A nearly perfect USB cable tester
blog.literarily-starved.com
blog.literarily-starved.com
But there is not standard for usb controllers to present this data to the OS. So it’s stuck in the low level firmware and never passed up. In theory we could have a popup box that tells you that both your computer and other device support higher speeds/more power, but your cable is limiting it.
Apple seems best able to do this since they control the hardware and OS, yet they aren’t doing it either. Users are just left to be confused about why things are slow.
I'm pretty sure my old Dell XPS laptop with Windows 10 had pop-ups just like this.
"This device can run faster" or something.
Dell XPS laptops (and some others) can also warn if the charger isn't providing the full wattage the laptop is rated for. This warning is an option that can be turned off in the BIOS settings.
I usually turn it off because I sometimes intentionally do day trips with a smaller/lighter portable charger which delivers 45w to my laptop which can need up to 65w due to having a discrete GPU. However, 45w is more than sufficient to charge the laptop during normal use on the Balanced power plan with iGPU. I only need more than 45w when gaming with the discrete GPU active.
Weirdest part was it was 100% charged, so could have booted with 0 Watts of charger but decided not to boot with 20 Watts more.
Sure, you or I would just unplug the charger and run on battery but bad UX decisions like that generate a support call to me from my 95 yr old mom. It should not only warn and continue to boot, it should use whatever power is on offer to reduce the rate of battery drain.
If I have a lower wattage charger connected on booth it shows me that information but I can just press enter to continue. It's just a warning.
Maybe it's a bios setting?
Workaround is of course to boot without a charger connected and then connect it later :)
* https://imgur.com/gallery/spoof-original-dell-laptop-power-s...
Since a >$1,000 automated lab cable throughput tester is overkill, my thumbnail test for high-speed USB-C data cables is to run a disk speed benchmark to a very fast, well-characterized external NVMe enclosure with a known-fast NVMe drive. I know what the throughput should be based on prior tests with an $80 active 1M Thunderbolt cable made for high-end USB-C docks and confirmed by online benchmark reviews from credible sources.
Regular people hate technology, both for how magical and how badly broken it is, but they've long learned they're powerless to change it - nobody listens to their complaints, and the whole market is supply-driven, i.e. you get to choose from what vendors graciously put on the market, not from what the space of possible devices.
They hate having to go through people that get them upset, in order to use their kit.
Not just tech (although it’s more prevalent). People who are “handy” can also be that way (but, for some reason, techies tend to be more abrasive).
I’ve learned the utility of being patient, and not showing the exasperation that is often boiling inside of me.
In general for the 40+ years I’ve been a programmer I have detested the practice of not surfacing diagnostic information to users when technology makes it possible to do so in a clear and unambiguous way.
"What did the error message say"
"I don't know."
'Segmentation fault. Core dumped.'
'Non-fatal error detected. Contact support.'
'An error occurred.'
'An illegal operation was performed.'
'Error 92: Insufficient marmalade.'
'Saving this image as a JPG will not preserve the transparency used in the image. Save anyway?'
'Saving as .docx is not recommended because blah-blah-blah never gonna give you up nor let you down.'
I can't blame any normal user from either not understanding nor giving a shit about any of these. If we'd given users actionable information from day 1, we'd be in a very different world. Even just 'Error 852: Couldn't reach the network. Check your connection to the internet.' does help those who haven't turned of their brains entirely yet.
"I don't understand, it says 'System Error Type 11', and no matter how many times I type 11, nothing happens!"
That one's a good example of why these things are hard. The user could have been running 5 different programs, any one of which caused this error, and MacOS can't point the finger at anyone. Not to mention that the problem could be MacOS itself, or the user being a dunce who misconfigured something. I'm not sure if that error can occur without 3rd party software being involved, but if it can, then that error message might need to be even more vague, helping the user even less. Not to mention it could just be faulty hardware.
A paper manual offering troubleshooting steps for each error would be really helpful. Just 'Error 11. Consult your manual.' and the manual actually telling you what the problem could be is also miles better than what we usually get.
It's still an example why it's worth giving your users a fighting chance. MacOS may not know enough to point the finger at anyone, but the user knows what they were doing at that moment, and even if they were not paying attention, they might start now. They'll realize if something is off. Or, after 10th time they get this error, they'll connect the dots and realize it's always happening when application X is running and they try to launch Y.
