I’ve got a 32 gig m1 MacBook Pro. how would I go about trying Gemma like you mentioned? Would it run at an acceptable speed, and what could I do? Coding?
It seems like this person is incredibly intelligent and resourceful. Why not just design a new PCB for this instead of modding one? Surely the component cost of 2 connectors and a little PCB could still be in the < $20 range and actually output good quality analog signals? $20 is the same as $3 if you’re also buying CRTs just to play Switch games. This is such a hobby level activity I’m sure people would love a kit + a guidebook, could even provide a 3d printed case.
Why is the EV better? Because electricity transmission is more efficient than gas? What about the losses in the electricity transmission and the batteries and the conversion to motive force in the motor? Is it way less than that 40%? And wouldn’t there be more than 0% losses because refinery -> power plant shipping?
I’m pro EV by the way, I just want to understand your point better. Being able to go all the way to transportation using clean energy is an obvious benefit of EVs. The “dirty electricity” angle is less obvious to me.
The US has no social safety net. Healthcare comes from your employer. Everything is centered around having a job. Opinions on AI diverge significantly and someone’s response to this question would be pivotal to me in a hiring role. The market is not great for job seekers. The hiring manager can wait for someone who aligns with their company’s perspective on this.
Yes, if you get a Toyota and maintain it, it would be expected to make it past 200k miles. They are by far the most reliable cars. Timing belt failures are only catastrophic for interference engines, and most cars use timing chains now, which have a much lower failure rate.
Got anything from your orchestrator you could share that’s usable by others? Sounds like how I’d like to work but is difficult to get going from scratch
Traffic is usually caused by adding inefficiencies across a system with little slack - someone brakes too hard or too early, and if all the cars are stacked up, that one brake event can ripple through hundreds of following cars, getting worse and worse because each person brakes more. Self driving cars can perfectly sync up and move like a train. Theoretically there could be no traffic on highways if all cars are self-driving. Rarely is a highway so full that there couldn’t be more cars (eg. The entrance ramps are backed up) which implies the issues are related to the driving flow and not the capacity of the street itself.
I have a fairly large house (2 story 3k sqft) with all cat5e. I iperf’d every run and they could all do 10gb negotiation and TCP, most of the runs could sustain very high UDP rates with low packet loss. There’s just one run (which is the one to the internet) which had a slightly higher UDP packet loss rate. So basically every run can do 10gb fine. Been running the whole network like this for a year. It’s been great! I just need a 10 gig capable NAS. My current one can only do 3.5 or so because it’s a usb 5gb/s which isn’t really 5 gb.
If I was running a physical business and I wrote down each person’s name and credit card number and the exact time and order they placed, that would be pretty invasive and “spying”. If I write down how many units I sold of each item per day, and the volume of transactions by credit card vs cash, it’s anonymized and I don’t think this would generally be considered “spying”, just normal business metrics. How’s the latter much different than anonymized product analytics?
Cloudflare already did that and it’s available now[1], although it’s billed as a “spiritual successor” and not a literal one (so probably not backwards compatible).
Does anyone believe this? Is it real? Windows laptops always claim crazy battery life figures but when I have the latest, nicest windows machine it ends up lasting like 3-4 hours every time, with a normal coding workload. My MacBooks in contrast will last 8+.
And what is using Confluence in the first place? Your MacBook Pro is faster than a supercomputer from 20 years ago. As we make compute cheaper, we find ways to use it that are less efficient in an absolute sense but more efficient for the end user. A graphical docs portal like Confluence is a hell of a lot easier to use than EMacs and SSH to edit plain text files on an 80 character terminal. But it uses thousands of times more compute.
It seems ridiculous right now because we don’t have hardware to accelerate the LLMs, but in 5 years this will be trivial to run.
This thread is about how a static text article loaded 500 megs in the background. How would someone prepare for that exactly? This is effectively malware as far as your bandwidth is concerned.
Contact me directly at lukepdev. The place my email is hosted is on live.com.