All posts by Ed Tittel

Full-time freelance writer, researcher and occasional expert witness, I specialize in Windows operating systems, information security, markup languages, and Web development tools and environments. I blog for numerous Websites, still write (or revise) the occasional book, and write lots of articles, white papers, tech briefs, and so forth.

Short-Lived CalDigit TS4 Hiatus

My first job out of college, I worked as a studio engineer in recorded sound. I remember one of the senior engineers telling me one day: “The hardest problems to diagnose are the intermittent ones.” Over the years, I’ve seen that revealed as a terse understatement. I was reminded of that principle last week when my $400-plus Thunderbolt 4 dock quit working. As I dickered with CalDigit tech support to try to get an RMA number for that momentarily dead device, it came back to life. Because of this Short-Lived CalDigit TS4 hiatus I never did get an RMA; instead it’s back at work. Sigh.

When Short-Lived CalDigit TS4 Hiatus Ends, Then What?

As you can see from the front and back views in the lead-in graphic, the CalDigit Thunderbolt 4 Station (aka TS4) is a port-laden beast of a dock. I purchased it last August because I wanted to test this top-of-the-line Thunderbolt 4 (TB4) unit against other TB4 docks from Belkin and Lenovo. Until last week, it has behaved flawlessly, and worked well under every test of its capabilities I could devise.

Initially, I explained my symptoms to CalDigit tech support: no power light, no DC pass-through to power plugged-in devices, no appearance in Intel’s Thunderbolt Control Center (TCC) app when plugged in (and ditto for inserted TB4 or USB4 devices, either). They didn’t seem to want to believe me. So, under their guidance, I tried the device by itself (no power light, TS4 box didn’t warm up as it previously did). Next, I tried the device with their TS4 cable into a laptop. Still nothing. I reported those results and asked again for an RMA.

The Waiting Is the Hardest Part…

While waiting for a reply from Tech Support, I unplugged the device and left it completely alone and unused. When I got a response from CalDigit a couple of days later, they had me try one more thing: hook it up to a different laptop, in a simple configuration (TB4 from CalDigit to laptop for power and connectivity, GbE and USB-C for a storage device on the dock). To my utter astonishment it worked! And it kept working, even when I switched it back to the original laptop.

As far as what happened, nobody knows. Now the power indicator works. Pass-through power has kept my Lenovo ThinkPad X12 Hybrid Tablet humming for the past 4 days without interruption. And the TCC has consistently reported the presence of the dock and the Konyead USB4 NVMe drive enclosure also plugged into an open USB4/TB4 port (see below).

I’ve gotten into the habit of checking things as I sit down with my first cup of coffee to start up my day. And since last Thursday, everything been peachy. No problems at all.

But gosh, doesn’t that just underscore the loathing and dread that an intermittent failure can inspire? Why am I checking this stuff every day? Because I’m waiting for the next failure to pop into view. CalDigit doesn’t seem concerned, and hasn’t issued an RMA. Why can’t I be as cheerfully indifferent to the possibility of impending doom? Because I bought and paid for the unit, the problem is mine, all mine, I guess!

And boy, isn’t that just the way things go sometimes, here in Windows-World? Stay tuned…

Facebooklinkedin
Facebooklinkedin

Newer USB Justifies Added Costs

I had a revelation via contrasting benchmarks yesterday. A friend returned a mid-range USB 3.1 NVMe drive enclosure after an extended loan. Thus, I popped it into my production desktop (an i7 Skylake Gen 4 PC) to see how fast it ran. Good enough. Then, just for grins I popped it into the 2021 vintage Lenovo P16 Gen 1 Mobile Workstation (an i9 Gen 12 PC). Much faster! Enough so, in fact, that it’s clear that newer USB justifies added costs of acquisition. Let me explain…

Why Say: Newer USB Justifies Added Costs?