Or maybe sometimes they won't. Maybe they'll form a story and maybe it'll be all bullshit, or maybe good enough. Either way, the important part is, the user retains agency in the process. Giving people information is how they can become self-sufficient users and trust technology more.
++?????++ Out of Cheese Error. Redo From Start.
+++ Divide By Cucumber Error. Please Reinstall Universe And Reboot +++
+++Whoops! Here comes the cheese! +++
A modal dialog is supposed to be for something damn near irreversible--like about to blow away your application because of error. You are supposed to STOP and go get the guru or you are about to lose, badly.
Unfortunately, UI designers throw them up for everything and people get used to simply clicking "OK" to make them go away so that they can get back to doing their task. So, when the user gets an actual error, they've already blown away the dialog box with information.
Your 'Saving this image as a JPG will not preserve the transparency used in the image. Save anyway?' line is a horrifically excellent example. That is a standard "Save As..." response, and it should NEVER have been. That should have always been under "Export..." as saving should never throw away information and it would be perfectly fine to regenerate a JPG as long as you have the full information still available in the original file.
This is the stuff that infuriates me about the UI designers. Your job is about interactions, first, and pixels, second.
This is because error messages have historically been bad, unintelligible, un-actionable, and hard to separate from soft errors that don't actually matter.
If you ever run into a developer who thinks "something went wrong" is an appropriate error message, have them killed. Then kill their entire family and pets, burn their house down, and plough salt into the ground where it stood. Finally, put up a sign that says "The person who used to live here thought 'something went wrong' is an appropriate error message to display when something goes wrong. Take note of their current situation when you next add an error message to your software".
I'm an electrician.
People will notice some things. For example, with USB if they are using it for local backup they might notice, but with a lot of devices they will not. When they do notice, they will feel powerless.
Even if we had a wider choice, they are not well placed to pick products. There is no way they will know about details of things such as USB issues (a cable is slow, the device will not tell you if it is) at the time of purchase.
I implored him to try a different cable (after checking cables with the Treedix mentioned in TFA), and the copy went from taking over a minute to about 13s.
Its not just normal people confused.
Oh, and pointy jab: these folks are also, in my opinion/experience, the most eager to vibecode shit. Make of that what you will.
Swore blind it couldn't be improved.
By next morning's stand-up, I'd found it was doing something pointless, confirmed with the CTO that the thing it was doing was genuinely pointless and I'd not missed anything surprising, removed the pointless thing, and gotten the 10 minutes down to 200 milliseconds.
I'm not sure if you're right or wrong about the correlation with vibe-coding here, but I will say that co-workers's code was significantly worse than Claude on the one hand, and that on the other I have managed to convince Codex to recompute an Isochrone map of Berlin at 13 fps in a web browser.
But I know stories like yours from a decade past as well. A tale old as time, but compounding in recent years - IMHO.
I don’t know if they check that via USB protocol, or if they are measuring the actual power draw on the USB port.
In order to use the device, I had to connect it via an externally powered USB hub.
There is. I used to use a KVM with USB 2 ports connected to my PC's USB 3 port, to which I connected a monitor with integrated USB 3 hub to drive my keyboard and mouse. Windows would show a popup every time telling me that I should use a faster cable.
There are also popups telling me that my laptop is connected to a "slow" usb-c charger.
They probably have to weigh potential new hardware sales against added complexity. I have counterpoints too but: I believe they try to protect users’ mental models of their ecosystem (which perhaps I appreciate when I don’t notice, and can’t stand when something is uncustomizable). Like there are enough variables they don’t trust us with as it is.
You jest but that notification (it's been a thing on Android for at least 8 years, and on thinkpads for at least 10) has been very helpful to me. Sometimes the negotiation just fails and being told is helpful. Sometimes the charger lies about its specs and once again it's helpful to have a hint, rather than expect everybody to systematically have usb testers on hand.
With data speed I think it could be a little more complicated. Like OP was saying it would need access to some level of hardware information where it can see which pins are used by the cable. Since the connection 'speed' is still variable even when you DO have a supported cable.
Using DevCheck might show 2.2A/9V as an example.
