Earl Cameron

Web Dev

[Earl Cameron @monstercameron]

@slimbuck7 @willeastcott @playcanvas I meant fully destructible, brick by brick physics, not baseline hard body physics

[Earl Cameron @monstercameron]

@thesamuelnam Panel latency

[Earl Cameron @monstercameron]

@mweinbach The value for $ is too poor for this to be a gaming product.

[Earl Cameron @monstercameron]

@VadimYuryev Most AMD docs are just like you say, strix halo might be different because it's an MCM package and the c2c interconnects might be too slow.

[Earl Cameron @monstercameron]

@kimmonismus No this makes no sense. You are better off renting GPUs, you will never amortize the price for these workstations unless you are extremely lucky with your research

[Earl Cameron @monstercameron]

@kimmonismus Yes the $100k workstation is the more interesting drop, what are you smoking?

[Earl Cameron @monstercameron]

@mzuhair123 But the arm designed cores are better so who cares?

[Earl Cameron @monstercameron]

@WuMacus @PatrickMoorhead This assessment makes no sense. The arm cores are faster than Intel/amd and cuda is mostly decoupled from CPU arch.

[Earl Cameron @monstercameron]

@Daniel_Rubino Mini led is a horrible decision point for what will be a $5k plus laptop. It needed a 4k OLED with HDR1000 and VRR support.

[Earl Cameron @monstercameron]

@surface NGL mini-led was an idiotic decision point.

[Earl Cameron @monstercameron]

@Sentdex @llm_wizard Linus' middle finger still haunts them to this day. The joke is now they make more money from Linux than windows.

[Earl Cameron @monstercameron]

@MorePerfectUS I'm actually ok with this. Make sure it aligns with the public interest by force

[Earl Cameron @monstercameron]

@never_released Next gen product will have a grok macro as an NPU

[Earl Cameron @monstercameron]

@ValigurskyM @playcanvas Wow, realtime lighting seems interesting, death to the triangle but how are you handling the issue that lighting is based on the spherical harmonics? Although this is a shadow demo, just curious

[Earl Cameron @monstercameron]

@kimmonismus Does it overthink like all the Chinese models?

[Earl Cameron @monstercameron]

@rezoundous Agreed, Claude gives more usage for the same $

[Earl Cameron @monstercameron]

@GeForce_JacobF When can buy?

[Earl Cameron @monstercameron]

@SebAaltonen Or perhaps they can make it work better with WSL2

[Earl Cameron @monstercameron]

@lafaiel They don't know that the ARM cores have higher IPC nowadays, x86 is dead