Take a look at the lead-in graphic. It shows the difference between older USB technology in the Skylake desktop vs. newer USB technology in the Gen 12 mobile workstation. Both are using USB 3.1 ports (though the older PC goes via USB-A, the newer goes thru USB-C) to the same hardware running the same benchmark. Why is the new so much faster than the old?

Short answer: UASP, aka the USB Attached SCSI Protocol. The newer PC supports it, while the older one does not. You can see there’s a driver difference in Device Manager when it comes to accessing the NVMe drive enclosure and its installed SSD: the older machine runs a driver named USBSTOR.sys, while the newer one runs UASPStor.sys. Plain as day.

The Deal With UASP

The Wikipedia article on UASP is a good place to find some explanation. To wit: “UAS [USB Attached SCSI] generally provide faster transfers when compared to the older USB Mass Storage Bulk-only (BOT) protocol drivers.” In a nutshell, that’s UASPStor.sys versus USBSTOR.sys.

As I learned about this technology in the period from 2016 to 2019, the word at TenForums.com ran something like “Speeds of 500 MBps mean USB bulk transfer; 1 Gbps or better means UAS transfer.” And that, dear readers, is the difference you see between the right-hand side in the lead-in graphic (USBSTOR.sys on the Skylake) and the left-hand side (UASPStor.sys on the Gen 12).

In practical terms, this translates into much, much faster IO on the newer PC vis-a-vis the older one. I think it’s incredibly worthwhile, given that backups complete 2-3 times faster on the P16 than the Skylake. Likewise for big, bulk file transfers (such as Windows ISOs, which I mess with frequently).

Retrofit and Replacement

Does this mean one has to toss older PCs and replace them with newer models? Maybe, but not necessarily. For between US$50 and 100, you can purchase UASP capable PCIe adapter USB cards. As long as you’ve got an open PCIe x4 port available on your motherboard (desktops only, so sorry) this could be a good solution. I’m a fan of this US$95 StarTech unit for that purpose.

Older laptops can be dicey and depend on support for USB ExpressCards. I mucked around with these on some 2012-vintage Lenovo ThinkPads in the 2014-2016 timeframe (an X1 and a T420). They work, but they’re cumbersome and expensive (see this Amazon Review for a great discussion).

For best results, it may be time to shell out for a new desktop or laptop PC. That way, the fastest USB (and even Thunderbolt) technologies are likely to come built-in and ready to go. Could be worthwhile!

 

 

Facebooklinkedin
Facebooklinkedin

Build 25300 Restores Taskbar Clock Seconds

OK, then, they’ve been gone for some time now. But Dev Channel Build 25300 restores Taskbar Clock seconds to its display capabilities. The lead-in graphic shows that Settings checkbox, next to Winver for the build.

Note: we’ve not had access to seconds readouts in the Windows 11 taskbar clock since Day 1 of the release. It popped in — and then out again — in a recent Insider Preview. And right now, it’s only available in the Dev Channel release fork. Just sayin…

Find this by clicking through Settings → Personalization → Taskbar. Then, open the Taskbar behaviors pane. That’s where you’ll find the checkbox labeled: “Show seconds in system tray clock…” Notice that it comes with this caveat: “(uses more power).” MS has long put this theory forward (it recommended against turning on the second hand in Vista-era clock gadgets for the same reason) but doesn’t really present actual data to report how much more power is used — or battery life lost — as a consequence of turning this on. Sigh.

If Build 25300 Restores Taskbar Clock Seconds, Then…

I can only interpret the MS caveat as a warn-off of sorts. I guess we should be grateful they’ve deigned to restore this capability to those bold (or stupid) enough to use it. Count me among that number, and decide for yourself its potential significance. Here’s what it looks it, after you turn seconds back on:

Build 25300 Restores Taskbar Clock Seconds.clock-showing

Even at the cost of a bit of power, glad to get those seconds back!

