ribit

18 ms·

91.
by ribit 3y ago
I think it should be as easy as not presenting a drawable if you detect that the previous frame is still rendering. Should be solvable by adding a single conditional guard to the command buffer completion handler. Never did that myself as I
92.
by ribit 3y ago
Metal allows you to present surfaces at exact times. You have access to the display refresh timing information and it's your job to synchronise your drawing rate to (potentially variable) display presentation interval. Vulkan presentat
93.
by ribit 3y ago
I am sure that bytecode representation of shaders will come in a future revision. It’s not a first priority. SPIR-V is a poor choice for portable shaders for reasons outlined by the WebGPU group on multiple occasions. And WebGPU shading lan
94.
by ribit 3y ago
> the Metal/Vulkan/OpenGL situation is unfortunate and I don't understand Apple's motivation there. I don't think it's hard to understand. Apple wanted an easy to use API that could be extended and updated e
95.
by ribit 3y ago
I am not convinced that Vulkan gives you more control. Metal is adaptive in the sense that it can manage some state and configuration for you, but that is strictly opt-out. You still get your manual memory management and low-level event syn
96.
by ribit 3y ago
Nonsense. Feature-wise, they are mostly equivalent (Metal has better support for GPU driven pipelines and shader authoring). Metal is much simpler and more flexible. The only way Vulkan is "better" if you measure lines of codes.
97.
by ribit 3y ago
Apple's strategy for cross-platform GPU is WebGPU, which they are actively involved in. OpenGL has been obsolete for a while and Apple has no interest in supporting Vulkan (for multiple reasons). The core philosophy of Vulkan — which i
98.
by ribit 3y ago
It’s a bit more complicated than this. Of course you can represent any information with any kind of coding schema but here the truck to find a schema that’s also reasonably efficient. As so often, this is about compression. In the context o
99.
by ribit 3y ago
The “only 120 words” is an intriguing idea (and illustrates just how far one can come with periphrasis alone), but if this were an actually actively spoken language the lexicon would likely quickly expand, first to include conventionalized
100.
by ribit 3y ago
Do this basically copies ARM’s Aarch64, but with a really awkward instruction encoding? Interesting move, Intel.
101.
by ribit 3y ago
And your comment might even be relevant if “left-leaning feminists” were a non-fringe political power threatening to take over legislative, executive, and judicial branches of power. We have real problems to deal with, not some imaginary “b
102.
by ribit 4y ago
Apple would certainly benefit from more transparency in these matters, no question. Decoupling Safari updates from OS updates and better communication/tracking of issues and release dates would be a big help At the same time, I don&#x2
103.
by ribit 4y ago
> Time Machine works with full files and uses hard links to link different versions They did change it recently. It uses snapshots now (I think from Big Sur on?)
104.
by ribit 4y ago
APFS does have snapshots and at least the local TM backups heavily use them (it takes a fraction of a second to perform a local TM backup). TM backups to disk also seem to be organised as snapshots and have become much much faster in the re
105.
by ribit 4y ago
Happy to hear that AMD takes these issues seriously. The power draw of current Intel and Nvidia chips is seriously getting out of hand. There is absolutely no justification for drawing 40-50 watts of power for rendering a website!
106.
by ribit 4y ago
For RAM that's around 3x slower. I mean, if it works, it works, but it's hardly something to boast about.
107.
by ribit 4y ago
I wouldn't be able to do it. I spend a significant part of my day in front of the computer and having to deal with a poor ergonomic of a budget tool would drive me very frustrated. I also have no desire to tinker with my computer or th
108.
by ribit 4y ago
What’s really interesting is that entire Chomskian syntax worldview is fairly pseudo-scientific in nature. Most of these papers are about taking an essentially Turing-complete computation system and tweaking it until it can solve a specific
109.
by ribit 4y ago
Sure, that's the most straightforward way to approach these things, but doesn't this mean your elements don't have persistent identity between two renders? Tracking elements between renders is what allows animations and other
110.
by ribit 4y ago
There are at least two reasonably popular frameworks that offer reactivity without virtual DOM: Svelte and Solid. And both of them are way more ergonomic than what is presented here. I believe VDOM-based approaches have one decisive advanta
111.
by ribit 4y ago
Are you saying that the world should just bend over to russian imperialism and merrily let them commit cultural genocide because global warming is bad? Cause I am at a loss what your comment is supposed to mean. If not for imperialism and l
112.
by ribit 4y ago
It’s interesting how pro-russian propaganda has massively spiked in the recent days. They are bringing pretty much every talking head with half a name to it they can get. Putin and his cronies must really feel the pressure. I have to say i
113.
by ribit 4y ago
There is a lot of credible motive for the fascist russia. First, make allies distrust each other, sabotaging their unity. Second, create panic among the EU population, hoping for increased pressure to lift the sanctions and stop support of
114.
by ribit 4y ago
They did say 20% faster CPU and 30% faster GPU than M1 series in the same thermal envelope. Intel doesn’t really seem to make any strides in low-power designs, they just throw more mid-power cores at the problem to achieve better multi core
115.
by ribit 4y ago
Performance will be comparable to top Zen4 mobile CPUs at 45W TDP, just that Apple will use 50-60% less power in single-core and 20-30% power in multi core. That’s about it.
116.
by ribit 4y ago
They just published an additional product introduction presentation on YouTube. I don’t think Apple did live events since Covid.
117.
by ribit 4y ago
Manufacturing such large volume of high-end OLED displays with parameters Apple needs is not economically feasible at the moment. Other brands can afford it because a) they ship much fewer OLED-equipped laptops than Apple does mini-LED equi
118.
by ribit 4y ago
I mean actually executing the operations at that precision with improved performance. Apple GPUs support both FP16 and FP32 as data types, but the ALU throughput for both is identical (my personal speculation is that ALUs are 32-bit only an
119.
by ribit 4y ago
It's pretty different alright. First, there is the tile size. For current crop of desktop GPUs, tiling is primarily about cache locality (if you keep your processing spatially local you are also less likely to trash caches), but they s
120.
by ribit 4y ago
Does Adreno really have a deferred mode? The documentation I could find only describes tiled immediate rendering. Edit: I just had another look, pretty sure this is standard Tile-Based Immediate Rendering. The documentation sometimes refers

More ›