正直、全くわかない

Well, in this case, how unprofitable a company is. OpenAI hemorrhages money like no other.

No, no, you see, when you go to buy medication now, the hospital pays you! Thanks Mr. Trump!

Honestly, I would love a lil Japanese kei truck in addition to my daily driver vehicle. They seem like the perfect lil truck for dump runs and transporting materials from Home Depot or what have you.

Serfs up, dude!

I couldn’t find the answer easily myself and ended up asking AI, so take this with a significant grain of salt, but supposedly a 3mL pen would be around $145 without insurance. If anyone can find a better source, I’d be all ears.

At least you can buy a human for only $85.21!

Windows 12 will be a realtime AI fever dream, generated as the user uses it.

I’m guessing if they’re too broke to switch their tech stack to Windows 11, a mostly similar OS to 10, the costs to switch to an entirely new paradigm is completely out of the question.

Found the Canadian

Please…please don’t give them ideas.

Oh, you still think they know how to read graphs?

Well, I believe the scientific literature.

Riiight, but like, I can make some shit up about a fancy ass apple “The surface is mostly red, mottled with large swaths of not yet ripe green and yellow patches. It’s waxed, shiny skin reflects rays of a nearby lightsource behind me. There’s a slight bumpiness to it, almost like goosebumps, but not as pronounced, with darker spots at the apex of each peak. My mouth is watering.”

Meanwhile, if I try to picture an apple I actually don’t see shit because I have aphantasia, so…

Not everyone is like you.

Either that or aphantasia, a well documented phenomena, is a thing.

I dunno, train bubble has a nice ring tho…

I don’t think you need the disclaimer. This is Lemmy. We’re all autistic here.

A lot of the online tests start with a negative offset to account for ping.

They absolutely do not do this. Where are you getting this from? These tests runs fully within your local browser; there is no ping. Whatever it shows you is what it is. There will always be added latency due to your setup (computer, mouse, monitor, etc), but it should be fairly negligible. Perhaps some crappy tests incorporate a blanket offset for this reason?

I just made my own very simple reaction tester web app running 100% locally and I get the exact same results as Human Benchmark.

Perhaps there were delays induced by your testing setup? Different output, different input methods (mouse click vs button press vs tap, etc), and different computers can all influence the results. I regularly hover around 150-175ms, depending on the day, with pretty much any reaction time test and computer setup I’ve tried, however.

Edit: I just made my own extremely simple reaction time test to account for potential test weirdness, same result:

Of course a test like this does not compensate for all variables. Every piece of the chain is going to have an effect, from mouse, to your specific computer hardware, operating system, all the way to your display output.

That said, I think your delta of 30ms was just variance that would be reduced by averaging several results. The maximum added latency by using 60hz is 16.66ms, while the maximum average for 75hz is 13.33ms, a far cry from the 30ms you experienced.

Even 240hz, like my monitor, is 4.16ms, which can induce a maximum difference of 12.5ms compared to someone on 60hz in the worst case scenario.