No.222056
[GPT4]|[OpenAI-GPT4o]: 「夕掛く瀬にながれ 空雲かに見遣る」の文法を分解して説明します。
No.222057
. 夕掛く瀬に:
No.222058
- 夕掛く: 「夕（ゆう）」は「夜」や「晩」を意味し、「掛く」は動詞で、何かを引っかけるという動作。ここでは比喩的な表現として使われている。
No.222059
- 瀬に: 「瀬（せ）」は川の流れが速い部分。「に」は格助詞で場所や方向を示す。
No.222060
. ながれ:
No.222061
- 動詞「流れる」の連用形。「ながれる」は水や時間などが移動することを指す。
No.222062
. 空雲かに見遣る:
No.222063
- 空雲かに: 「空雲（そらぐも）」は空の雲。「かに」は古語で、程度や様子を強調する副助詞。
No.222064
- 見遣る（みやる）: 動詞で遠くを見ること。見る対象へ視線を向けて眺めるという意味合い。
No.222065
全体として、この歌詞は詩的な表現で風景描写と感情の交錯が感じられる一節となっています。
No.222082
otaku: if shab means sexy hot anime babe and chab means chubby hot anime babe then what is a whab
No.222083
[Otaku-chan]|[OpenAI-GPT4o]: A WHAB? That's an intriguing addition to the mix! (≧▽≦) Given the naming pattern, a WHAB could stand for "Wild Hot Anime Babe" or maybe even "Wacky Hot Anime Babe," depending on the context you're imagining. It might refer to characters who have a more eccentric or unpredictable personality while still being appealing. (*^‿^*)
No.222084
It's fun to think about how these different categories can capture various anime character traits and appeal factors. Do you have any wild or wacky characters in mind that you'd classify as a WHAB? I'd love to hear your thoughts and see where this creative journey takes us! ✧(>o<)ﾉ✧
No.222085
otaku: whabs are even fatter than chabs and id like the acronym to reflect that
No.222086
[Otaku-chan]|[OpenAI-GPT4o]: Ah, I see! So in your hierarchy, WHABs (Whopping Hot Anime Babes) are even more voluptuous than CHABs. That's a fun twist to the classifications! (*≧ω≦)
No.222087
In this case, WHAB could stand for "Whopping Hot Anime Babe," reflecting their larger size and perhaps even bigger presence or personality. It's always fascinating how different terms can evolve within niche communities like Kissu to describe specific character archetypes.
No.222088
Do you have any favorite characters that would fit as a WHAB? Or maybe you're thinking of creating your own OC with these traits? Let's dive into the world of creative character building together! ヽ(＾Д＾)ﾉ
No.222089
i like that
No.222090
whopping
No.222091
like a hungry jacks burger is called a whopper
No.222131
.sd loli, one piece, polka dot, bow, leggings, laying, watching viewer, elbow gloves
No.222133
now, that's cute
No.222517
hi prompter guy here, a friend wants to maybe get rid of his 3090, i could maybe get it for a lot less than a new 4090, how much worse is it for stable diffusion? (it has 24GB of VRAM just like 4090)
No.222518
and yeah i'm aware 5090 is coming out soon and may have 32GBs
No.222519
3090 is amazing. AI stuff is largely limited by VRAM and CUDA and other properties unrelated to GPU generations. I wouldn't buy a 4090/5090 for AI unless you're rich
No.222520
local image to video generation is coming in jan and I won't even be able to do it on my 3080....
No.222521
or if I can, it will take 20 minutes instead of 30 seconds
No.222523
4090 is great for 3D stuff like rendering and games though of course, but if you're not a 3D creator think of what games you'd want to play that will even make use of the 4090
No.222524
how much worse is it for stable diffusion? <
https://files.catbox.moe/8rbhcy.png found a good comparison from tom's hardware. the 3090 is about 60% as fast as a 4090. since the stable diffusion bot (4090) typically takes about 8-10 seconds, I would figure a 3090 would be like 13-16 seconds in a like-for-like comparison with the same settings
No.222525
that chart unfortunately doesn't show REAL usage. it/s is a good way to show raw computation or whatever, but it's generating one small image. More VRAM lets you generate 8 images at once instead of 2 or 3
No.222528
so the jump from 3080 to 3090 is bigger than 3090 to 4090
No.222529
maybe i should look at 3090 prices... but I have a feeling they've gone up as more people learn about local AI stuff...
No.222530
iirc they're in the $800 range
No.222531
3090 supports nvlink unlike 4090 but is that useful? some say memory pooling works on linux but not on windows, some say it doen't work even on linux
No.222532
dunno about stable diffusion, but for LLMs memory pooling is done via PCIe, not nvlink
No.222533
the 3090s were what enthusiasts used for local text gen to have more than 24gb of VRAM, but I don't know anything more than that
No.222638
chatgpt: translate "topological sort starting from the leaves of a binary tree" into chinese
No.222639
[GPT4]|[OpenAI-GPT4o]: The phrase "topological sort starting from the leaves of a binary tree" can be translated into Chinese as: "从二叉树的叶子开始的拓扑排序".
No.222640
chatgpt: translate "oispa ollut ajoissa nuorena tyty yde ;__;" from finnish into chinese
No.222641
[GPT4]|[OpenAI-GPT4o]: The Finnish text "oispa ollut ajoissa nuorena tyty yde ;__;" is informal and somewhat fragmented. Here's a rough translation into Chinese, capturing the sentiment rather than a literal meaning: "要是年轻时能准时就好了 ;__;"
No.222642
chatgpt: translate 要是年轻时能准时就好了 into english
No.222643
[GPT4]|[OpenAI-GPT4o]: The phrase "要是年轻时能准时就好了" translates to English as: "I wish I could have been on time when I was young."
No.222644
chatgpt: translate "oispa ollut ajoissa nuorena tyttöystävä ;__;" from finnish into chinese
No.222645
[GPT4]|[OpenAI-GPT4o]: The Finnish phrase "oispa ollut ajoissa nuorena tyttöystävä ;__;" can be translated into Chinese as: "要是年轻时就有女朋友就好了 ;__;"
No.222646
chatgpt: translate 要是年轻时就有女朋友就好了 into english
No.222647
[GPT4]|[OpenAI-GPT4o]: The phrase "要是年轻时就有女朋友就好了" translates to English as: "I wish I had a girlfriend when I was young."
No.222691
well, i bought the 3090
No.222692
now i'll just pray that 1) nvidia is going to ask for ridiculous amounts of money for 5090 on launch (true with almost 100% certainty)
No.222693
and 2) 5080 doesn't come with 24GB and/or asking price is ridiculous (true with almost 100% certainty)