Small though this change may be, I am glad to have the choice as to whether or not I get seconds with my time readout on Windows 11. It’s been that way in Windows as far back as I recall. And now, it’s back again.

Sometimes, those little things do make a difference. I count this as a minor victory for the small people, here in Windows-World.

Facebooklinkedin
Facebooklinkedin

Surface Pro 3 Dock Fail

Oh boy! For more than a few minutes yesterday, I thought I’d completely lost my now-ancient Surface Pro 3 hybrid tablet. It took me a while to diagnose, but it was actually a Surface Pro 3 dock fail, not the PC itself. Seems that the brick that provides power to the dock is no longer working. It wasn’t charging the battery anymore, so once the battery died, so did the PC.

As you can see in the Speccy motherboard info screencap above, this unit goes all the way back to the Haswell CPU days. That makes it a Gen 4 Intel CPU. According to Intel, this model launched in Q3 2013, about 9.5 years back. That’s a long run for any PC, if you ask me.

Surface Pro 3 Dock Fail.dockshot

After a couple of tests, I determined that power to the dock itself wasn’t working.

Diagnosing Pro 3 Dock Fail

At first I couldn’t get the SP3 to keep running. It would start up, then immediately fall over. I checked the battery and saw it had 0% charge. Upon leaving it alone and plugged into the dock for a couple of minutes, the charge level remained the same. “Aha!” I thought “No power to the charger, no power in the battery.”

And so it proved to be. I still had the standalone charger for this unit. Upon plugging it into the wall and the SP3, the battery charge level started to climb. It took almost 2 hours but it eventually reached a full charge, to wit:

Surface Pro 3 Dock Fail.battlevel

Given sufficient time, the SP3 returned to full charge.

Here’s the Question: Do I want to spend $41?

I can replace the AC adapter charger for the dock for the aforementioned price. Do I want to do that? I’ve been thinking about retiring this machine for more than year now. I’d been keeping it to ride Windows 10 to its retirement date with a machine likewise fated. But now I’m wondering if it’s worth it. $41 ain’t much, so maybe I will. Let me think on it, and I’ll post again…

Note added 20 mins later: I found a cheaper replacement on Amazon. For under $19 (including tax) I’ve ordered a new AC adapter for delivery next Monday. I’m hoping it will restore the dock to operation upon plug-in. I’ll follow up…

Note added Saturday AM, February 19: The El-Cheap AC adapter showed up at our front door late last evening (thanks Amazon Prime!). I removed the old unit and replaced it with the new one this morning. It works: as you can see in the next screencap the Surface has its wired GbE connection back, courtesy of the powered-up dock.

With power to the dock restored and Surface re-seated; Ethernet now works!

That was definitely worth the near-sawbuck expended for the replacement part!

Facebooklinkedin
Facebooklinkedin

Figuring Out Intel Arc Iris Xe Drivers

For a long, long time Intel has made newer drivers available for its various integrated graphics circuitry. I’m talking older stuff like its UHD graphics, as well as newer Arc and Iris Xe graphics. Until last year, laptop operators were warned off these drivers because they could overwrite OEM extensions and customizations. I’ve been installing and figuring out Intel Arc Iris Xe drivers lately because that warn-off has been modified.

Here’s an “exception” of sorts that now appears in the Intel Driver & Support Assistant‘s cover language for the Intel Arc & Iris Xe Windows Drivers:

If you have a 6th Generation Intel® processor or higher, your computer manufacturer’s customizations will remain intact after upgrading to this graphics driver. To identify your Intel® Processor generation, see How to Find the Generation of Intel® Core™ Processors.

For the record, my test PC is a Lenovo ThinkPad X12 Hybrid Tablet. It runs an 11th Gen Intel CPU (i7-1180G7), with onboard Irix Xe graphics.

