![SanDisk 32GB (Pack of 2) Ultra microSDHC UHS-I Memory Card (2x32GB) with Adapter - SDSQUA4-032G-GN6MT [New Version]](https://m.media-amazon.com/images/I/614AdIvk15L._AC_SS57_.jpg)
![[Gigastone] Micro SD Card 32GB 10-Pack, Camera Plus, MicroSDHC Memory Card for Video Camera, Wyze Cam, Security Camera, Roku, Full HD Video Recording, UHS-I U1 A1 Class 10, up to 90MB/s, with Adapter](https://m.media-amazon.com/images/I/51j-vaCZ35L._AC_UF480,480_SR480,480_.jpg)
For funsies, I bought a couple of microSD cards that were obviously "fake flash" (where the card misrepresents itself as a higher capacity than it actually is) and ran them through some tests to see what capacity they really were and how reliable they were. I bought this card as my control.
My rig for doing these tests was an Orange Pi 4 running Armbian (a Ubuntu-based distro tuned towards single board computers). I used AltraMayor's f3 suite of tools, along with ncw's stressdisk tool (both of which you can find on Github).
Step 1 was to see how the card represented itself to the operating system. You can see from the dmesg logs that it presented itself as a 256 GB card to the OS. (It technically presented itself as having 500,695,040 512-byte blocks -- which would come out to 256,355,860,480 bytes. Most computer programs measure a gigabyte as 1024^3, whereas most products are advertised using 1000^3. So by the latter definition, this card is 256 GB; but by the former definition, it's only 239 GB. It's annoying, but not unexpected.)
Step 2 was use to use f3probe to see what the card's actual capacity is. (This program works by writing pseudorandom data to different places on the card, then reading it back to see if it matches what's expected. Genuine flash will always return the data that's expected, whereas fake flash will return the expected data up to a point -- after which it'll start returning bad data.) The good news here is that it only took f3probe 29 seconds to tell me that yes, this card is actually 256 GB. (For comparison, it took it over three hours to tell me that my two "fake flash" cards -- which were advertised as 1024 GB -- were actually only about 8 GB in size.)
Step 3 was to use f3write and f3read. (These programs fill up all of the available space on the card with pseudorandom data -- as opposed to f3probe, which just writes data to strategic points on the card -- and the read it back to see if it matches what's expected, to check that the flash cells are working properly.) Again, nothing but good news here -- this card passed with flying colors.
Step 4 was to run a stress test on the card using stressdisk. (This program fills up the card with pseudorandom data, then reads it back a couple of times to check to see if it matches what's expected. It then writes new data to the card, reads it back a couple of times, and repeats this process over and over again for 24 hours -- watching to see if any errors crop up.) Once again, this card passed with flying colors -- not a single error during the entire 24 hours. (One of my "fake flash" cards started showing errors about halfway through the test; the other one started showing errors before the 2-hour mark.)
The last thing to look at is read/write speeds. Both f3read/f3write and stressdisk will tell us what read/write speeds it was able to attain -- so all we have to do is look at the output from those programs. This card carries the UHS-3 mark -- which means that it must be able to sustain read/write speeds of at least 30 MB/sec -- and this card blew that out of the water. Both tools said that it was able to attain over 67 MB/sec write speed and over 87 MB/sec read speed. Heck, that's enough to qualify for the next highest speed rating, V60 -- which requires read/write speeds of at least 60 MB/sec. (Samsung advertises that this card can get over 90 MB/sec write speed and 100 MB/sec read speed -- I'm going to chalk up the difference to limitations in my rig.)
So if you're looking for a good quality SD card...in my opinion, this is it!
Reviews with images
It's a winner!
