Tag Archives: featured

Thinking About Windows 10/11 SSDs

I’m still busy benchmarking away on the two Thunderbolt4/USB4 PCs that Lenovo has recently sent my way. But as I’ve been doing so, I’ve been thinking about Windows 10/11 SSDs in general. On that path, I’ve realized certain principles that I’d like to share with you, dear readers.

I’m spurred in part to these statements from a sponsored (and pretty contrived) story from MSPowerUser entitled “Is NVMe a Good Choice for Gamers?” My instant response, without reading the story — which actually focuses on storage media beyond the boot/system drive — was “Yes, as much as you can afford.” Spoiler alert: that’s what the story says, too.

Where Thinking About Windows 10/11 SSDs Leads….

Here are some storage media principles that flow from making the most of a new PC investment.

  1. The more you spend on a PC, the more worthwhile it is to also spend more on NVMe storage.
  2. Right now, PCIe Gen4 drives run about 2X the speed of PCIe Gen3 drives. They don’t cost quite twice as much. Simple economics says: buy the fastest NVMe technology your PC will support.
  3. Buy as much NVMe storage as you can afford (or force yourself to spend). For pre-built PCs and laptops, you may want to buy NVMe on the aftermarket, rather than get the drives pre-installed. Markup on NVMe drives can be painful. Hint: I use Tom’s Hardware to keep up with price/performance info on NVMe SSDs and other PC components (it’s also the source for the lead-in graphic for this story, which still prominently displays the now-passe Intel Optane as an SSD option. Caveat emptor!).
  4. Corollary to the preceding point: fill every M.2 slot you can in your build. For both my recent Lenovo loaners — the P360 Ultra and the P16 Mobile Workstation — that means populating both slots with up to 4TB each. Right now, the Kingston KC3000 looks like a 4TB best buy of sorts.

Thinking Further (and Outside the Box)

More thoughts in this vein, with an eye toward external drives and multi-tiered storage (archives and extra backups):

  1. If you’re going to put an NVMe SSD in an external enclosure, you will be OK for the time being in a USB 3.2 rather than a USB 4 enclosure. Right now, the newer enclosures cost more than twice as much but don’t deliver anywhere near 2x the speed (except on synthetic benchmarks — I used C: imaging times as a more reliable indicator). Over time this will no doubt change, and I’ll keep an eye on that, too.
  2. I don’t consider spinners (conventional mechanical hard disk drives, or HDDs) any more, except for archival and inactive storage. If I need something for work or play, it goes on an SSD. If I might need something, someday (or to restore same) then it’s ok on an HDD.

I used to restrain spending on NVMe SSDs because of its high price differential. I’m now inclined to believe that restraint is a false economy and forces less productivity as a result. That’s why I’m rethinking my philosophy. I haven’t quite yet gotten to Les Blanc’s famous dictum (“Spend It All”) but I am coming around to “Spend As Much as You Can”…

Remember This Fundamental Assumption, Tho…

My reasoning aims at high-end PCs where users run data-, graphics-, and/or compute-intensive workloads. It does not apply, therefore, to home, hobbyist, and low-end office users. For them typical productivity apps  (e.g. MS Office or equivalent), email, web browsing and so forth predominate. They wouldn’t need, nor benefit much from, buying lots of fast NVMe storage. That said, a 1 TB fast-as-possible NVMe for the boot/system drive is the baseline. Other storage options will balance themselves against budget to dictate other choices and PC builds for such users.

In different terms, if you’re not maxing out your PC running data analytics, 3D models and other high-end graphics rendering, or AI or machine learning stuff, this advice is most likely overkill. Too, too costly. But for this user community, more spent on NVMe (and GPUs and memory as well) will repay itself with increased productivity. ‘Nuff said.

Facebooklinkedin
Facebooklinkedin

Lenovo P16 Gen1 Gets Unboxed

Just over a month ago, I reached out to my contacts at Lenovo. I’d been wanting to lay hands on some newer PCs so I could dig into Thunderbolt 4 and USB 4 to understand its workings. A couple of weeks ago, I received a P360 Ultra SFF PC equipped with 2 each TB4/USB4 ports. Last Friday, unannounced and unexpected, another so-endowed laptop arrived at my door. Here, I’ll report on my initial findings as this Lenovo P16 Gen1 gets unboxed and set up. It’s a doozy!