Of course they are advertising their own new USB cable, but as someone who didn't know much about USB cables I find it quite interesting.
It’s well worth the hype, I used it to audit all my cables (both for home and work) and it’s amazing how many thick and unwieldy cables are actually terrible for data.
For example I purchased a pair of B&W Px8 S2 noise cancelling headphones, which boast a DAC if you connect via USB-C directly, the cable it came with though was thick but only rated for USB 2.0 speeds. These headphones cost more than AirPods Max, which are themselves considered overpriced, and include comforts like nappa leather; so shipping with a chunky cable that doesn’t even carry decent data feels like a bizarre oversight. Apple’s own USB-C cables manage the same power delivery at less than half the thickness with a woven shell. You’d assume a premium product would at least match that.
Honourable mention to the USB-C cables that ship with Dell Ultrasharp monitors (both pre-USB4 and post). Those support basically everything except Thunderbolt 4 despite being unmarked.
It’s rigid and thick, like a Thunderbolt 3 cable, yet only supports USB 2.0 speeds and fast charging for a device that doesn’t need fast charging.
Compare that to Apple’s iPhone USB-C cable which is thin, flexible, and supports the same features.
That matters because someone might grab that cable assuming it’s a “better cable”: it came with a £629 product, it’s thick and feels serious, so surely it’s capable. But it isn’t. And there’s nothing marked on it to tell you otherwise.
The whole system ends up relying on presumption, which is exactly the problem the device in the article is solving.
Or, why Apple manages the same in half the footprint?
Or, why someone would expect that a cable that came with a pair of headphones actually charges things at over 65w?
I don’t know how to fix the market especially when consumers keep rewarding these practices, and I think the effectiveness of TikTok style influencer marketing will make it worse.
The problem is the opposite of what you’re describing, it’s not a cynical design choice, it’s a lazy one. They probably just purchased a cable for capabilities irrelevant to the product and the result is worse ergonomics and misleading physical cues about what the cable can actually do.
I think you are underestimating the importance of perceived premium combined with the pressures of cost accounting, but I do think that is pretty normal for ‘audiophiles’ which is their target market.
If the argument is that B&W deliberately chose a thick cable to seem premium, it doesn’t square with them actively slimming down the headphones. B&W are primarily a speaker company, their USB-C product range is basically just a few headphones and earbuds.
More likely they just sourced a generic cable that happened to support high wattage and didn’t think about the mismatch.
Either way, we’re deep in the weeds on B&W’s cable procurement now. The root point is that USB-C is a mess. You can’t tell what a cable supports by looking at it, and even premium manufacturers are shipping cables that don’t do what you’d reasonably expect.
That’s exactly the problem the Treedix from the article solves.
You are using circular reasoning in your logic, you assume the premise is true and from there you derive your evidence.
I would contend that someone thought about it and decided to go with the cheaper option because they could get away with it. I would consider my assumption to have more grounding given my experience with manufacturing and cost accounting.
My example of weights is that the steel weighs are cheaper than the alternative of using heavier drivers, by adding weight they are signaling premium without delivering it. Similarly with the USB cable, consumers assume such cables are thick because of thicker wires and better shielding, it’s cheaper to make a thick cable without those those features, once again signaling premium without actually providing it.
The vast majority of high volume consumer manufacturers use cost accounting practices which would absolutely be tracking and attributing the usb cable costs and the whole point of that accounting practice is to constantly be thinking about minimizing costs of even the smallest inputs, all the way down to the individual screws used. Yes, they’re thinking about how to save 1/100ths of a cent from each screw.
The purpose of the heavy construction is to make it durable, not to carry 5 Gbps data streams to your headphones.
Unlike most USB peripherals like your printer and keyboard that get plugged in and then don’t move around, headphone cables go to your head and move around constantly. They can get pinched in drawers or snagged on corners.
Hence the more durable construction.
Apple’s USB iPhone cables wearing out prematurely is so common it’s a meme.
Maybe Apple's changed their cables recently, but the fragility is the reason I avoid Apple cables.
Especially in headphones. The number of times those broke during a bike ride or run was way to high for me to keep wasting money on them knowing full well they weren't going to last more than a few months just like every other Apple headphone I've ever had.
https://www.techgearlab.com/topics/electronics/best-usb-c-ca...