The genuine Samsung Evo Select line of Micro SD cards are great. They are reliable and they live up to the speed and capacity classifications they claim to. I've been using Samsung cards for years and they are the only brand that consistently doesn't give me any issues year after year.This 32GB card is no exception. However, by default the card is formatted as FAT32, but I prefer exFAT as FAT32 won't allow you to transfer a folder containing more than 4GB to the card. That's annoying if you are moving data from an old card to a new one. FAT32 has 4GB file size limitations, but on a transfer it recognizes a directory containing lots of files as 1 thing, and applies the 4GB limit. This will give you an error (in Windows anyway) that there isn't enough space available for the transfer. But if you copy the files individually it works fine. Or just format the card to exFAT and it will accept the folder that exceeds 4GB easily. exFAT will work just fine in an Android phone as well (Android 5.1 or higher. Earlier versions may have some issues with the exFAT formatting)This card handles full HD Video at 60FPS without faulting. It's write speeds are fine to use for HD video.But be sure to authenticate the card upon arrival to make sure it's a genuine Samsung card with the stated capacity. Micro SD cards are a favorite thing for fraudsters to counterfeit because it's cheap and easy to do it. In the past the regular Evo line of Samsung cards (the white and orange colored ones) where the popular cards to counterfeit. Though with these higher grade cards (Evo Select line with the white and green color scheme) starting to be more popular......well it behooves anyone buying a new SD card to check it's authenticity upon arrival. The 2 tone fade of the green coloring adds an additional anti-counterfeiting measure over the previous solid colors. This makes it more expensive to counterfeit a Samsung card convincingly.What happens is that fraudulent card makers will take a 4GB Class 1 card that costs $1 to make, and hack the control chip that tells the computer what the capacity of the card is so your computer thinks it's a much larger card. Your computer will even try to write well over 4GB of data, but the data doesn't actually go anywhere. It's the same idea as putting a 100 gallon label on a 10 gallon bucket. You can pour 100 gallons into it, but 90 of those gallons are just going to flow out over the top and be lost forever.Below are some steps to help you ensure that you get a legitimate Samsung card.❖ For the last several years Samsung has made it's micro SD cards with white plastic. That means the actual card itself behind the colorful label on front is white. You can see the white plastic on the sides. Fake cards are almost always all black, because they are generically produced low cost cards that are given a fancy Samsung looking label on the front. If the card you get it black ON THE SIDES, it's not a Samsung. White plastic is more expensive to make, and fraudsters are all about doing it cheap. (See my images below of a real Samsung card with the white sides and back showing) The back should be black. The sides should be white. And the front should be colorful with the 2 tone green fade.❖ Buy only from reputable vendors. Amazon is a reliable vendor. And if there is a problem, they will take it back without hassle. There are dozens of other 3rd party vendors who are just as trustworthy. But make sure you are buying from one of those. Read their seller feedback. If there are numerous complaints from people who bought fake capacity cards or even cards that didn't work properly, find a different seller.❖ Make sure the listing looks right. This one is fine. It's a legitimate Samsung card. (I've personally bought it and tested it) But scammers will post nearly identical looking listings on Amazon that have cards that look nearly identical buy don't have the Samsung logo at the top. (Recently another scammer replaced Samsung with Sanshen on fake cards. Other than that they were identical. They used the same font any everything, including the A in Samsung that doesn't have the bar in the middle)❖ Be careful of the price. If the price of the card you are looking at is significantly below the price of other listings for similar capacity cards, something is wrong. One of the biggest tricks scammers play is to make the price WAY lower than it should be in hopes of getting you to spend more time looking at the price tag than looking at the other things that don't add up right on the listing. Unusually low priced flash storage almost never works out well.❖ Test the card once you buy it! Regardless of who you buy your Micro SD card from or what brand it is, test it out. A thorough way to test it is with a utility app called "h2testw". You can download it free. Just do a search for it on any search engine and you'll find it. That app writes single bit data to the drive until it's full, then it reads that data back to make sure it's readable. It will tell you what the actual capacity of this disk is. The default language on that utility is German, but you can switch to English easily. It is not fast. It will take about 45 minutes to test a 32GB card, and several hours to test a 256GB card. Just run it overnight and check the results in the morning. This is smart to do even if you have no doubts that you have an authentic Samsung (or other brand name) product. Sometimes things go wrong in manufacturing and quality control and a dud gets through. An app like h2testw will verify that your legitimate brand name card is working properly. And if it's not, you'll know right away while you are still in your 30 day exchange window.
Top reviews from the United States
There was a problem filtering reviews right now. Please try again later.
Reviewed in the United States on January 24, 2021
Capacity: 256GBVerified Purchase
For funsies, I bought a couple of microSD cards that were obviously "fake flash" (where the card misrepresents itself as a higher capacity than it actually is) and ran them through some tests to see what capacity they really were and how reliable they were. I bought this card as my control.
My rig for doing these tests was an Orange Pi 4 running Armbian (a Ubuntu-based distro tuned towards single board computers). I used AltraMayor's f3 suite of tools, along with ncw's stressdisk tool (both of which you can find on Github).
Step 1 was to see how the card represented itself to the operating system. You can see from the dmesg logs that it presented itself as a 256 GB card to the OS. (It technically presented itself as having 500,695,040 512-byte blocks -- which would come out to 256,355,860,480 bytes. Most computer programs measure a gigabyte as 1024^3, whereas most products are advertised using 1000^3. So by the latter definition, this card is 256 GB; but by the former definition, it's only 239 GB. It's annoying, but not unexpected.)