What Figuring Out Intel Arc Iris Xe Drivers Buys

Much of the Arc and Iris capability in the Intel ARC Control app is oriented to games, especially its “Studio” functions, designed to let an ARC device broadcast, capture, share highlights, or set up and use a virtual camera, all within the game-play context. Because I’m not a gamer (and have no actual ARC GPUs only on-CPU graphics subsystems) this doesn’t really signify much for me. I did, however, learn that ARC was looking out of the camera on the back of my X12 Hybrid at my desktop cubbyholes. I promptly turned that off.

What I’m concerned about is usability, stability, and everyday performance. By this I mean that the new driver doesn’t impact usability or stability. I also mean that it has no negative impact on performance, either.

The Verdict So Far

In working with my test system locally and remotely, I’ve noticed nothing different or unusual about the graphics driver. Usability, stability and performance all seem the same.

Reliability Monitor tells a different story, though. Over the past 6 days, it shows 3 APPCRASH events all aimed at “ArcControlAssist.exe.” Each seems to fall around when I open or update the ARC Control Assist app.

Thankfully, the everyday behavior of the system remains rock solid. I’m guessing there may be some teething pains involved here. I’ll say that the new drivers are worth testing, but don’t seem entirely ready for production at this time. At this point, I’m more inclined to blame flaky software (the Intel ARC Control Center app itself) rather than a flaky driver (no other behavior to indicate problems). I’ll keep an eye on this stuff, and let you know how it plays out. Stay tuned!

 

Facebooklinkedin
Facebooklinkedin

Phased Updates Norton 360 Strike

It’s not like I’m unfamiliar with this sensation. No, not at all, based on oodles of history with phased feature roll-outs for Insider Previews on Windows 10 and 11. But Friday, I got bit for the first time on a Norton 360 upgrade. That translates into what I call a phased updates Norton 360 strike. Let me explain…

What Phased Updates Norton 360 Strike Means…

I found a Norton update notification dated February 4, 2023 online. It includes a telling sentence that reads in part as “…this version is being released in a phased manner.” Alas, SUMo doesn’t care: it thinks the release is generally available. So here’s what it tells me about my production PC:

Notice the second (orange) item indicates that 22.23.1.21 is available. True enough in some general sense, but not true for my PC. Here’s what Norton tells me when I attempt to update it through the Norton Update Center:

Just for grins, I positioned the open Norton 360 home window underneath the results of the Webscan that says “no product found.”

Trip a Little Reboot, and…

Of course, one can’t have the 360 window open without running said software. So I knew something was flaky about the Norton AutoDetectPkg.exe that I’d just downloaded. A quick reboot later, and the following message appears instead (and confirms my preceding hypothesis):

Am I Frustrated, or What?

Yes, indeed I am. But because I have long, sad and weary experience behind the curve on phased releases I know this means my turn has not yet come for release 22.23.1.21. I’ll wait as patiently as I can for same, since Norton won’t make it available any other way. Otherwise, I’d just force install it, and make the SUMo warning go way.

And boy howdy, is that ever the way things go sometimes, here in Windows-World … with a bullet, this time!

Note added February 22

In the past little while (it’s been 9 days since I wrote this post), Norton has quietly updated itself to the new release. Here’s what the about information says now (notice the version number: indeed, it is now at 22.23.1.21).

My phase has come in the form of a recent and silent upgrade to 22.23.1.21.

‘Nuff said!

Facebooklinkedin
Facebooklinkedin

The Big Ethernet Dock Sleep

Corny title, I know, but eerily accurate. In the wake of some recent update, my Lenovo Thunderbolt 4 dock keeps losing its wired Ethernet connection. This is particularly vexing when the RJ-45 is plugged directly into the Lenovo P27u-20 integrated dock. Why? Because the Cat-6 cable I’m using is hard to plug and unplug. Sigh. In perverse homage to Raymond Chandler: I call this the “Big Ethernet Dock Sleep.”