So, no: I wouldn't expect the cable for a pair of headphones (of any price) to support USB 3. That represents extra complexity (literally more wires inside) that is totally irrelevant for the product the cable was sold with. (The cables included with >$1k iPhones don't support USB 3, either.)
Meanwhile: Fast charging. All correctly-made USB C cables support at least 3 amps worth of 20 volts, or 60 Watts. This isn't an added-cost feature; it's just what the bare minimum no-emarker-inside specification requires. A 25-cent USB C-to-C cable from Temu either supports 60W of USB PD, or it is broken and defiant of USB-IF's specifications.
---
Now, of course: The cable could be thinner and more flexible and do these same things. That'd probably be preferred, even: Traditional analog headphones often used very deliberately thin cables with interesting construction (like using Litz wire to reduce the amount of internal plastic insulation) to improve the user's freedom of movement, and help prevent mechanical noise from the cables dragging across clothes and such from being telegraphed to the user's ears.
Using practical cabling was something that headphone makers strived to be good at doing. I'm a little bit annoyed to learn that a once-prestigious company like B&W is shipping cables with headphones that are the antithesis of what practical headphone cables should be.
---
But yeah, both USB C cables and the ports on devices could be better marked so we know WTF they do, to limit the amount of presumption required in the real world. So that a person can tell -- at a glance! -- what charging modes a device accepts or provides, or whether it supports video, or whether it is USB 2 or USB 3, or [...].
Prior to USB C, someone familiar with the tech could look at a device or a cable and generally succeed at visually discerning its function, but that's broadly gone with USB C. What we have instead is just an oblong hole that looks like all of the other oblong holes do.
After complaining about this occasionally since the appearance of USB C a decade or so ago, I've come to realize that most people just don't care about this -- at all. Not even a little bit. Even though these things get used by common people every day, the details are completely out of the scope of their thought processes.
It doesn't have to be this way, but it's not going to change: Unmarked ports are connected together with unmarked cables and thus unknown common capabilities are just how we roll.
Your last paragraph is depressingly accurate though. I think that's exactly why devices like the Treedix exist: the standards bodies and manufacturers clearly aren't going to fix the marking problem, so now we need test equipment to figure out what our own cables do.
"I heard what you guys are planning and I talked to my financial guy. He said I have enough to put a manufactured home on some land in some desolate place like the Dakotas or central Wisconsin, as long as I keep a bit of supplemental income and live a little lower. So I'm going to do that, and take my chances on growing artisanal rutabaga to sell at farmers markets.
I've already packed up the Prius. I just stopped by to wish you kids luck with your new headphone project and tell you that I won't be back."
USB 2.0 can support up to 480 Mbps. It’s more than fast enough for any audio stream you can send to a DAC.
Your headphones don’t need USB 3.0 5 Gbps speeds. USB 3 requires extra wires with different properties that need to be controlled more tightly, which can impact cable flexibility. If your headphones used USB 3 when they didn’t need it that would be one more thing to break and more failure modes for the cable.
A USB 2 cable with fewer conductors was the right choice for this product. The fact that you only got miffed about it when plugging the cable into a tester, not from actually using the product or cable, is good evidence that a USB 3 cable wasn’t needed.
Apple’s iPhone cables are not known for their durability. They serve a mostly stationary purpose, unlike headphones you wear on your head.
I keep a few converters for older devices and servers that don't have (m)any C ports, but as far as a consumer "forever cable" goes, TB5 feels close. Certainly the cable's bandwidth is beyond what most people need, unless you're editing 8k video or continually shuffling hundreds of GBs between external disks.
It alleviates the anxiety of knowing what cable does what.
I use Apples Thunderbolt 4 or USB-C cables exclusively: if its white its for charging and low data, if its black its for high data.
I’ve been doing this for a few years, but its really costly as those Apple Thunderbolt cables are crazy expensive.
LTT or another big YouTuber made a cable and made sure to get it labeled. Also complained how difficult it was to find a supplier willing to make a better cable than usual.
