Tag Archives: ssd

Twice the Speed? USB C-ing Things!

During this past weekend, I heard from my client from May, on whose Dell PC I had performed a clean Windows 10 installation. Since I last reported, we’ve done a RAM upgrade (from 12 Gigabytes to 16); and we’ve added a spiffy new graphics card. These upgrades have extended the usefulness of the now nearly seven-year-old PC. And it continued to run swimmingly, until a recent video editing assignment took things to a new level.


A new graphics card can make a huge difference for a PC. Source: MSI.com

The client was handed two portable hard drives, each holding two Terabytes (TB) of footage. This was too much to transfer to the PC’s internal hard drive, so she had to run directly off the externals. That’s not an entirely unreasonable task, as we had installed a USB 3.0 card some time earlier, and that allowed her to input data at five Gigabits per second (5 Gbit/s). But she was still experiencing a frustrating lag when trying to play video directly from the drives. When she called, I asked her to provide me with the model numbers of the drives, so I could get a better sense of what we were working with.

“MU-PT2T0B,” she texted me. This is the model number for Samsung’s T3 Portable Solid State Drive (SSD).


Samsung’s T3 Portable SSD. Note the shape of the port on the left. Source: Samsung.com

Checking out its specs on Samsung.com, I saw one line that intrigued me:

“USB 3.1 Interface”

I thought to myself, “Three… point… one?

No, just 3.1!

I remembered the blog post I did last November about the advent of Thunderbolt 3, and how it shared the rounded, “USB-C” connector shape with the nascent USB 3.1 format.

Okay, there it was, bottom of the chart:


This is where things get a little messy. On their website for the  portable SSD, Samsung doesn’t get into which USB 3.1 standard the drive uses: 3.1 Gen 1 (formerly 3.0) with a top speed of 5 Gbit/s; or Gen 2, which, as you can see in the chart above, is twice that speed. All they say as far as speed is, “450MB/s Transfer Speed.” If that seems slow, note the capital “B.” That indicates Megabytes, as opposed to Megabits. It’s a subtle difference, but a Megabyte is equal to 8 Megabits. That means 450MB/s (Megabytes per second, mind) is 3,600 Mbit/s (Megabits per second). Applying the metric system, this becomes 3.6 Gigabits per second (Gbit/s). It’s still within the threshold of USB 3.0’s 5Gbit/s speed, but I wanted to try something nonetheless.


This chart shows how USB 3.0 became 3.1 Gen 1. Honestly, this whole mess could have been avoided if “Gen 2” had just been called, “USB 4.0.” Source: Kingston.com

Adding USB 3.0 ports had been a simple, effective upgrade. Would it be just as helpful to add 3.1 ports? I went to my local Fry’s Electronics to look into it. If the upgrade was too expensive (or unavailable), then we’d just carry on at 3.0 (or 3.1 Gen 1, whatever you want to call it) speeds.

As it turned out, Fry’s had a wide variety of affordable 3.1 cards. But a closer look indicated that many of them were indeed “Gen 1,” and thus offered no speed boost over the “3.0” card we had installed a while back. I had to keep my eyes peeled for that essential piece of fine print.

I did find a few that offered 10 Gbit/s speeds, but the connectors gave me pause. You see, most 3.1 cards either use just the old, rectangular “USB-A” ports; or they go 50/50, with one rectangular port, and one rounded “USB-C” port. Both potentially work at the full Gen 2 speed, but I was thinking about my client’s Samsung drives. I wanted to go directly from each of their native USB-C ports directly into the computer’s, without using an adapter cable, or a hub (and more on that in a moment). We had two drives, so I wanted two USB-C ports (at that full Gen 2 speed, remember).


A typical USB card with both shapes of the USB 3.1 interface. Source: Frys.com

Luckily, after a little digging, I found the only card in the store that met my needs: The Xtrempro 11107 PCI-E 2Ports USB3.1 Type-C Card (just rolls off the tongue, don’t it?). It met both my needs: 10 Gbit/s transfer speeds, and two USB-C ports. All that, at less than $30.


