Elliot Arledge ✓
@elliotarledge · 4 x RTX PRO 6000 · joined 25 Nov 2022
@infinity_ai_ https://t.co/82V5vFob7t | made the 12 hr CUDA course | https://t.co/2hzezpo5Hi | @shipfr8 alum
43 369Followers
345Following
16 626Posts total
1.8MViews on collected posts
โพสต์ล่าสุด
กำลังค้นหาโพสต์ใหม่…
gemini 3.8: Bottleneck is the 600-watt board power
ceiling.
126 views · 0 likes
· 0 reposts · 0 replies
Open on X →
Astra cooking delegating groks for execution while I'm on starlink wifi (coming back to canada! more timelapses coming) https://t.co/RVNar6Zrx2
496 views · 4 likes
· 1 reposts · 0 replies
Open on X →
astra reading gemini 3.8s slop https://t.co/3Lv1tlnMWG
513 views · 2 likes
· 0 reposts · 0 replies
Open on X →
the strat of buying all the coding plans across all providers (maybe 3 to 7 of them) and having a solid "delegate" skill for spreading token usage evenly is the more economical way to move faster.
761 views · 10 likes
· 0 reposts · 0 replies
Open on X →
Numbers (RTX PRO 6000 / 9950X3D, methods + harnesses in the repo):
> exact render, 1080p: 35.9 fps GPU, 4.5 CPU
> batched sim: 3.0M env-ticks/s at N=8192, 756k agent steps/s
The CPU one is bit-identical. The slow one is the oracle.
40.8K views · 135 likes
· 1 reposts · 1 replies
Open on X →
C, CUDA, no engine, no game jar at runtime. You bring your own copy of the game; every Mojang-derived input regenerates locally, byte-identical.
I'm excited to see any and all constructive criticism and what the community can do with this.
kernels + source + oracle mod:
40.1K views · 384 likes
· 3 reposts · 10 replies
Open on X →
Training mode throws the full rasterizer away. The policy sees a 64x36 block-id camera with depth and edge planes - right pane is the whole observation.
That is why thousands of worlds fit on one card. https://t.co/x29AxnQPoB
44.4K views · 227 likes
· 0 reposts · 1 replies
Open on X →
2,058 actions, one unbroken chain, nothing scripted: spawn empty-handed -> logs -> planks -> sticks -> crafting table -> wooden pickaxe -> stone -> coal.
This is the middle of it, unedited, real speed. https://t.co/mx3SGf6phh
50.7K views · 157 likes
· 0 reposts · 2 replies
Open on X →
Why exactness: it makes an RL sim you can trust. 23/23 recorded tapes replay with physics clean to 1e-9, CPU and CUDA bitwise-identical every tick (all the way to defeating the ender dragon).
Train here, replay the same actions in the real Java game, get the same result. There
52.3K views · 245 likes
· 0 reposts · 2 replies
Open on X →
Left is Mojang's game. Right is mine, replaying the same inputs.
Validated on terrain, physics, and pixels - the gates ship in the repo. Can you spot the difference? https://t.co/LeLwt572VT
85.9K views · 429 likes
· 4 reposts · 5 replies
Open on X →
The receipts: my inventory screen minus Minecraft's, amplified 80x so you can see anything at all.
112,796 pixels. 442 differ, each by 0.3% https://t.co/AjbaMrdO2h
62.2K views · 324 likes
· 1 reposts · 2 replies
Open on X →
Introducing Netherite!
Minecraft 1.11.2, rewritten from scratch in C and CUDA, bit-verified against the real game.
This is one trained agent playing in it - then the renderer it actually trains through - then 7,200 live worlds stepping in lockstep on one GPU. https://t.co/eg2gX
1.4M views · 4.8K likes
· 197 reposts · 160 replies
Open on X →
เทียบกับบัญชีขนาดเดียวกัน
4 โพสต์จาก 90 วันที่ผ่านมา เทียบกับช่วง 10K–100K ผู้ติดตาม ต่ำกว่าบัญชีรุ่นเดียวกันทั้งการเข้าถึงและการมีส่วนร่วม.
ยอดดูมัธยฐาน504บัญชีนี้924ค่ามัธยฐานของ 10K–100K
การเข้าถึง, %1.16%บัญชีนี้3.62%ค่ามัธยฐานของ 10K–100K
การมีส่วนร่วม, %0.70%บัญชีนี้1.52%ค่ามัธยฐานของ 10K–100K
| ตัวชี้วัด | บัญชีนี้ | ค่ามัธยฐานของ 10K–100K | อัตราส่วน |
|---|---|---|---|
| ยอดดูมัธยฐานต่อโพสต์ | 504 | 924 | 0.55× |
| การเข้าถึง (ยอดดู ÷ ผู้ติดตาม) | 1.16% | 3.62% | 0.32× |
| อัตราการมีส่วนร่วม | 0.70% | 1.52% | 0.46× |
บัญชีอื่นในช่วงนี้ → เปรียบเทียบกับบัญชีอื่น → ค่าอ้างอิงเหล่านี้คำนวณอย่างไร →
Growth & engagement
How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.
Views per post
1.4M29 Jul
62.2K
85.9K
52.3K
50.7K
44.4K
40.1K
40.8K
12.3K
7614 Sep
513
496
126
Last 13 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.
Engagement rate per post
0.37%29 Jul
0.53%
0.51%
0.48%
0.31%
0.52%
0.99%
0.34%
1.00%
1.31%4 Sep
0.39%
1.01%
0.00%
Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.
What the audience does
Likes69.2%6 815 in total
Reposts2.1%208 in total
Replies1.9%186 in total
Quotes0.8%76 in total
Bookmarks26.1%2 567 in total
Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.
The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.
บัญชีที่คล้ายกัน
Fiscalía San Luis Potosí
@FiscaliaSLP
86.7K followers
· 42.1K posts
Allister Heath ✓
@AllisterHeath
86.7K followers
· 12.1K posts
Andrés Repetto ✓
@andresrepetto
86.7K followers
· 77.4K posts
Matt LaFleur
@CoachMLaFleur
86.7K followers
· 35 posts
Cafe Coffee Day
@CafeCoffeeDay
86.7K followers
· 30.9K posts
レオンチャンネル ✓
@LEON1ch
86.7K followers
· 18.7K posts
kelsey mckinney
@mckinneykelsey
86.7K followers
· 6.2K posts
TechLead ✓
@techleadhd
86.7K followers
· 67 posts