I have one of those. They are thick and unwieldy af. Since I've borked the usb connection on my monitor because of static discharge, I no longer use it and figured I'd repurpose it for my digital camera, for which I used to have a short cable that was sometimes annoying. This cable is so freaking think and hard that it'll move my (admittedly somewhat light) camera on the table.
After the publicity a few years ago of bad USB-C cables they've been mostly fixed, but what hasn't been fixed is the infinite number of broken downstream USB-C implementations. So your charging problems aren't due to the cable, which is most likely fine by now, but because the downstream device is telling the upstream one that it can't take more than 5V 1A. One sure way to tell the vendor has screwed up is when your USB-C device comes with an A-to-C cable to charge it.
It is difficult to send 10Gbps, 20Gbps, 40Gbps or even 80Gbps, then receive it, then validate it.
[0] - https://www.totalphase.com/products/advanced-cable-tester-v2...
I can't believe I'm still swapping cables and ports, trial and error, trying to determine if the cable, the port (eg iPhone's charging port), or both, are failing.
Not knowing the USB specs; I assume each head of a cable can do a loop back test, determine if can even see its other head, etc.
For home use that doesn't matter usually, but I for example run events where I need the cable to work also after 10 people stepped on it and then this can become a significant thing.
Not in terms of quality, but reliability.
The quality degrading is not something you will see, as it's a digital protocol.
"Audiophile grade" HDMI cables are likely to just be a Shenzhen bargain-bin special with some fancy looking sheathing and connectors. I would trust them less than an Amazon Basics cable.
If you send bits across a line fast enough you're grtting into the territory of RF electronics, with wrong connector or conductor geometry you will get echos on the line and all kind of signal loss. A good digital protocol should keep this at bay with error correction and similar mechanisms, but if you want to know what the good cable is on a better than binary scale of works/does not, you need to look at these things.
This is a big part of what makes any pro gear expensive: reliability. If you just connect your home hifi to your speakers in an acoustically untreated space, you could also just use a bunch of steel wire coathangers and get an indistinguishable result. Even a el-cheapo store brand music shop cable will do the trick for years if you don't habitually change your setup four times a week (most people don't).
But if you need reliability and predictability in a studio or live context giving a damn about cable quality is mandatory since a broken cable in the wrong place can ruin your day and reputation. But it is an absolute myth that they will affect the sound in any meaningful way.
Exeption: guitar cables. The capacitance of guitar cables can shift the resonance frequency of the pickup up or down leading to audibly different results. But that id no magic either, you could just take a low capacitance cable and add in arbitrary capacitor for 10 cents as needed.
Gauge is only important if the driving or receiving sides suck or you're using ridiculously small gauges or driving high currents (e.g. speaker signals) for long distances. In my lab tests I drove line signals over a spool of hair thin wire without audible difference. A famous experiment that gave me a good chuckle came to similar results: https://www.diyaudio.com/community/threads/copper-wire-vs-ba...
You could literally use CAT 6 ethernet cables and call it a day. There have been tests running balanced audio signals over a kilometer of ethernet cable without audible loss. And since you got 4 twisted pairs you could run 4 channels over one cable (commercially available here for example: https://store.monkeywrenchpro.com/thomann-5fc-cat-snake-spli...). I use this in combination with an ethernet patchbay to route signals around our event spaces and it works well even for unamplified mic level signals.
There's cheaper generic VNA's that dont claim "yup, this is HDMI Ultra High Speed" that might work in the right hands.
It seems to be a more comprehensive "Make sure the lines go where they are supposed to" tester. Looks pretty good.
But the devices that test things like transmission speed, are a lot more expensive.
I think that many of the issues that this device tests, can be mitigated by simply buying cables from reputable sources.
Even in my reputable cables, there are a couple with suspicious continuity issues. I wonder if this could find them.
You could probably build a data transfer tester using an FPGA and some signal processing.
Don't forget the speeds at which modern serial interfaces go. Being able to look at the data, at that speed, requires some serious kit.
Keep in mind that's just for the 480Mib/s USB 2.0. If you want to test USB 5Gbps, USB 10Gbps, USB 20Gbps, USB 40Gbps, or USB 80Gbps you'll be spending much more money. The USB4 V2 GEN4 Electrical Compliance Test Specification⁴ requires a 25GHz bandwidth realtime oscilloscope with ≥80Gs/s sample rate or better. Not to mention a ≥25.6Gbps pattern generator/BERT, 20GHz network analyzer, & RF signal generator, all of which are "call for quote" but likely to be in the 6-digit price range. You're easily looking at the better part of half a million dollars to test the 80Gbps cables fully.
