Isn’t the whole point of these things the “bloated” (CI/CD, issue tracker, merge requests, mirroring, etc) part? Otherwise we’d all be using bare git repos over ssh (which works great btw!)
It’s like complaining about IDE bloat while not using a text editor. Or complaining there’s too many knives in a knife set instead of buying just the chef knife.
I’m not familiar with the topic but couldn’t they cut straight to the source and directly contact Corning? Or alternatively, one of those Chinese high end OLED knock offs? I’ve heard they’re basically less than 1 generation apart in terms of quality.
edit: alternatively, I assume all cables/connectors are standard. What’s preventing Jim next door from starting a group buy to manufacture replacement OLED screens/upgrade kits?
Starfields is one of the biggest games of 2023 – but it’s joined other recent games like Baldurs Gate 3 in being boycotted by conservatives because of the way it interacts with gender.
Don’t have a steam deck, but what I used to do on Linux when upgrading drives is just dd the old drive over. Sure it’s inefficient as hell, but with pcie 4.0 speeds You’re looking at 20 min to make a byte-for-byte copy of your old 1 tb drive, which you can then extend your partitions to make them bigger.
Like, I get comments from people telling me it’s weird I always try to peel potatoes like I am trying to make the worlds longest 1-piece potato peel. To me it feels way for efficient and fun to continu down a potato in 1 peel, while circling around it, instead of randomly scraping a hundred different pieces of peel off and...
I’m just going to put this information here: the use case for 46Gb WiFi is going to be extremely niche. There is nearly no legitimate use case where you can achieve that speed on your phone.
The problem here is that:
The majority of internet traffic is TCP
TCP protocol processing is atomic (i.e. your speed is bottlenecked by a single CPU)
The bottleneck is the receiver (i.e. downloader)
TCP is too complex for efficient receiver-side hardware offloads (i.e. can’t workaround this issue by adding more special hardware)
What does this mean?
Your connection speed on a wifi 7 device WILL be bottlenecked by your single-core CPU speed, even if you are doing absolutely nothing except transmitting data. This assumes you are only using a TCP single connection (e.g. downloading a file from a website). But that’s the majority of use cases unless you are running a server (in this case on your phone).
I haven’t checked what CPU the Pixel 8 uses. But my Pixel 7 has a Cortex A-78. I also don’t have the raw data handy for the 3Ghz A-78, but I do have data from the 2Ghz A-53 connected to a 100Gbps Ethernet NIC which is around 8-9Gbps. The A78 generally outperforms the A53 by 1.5x (At least that’s the characteristics on the Nvidia Bluefield DPUs). So we can assume 12-14Gbps max for a single connection with Wifi 7 running on a state-of-the-art ARM CPU.
That is still nowhere near 46Gbps. It’s like mounting a Vulcan Minigun on a bicycle.
To use the full wifi bandwidth, you would need to have multiple connections running on different cores. That’s also not including the switches/servers connected to the wifi AP. Unless you are running a Redis server on your phone, I see no reason why Wifi 7 would be needed unless the remaining hardware is upgraded significantly.
It’s overkill, but 50Gbps is not enough for a Public University these days. A single core switch in NYC can see an average of 2.6Tbps of traffic. You’ll probably need at least 200-400Gbps for an entire university.
Didn’t know one is in-order and the other is OoO. The A53 is still being used for new products by Nvidia in 2020 (Bluefield-2). So there must be some merit to it or Nvidia is cheaping out on stuff
The BlueField-3 uses the A78 and unfortunately I don’t have one to test. I’m basing everything I know based on conference talks. I do know apparently the A78 does not have working performance counters for perf which makes it a pain to debug.
That being said, a 2023 Mid-end Xeon gets you up to 60Gbps TCP single flow (100Gbps ConnectX-6 NIC) So maybe that’s a better comparison? Might need to account for all the other x86 optimizations
Also, I think the bottleneck for TCP processing is branching, not memory access. So I’m not sure if OoO execution would help much. Would the A78 have improved branch predictors?
It’s the AVX-2/AVX-512 instructions that have issues. In most cases unless you’re running a server CPU (or extremely recent consumer CPU) you’ll be fine.
Scary for HPC/AI? Yes. For most people? Not really.
I would say you’ll be fine. Most games don’t compile with avx-2 anyways since it’ll crash if you run it on something that doesn’t have them (which is a lot of CPUs) and AVX-512 is straight up only available on Xeons, Epyc and zen 4. Nobody is going to use that for consumer software.
The only game I can think of using AVX is a Skyrim mod for realistic physics, where the author provided binaries for AVX-2/AVX-512. So it won’t affect most compiled applications much since you need to compile with it first (which almost nobody does).
