Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
wtallis
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
28 ms
·
661.
▲
by
wtallis
1y ago
If they're running hardware that old, and it's causing them software compatibility problems, then we can infer that their infrastructure is unmaintained, because the cost of moving to newer hardware is so low that the cost of ne
662.
▲
by
wtallis
1y ago
Yes, a used laptop would be an upgrade from server hardware of that vintage, in performance and probably in reliability. If they're really using hardware that old, that is itself a big red flag that F-Droid's infrastructure is fra
663.
▲
by
wtallis
1y ago
> However at the same time, not even offering a fallback path in non-assembly? There's probably not any hand-written assembly at issue here, just a compiler told to target x86_64-v2. Among others, RHEL 9 and derivatives were built w
664.
▲
by
wtallis
1y ago
It seems quite implausible that F-Droid is actually running on hardware that predates those instruction set extensions. They're seeing wider adoption by default these days precisely because hardware which doesn't support them is g
665.
▲
by
wtallis
1y ago
Far better would be to tweak the time constants to your liking, so that you can use the full clock range of the chip, but constrain its sustained power draw for quiet and long battery life.
666.
▲
by
wtallis
1y ago
The Type-C DP input appears to be using a DP to LVDS converter, so the FPGA is already accepting LVDS, it just doesn't expose a connector for it. But for interfacing with laptops, these days it's eDP that matters, not LVDS. So the
667.
▲
by
wtallis
1y ago
> Apart from some models of Alder Lake That bit actually still applies. Intel may have branded the 14100F as Raptor Lake , but it is almost certainly Alder Lake silicon, just a higher speed bin of the 12100F. See https://
668.
▲
by
wtallis
1y ago
> So, when powered by AC power, schedule everything on P cores when possible, schedule processes that eat a lot of CPU on P cores, same for any process with a negative nice value. Even when plugged in, you may have thermal limitations. P
669.
▲
by
wtallis
1y ago
Intel's Arrow Lake platform launched in fall 2024 is the first to support CUDIMMs (clock redriver on each memory module) and as a result is the first desktop CPU to officially support 6400MT/s without overclocking (albeit only r
670.
▲
by
wtallis
1y ago
I think it's important to make a distinction between secure boot features that are local-only, and remote management features. The "Remote Device Management baked into firmware" claim above carries with it some pretty importa
671.
▲
by
wtallis
1y ago
The workloads that people care most about today that need high RAM capacity are workloads that also need very high memory bandwidth. Old server hardware from eBay doesn't do a good job of satisfying the bandwidth side of things.
672.
▲
by
wtallis
1y ago
> Does mystify me that x86 has a hard time matching even a mac mini pro on bandwidth, let alone the models with 2x or 4x the memory bandwidth. The market dynamics are pretty clear. Having that much memory bandwidth only makes sense if yo
673.
▲
by
wtallis
1y ago
I don't think it's fair to describe MDM as a firmware-level feature. I think it's entirely implemented and enforced within the environment of a booted macOS; the firmware isn't going to be bringing up a whole network sta
674.
▲
by
wtallis
1y ago
> Keep in mind the ram is inside the strix halo package, not something HP has control over. It's not in the package, it's on the motherboard spread around the SoC package: https://www.hp.com/content/dam
675.
▲
by
wtallis
1y ago
> Mac Mini with 32GB of RAM, 512 GB storage: $1,199.00 You're looking at the wrong Mac Mini. The model with the M4 Pro is the right comparison, on account of also having a 256-bit memory bus giving substantially higher bandwidth tha
676.
▲
by
wtallis
1y ago
With a dial-up connection, you really want a native app that only hits the network for functionality that fundamentally needs to be done over the network. And you want the app to be hitting stable APIs, rather than requiring weekly upda
677.
▲
by
wtallis
1y ago
Not possible anymore is many areas, where 2G and 3G networks have been shutdown to re-use spectrum for newer standards. The last time I was in a rural area with minimal signal strength, my phone was alternating between satellite-only messag
678.
▲
by
wtallis
1y ago
Is that an x86 translation issue, or a graphics issue? The chips Qualcomm has shipped so far for ARM PCs have bad graphics hardware and worse drivers. And that's the best case, where the ARM system is at least running Windows and nomin
679.
▲
by
wtallis
1y ago
You're glossing over the distinction between voting without ID, and registering to vote without having your citizenship confirmed.
680.
▲
by
wtallis
1y ago
4TB is what currently fits easily into the M.2 form factor, and 8TB has been possible for years without much extra expense in custom packaging. I'm quite confident that it is fair to say the "vast majority" of consumers
681.
▲
by
wtallis
1y ago
They're (over)due to start shipping miniPCs aimed at being AI/ML workstations, using a SoC that's a collaboration between Mediatek and NVIDIA, with 128GB of DRAM. They have not yet announced intentions to put those chips in l
682.
▲
by
wtallis
1y ago
Definitely don't think of enterprise vs consumer SSDs as good vs bad ; they're optimizing for different use cases. Enterprise SSDs get you power loss protection and firmware QA aimed at server workloads and operating systems, a
683.
▲
by
wtallis
1y ago
> But in a real world usage scenario where we subject a consumer SSD and an enterprise SSD to the same number of P/E cycles, would they have different data retention? Probably not, assuming they're using the same underlying med
684.
▲
by
wtallis
1y ago
The layers largely aren't different. The end result is a vertical string of (theoretically identical) NAND memory cells filling the holes that were etched through the stack, with contacts at the top and bottom. At the edge of the sta
685.
▲
by
wtallis
1y ago
The reports in that thread you linked to are clustered around several models that do have substantial shared technology: a DRAMless design using an in-house SanDisk controller, and difficulties arise with the NVMe Host Memory Buffer featu
686.
▲
by
wtallis
1y ago
> Does this serialise the disk io or does it use multiple lanes of the pci bus? That's not how PCIe lanes work. Regardless of how many PCIe lanes are used to connect the CPU to the HBA, only one packet will be traversing that link a
687.
▲
by
wtallis
1y ago
DRAM and NAND require entirely different fabs.
688.
▲
by
wtallis
1y ago
The whole reason why 3D NAND happened and higher layer counts enable cheaper, higher density storage is that it doesn't require an etch step per layer to build the memory cells. It's just deposit several dozen layers then etch h
689.
▲
by
wtallis
1y ago
You can also hide an entire application rather than minimizing specific windows. It'll unhide when you switch back with cmd-tab.
690.
▲
by
wtallis
1y ago
> Despite that, purists were unhappy that Firefox was doing literally anything at all with a third party That's a horribly dishonest explanation. The way that Pocket was integrated into the browser was obviously shady. Most clearl
More ›