¹https://www.keysight.com/us/en/product/E2667B/usb-2-0-hi-spe...
²https://www.keysight.com/us/en/product/N2752A/infiniimode-ac...
³https://www.keysight.com/us/en/products/oscilloscopes/bencht...
⁴https://www.usb.org/document-library/usb4-electrical-complia...
One of these devices for approximately $100 would sell all day long.
Talking about near perfect wishlist:
Some standard hardware with fail-safe power connections a set of fully wired ports like in display-less Treedix USB platine tester version.
On the other side powered and extended diagnostics through USB-C can then be done from any smartphone or pc providing the display and updateable extended software layer.
Bonus for connectivity to some brother label printers.
That cable has one power input (that is only an input), and two outputs (that are only outputs), and a brainbox in the middle to direct the circus.
If we label the connectors as A, B, and C, then it works like this: A charges B and/or C, and other charging directions are no-op.
The less-complex way is to use a USB A to C cable, if that's appropriate. With these, the A side is always the source and the C side is always the sink.
---
And yeah, it's annoying. I got a cheap lithium car jump starter several years ago with some neat power bank features (like 60W USB PD in/out, on one port). So I plugged it into my phone with USB C at my desk, and discovered that they'd charge eachother seemingly randomly. While changing nothing, I'd look over and sometimes the jump starter would charge the phone, and sometimes the phone would be charging the jump starter. The conglomeration formed a heater, with more steps.
(Back and forth with the same poop, forever.)
--- (I remember. The poop.)
You need a >5A output power supply, two voltmeter channels (for source & sink), one ammeter channel (to sense applied load), an electronic DC load (actively cooled FET that uses the ammeter to set a constant current), a microcontroller, screen, some buttons, and software to run the whole thing. Or the manual version: Lab power supply, some USB connector breakout boards, a DC load, optionally a multimeter, a pencil, and some graph paper. Set the power supply current limit over 5A, voltage to 5V, set the DC load to 500mA, and measure the voltage at the power supply & DC load every 100mA as you increase it up to 5A load (or 3A if the cable isn't marked for 5A capability). If the sink drops more than 0.6V below the source, the cable is not compliant.
They can still provide some power, but between a MacBook and official Apple power supply, they’ll keep flicking between charging and not charging.
I didn't know there were cable testers like this, thank you.
Music production peripherals, until the recent USB C transition, commonly used USB B for some reason. Anyone know why?
> Access Denied
> Sorry, you do not currently have the necessary permissions to access this site, or this site may not be available in your region.
Are they geoblocking the USA from even viewing their site for some reason?
(Must say I'm not a fan of how, increasingly, taking any steps to preserve privacy is seen as deviant, or justified because bots)
I imagine at that length and speed, signal integrity becomes difficult.
What would work better is a flexible 100w+ usb3 cable. You can’t do thunderbolt on it but it’s a tiny fraction of the cost and does everything you’d actually need on the go.
If you actually do want it, this is the do everything cable https://www.apple.com/au/xc/product/MW5H3ZA/A
And in my tiny 'go bike bag' for day trips I need one 2M cable that's thin, coils into a tight ball and weighs nothing yet will charge up to 45w and reliably xfer data at up to 5Gbps (USB 3.1) for quick uploads with optional USB-A and Micro-USB adapters at either end (because I still know people with Micro-USB (though it obviously drops to USB2 speeds)).
https://iaohi.com/products/aohi-the-future-adonis-usb4-2-0-2...
https://news.ycombinator.com/item?id=47561827
(these are what I would buy from a sea of cables, not the cheapest, but far from the most expensive)
I speculate USB B wasn't included because there are only really two types, 2.0 (regular size) and 3.0 (has an obvious extension on the connector). There also don't tend to be power-only A-B cables because they are usually found on printers, Arduino s, ... And not for charging devices.
Fun fact: A Xiaomi fast charge cable (with orange plugs) has an extra contact on the A end to support USB C PD out of a USB A charger.