Ending the Big Ethernet Dock Sleep

Upon investigating the issue, it seems to be an endemic dock problem. It appears to be related to dropped connections following sleep. In fact, from what I can see, it affects not just PC docks (I saw posts from Dell, HP and Lenovo users) but also Mac docks (I saw some Apple Support posts as well).

What’s the issue? Take a look at the lead-in graphic. Basically it shows that, by default, the GbE port on the dock can be turned off “to save power.” That’s pretty much a given when a PC goes to sleep.

The fix is dead simple, though. Simply uncheck the first checkbox and everything goes blank. I can’t say for sure that this absolutely, postively fixes the issue. But I can say for sure that the affected PC has gone to sleep, then been awakened, and the Ethernet connection stayed up the whole time.

I’ll post back if it recurs. But from what I see online, this fix has worked for others likewise affected. Thus, I’m optimistic that it will also do the trick for me. If not, I’ll post back here again.

Fingers crossed…

Note added next morning 7:30 AM

 

The X12 Hybrid slept peacefully, all night long. And when I just RDP’ed into it now, it awoke with a working Ethernet connection. Problem still not conclusively solved, but close!

 

Facebooklinkedin
Facebooklinkedin

Win10 Enterprise Image Repair Mismatch

I’m flummoxed. As part of my production PC repairs the other day, I ran an in-place upgrade install. It didn’t fix my problem, but it ran to reportedly successful completion. Here’s the thing: I used a Windows 10 Pro image for build 19045.2546 (from UUPdump.net) to make those repairs. I’m surprised it worked!

Why Say: Win10 Enterprise Image Repair Mismatch?

As you can see in the lead-in Winver graphic, this PC is clearly running Windows 10 Enterprise (2nd text block). Yet the filename and download info at UUP Dump clearly identifies the Windows version as Pro:

19045.2546.230118-0641.22H2_RELEASE_SVC_PROD3_CLIENTPRO_OEMRET_X64FRE_EN-US.ISO

Targeting install.wim from the Sources directory, DISM unambiguously identifies the Windows version as Windows 10 Pro.

Yep: it definitely says the image is Windows 10 Pro.

IDKYCDT

IDKYCDT = “I didn’t know you could do that.” But apparently, you can. Indeed the MS Answers advice on this technique says only that one must

download the latest .ISO file available for Windows 11 or Windows 10.

It says nothing about version. Likewise, the TenForums tutorial on this topic simply says

  • If you have a 32-bit Windows 10, then you must use a 32-bit ISO or USB.
  • If you have a 64-bit Windows 10, then you must use a 64-bit ISO or USB.

Again, there’s nothing here about version, simply that a valid ISO is required. I don’t where I got the idea that the version and kind of ISO used for repair had to match the repair target. But it does NOT have to match. I got explicit evidence to the contrary earlier this week with my own eyeballs, on this very PC.

Thus, I learned something useful and can pass it on to you, dear reader. Any valid Windows 10 ISO works for Windows 10; ditto for Windows 11. Cool!

This is actually pretty handy because you can use UUPdump.net to cobble together an image for the current build number for Windows (10 or 11) including all recent updates and CUs. Then, when you repair the image it should work for Home, Pro, Education and Enterprise even if you — as I did — download the Pro-only ISO. No further updates will be required, when that repair completes.

Facebooklinkedin
Facebooklinkedin

Why Give PowerToys Admin Access?

I found myself looking at a suggestion from PowerToys on a test PC yesterday. It popped up when I opened Windows Terminal as Administrator, as per usual practice. It warned me that Fancy Zones and other PowerToys tools might not work properly unless I gave PowerToys admin access, too. Hence the question: Why Give PowerToys Admin Access?

Why Give PowerToys Admin Access?
Because other apps use it…

I turned to MS Learn. There I found an item entitled PowerToys running with administrator access. It pretty much explained everything. Here are the salient points from its second heading:

PowerToys only needs elevated administrator permission when interacting with other applications that are running in administrator mode. If those applications are in focus, PowerToys may not function unless it is elevated as well.

