This dev bought a MacBook Neo just to make sure his game (Petafab) runs flawlessly on low-end hardware
5% battery in 30m (36.5Wh) means ~3.65W average draw 🤯 (insane even with margins)
And Hitting locked 60fps with large-scale physics & destruction is a technical masterclass
5% battery in 30m (36.5Wh) means ~3.65W average draw 🤯 (insane even with margins)
And Hitting locked 60fps with large-scale physics & destruction is a technical masterclass
MacBook Neo: Played Petafab for 30 minutes straight. 60 fps locked. No dips. Only 5% battery used. Device stays cool. Feels like magic.
Spent so much time optimizing everything, but the real reason for lag was Steam overlay fence (fixed now).
3
1
70
@LelacAnnecy Even in Blender's official benchmarks, results more than twice as slow are observed.
Because I know things like this, I don't jump to the conclusion that it's wrong. The more you know, the more you realize that all kinds of oddities are possible.
Because I know things like this, I don't jump to the conclusion that it's wrong. The more you know, the more you realize that all kinds of oddities are possible.
2
0
3
Apple chips pulling off that performance at just 36W, 15W, and 9W is borderline wizardry. Especially in path tracing renders where Nvidia used to dominate.
can't explain that with just node differences.
And M6 isn't even on the chart, which is gonna be even more efficient
can't explain that with just node differences.
And M6 isn't even on the chart, which is gonna be even more efficient
23
24
695
RTX Spark delivers RTX 3060-level gaming perf, Told ya
Honestly, even worse than expected
A $2,599 laptop losing to a $1,200 one in gaming is gonna be a tough sell for gamers.
Honestly, even worse than expected
A $2,599 laptop losing to a $1,200 one in gaming is gonna be a tough sell for gamers.
Don’t expect N1/N1X to be for gaming, it’s basically the laptop version of the DGX Spark. The compute performance is decent, but the memory bandwidth is only 273 GB/s. The RTX 5060 mobile does 384 GB/s.
It’ll probably be worse for gaming than a $1600 5070 laptop. And considering the DGX Spark was $4K, it’ll probably be pretty expensive too.
21
15
259
@thenewpersone FR, thought we'd see something like this within 3 years, but these idiots at LG have no clue what to make unless Apple tells them.
Even the door lock they put out for 2025 looks like complete crap.
Do they just have zero taste?
Even the door lock they put out for 2025 looks like complete crap.
Do they just have zero taste?
1
0
3
Been waiting for something like this.
Exactly, Apple. Stop just setting standards and protocols and waiting for third parties to make quality stuff, just collaborate like this, because so many companies have no clue how to make it unless you tell them.
Hope more door locks simple and sleek.
Exactly, Apple. Stop just setting standards and protocols and waiting for third parties to make quality stuff, just collaborate like this, because so many companies have no clue how to make it unless you tell them.
Hope more door locks simple and sleek.
7
3
51
@dev_zeroday0619 실제로 피해를 주는, 현실에서 벌어지는 범죄는 CVE와 관계 없는 경우가 대다수입니다.
애플은 사용자에게 정말로 치명적 피해를 줄 지점이 뭔지 맥에서 뼈저리게 깨닫고 iOS를 처음부터 저렇게 설계한 것이지요.
애플은 사용자에게 정말로 치명적 피해를 줄 지점이 뭔지 맥에서 뼈저리게 깨닫고 iOS를 처음부터 저렇게 설계한 것이지요.
2
2
3
Yep. And Google and Samsung are both trash at this, and Apple is no different. Seamless AI upscalers do exist, but they're way too heavy for on-device processing, so phones only get tiny, lightweight models.
It’s just sad how people can't tell AI-upscale from actual detail restoration
It’s just sad how people can't tell AI-upscale from actual detail restoration
2
0
3
@xslvrx333 Spot on. iPhones always had better actual resolving power than other sensors with the same resolution, so squeezing even more quality out of it is wild.
0
1
12
Ok, seeing Max Tech's test, the 18 Pro really is a massive jump in main camera quality over the 17 Pro!!
Never seen this much improvement when the sensor size and resolution are identical.
Why didn't Apple mention this??
Never seen this much improvement when the sensor size and resolution are identical.
Why didn't Apple mention this??
WTF? The iPhone 18 Pro resolution looks waaay better than the 17 Pro in 48MP. How did no other reviewers catch this?
Not an aperture thing either, it still looks better at the same f/1.8. Doesn't seem to be focus either, it's consistent across multiple shots.
25
25
794
The iQOO 16 results with the N2P Qualcomm 8EE6 dropped today, and it's pretty disappointing. Power efficiency gains in actual gaming seem way smaller than expected.
Qualcomm has benefited indirectly by sharing the process Apple uses, but it's an interesting choice to select N2P this time to gain an edge over Apple's chips.
5
5
69
The most exciting project in edge inference everyone should keep an eye on:
CPU: 1.9W / GPU: 0.4W
Running a 27B model at 20-40 tok/s on the ANE
while keeping SoC power in single digits
ON AN iPad CHIP!!!
CPU: 1.9W / GPU: 0.4W
Running a 27B model at 20-40 tok/s on the ANE
while keeping SoC power in single digits
ON AN iPad CHIP!!!
New update to our 27B model for Apple Neural Engine: ANEMLL Forge v0.2.
An ANE research project. Goal: match UD-IQ3_XXS at roughly the same weight size.
Video: Pi coding Tetris. 16-minute session at 48× speed.
github.com/Anemll/anemll-…
1/8
7
9
263