And this is the FRONT of the box! But at least it doesn’t leave anything to the imagination!

I did, though, have to buy some cables: an internal power cable to run from the PC’s power supply to the USB card; and of course, two USB-C cables to connect those Samsung drives to this new card. When shopping for the USB-C cables, I was just as diligent about reading their specs as I had been about the card’s. I didn’t want the cables to choke on the full data rate promised by Gen 2, after all! Several USB-C cables I found were indeed only rated at 5 Gbit/s, so I’m glad I didn’t fall for the old “they all look the same” trap. I eventually grabbed a pair of PPA Int’l cables, after reading on their package, “Up to 10 Gb/s.”


PPA’s USB-C cable. Source: ppa-usa.com

So I brought everything to my client, and we opened up her PC. The next consideration was where I was going to put this card. You see, a motherboard’s expansion slots can be just as prone to the “they all look the same” trap as the cables. And that can make a huge difference. Her motherboard contained six slots, all based on the Peripheral Component Interconnect (PCI) standard. One slot (labeled 25 in the diagram below) uses the original PCI format, capable of speeds between 133 and 533 MB/s (note that capital “B,” as in Megabytes). The other five use the newer PCI Express format, capable of anywhere between 250 MB/s and 63 GB/s. And that’s a pretty wide range, so let’s narrow it down a bit.


Dell’s schematic for the motherboard. Note the PCI slots toward the lower-left. Source: Dell.com

Slots 26, 29, and 30 are what’s known as PCI Express x1. This is a “single-lane” link, which means the following, according to howstuffworks.com:

Each lane of a PCI Express connection contains two pairs of wires — one to send and one to receive. Packets of data move across the lane at a rate of one bit per cycle. A x1 connection, the smallest PCIe connection, has one lane made up of four wires. It carries one bit per cycle in each direction.

Slot 28, an x16, was already in use by the snazzy new graphics card we got—and rightly so, as that’s the fastest PCI connection on the motherboard, and where better to put the indispensable graphics processor!

This left #31, which I’ll call, “Goldilocks.” It’s neither too slow, like its x1 siblings; nor too fast, like the x16 slot. This slot runs at x8, which—I checked—the new USB-C card supports.

So I installed the card into that x8 slot; and after running a power line to it from the internal power supply, we were good to go. We connected the Samsung drives directly to the card, and suddenly video that couldn’t even open was now coming up, buttery-smooth. We were no longer at the mercy of a 3.0 card in an x1 slot. We were now coasting at 3.1, Gen 2, via x8. And apparently, those numbers make all the difference.


I thought about the prospect of adding a third USB-C component in the future, like another Samsung drive, via a hub. To my surprise, as of this writing, nobody has manufactured a hub that supports the USB-C shape and the Gen 2 speed of 10 Gbit/s. I have reached out to the USB Implementers Forum at usb.org, to see if they know of anyone who has built such a device. I’ll post if and when I hear back from them. Until then, it appears if you want to connect more than two USB 3.1 (Gen 2) devices to a PC, you’ll need to use adapter cables. So make sure they don’t slow you down.


A typical “USB-C Hub,” from Macally. Except it only connects (right) via USB-C, and the USB-C port on the left is only for passing through a charge into a laptop. The USB ports included are both the older USB-A shape, as well as the slower 3.0 speed! Hopefully, newer, better hubs are on the way.


How I Learned to Stop Worrying and Love Nuking a Computer


I assure you, what I did was perfectly sane.

On Sunday, a client contacted me to recruit my help with her Windows computer, a Dell tower. During an overnight video render, it froze up and became unusable. When I came over to check it out, the only thing visible on the black screen was a white cursor. It moved, but there was nothing to click.


Yep, that’s what it looked like.

She had already attempted a Windows System Restore, but the nearest restore point was from about two years ago. She believed it could have been a virus that trashed her system, restore points and all.

