PIVlab - particle image velocimetry (PIV) tool with GUI
PIVlab - particle image velocimetry (PIV) tool with GUI
2 weeks ago
mathworks.com
Multi-session/connection support for PGlite landing soon! 🚀We have a POC that passes almost all the Postgres regression tests:- Concurrent connections- Parallel scans- Locks, cancellation & notifyThis has been a major ask & we're getting close!Draft PR with details in 🧵
0:26
Multi-session/connection support for PGlite landing soon! 🚀We have a POC that passes almost all the Postgres regression tests:- Concurrent connections- Parallel scans- Locks, cancellation & notifyThis has been a major ask & we're getting close!Draft PR with details in 🧵
764 views2 weeks ago
x.comSam Willis
Elflein et al., "VGG-TTT: Offline Feed-Forward 3D Reconstruction at Scale"VGGT, but with Test-time training to compress KV space. I finally understood TTT today thanks to a collaborator. Very much reminds me of scene coordinate regression networks.
0:16
Elflein et al., "VGG-TTT: Offline Feed-Forward 3D Reconstruction at Scale"VGGT, but with Test-time training to compress KV space. I finally understood TTT today thanks to a collaborator. Very much reminds me of scene coordinate regression networks.
7.8K views1 week ago
x.comKwang Moo Yi
Some people ask me why I charge so much for my assets when AI can make a half-decent water shader these days.This example is a good demonstration of where your money goes, and why AI isn't going to make premium-level assets anytime soon.AI is phenomenal at solving focused, technical problems. Better than me for sure. But it's not as good at broad, ambiguous problems.In the video, I'm submerging a test grid of semi-transparent squares underwater using different blending methods.I need to make sur
0:08
Some people ask me why I charge so much for my assets when AI can make a half-decent water shader these days.This example is a good demonstration of where your money goes, and why AI isn't going to make premium-level assets anytime soon.AI is phenomenal at solving focused, technical problems. Better than me for sure. But it's not as good at broad, ambiguous problems.In the video, I'm submerging a test grid of semi-transparent squares underwater using different blending methods.I need to make sur
420 views2 weeks ago
x.comDan Greenheck
🚨 BREAKING: THE LARGEST IVERMECTIN CANCER STUDY IN HISTORY JUST DROPPED A BOMBSHELL! Nearly HALF of desperate cancer patients achieved COMPLETE REMISSION or dramatic TUMOR REGRESSION in just 6 months—while the Chemo Cartel scrambles to bury the truth!• Our groundbreaking research, now published in Anticancer Research, exposes what Big Pharma doesn’t want you to know: safe, accessible drugs delivering jaw-dropping results against the deadliest disease on Earth.• Share this NOW if you’re ready to
0:37
🚨 BREAKING: THE LARGEST IVERMECTIN CANCER STUDY IN HISTORY JUST DROPPED A BOMBSHELL! Nearly HALF of desperate cancer patients achieved COMPLETE REMISSION or dramatic TUMOR REGRESSION in just 6 months—while the Chemo Cartel scrambles to bury the truth!• Our groundbreaking research, now published in Anticancer Research, exposes what Big Pharma doesn’t want you to know: safe, accessible drugs delivering jaw-dropping results against the deadliest disease on Earth.• Share this NOW if you’re ready to
19.2K views2 weeks ago
x.comDr. Jeremy Scott
AN ANTHROPIC ENGINEER LEFT HIS LAPTOP CLOSED FOR 9H 18M WHILE FABLE 5 REWROTE 214 ARCHITECTURE FILES ON 8×H100S AND CAME BACK TO A ~6X IMPROVEMENT.no prompt tuning mid-run. no human review. one Haiku 4.5 sub-agent watching every move.the shift: you stop asking which model is smarter. you start asking which model survives being graded by another model that doesn't share its context.Opus 4.7 ran the same experiment without a grader. one small scalar win in hour one, then copy-paste the template fo
0:29
AN ANTHROPIC ENGINEER LEFT HIS LAPTOP CLOSED FOR 9H 18M WHILE FABLE 5 REWROTE 214 ARCHITECTURE FILES ON 8×H100S AND CAME BACK TO A ~6X IMPROVEMENT.no prompt tuning mid-run. no human review. one Haiku 4.5 sub-agent watching every move.the shift: you stop asking which model is smarter. you start asking which model survives being graded by another model that doesn't share its context.Opus 4.7 ran the same experiment without a grader. one small scalar win in hour one, then copy-paste the template fo
18.8K views2 weeks ago
x.comexpemilly
I'm excited to see what Logan Reddemann can do in Colorado (2nd round, pick 38) with a revamped front office and pitching development group. Strikes (4.7% BB) + velocity is a good plan for mitigating the effects of Coors, but it's dependent on Reddemann maintaining the improved velocity (t98.6, sat 95-97) he showed before missing the final six weeks of the season with arm fatigue. From a slightly low release (5.8 ft, -4.55 VAA), Reddemann's fastball should play effectively on the road but will l
0:10
I'm excited to see what Logan Reddemann can do in Colorado (2nd round, pick 38) with a revamped front office and pitching development group. Strikes (4.7% BB) + velocity is a good plan for mitigating the effects of Coors, but it's dependent on Reddemann maintaining the improved velocity (t98.6, sat 95-97) he showed before missing the final six weeks of the season with arm fatigue. From a slightly low release (5.8 ft, -4.55 VAA), Reddemann's fastball should play effectively on the road but will l
8.1K views2 weeks ago
x.comMatthew Knauer
0:28
10 run game riding a 10 game win streak man this regression is going to be incredible I genuinely cannot wait
1.4K views1 week ago
x.comFreddie🫡
1:15
What is up with this recent regression in Deltarune fans viewing Berdly as a purely joke character? All while downplaying him defending Noelle and his importance to her breaking point in the WR like that isn't just horrendously insincere and irony-poisoned.
2.5K views2 weeks ago
x.comkaleidi
See more
Static thumbnail place holder