These are the two scenarios PowerToys will not work in:

  • Intercepting certain types of keyboard strokes
  • Resizing / moving windows

Seems pretty straightforward to me, and makes perfect sense. Here’s how to get to admin mode in PowerToys from its default “Running as user” mode.

Making the Switch: User to Admin

You must open PowerToys in admin mode to switch to admin mode. If PowerToys is running, right-click its taskbar icon and select exit to terminate its runtime instance. Next, right-click the PowerToys icon in the start menu, and select “Run as administrator.” In Settings, Administrator mode, move the “Always run as administrator” slider from off (as shown in the lead-in graphic) to On. That’s it!

Now, you can run some of your tools and programs in admin mode without warning messages from PowerToys (or concerns that its tools might not work as they’re supposed to). I like it, and I like ready access to simple, intelligible explanations as to why things must change to work properly.

One More Thing: v0.67.1 Is Out

As I write this item, MS has just released PowerToys update to v0.67.1. While you’re poking around inside Settings/General click the “Check for updates” button. If your PC isn’t yet caught up to this latest version, it’ll take care of it for you. Or, try this command

Winget upgrade Microsoft.PowerToys

if you prefer. Cheers!

 

Facebooklinkedin
Facebooklinkedin

Unscheduled Restore Drill Ends Well

Oh boy! I found myself fighting several interesting Windows issues yesterday. Long story shot: my unscheduled restore drill ends well at the final conclusion. But first: I have to rebuild my boot configuration data and create new Macrium Rescue Media before I can actually boot into the restore environment. It’s always something, right? Let me tell you what that meant yesterday…

After Initial Obstacles, Unscheduled Restore Drill Ends Well

First, a benediction:  I remain grateful that I take a scheduled backup at 9AM every morning, 365 days a year. If it weren’t for that backup, I would have been in real trouble. Indeed, I tried an in-place upgrade repair install during my recovery sequence. That’s when I observed that while it fixes OS files perfectly, it does not fix self-inflicted file system or Registry issues.

My first hint of difficulty came when I tried to restore my 9 AM backup. The timestamp on that file is actually 9:52 so that tells you how long it took to complete as a background task. Turns out my Macrium Rescue Disk wouldn’t or couldn’t boot successfully. It would simply get as far as the bootloader, spinning balls against a black screen, and never get any further. Yikes!

BCD Complications

Next, I unthinkingly added a Macrium Rescue Media element to the boot menu. This did neither harm nor good. But alas, it extended the time it takes to reboot by a good 2 minutes. In a situation where I was rebooting a LOT, that really didn’t help things much. Sigh.

Note: when I did eventually restore the 9AM backup it perforce rewrote the whole boot/system disk, so that little BCD change got undone. Yay! I really don’t need it anyway…

Restore Requires Working, Bootable Rescue Media

Eventually, I turned to the excellent Macrium user forums to find insight on my hung restore. I switched from a faster, bigger NVMe drive in a USB enclosure, to an older, slower, and smaller USB flash drive. That ultimately did the trick.

For some reason, I had to build the rescue disk twice to get it to work. I may have forgotten to check the “Check for devices missing drivers on boot” checkbox the first time around. I was sure to check it on the second try (see lead-in screencap: it’s unchecked by default). After that, the Macrium restore operation proceeded automatically and the process ground through to its lengthy conclusion.

It had been long enough since I did a restore that the time required came as a quasi-revelation. All told, it took about 70 minutes from firing off the restore inside the damaged Windows 10 installation to return to the desktop inside the 9AM image snapshot. That’s quite a bit longer than the 14 minutes or so it takes to backup in single task mode (52 minutes as a background task).

But wow, was I relieved to get over that hump! And now, I have working proof that my backup/restore regime produces its intended outcome. It was an interesting ride in the meantime…

Facebooklinkedin
Facebooklinkedin