Details: Lenovo P16 Gen1 Gets Unboxed

I’ll provide a recitation of facts and figures for this powerful portable workstation PC. In fact, it’s the most expensive personal computer I’ve ever worked on. Indeed, its website price, as configured, is a staggering US$9,719! It’s a big heavy sucker, too: 30.23mm x 364mm x 266mm / 1.2″ x 14.3″ x 10.5″, and 6.6 lbs/3.0 kg.

Here’s a selective list of what’s inside this beast of a Widows 11 Pro laptop. (Find all details on its product page under “Tech Specs”):

CPU: i9-12950HX (16 cores, 24 threads)
RAM: 128 GB (4 x 32GB  4800 MHz DDR5)
GPU (built-in): Intel UHD 770
GPU (discrete): Nvidia RTX A5500 (16 GB VRAM)
Display: 16.0″ WQUXGA (3840×2400) OLED touchscreen
SSD: 2 TB Kioxia KXG7APNV2T04 (PCIe 4.0 Gen4 NVMe)
Biometrics: Fingerprint reader and Hello IR Camera

As cool and impressive as all this stuff is — and it is all that for sure — the real reason I’m using this monster appears in the next image, enumerating the unit’s various ports:

My real reason for using this laptop is item 10, boxed in red.
[Click image for full-sized view.]

I’m jazzed, of course, by the panoply of features and stuff on this giant luggable PC. But I’m most interested in working with its two rear USB-C ports, both of which support Thunderbolt4 and USB4. And indeed, I’ve confirmed that both work as claimed. That’s not always the easiest or most obvious thing, as I’ll explain next.

Getting to TB4/USB4

As I’m learning, it takes some diligence to get either or both of these fast bus technologies to work. The PC port has to support these technologies, as does the target device, and the cable between the two. This is not always the easiest thing in the word to ensure or arrange. But as the following screenshot shows, I’ve gotten both working on the ThinkPad P16 Gen1 Mobile Workstation:

Intel TB Control Center: Above, the CalDigit TS4 dock; Below: an NVMe drive inside the Konyead USB4 enclosure.
[Click image for full-sized view.]

Both TB4 and USB4 remain cutting edge connection types. Everything about them is expensive right now. The CalDigit TS4 dock goes for over US$350 when you can find one for sale. The Konyead M.2 USB4 enclosure costs US$130, which is about what I paid for the Sabrent 1TB Rocket 4 Plus I put inside.

And then, one MUST use TB4/USB4 cables which aren’t cheap either (I got mine with the CalDigit) but they routinely go for US$20-40 for 1 M. Cables are not always well-labeled. It’s a good idea to go for those explicitly specced out for 40Gpbs data and marked as such. I’ve had lots of interesting issues from using lower-spec cables. Mostly, USB4/TB4 simply doesn’t work as promised and the device drops to UASP/USB 3.1/2 levels of performance.

Tomorrow, I’ll follow up and explain what all that means… Stay tuned!

Facebooklinkedin
Facebooklinkedin

Old School Driver Repair Still Works

Whoa! I’ve had the Lenovo P360 Ultra SFF PC for a week now, and I FINALLY got the discrete Nvidia RTX A2000 GPU working. It showed only a black screen with the Acer XR382CQK monitor. With a Dell 2717 from my wife’s PC as a stand-in, it would run (briefly) then fall over (AppCrash on NvidiaContainer.exe). My suspicion of driver issues were confirmed by the ace Lenovo engineering team. And I was happy to learn that an old school driver repair still works.

What Old School Driver Repair Still Works?

Good question! Having just written a story for TechTarget about fixing black screens, this was chapter and verse for me. If the current GPU driver falls over, received wisdom goes “roll back a version. Keep going till it works…” I’m actually not sure how far that would have gotten me.

But what the Lenovo engineering folks told me falls in line with that approach. They simply said “install version 511.65” and furnished me with a Lenovo download link for same.

Long story short: I installed the older driver. When I rebooted the machine, the previously non-functional XR382CQK monitor worked like a champ in the miniDP port. I didn’t even have to lug my wife’s Dell 2717 into position instead.

