[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/g/ - Technology


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


File: images (17).jpg (34 KB, 628x488)
34 KB JPG
>rtx spark
what the fuck do gemeral availability mini pcs need qsfp for? and why do they cost more than a top spec mac?
>>
shut up and consume product
>>
>>108966793
Mini PCs won.
>>
File: 1779639262005100.jpg (71 KB, 1000x917)
71 KB JPG
>>108966793
>qsfp
unfathomably based
>>
I can't afford anything anyway
>>
>>108966793
So you can daisy chain them together into an AI cluster with hundreds of GB of ram.
>>
>>108966793
so you can larp having a compute cluster
>>
>>108966793
A MacBook with that much RAM cost like $1-2000 more than a DGX at 4k. On top of that the Mac M chips will be much slower on inference.
>>
File: 1773873329172835.png (288 KB, 1409x1982)
288 KB PNG
>>108966793
why does a decade old budget cpu destroy a top spec mac?
>>
File: 1779919297639940.png (474 KB, 1798x1200)
474 KB PNG
>>108966793
Nvidia is an AI company, so their general availability mini PC's are for AI. The QSFP is for them to be able to form clusters to take advantage of all the RAM available to run huge models. Nvidia made a smart move by buying out Mellanox and developing their own networking solutions, they're basically utilizing standard Ethernet as a direct link between the machines, faster than Thunderbolt even. Meaning that you can get something like a Mikrotik CRS804 that they've likely designed specifically for this use case and chain together eight of these mini PC's.

Basically, if you're not into AI, it's not for you. Intel is gonna give you nice consumer level chips for your web browsing and vidya, while Nvidia will keep on AI'ing.
>>
There's no fucking available memory for this consumerslop it's purely a halo product.
>>
>>108968352
Was the binary compiled for arm or did you use amd64 to arm translation
>>
File: 1773017332204602.png (331 KB, 750x600)
331 KB PNG
>>108968432
Not x86's fault your currycorp decided to switch their entire product line to some jeet shit ISA that no one fucking uses to pinch pennies on production costs and requires some shit translation layer.
Maybe if crapple pajeets learned how to designed better laptops than $200 laptop makers in China, because even their walmart junk is better engineered than crapple currynigger trash that boils water when opening a webpage while PCs with the exact same CPU remained ice cold.
>>
>>108968703
I'm well aware that apple computers where a laughing stock before m1
>>
File: 1767862856711056.png (812 KB, 1671x1682)
812 KB PNG
>>108968756
crapple computers are an even bigger laughing stock after m1
>>
>>108966793
How is that not a hard-line adapter? That shit ain't a purse.
>>
>>108968410
there's atleast a million goycattle that will buy this according to the Steam hardware survey
>>
>to some jeet shit ISA that no one fucking uses to pinch pennies on production costs and requires some shit translation layer
nigger you're in a thread about an NVIDIA prosumer product that is using an ARM CPU



[Advertise on 4chan]

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.