That sounds marginally less effective than handing out maps of China with appropriate sized holes in them to incels so that they could fuck China.
Miscellaneous News
- Thread starter bd popeye
- Start date
That sounds marginally less effective than handing out maps of China with appropriate sized holes in them to incels so that they could fuck China.
You are baad ... you desperately need your own DeepSeek server in your garage.
Interesting, i forgot about the stuff being discontinued because i never really heard people talk about it. I could only remembered the announcement because it was some sort of holy grail back then. But by today's standard the Optane stuff seems on par with modern day .m2 nvme ssd in the 5~8GB/s mark when it comes to bandwidth. That's too slow to replace memory, i though they were hyping it up because it was like 20~35GB/s or so.
When shit goes down non of it matters. You’d think that Trump won’t annex Canada? Even if he doesn’t they’ll still have extradition treaty with the US.
manqiangrexue
Brigadier
LOLOL Somebody tell this guy America's already on that drug. It is only by luring Chinese scientists that America is at all present at the edge of technology. Without them, it'd be just a bunch of drunken fratboy blacksmiths trying to hammer out some semiconductor chips.
Yes further development in NAND flash and in PCIE interface since Optane's production stopped have made it so M.2 NVME can now exceed Optane in terms of sequential read/write. However to this day Optane is still a lot faster than NAND in low queue depth random read/write. Here's the difference with my C drive (a Samsung 990 Pro) with the Optane based cache turned off and on:
Off:
On:
So when accessing lots of small files constantly, as an OS (and AI) does in it's operation an Optane cache makes it much faster.
With AI model it would be interest to test out. If it uses say 700GB of memory to fit the whole R1 671b model but at any one time it's only reading/writing to a small subset of that 700GB than a two tier memory with Optane as the bigger slower layer would work well for a "consumer grade" setup. But if it's constantly reading/writing the entire dataset all the time then Optane as memory would be very slow.
Last edited:
We will just hit back with a true international law that states no wage earning American or American aligned person will be able to have stable housing, healthcare, nor education. That way they'll permanently stay down.
Oh wait it's already becoming de facto reality because of the ongoing economic war.
US has a lot to catch up on in escalation.
PiSigma
"the engineer"
They have easier standards than this forum. Once again chinese standards are higher.BTW, I am looking for an AI agent to keep posting here so that I can become Senior Member soon. It is a shame I am stuck as Junior Member after so many years here. Heck, I was promoted from junior engineer to senior engineer in corporate America much faster than this forum
I don't think RFK Jr. will be confirmed as the secretary of health, but damn it feels good for someone to call out Bernie the sellout to his face live on TV.
Last edited: