I never got an issue with copilot, at least in Visual Studio (only one I use).
Make repetitive tasks like unit test a breeze. And very useful for debugging issues with tools like Entity Framework.
I switched to full Linux on my personal hardware though. I only use Windows for work.
The idea of that webcomic was basically “what if god was just a drunk overpowered loser who created things out of drunk stupor”. That’s but an episode of it, so without the context it is hard to understand.
LED do not have a 100% efficiency, and do produce waste heat. A lot less than an incandescence one, sure, but enough for that answer to be valid.
Well, maybe you’d better wait 10min instead of one, to make sure the led lightbulb heats enough, but still…
Same here, the Steam Deck changed my life 😆. Less AAA, more AA and indie games, especially at work.
There are not many cards you need these days, especially not that doesn’t have an USB equivalent. USB capture cards are now decent, same for Wi-Fi+Bluetooth ones (provided you buy one with deported antennas). Other than storage related ones (for moar m.2!, or for sff SAS ports), I don’t see that much uses these days.
Even my NAS, which uses a micro-ATX MB, only uses one slot on the 3 available. And all its 4 m.2 ports are used (2 for redundant system discs, one for an AI accelerators (for Frigate object detection), the last one being an old SSD used as ZFS cache for my main disk array (will probably be replaced by another AI accelerator once I find another use which would need one).
IGPs are very dependant on memory speed. Many uses gains from faster memory speed (but I can’t give you one out of memory), but most games gains more from memory latency than raw speed (with some exception, like Stellaris).
Usually, if your app is CPU heavy, you’ll gain from RAM speed.
More like for the price of 64GB.
I didn’t even register the change before it got pointed out, my brain automatically translated it to “y” 😂.
Strange how brain works at time.
Funny experiment to do, read a sentence in which all word are scrambled, except the first and last letter. You’ll be surprised to realise you still can read it thank to the brain automatically decoding the word from him memory of it.
And suddenly 90% of both X, Facebook and Reddit fell silent.
To be honnest, Nextcloud is kinda a piece of shit software. Every major version break something, an error in an extension can make it crash entirely without any failsafe preventing it. I already lost the database twice (in what… 5 major versions), and the only reason I still use it is because I’m too lazy to use anything not in the TrueNAS apps repository.
On the other hand, they already sell it so high I ended up buying mine without RAM. I found 2x16Go for 125€ vs 180€ sold by Framework 😅. Same speed, but I can’t compare CAS as Framework do not list them on their page.
Edit : some source claims DDR5 5600 C46 for those sold by Framework, which is the same as the ones I bought. So I consider myself lucky, despite the price hike (the same kit was 65-70€ last summer)
Exactly. That what I added in an edit minutes after posting my answer 😆. There are some way to add pressure (a pump, which is often used in high buildings), but it is rarely used for hot water.
I’m not a plumbing expert (I got an electronic degree before going for CS), but I can extrapolate from my own experience.
If it does a bit like electricity, it comes from how much water the main line can let through. If it is sufficiently sized, and can accept enough water for both your tap and shower, all you’ll see is a pressure fluctuations (it will be divided per open tap). But if it is not big enough, the water will go to the shorter route first, which depending on the configuration may be the tap or the shower.
At my old man home, we are in the former case, which mean he often used what this comic shows to make sure we finish our shower if it gets too long to his liking (he is an asshole sometimes 😂).
Meanwhile, at my current home, the boiler is just next to the shower, so the shower takes priority.
The biggest difference with electricity though, is that you can put far more electricity than the wiring is made for, which can cause fires, which you cannot with water. 😅
The good old “parallel vs serial” patterns. The later being cheaper, it is used everywhere but in those places where the owner specifically ask for parallel because of knowing its advantages.
The robot was trained on Russian social media videos.
You can now use thirds parties APIs for Passkey. I use ProtonPass on my part, it works great most of the time, but there are still some apps that have Google provider hard-coded.
For a normal user, 16G is still enough.
But for a power user like me, a Dev, with multiple IDE open, multiple browsers, a database manager with a dockerized DB, I’m basically at 96-98% RAM all the time.
But even then, if you use a local AI, prepare to loose multiple Gigs to it.
I already used more than 30 Gigs on one LLM (dolphin-mixtral:8x7B for the curious), when I tried testing it, and a more reasonable Mistral:7B still took me like 4Gigs just to run.
It always does. I remember buying my DDR4 RAM 200€, and two months later the same kit was 550€. Some month later it was back at 220€.
Removed by mod
Firefox vertical tabs are better IMO. At least you can close them when minimized, which Edge cannot.