A Further Bulletin from Engineering…

Here’s what one of the engineering team emailed to the group assembled to help me over this hump:

 I checked with our lab and there is a known recent issue with Nvidia’s latest driver 513.12 and later. There will be a P360 Ultra BIOS release by end of month to address the issue. However, the workaround in the meantime is to use driver 511.65.  The symptoms are similar to what Ed is seeing – driver crashes.

Given that insight, a quick confirmation that I was running 516.94, and a link to the download for that older driver version, I got straight to work. Problem solved! Nice to know the old school repair still works. Even nicer to get pointed at the last known working version by the Lenovo team.

Facebooklinkedin
Facebooklinkedin

Accidental Pause Kills In-Process Updates

I just learned something I didn’t really want to know. I “oopsed” my way into pausing updates on a Dev Channel test PC this morning. As I did so, the download for Build 25201 was underway, as was the install for KB5017257 (CU for .NET 3.5 and 4.8.1). Alas, this accidental pause kills in-process updates. Thus, I had to restart to apply all the other stuff that had finished, then un-pause updates. Next, I had to redownload Build 25201. Both installed correctly, and another reboot finished the job.

Living with Accidental Pause Kills In-Process Updates

Oh well. If that’s the worst thing that happens to me today, it will still be a good day. What I didn’t know was that in-process items would come to a screeching halt. That’s because I’d never accidentally clicked “Pause for 1 week” during the update process before. Sigh.

Hopefully, alert readers can profit from my mistake without having to learn the hard way for themselves. Tip: stay away from the “Pause…” button while updates are in process. That’s the best way I can think of to skip the whole learning experience entirely.

Compounding the Mistake…

Because I hadn’t yet applied last week’s Patch Tuesday updates to the affected machine, as well as pending Dev Channel build 25201, this was a pretty big update cycle for that machine. I count 1 driver update, 2 Definition updates, and 3 “Other” updates among that number, as well as the items already recited.

But alas, that’s the way things sometimes go in Windows World. Fumble fingers got me pretty good this time. Hopefully, we’ll all be exempt from this particular gotcha going forward. Sigh.

Facebooklinkedin
Facebooklinkedin

Backblaze Data Confirms SSD Trumps HDD Reliability

It’s always made sense on an intuitive basis. Hard Disk Drives (HDDs) include spinning platters, moving arms with read/write heads, motors to power things, and gears to control action. SDDs are made entirely of circuitry: no moving parts. Thus, it’s compelling to assert that SDDs should be more reliable, and less prone to failure than HDDS. And indeed, the latest 2022 Drive State report from online backup and storage provider Backblaze weighs in on this topic. As I read it, that Backblaze data confirms SSD trumps HDD reliability.

The lead-in graphic shows 4 years’ worth of SSD data vs. 8 years for HDDs for boot drivers in their thousands of datacenter based servers. Whereas there’s a dramatic upward knee in the curve for HDDS between years 4 and 5 (from 1.83% to 3.55%), failures actually dipped for SDDs during that interval (from 1.05% to 0.95%). Interesting!

How Backblaze Data Confirms SSD Trumps HDD Reliability

The afore-linked report explains that boot drives function in multiple roles on the company’s plethora of storage servers. They store log and temprorary files; they maintain storage holdings based on each day’s storage activities and volume. The disparity in the number of years for which data is available comes from later adoption of SDDs as boot drives at BackBlaze. That practice started in Q4 2018. Today, all new servers boot from SSDs; older servers whose HDD boot drives fail get SSD replacements.

The numbers of SSDs keep going up, too. The end-of-year 2021 SSD report encompassed 2,200 SSDs. By June 30, 2022, that count grew to 2,558. Failure rates for such devices show much lower numbers than for HDD (see the tables labeled Backblaze SSD Quarterly Failure Rates in the latest report for more detail). Models included come from the following vendors: Crucial, Dell, Micron, Seagate and WDC.

Note: the report itself says:

For any given drive model in this cohort of SSDs, we like to see at least 100 drives and 10,000 drive-days in a given quarter as a minimum before we begin to consider the calculated AFR to be “reasonable”.