The good news was that in this tower, Windows was installed on its own drive, and the client’s documents and other critical data were held on a separate hard drive. This meant that if there were a software problem (such as a corruption in Windows or a virus); or even if the drive’s hardware was physically malfunctioning, most of what my client needed wasn’t necessarily at risk. But we still needed Windows to work.


Dell’s Studio XPS tower. Source: Dell.com

I proposed two options:

1. A Fresh Start

When a hard drive fails, data recovery can be an arduous–and, depending on the extent of the crash, pricey–process. But if the only data lost is the Operating System (OS), it can be a much faster process just to buy a fresh new hard drive and reinstall the OS. Some computers don’t lend themselves to having their hard drives removed and replaced. Luckily, this Dell tower allows for its drives to be removed, so this was indeed an option.

A new hard drive is nowhere near as expensive as they used to be when PCs first entered the home market. If my client were to opt to get a new drive, she could pick up a new 500 Gigabyte (GB) Hard Disk Drive (HDD) for under $50 (as of this writing—I can only imagine the price in the future). If she preferred speed over size, she could get a Solid State Drive (SSD) starting at a little over $50 for 120GB.


Click the above picture for the LinkedIn article, “SSD vs. HDD: What’s the Difference?”

The only other consideration would be the cost of the OS itself. This is one distinction Windows users must concede to their Mac-using brethren: Mac OS has been available for free to all Mac users since 2013’s version 10.9 “Mavericks.” Windows, on the other hand, only offered a brief, free Windows 10 upgrade download to current users of Windows 7 or 8. After July 29, 2016, Windows 10 was only available for $119 for the Home Edition, and $199 for the Pro Edition. Unlike prior versions of Windows, neither version of Windows 10 is available on an optical disc; instead, it comes on a tiny USB flash drive.


This photo demonstrates how tiny a new copy of Windows is, as well as the importance of a good manicure. Source: Speedtest.net.in.

Luckily, I had a Windows 10 Home Edition USB stick, so my client didn’t need to buy a new copy. The way Microsoft had it set up was the installation would only proceed if the user entered a valid license key. A new key comes with each copy of Windows, of course, but I had already used the one that came with mine. Fortunately, Dell was good enough to include a label with the license key on the back of their tower. It was the key for the copy of Windows 7 with which the Dell tower shipped; but Microsoft doesn’t mind that it’s replacing Windows 7, as long as it’s legitimate (which this was, of course).


An example of a Windows license key label. Source: techingiteasy.wordpress.com.

So we were ready to go with a new copy of Windows 10. I was prepared to go pick up an SSD (because when booting Windows, speed matters more than space), and to get started. I opened the case, and guess what I found: an SSD!

About a year prior, I had upgraded this Dell tower with a 500GB SSD, onto which we had reinstalled Windows 7 from its official DVD. My client then participated in the free Windows 10 upgrade program I mentioned earlier. The upshot is, there was no need to rush out and get an SSD; we had one already. Faced with this reality, I agreed that we should pursue…

2. The Nuclear Option

This is the more cost-effective option of the two, as it uses the existing hardware and saves the step–and the money–involved in buying a new drive. I extracted the SSD, connected it to a “spare” working PC, and ran a barrage of drive integrity scans. The good news was that the SSD was “healthy,” so I could confidently recommend continuing to use the drive. But we were going to want to “nuke” it, first.

Opening Windows’ Disk Management utility, I had at my fingertips the tools to wipe the SSD entirely. Confirming that my client really, for sure didn’t need any files from the SSD, I deleted its Windows partition. I had to use a third-party program to scrape off the last remnant of a “recovery partition;” but once I did that, the SSD was as blank as the day I picked it up at the store. One more disk integrity check, and it was ready to be reinstalled in the Dell.


Windows Disk Management. Proceed with caution. Source: technet.microsoft.com.

