Looking to buy a new laptop, and of the model I'd like (thinkpad carbon x1), it seems like I can choose one with or without an NPU.

My surface-level understanding is they make built in AI processing and a couple other niche uses more efficient. Flip side, they're maybe a marketing gimmick?

Price aside (price difference doesn't seem too great), I'm wondering: should I buy a model with an NPU to help potentially future proof? Are there potential downsides to an npu model? Upsides?

I know there are lots of technical people here, I'd love to hear your thoughts or experiences!

I have decent internet at home.

I have great internet at work.

Despite the speeds of those though, seemingly every website out there feels laggy and heavy. You click, you wait, you get a skeleton of the page, with different elements that rapidly pop in until you're staring at the full site. You see the little loading animation on the tab for one, two, three seconds. It isn't exactly "slow" by any means, but it's far from instantaneous either.

Clicking around the web these days feels like I'm playing a game with unignorable input lag.

And I get it. The modern web is complex. It's genuinely a miracle that this is possible in the first place, so I really shouldn't be complaining that the bits traveling through the internet from dozens of servers thousands of miles away aren't getting here immediately.

I get that high resolution screens require large images, and the ubiquity of video these days adds even more weight. I get that many websites are closer to applications than they are static pages.

I'm not trying to take away from the awesome magic that is our modern miracle of connectivity in the slightest, and I'm appreciative to all the people here who spend their livelihoods working on it. Y'all are awesome.

I'm just trying to say that, well, sometimes moving around on the web can drag. And when you've been using it for a long time, the dragging can get under your skin a little bit.

However, my real point lies not in the rest of the internet, but here. I'm talking about this "heavy web" baseline as a contrast for one of the things I love about Tildes:

it. is. so. snappy.

I click, and BAM, the page is there. Immediately.

It's sharp. It's crisp. It's no-nonsense. No waiting for elements to pop in. No subconsciously watching for the loading animation to stop so that I know I can start to interact with site.

For general design reasons, I've always loved that Tildes is text-only, but more and more I appreciate that aspect simply because Tildes feels good to use because it is so quick and responsive. I don't know how much of that is due to the text-only part of things and how much of it is Deimos being a genius code wizard who made an amazing platform, but I'm happy about it regardless.

This site has got zero input lag.

And that feels great.

[SOLVED]

... well, this is in many ways very unsatisfying, because I have no idea why this worked, but I seem to have fixed it.

Server A has two Ethernet ports, an Intel I219V and a Killer E3100. Several months ago, when trying to debug sporadic btrfs errors (I had my RAM installed incorrectly!), I had disabled some unused devices in BIOS, including the Killer Ethernet port.

Since I had no other ideas, and it seemed like this was somehow specific to this server, I just re-enabled the Killer port and switched the Ethernet cable to that port. I'm now getting 300 Mb/s transfers from my wireless devices to my server, exactly as expected.

I'm gonna like... go for a walk or something. Thank you so much to everyone who helped me rule out all of the very many things this could have been! I love this place, you all are so kind and supportive.

Original:

I'm trying to debug a perplexing networking situation, and I could use some guidance if anyone has any.

Here's my setup:

• UniFi Security Gateway
• UniFi Switch Lite
• Two UAPs
• Two servers, A and B, connected to the USW-Lite with GbE
• Many wireless devices, connected to the UAPs

Here's what I'm experiencing:

• Network transfers from the wireless devices to server A (as measured by iperf3 tests) are very slow. Consistently between 10 and 20 Mb/s.
• Network transfers from server A to all devices are expected speeds. 900-1000 Mb/s to server B, 350-ish Mb/s to wireless devices.
• Network transfers between server B and all devices (in both directions!) are expected speeds.
• Network transfers from the USG to server A also seem slow, which is odd. Only about 60 MB/s.
• Network transfers from the USG to server B and the wireless devices is about 300 MB/s

So, specifically network transfers from any wireless device to server A are slow, and no other connections have any issues that I can see.

Some potentially relevant details:

• Server A is running Unraid
• Server B is running Ubuntu
• Wireless devices include a Fedora laptop, an iPhone, and a Macbook Pro
• UniFi configuration is pretty straightforward. I have a few ports forwarded, a guest WiFi network (that none of these devices are on), a single default VLAN, and two simple "Allow LAN" firewall rules for Wireguard on the USG. No other firewall or routing config that I'm aware of.

If anyone has any thoughts at all on how to continue debugging, I would be immensely grateful! I suppose the next step would be to try to determine whether it's the networking equipment or the server itself that is responsible for the throttling, but I'm not sure how best to do that.

I’d like your help – and your stories!

I’m working on an article with a tentative title of “Tales of the Crypt: Horror stories (about your past) where database performance caused a real problem.” It’s meant to be schadenfreude nostalgia, about your late nights coping with a performance issue (with, hopefully, a happy ending of “…and this is what we did to fix it”).

So, what happened? Tell me about it.

I do want to quote you, but we can be oblique about the attribution – especially because sometimes these stories are from a previous employer and do not represent any current affiliation. But I do want the verisimilitude that demonstrates that these tales-of-woe come from real people. As a result, I’m fine with writing, “Kim’s first job was as a mainframe programmer at a hotel chain, where database transactions required tape changes. ‘Yada yada story,’ says Kim, who now is CIO of a Midwest insurance firm.” Real person, but you don’t need to worry about getting anyone to approve your words. (Though if you’re happy with full name, company, and role, I’m even happier; send in a private message if you prefer.)

I used an ancient example above, but I’m hoping for more recent database performance stories. Ideally some of the “here’s how we fixed it” become practical suggestions for developers who are enduring such a situation today.