The real news, of course, is that quarterly, annualized and lifetime failure rates for SSDs are significantly lower than for HDDs, based on Backblaze’s own long-running data collection. Thus their conclusion comes with the weight of evidence “…we can reasonably claim that SSDs are more reliable than HDDs, at least when used as boot drives in our environment.”

Good stuff! As for me, I like SSDs not just because they’re less prone to failure. They’re also FAST, if more expensive per storage unit than spinners.

 

 

Facebooklinkedin
Facebooklinkedin

Exploring TB4/USB4 Backup Speeds

OK, then. I’m starting to dig into the capabilities of my new loaner SFF Lenovo P360 Ultra PC. It’s a beast, especially for such a small package (3.4 x 8.7 x 7.9″, 87 x 223 x 202 mm, weight 4.4lb/2.0 kg). Right now I’m giving the front USB-C ports a workout, and exploring TB4/USB4 backup speeds. They’re amazing.

Exploring TB4/USB4 Backup Speeds.f&rview

About the preceding graphic. It shows a front and rear view of the P360 chassis. Here’s what those numbered items convey:

1. Power switch (on/off)
2. Audio/headphone jack
3. USB 3.1 Gen 2 Type A port
4. 2 x Thunderbolt4/USB 3.1 Gen 2 Type C ports
5. Wi-Fi antenna mount
6. 2.5 GbE wired network (RJ-45)
7. 1.0 GbE wired network (RJ-45)
8. 4 x miniDP GPU (connects to Nvidia GPU)
9.  Chassis latch release
10. 3 x full-size DP GPU (connects to on-chip Intel GPU)
11. 4 x USB 3.1 Gen 2 Type A ports
12.  Optional PCIe card slot/port
13. Power in from 300W power brick

What Exploring TB4/USB4 Backup Speeds Says

First things first: I ran comparatives using CrystalDiskMark on a set of different NVMe enclosures with their own drives, as follows:

Enclosure                NVMe SSD              Price (Date)
======================   ==================    ==============
Sabrent NVMe PCIe x1.3   ADATA XPG 256GB       US$ 60  (2019)
Puhui USB 3.1 USB-C      Samsung OEM 512GB     US$ 30  (2022)
Konyead M.2 TB4/USB4     Rocket 4 Plus 1TB     US$162  (2022)

I didn’t get a lot of useful data out of that comparison, though the numbers for all three devices increase their readings down the preceding list. The final item shows most readings between 2x and 3x those for the first item. However, I decided to compare backup results for all three setups, working through a brand-new Belkin Pro Thunderbolt 4 Dock.

The results turn out to be a bit of a good new/bad news scenario.  New TB4/USB4 NVMe enclosures are still punishingly expensive. Performance results from backup show them not yet worth the $132 differential vis-a-vis a cheap0 USB 3 3.1 Gen2 version. About the only thing they can do right now, as far as I can tell, is bring up the “USB 4.0 SSD” label in the Thunderbolt Control Center, as shown in the lead-in graphic.

Big Price Diffs Don’t Translate to Performance

Here’s a table of backup times from Macrium Reflect Free to the three drives, listed by Enclosure name (consult previous table for more info on innards):

Enclosure                Backup (times)
======================   ==============
Sabrent NVMe PCIe x1.3     162 (2:42)
Puhui USB 3.1 USB-C        131 (2:11)
Konyead M.2 TB4/USB4       132 (2:12)

While there’s a 31/32 second difference (about 20%) between the older Sabrent enclosure and the two newer ones, there’s so little difference (1 second) between the other two that I’m sure that falls in the margin of measurement error one would expect.

What’s interesting here is that these backup speeds — even on the slowest/oldest device — are about twice as fast as on my other, similarly loaded test machines (which top out at USB 3.1 Gen 2). That tells me for those who do a lot of backing up, video editing, or other data intensive stuff there’s some real benefit to be gained from investing in TB4/USB4 ports and devices.