ELI5, or ELIAFYCSS (Explain like I’m a first year CS student): modern x86 CPUs have lots of optimized instructions for specific functionality. One of these is “vector instructions”, where the instruction is optimized for running the same function (e.g. matrix multiply add) on lots of data (e.g. 32 rows or 512 rows). These instructions were slowly added over time, so there are multiple “sets” of vector instructions like MMX, AVX, AVX-2, AVX-512, AMX…
While the names all sound different, the way how all these vector instructions work is similar: they store internal state in hidden registers that the programmer cannot access. So to the user (application programmer or compiler designer) it looks like a simple function that does what you need without having to micromanage registers. Neat, right?
Well, problem is somewhere along the lines someone found a bug: when using instructions from the AVX-2/AVX-512 sets, if you combine it with an incorrect ordering of branch instructions (aka JX, basically the if/else of assembly) you get to see what’s inside these hidden registers, including from different programs. Oops. So Charlie’s “Up, Up, Down, Down, Left, Right, Left, Right, B, B, A, A” using AVX/JX allows him to see what Alice’s “encrypt this zip file with this password” program is doing. Uh oh.
So, that sounds bad. But lets take a step back: how bad would this affect existing consumer devices (e.g. Non-Xeon, non-Epyc CPUs)?
Well good news: AVX-512 is not available on most Intel/AMD consumer CPUs until recently (13th gen/zen 4, and zen 4 isn’t affected). So 1) your CPU most likely doesn’t support it and 2) even if your CPU supports it most pre-compiled programs won’t use it because the program would crash on everyone else’s computer that doesn’t have AVX-512. AVX-512 is a non-issue unless you’re running Finite Element Analysis programs (LS-DYNA) for fun.
AVX-2 has a similar problem: while released in 2013, some low end CPUs (e.g. Intel Atom) didn’t have them for a long time (this year I think?). So most compiled programs wouldn’t compile with AVX-2 enabled. This means whatever game you are running now, you probably won’t see a performance drop after patching since your computer/program was never using the optimized vector instructions in the first place.
So, the affect on consumer devices is minimal. But what do you need to do to ensure that your PC is secure?
Three different ideas off the top of my head:
BIOS update. The CPU has a some low level firmware code called microcode which is included in the BIOS. The new patched version adds additional checks to ensure no data is leaked.
Update the microcode package in Linux. The microcode can also be loaded from the OS. If you have an up-to-date version of Intel-microcode here this would achieve the same as (1)
Re-compile everything without AVX-2/AVX-512. If you’re running something like Gentoo, you can simply tell GCC to not use AVX-2/AVX-512 regardless of whether your CPU supports it. As mentioned earlier the performance loss is probably going to be fine unless you’re doing some serious math (FEA/AI/etc) on your machine.
But my WiFi is just fine! (lemmy.dbzer0.com)
Harness launches Gitness, an open-source GitHub competitor (techcrunch.com)
Nearly 500 smartphone brands have left the market since 2017 (www.techspot.com)
Microsoft - keep your filthy hands off Valve, leak shows MSFT would buy Valve (www.gamingonlinux.com)
Valve: don’t expect a faster Steam Deck ‘in the next couple of years’ (www.theverge.com)
Starfield is the latest game to be boycotted by conservatives. This time because of pronouns (theconversation.com)
Starfields is one of the biggest games of 2023 – but it’s joined other recent games like Baldurs Gate 3 in being boycotted by conservatives because of the way it interacts with gender.
WD unveils massive 2TB SSD for Steam Deck and PC handheld owners (www.pcgamesn.com)
What's the most efficient way for peeling potatoes by hand? (sh.itjust.works)
Like, I get comments from people telling me it’s weird I always try to peel potatoes like I am trying to make the worlds longest 1-piece potato peel. To me it feels way for efficient and fun to continu down a potato in 1 peel, while circling around it, instead of randomly scraping a hundred different pieces of peel off and...
Google Pixel 8 and Pixel 8 Pro confirmed to lack Wi-Fi 7 connectivity (www.notebookcheck.net)
So according to an FCC filing the Pixel 8 and Pixel 8 Pro will support Wi-Fi 6E but not the newer Wi-Fi 7....
Intel 'Downfall': Severe flaw in billions of CPUs leaks passwords and much more (www.pcworld.com)
Oh no.
“We just lost 3TB of data on a SanDisk Extreme SSD” - The Verge (www.theverge.com)
Anyone else have a similar experience with one of these drives?
Me, trying to step into the world of linux (feddit.de)
Gnome is Rethinking Window Management (blogs.gnome.org)
I feel called out (sh.itjust.works)
What types of services are you not willing to self-host?
For example, something that is too complex for your comfort level, a security concern, or maybe your hardware can’t keep up with the service’s needs?
What is your favorite magic school?
Can be anything from Hogwarts (the most obvious one) to Scholomance, Brakebills, anything else you can think of. Would you attend it yourself?