At about 15 minutes beginning-to-end, installing Windows 10 from its USB stick onto this SSD was one of the fastest OS installations I’ve ever encountered. If you ever have to install Windows fresh onto a blank drive, this is definitely the way to do it.

Once Windows was up and running, I installed the latest drivers from Dell and AMD (for the graphics card); as well as the popular suites from Adobe and Microsoft. Since these were purchased online for download as opposed to disks from a store’s shelf, re-installing them with the relevant licenses was an easy, swift process. I enjoy any process that doesn’t require hunting for disks or slips of paper with license keys on them.

Confirming that Windows was now running at “Day One” speed and efficiency, it was time to reinstall the large HDD with all my client’s documents and data on it: a 2-Terabyte (TB) beast we nicknamed “BIGBOY.” After its own antivirus sweep and drive integrity check, I installed it back in the tower, and Windows Explorer found it without a moment’s hesitation.

Everything was as good as, or even better than, new. The computer now contained only the software my client wanted, and there were no trial programs or Dell pack-ins to be found.

Perhaps after hearing how well this went, you’re considering “nuking” your own computer. Maybe your apps aren’t running as smoothly as they used to. Maybe you’re running out of space. Maybe you miss how clean everything was when you first turned on your PC, before years of downloads bogged everything down.


This is what it sounds like when drives cry.

I still prefer installing a brand new drive, because there really is no substitute for new hardware. It doesn’t have the “miles” on it that are bound to age the drive that came with your computer. Also, when you remove an old drive, you can keep it as an “archive” of the system as it was to that point. You may realize later on that there are files on that drive that you may need, after all.

If you’re just not in the position to get a new drive, then go forth with the Nuclear Option. Just make sure you have a valid copy of whatever OS you plan to install (along with the relevant license key, if necessary); and that before you do anything, that you double- and triple-check that you have copied (or don’t need) the data on your boot drive. You’ll also want to make sure you have the ability to re-download your important programs—unless you have them on their original disks, close at hand.

A spare computer is also a great thing to have, to test the health and integrity of old and new drives before installing or re-installing them in your primary computer. If you don’t have access to another computer, you can continue as planned, of course; but you’ll want to proceed that much more cautiously through each step.

Finally, when it’s all done, take this time to take stock. How quickly do you want to fill up this clean hard drive? Do you really need to download every new app that comes down the pike? How many desktop wallpapers is enough? I’m not saying you have to go completely Spartan, but the idea should be to avoid having to “nuke” your computer again for a while.


Tempting, isn’t it?

On Language and Understanding (and Misunderstanding)

Last week, I discussed what happens when I’m called in to fix a problem and it either disappears before I can solve it, or when it gets worse before it gets better. In the scenarios I described in that post, the clients were able to express their issues clearly–it just got complicated when the time came to resolve them. But what happens when the language gets in the way?

I recently saw the film Arrival, the central conflict of which concerns communicating with extraterrestrial visitors. First our human protagonists need to learn the language the aliens use, so the earthlings can speak back to the visitors on common ground. I realized that I encounter a similar struggle all the time, just trying to solve computer and other technological issues.

1. “Are You Thinking What I’m Thinking?”

I was recently called in to help with a client who couldn’t get her email. “My Outlook isn’t working,” she said. I’m pretty familiar with Microsoft Outlook, the e-mail/address book/calendar program, so I came in prepared to troubleshoot the program and get her back up and running.

She didn’t have Outlook installed on her computer.


Jamie and Adam from Mythbusters would make short work of some of my clients’ issues, I’m sure. But they’d probably use explosives.

As it turns out, she was using a web browser to go online to Outlook.com to check her email on the web. The browser was having some issues, so she couldn’t get to that website. I couldn’t have solved the problem until I saw what she was experiencing; because in my mind, it only made sense that her “Outlook” was one thing; when to her, it was something else altogether. So now, in addition to my litany of questions starting with, “Is it plugged in,” now I make sure to ask just what program, exactly, is causing the problem… even if I think I know which one it is.