Lessons Learned

What lessons do I draw from this experiment? Glad you asked! Here’s a list:

  • It’s definitely worth adding an interface to older desktops to support TB4/USB4 for the speed bump it provides.
  • This new technology provides a “speed reason” to consider buying in on a newer laptop or PC.
  • Newer, more expensive TB4/USB4 NVMe enclosures may not be worth the added cost as compared to USB 3.1 Gen 2/TB3 counterparts.
  • From what I’m reading, it’s a good idea to use as short a USB4/TB4 rated cable as possible.
  • It’s also best to hook the NVMe enclosure directly to the PC if you can (going through the dock reduced performance by about 5% overall)

A terrific experiment, and a  great learning lesson, too. Thanks to the nice folks at Belkin and Lenovo who made their gear available to me.

Facebooklinkedin
Facebooklinkedin

RDP Mystery Finally Resolved

For years now, I’ve wondered why some of my RDP connections work only some of the time. Now I know why, and it’s provokes a Homer Simpson response “Doh!” Now that I’m done laughing at myself, let me explain how I got that RDP mystery finally resolved.

I was working with my various Thunderbolt docks this weekend, and I noticed that a previously not-working RDP connection started working again. Turns out that of necessity devices with both GbE and Wi-Fi adapters have separate and distinct IP addresses for each such adapter. Therein lies the key to the mystery, as shown in the lead-in graphic.

Explaining How RDP Mystery Finally Resolved Itself

This all started when I had to move my X12Hybrid dock upstairs. When I disconnected from the dock, and its GbE connection, the RDP connection defined for X12Hybrid (also the machine name used in RDP) resumed working. Then it hit me: because the GbE connection uses a different IP address from the Wi-Fi connection, my RDP definition works only when the IP address it knows about matches the address actually in use. Again: “Doh!”

You can see this clearly in the dual windows shown in the lead-in graphic. PowerShell is in top position, and shows that nslookup stores the IP address associated with the Wi-Fi interface (192.168.1.20). But when I plug the GbE interface in, it takes a different address instead. That private IP address ends with .39, as shown in the NetBScanner window at bottom of the same graphic.

How to Adapt to Changing Connections

The primary router on my local network is an Arris model: it comes from Spectrum as part of its Internet connection and services. I’ve not figured out how to forcibly reset its address tables for DNS lookups on Windows machine names. Instead, I use NirSoft NetBScanner when an RDP connection fails and enter the correct IP address instead of machine name in its “Computer” data field. This works every time.

By observation, it looks like this data updates every 24 hours or so. If I leave the RDP connection unchanged (no switch from Wi-Fi to GbE, or vice-versa) over that interval, nslookup eventually matches the current address. But at least I now know why this is happening, and why using the IP address for the NIC in use fixes the issue.

That’s the way things go here in Windows World, where I still manage to surprise myself by  relearning the obvious. Sigh.

Facebooklinkedin
Facebooklinkedin

CarPlay Cable Connections Are Key

Here’s another recent lesson learned from our just-completed trip to New England. On our reconnaissance mission in late July, we quickly figured out that a Lightning cable between iPhone and in-car USB makes connecting simple and fast. This time around, we learned that the cable itself also matters. Though I packed 3 such cables in our cable bag, only one of them worked well to support CarPlay. Hence my title: CarPlay cable connections are key. Let me explain…

Why CarPlay Cable Connections Are Key

One of the cables was probably shorted: the charge indicator kept turning on and off when it was in use in the car. That simply won’t do.

The second cable was an old — iPhone 6 vintage, at least — Apple-provided charging cable. Clearly, it couldn’t handle the bandwidth requirements needed to ferry comm traffic between the iPhone and the car’s built-in display. It simply didn’t work reliably or well.

The third cable proved to be the charm. It was a 10-foot Amazon Basics USB A for iPhone and iPad cable purchased in 2019. This item is no longer in stock, but something like this iPhone 11 model (US$16.99) would undoubtedly work. I gave one to my son when he went off to school, so I’m ordering 2 more right now.

Underlying USB Support in CarPlay

As I understand it, Lightning cables support USB 2.0 more or less uniformly (here’s an interesting discussion from Volvo, and an informative Reddit thread). My guess is that both of my old cables were sufficiently “used” that they simply couldn’t provide full USB 2.0 capability/bandwidth. The newer cable — despite its 10ft (~3M) length — worked just fine.

