver only in control when municipal control is not available, and the cameras control used when there is a conflict between municipal control and driver actions. (Eg cameras indicate an obstacle when the municipal control indicates the lights are green.)
 
A lot of the over-hype again, is with the idea that AI might replace creatives. This is not going to happen, nobody is going to pay for machine-generated works that they can produce themselves. AI can't predict the future, it can't imagine, it can only hallucinate, and that hallucination is largely a product of it consuming poisonous discourse from the internet and losing it's own context. 
 

I guess we can't help ourselves but copy every stupid idea Apple has 😛
 
The AI stuff seems to be... little more than yet another marketing gimmick. Voice transcripts aren't new, tone change may be somewhat useful in some scenarios but it almost sounds like they expect you to be illiterate. The selection search is, well, just a slightly better reverse image search and OCR... likely with some baked in "sponsored" templates to better recognize products from brands that worked with samsung/google.
Getting xbone "TV" vibes lmao

With 7950X3D specifically it's necessary to have 1CCD asleep in games or you'll lose performance

There is no performance degradation.
 
There is physical degradation... but it heavily depends on the conditions the card was in. Same applies for cards used for gaming as well so it's not like mining will do extra "damage" to the card just because it was used for that purpose.
 
More often than not a mining card was heavily undervolted to save on power as power is a big cost for miners, that means the card was likely running a lot cooler But if there were many cards next to each other in a mining farm then the temps were likely not that much lower but still way within spec.
Usually mining puts heavy load on VRAM and not the GPU core itself so something like mining RTX 3090 is about the only card I can think of which may actually be a risky purchase simply from the fact that half of it's VRAM chips are on the backside with barely any cooling and these chips may have heavily degraded life span. By how much? Difficult to say, the card still may last you for years or it may break within a week.
 
edit:
 
I think it's also worth mentioning a thermal expansion and compression of materials and it's wear down effect. A mining card that constantly runs at 80 C for a year will likely be better off than a gaming card that keeps jumping between 30 C and 80 C many times a day for a year.
 
Only thing that is definitely degraded and affected are fan bearings so if you get a mining card the fans may be very loud or barely functioning and may need replacing.
 
 
A mining card that was used in bad conditions will likely be a risky purchase too, but usually miners take a good care of their cards... same goes for a gaming card in bad conditions though.

Google it, ltt did a video and I think GN. The miners often even undervolt the cards as its more profitable. It's the same performance but you'll prob want to change the thermal paste just to be safe. 

I wish for Nicky V to go solo, oh wait, he already did.
Then maybe Taran, oh, he also already went.
Surely Ivan hasn't left yet, oh, he is also gone.
 
Out of the still working there crew, James, Alex and Riley I guess.

I wish for Nicky V to go solo, oh wait, he already did.
Then maybe Taran, oh, he also already went.
Surely Ivan hasn't left yet, oh, he is also gone.
 
Out of the still working there crew, James, Alex and Riley I guess.

Posting just to drum up more drama.
 
Hiding behind a new account as well.  Strong work 🙂
 
 

As I understand some Mac SSDs are proprietary and you will need an adapter to make them work in other machines. However, I suggest you try diskpart in CMD first.
Open CMD as administrator, type in "diskpart", then:
Type "list disk". If you see these drives in the list that you plugged in, then:
Type in "select disk x" (x signifies the number of a drive, which you can see in the list, for example: select disk 2), then:
Type "clean", after that, if process completed successfully, type "create partition primary";
Then, you can assign a letter to a drive, so type "assign letter H".

Hope this helps.

Whats your definition of kids? I know an 11 year old who watches LTT. When I was 12 I started watching LTT.

I can't be the only one seeing it.

I just made the same mistake back when I was starting too, went looking for the best fans for my radiator, spent a lot of money, only to discover the different in temps was 1-2celsius, so all that money wasted for nothing. I could have gotten an extra 32gb of ram for that money, or a better motherboard, more storage, anything. Sometimes the best option is just to stay reasonable and invest that money wisely into something else. For the price of a few high end fans you can already get a 1tb GEN4 m.2 for example