Step 2 was use to use f3probe to see what the card's actual capacity is. (This program works by writing pseudorandom data to different places on the card, then reading it back to see if it matches what's expected. Genuine flash will always return the data that's expected, whereas fake flash will return the expected data up to a point -- after which it'll start returning bad data.) The good news here is that it only took f3probe 29 seconds to tell me that yes, this card is actually 256 GB. (For comparison, it took it over three hours to tell me that my two "fake flash" cards -- which were advertised as 1024 GB -- were actually only about 8 GB in size.)
Step 3 was to use f3write and f3read. (These programs fill up all of the available space on the card with pseudorandom data -- as opposed to f3probe, which just writes data to strategic points on the card -- and the read it back to see if it matches what's expected, to check that the flash cells are working properly.) Again, nothing but good news here -- this card passed with flying colors.
Step 4 was to run a stress test on the card using stressdisk. (This program fills up the card with pseudorandom data, then reads it back a couple of times to check to see if it matches what's expected. It then writes new data to the card, reads it back a couple of times, and repeats this process over and over again for 24 hours -- watching to see if any errors crop up.) Once again, this card passed with flying colors -- not a single error during the entire 24 hours. (One of my "fake flash" cards started showing errors about halfway through the test; the other one started showing errors before the 2-hour mark.)
The last thing to look at is read/write speeds. Both f3read/f3write and stressdisk will tell us what read/write speeds it was able to attain -- so all we have to do is look at the output from those programs. This card carries the UHS-3 mark -- which means that it must be able to sustain read/write speeds of at least 30 MB/sec -- and this card blew that out of the water. Both tools said that it was able to attain over 67 MB/sec write speed and over 87 MB/sec read speed. Heck, that's enough to qualify for the next highest speed rating, V60 -- which requires read/write speeds of at least 60 MB/sec. (Samsung advertises that this card can get over 90 MB/sec write speed and 100 MB/sec read speed -- I'm going to chalk up the difference to limitations in my rig.)
So if you're looking for a good quality SD card...in my opinion, this is it!
My rig for doing these tests was an Orange Pi 4 running Armbian (a Ubuntu-based distro tuned towards single board computers). I used AltraMayor's f3 suite of tools, along with ncw's stressdisk tool (both of which you can find on Github).
Step 1 was to see how the card represented itself to the operating system. You can see from the dmesg logs that it presented itself as a 256 GB card to the OS. (It technically presented itself as having 500,695,040 512-byte blocks -- which would come out to 256,355,860,480 bytes. Most computer programs measure a gigabyte as 1024^3, whereas most products are advertised using 1000^3. So by the latter definition, this card is 256 GB; but by the former definition, it's only 239 GB. It's annoying, but not unexpected.)
Step 2 was use to use f3probe to see what the card's actual capacity is. (This program works by writing pseudorandom data to different places on the card, then reading it back to see if it matches what's expected. Genuine flash will always return the data that's expected, whereas fake flash will return the expected data up to a point -- after which it'll start returning bad data.) The good news here is that it only took f3probe 29 seconds to tell me that yes, this card is actually 256 GB. (For comparison, it took it over three hours to tell me that my two "fake flash" cards -- which were advertised as 1024 GB -- were actually only about 8 GB in size.)
Step 3 was to use f3write and f3read. (These programs fill up all of the available space on the card with pseudorandom data -- as opposed to f3probe, which just writes data to strategic points on the card -- and the read it back to see if it matches what's expected, to check that the flash cells are working properly.) Again, nothing but good news here -- this card passed with flying colors.
Step 4 was to run a stress test on the card using stressdisk. (This program fills up the card with pseudorandom data, then reads it back a couple of times to check to see if it matches what's expected. It then writes new data to the card, reads it back a couple of times, and repeats this process over and over again for 24 hours -- watching to see if any errors crop up.) Once again, this card passed with flying colors -- not a single error during the entire 24 hours. (One of my "fake flash" cards started showing errors about halfway through the test; the other one started showing errors before the 2-hour mark.)
The last thing to look at is read/write speeds. Both f3read/f3write and stressdisk will tell us what read/write speeds it was able to attain -- so all we have to do is look at the output from those programs. This card carries the UHS-3 mark -- which means that it must be able to sustain read/write speeds of at least 30 MB/sec -- and this card blew that out of the water. Both tools said that it was able to attain over 67 MB/sec write speed and over 87 MB/sec read speed. Heck, that's enough to qualify for the next highest speed rating, V60 -- which requires read/write speeds of at least 60 MB/sec. (Samsung advertises that this card can get over 90 MB/sec write speed and 100 MB/sec read speed -- I'm going to chalk up the difference to limitations in my rig.)
So if you're looking for a good quality SD card...in my opinion, this is it!
Images in this review
Top reviews from other countries
Luis Angelo Teixeira da Silva
Otimo
Reviewed in Brazil on July 22, 2021Capacity: 32GBVerified Purchase