Yes, but which Outlook?

2. Memory vs. Storage, and the Fat Phone Cord

Terminology in any field can cause communication issues (just ask your local doctor or auto mechanic). Computer jargon can be especially arcane, because some of the terms are too easily confused for one another.

For example, all too often I’ll have a client mistake memory for storage. In computers, both can be measured in gigabytes (GB). Random Access Memory (RAM) or just “Memory,” is made up of chips that (as of this writing) typically store around 8 or 16GB, but only while the computer is on. Storage, on the other hand, either in the form of a Hard Disk Drive (HDD) or Solid State Drive (SSD), can run into the thousands of GB (and a thousand GB is a Terabyte, or TB) and stores data even while the computer is off.

Techtarget.com goes into some detail about the distinction:

Both terms are used to refer to internal storage space on a computer. Memory, usually referred to as Random Access Memory (RAM), is the place where an application loads its data during processing, while a hard disk drive (HDD) is usually the place where data is stored for long or short term retention.

And so, if a client is getting “low memory” errors, deleting files from his hard drive won’t solve the issue, even if it does indeed free up “gigabytes.” Usually, quitting all the open programs is the best fix for a low memory issue.

Another scenario in which terminology can either help or hinder involves networking cables. When I try to talk clients through their issues over the phone, I try to visualize their setup so as to adequately guide them through the solution. If a client is having network connectivity issues and I know they’re not using wi-fi, I’ll ask them to inspect the cord from their modem or router to their computer. Remember, I always try to confirm if it’s plugged in.

A typical network cable uses an RJ-45 connector, which clicks in to the jack on each end, much like how a telephone cord’s narrower RJ-11 connector does. Based on its similarity, I’ll often describe a network cable as a “fat phone cord.” Of course, if I’m dealing with more advanced clients, I can refer to “Ethernet,” “Cat.5 (or 6),” or the aforementioned “RJ-45.” All of these terms apply to the same thing, and that multitude of choices tends to cause more problems than it solves.


A typical telephone cord (left) with its RJ-11 connector, and the larger Ethernet cable with an RJ-45 plug above it.

3. “Sokath, His Eyes Uncovered!”

Finally, one of the most valuable tools in my communication arsenal is metaphor. I think of the “Darmok” episode of Star Trek: The Next Generation, when Capt. Picard has to communicate with an alien using the language of metaphor. It’s a fascinating concept, but it’s not just the stuff of science fiction: I use metaphors all the time to make technical concepts easier for my lay clients.

For more on Darmok, The Atlantic has a great essay on the episode here. The episode itself is available to stream or download at several sites, shown here.

Remember the client who confused memory and storage?

Picture a juggler, with a backpack full of beanbags on his back. He’s juggling a few beanbags. Your computer’s storage (be it HDD or SSD) is that backpack. It’s got everything you need, whether you’re using it or not. Your computer’s memory, on the other hand, are the free hands with which you can juggle. The more “hands” you have, the more you can juggle (and some computers are trying to juggle “one-handed!”)

I was speaking metaphorically, Ralston!

I had another client who decided she needed a new computer, because hers was going much too slowly on the internet. A slow computer is a genuine concern, but it’s always important to run a speed test on your internet connection, using a few different devices, before blaming the computer. I use a combination of Netflix’s fast.com and Ookla’s speedtest.net. If the connection is slow across the board, it doesn’t matter how new or fast your computer is. Think of it like this: if your driveway isn’t paved, it doesn’t matter how new or fast your car is; it’s still going to be a bumpy ride. Just remember when you get angry at your computer for being slow: “It may not be the car at fault, but the driveway.”

A view of Hall Street  from Institute Ave.

By all means, get a new car if it’s time. But for goodness sake, get that driveway paved, first! (Source: Worcestermag.com)

I may never be asked to communicate with aliens in a first contact scenario. But I take comfort knowing that I’m doing what I can to help my fellow earthlings communicate with their devices–and each other–a little more easily. ■