Hint/tip: before you take off on a road trip, it’s probably a good idea to test your chosen Lightning cables (listening to music is a fair method) to make sure they can carry the load. I’d also recommend taking a spare — I always do — just in case you lose or damage one while traveling.

Facebooklinkedin
Facebooklinkedin

USB-C Port Choice Really Matters

Here’s an interesting discovery. Or maybe it should be called a “realization.” Yesterday, upon trying out my new Belkin and CalDigit Thunderbolt 4 docks, I learned that USB-C port choice really matters. In fact, my reported GbE issues with the Belkin Thunderbolt 3 port are probably related. Please: let me explain…

Why USB-C Port Choice Really Matters

Simply put, if you plug a dock into the upper USB-C port in the Lenovo X12 hybrid laptop it works as it should. Plug it into the lower USB-C port and the GbE connection disappears. Also, the device does not show up in the Thunderbolt Control Center app, either.

More details:

  • The Ethernet controller built into the CalDigit unit depicted in the lead in graphic is an Intel I225-LMvP. When the unit is plugged into the upper USB-C port it appears in Device Manager. If plugged into the lower USB-C port it does not.

  • When I plug the dock into the lower USB-C port, it vanishes from Thunderbolt Control Center, which then shows no attached devices. Interestingly, Windows still finds attached storage devices. But wired networking through the dock no longer works.

Extremely interesting!

What Does It All Mean, Mr. Wizard?

What it means is that on this Lenovo model, only one of its two USB-C ports also supports Thunderbolt (and it’s version 4, interestingly enough). Here’s my clue from the product family specifications page, which reads as follows under “Ports/Slots”:

    • USB 4 Type-C with Thunderbolt™ 4 (DisplayPort, Power Delivery and Data Transfer)
    • USB 3.2 Gen 2 Type-C

The reason why storage keeps working, but why networking and video — and presumably  other high-bandwidth connections — do not, is because Thunderbolt support is required for such things. If I’d still had a monitor attached to the X12 (I sent it off to school with my son) I might have figured this out faster. But now I know . . . and so do you! And it goes to show that sometimes, where you plug in really matters, even if the “gozintas” look the same.

Facebooklinkedin
Facebooklinkedin

Appreciating Apple CarPlay

Hello there! We’re just back from a combination trip to the Northeast. On August 25, we flew to Boston with son Gregory to move him into Emerson College. On August 31, we took an Uber to the Boston Logan rental car pickup to head north to Maine. This latter half of the trip has me appreciating Apple CarPlay greatly and enthusiastically.

Why I Am Appreciating Apple CarPlay

I’ve owned and used various in-car GPS systems for years, including those available in Mercedes, Volvo, and Toyota models. Each one is a little different. Each one has its decided UI quirks and foibles. That leads me to my number one source of CarPlay gratitude: a single, consistent and pretty intuitive UI for navigation via Google Maps. It’s great!

Item number two may be just as important. I can remember paying $300 to $500 to access GPS capability in several cars. CarPlay comes as a standard (no added cost) feature on many cars these days. That’s a nice savings, in addition to the benefits of a familiar and standard UI.

My third source of gratitude is a YMMV thing, or it may be a matter of personal preference. I find the voice instructions in Google Maps to be easier to understand and follow than those in the Mercedes GPS (that’s the built-in one I know best thanks to using it in 3 cars over the last 10 years or so).

When driving in unfamiliar places, that’s a benefit that’s hard to overstate. I’ve learned to turn around or re-route to rectify my mistakes while driving. But gosh! I’d rather not have to rectify mistakes that result from misunderstanding voice instructions. Google Maps works better for me in that regard.

All My Future Vehicles Will Have CarPlay

Because we’re an iPhone family (wife, son and myself all have 13 models right now), it makes sense to integrate phone and vehicle. There are lots of other benefits, too — including music, phone calls and SMS messages, the Waze app, and more. Frankly, I just can’t see paying extra for a feature from a carmaker that doesn’t work as well for me as the iPhone running Google maps (or some other equivalent). One more thing: here’s an interesting MakeUseOf story that explains the best of the CarPlay apps, and what new stuff lies ahead in this category. Check it out!

‘Nuff said!

Facebooklinkedin
Facebooklinkedin