CKBench is the bench we built because the numbers on the package answer almost none of the questions that matter. Every test byte is committed to the device and confirmed, every byte is read back and checked, and the result gets published whether it flatters the product or not. Its first module is CardCheck, for memory cards, and a microSD is a good place to start: it is the cheapest component in your dash cam and the one most likely to quietly ruin it.
Same bench, same reader, same suite, with one exception the table states outright: a microSD Express card carries PCIe on its second row of pins and cannot reach that bus through any ordinary reader, so it was run on both and appears twice, once per bus. Every other row shares one measurement chain. Write speed is committed sequential, which reads lower than burst benchmarks because every byte had to land before the clock stopped. Capacity is what the card actually delivered against the number on its label, verified byte by byte.
| Card | Committed write | Sustained floor | Capacity delivered | Peak temp | Score |
|---|---|---|---|---|---|
| SanDisk High Endurance 128GB | 89.7 MB/s | 88.0 MB/s, V60 | 100.0% of label | 151.4 F | 5/5 |
| Kioxia Exceria G2 128GB | 65.3 MB/s | 62.6 MB/s, V60 | 96.7% of label | 143.3 F | 4.5/5 |
| VIOFO High Endurance 128GB | 57.0 MB/s | 55.8 MB/s, V30 | 97.4% of label | 160.8 F | 3.5/5 |
| VIOFO High Endurance 32GB | 35.3 MB/s | 34.3 MB/s, V30 | 99.1% of label | 153.0 F | 4/5 |
| ProGrade SDXC UHS-II V60 128GB | 108.4 MB/s | 102.8 MB/s, V90 | 100.1% of label | 145.1 F | 4/5 |
| Lexar Professional 1667x 128GB | 119.3 MB/s | 109.5 MB/s, V90 | 100.0% of label | 142.7 F | 4.5/5 |
| Samsung PRO Ultimate SD 512GB | 137.3 MB/s | 133.1 MB/s, V90 on this reader V60 in a camera |
100.1% of label | 144.4 F | 4.5/5 |
| Samsung P9 Express 256GB (SD Express) | 212.9 MB/s | 161.3 MB/s, V90 | 100.1% of label | 194.8 F | 4/5 |
| Samsung P9 Express 256GB (UHS-I fallback) | 87.7 MB/s | 87.0 MB/s, V60 | 100.1% of label | 140.9 F | 4/5 |
| Samsung PRO Plus 512GB | 142.2 MB/s | 141.6 MB/s, V90 | 100.1% of label | 169.6 F | 4.5/5 |
| Botslab 128GB (bundled) | 59.0 MB/s | 53.9 MB/s, V30 | 97.4% of label | 135 F | 4.5/5 |
| FitcamX 64GB (bundled) | 47.1 MB/s | 43.8 MB/s, V30 | 97.3% of label | 131 F | 4/5 |
| Newegg TeamGroup 128GB | 39.8 MB/s | 35.5 MB/s, V30 | 96.7% of label | 133 F | 3.5/5 |
| SanDisk ImageMate 128GB | 33.2 MB/s | 31.8 MB/s, V30 | 26.8% of label | 150.1 F | 1/5 |
| SanDisk Extreme 512GB | 135.8 MB/s | 127.6 MB/s, V90 | 100.0% of label | 172.3 F | 4.5/5 |
| Lexar Silver Plus 128GB | 182.7 MB/s | 187.3 MB/s, V90 | 98.8% of label | 158.0 F | 4.5/5 |
| Samsung EVO Plus 128GB | 82.0 MB/s | 83.6 MB/s, V60 | 100.0% of label | 135.0 F | 5/5 |
| SanDisk Ultra 128GB | 60.7 MB/s | 59.7 MB/s, V30 | 99.9% of label | 155.0 F | 4.5/5 |
| Lexar Blue 128GB | 152.6 MB/s | 155.1 MB/s, V90 | 98.2% of label | 162.0 F | 4.5/5 |
| Generic PRO PLUS 128GB | 12.5 MB/s | 11.9 MB/s, V10 | 51.9% of label | 110 F | 1/5 |
| Generic MICRODRIVE 128GB | 13.8 MB/s | 12.8 MB/s, V10 | 37.5% of label | 118 F | 1/5 |
| Generic LD 128GB | 15.5 MB/s | 14.6 MB/s, V10 | 50.7% of label | 111 F | 1/5 |
| Vantrue Endurance PRO 128GB | 84.7 MB/s | 83.4 MB/s, V60 | 98.8% of label | 170 F | 4.5/5 |
| SanDisk High Endurance 512GB | 135.0 MB/s | 68.0 MB/s, V60 | 100.0% of label | 174 F | 4/5 |
| VIOFO High Endurance 64GB | 61.2 MB/s | 59.5 MB/s, V30 | 97.4% of label | 174.4 F | 4/5 |
| Gigastone High Endurance Pro 128GB | 58.5 MB/s | 31.0 MB/s, V30 59.6 fresh |
97.7% of label | 132 F | 4.5/5 |
| Amazon Basics 128GB | 189.4 MB/s | 188.1 MB/s, V90 | 98.2% of label | 171 F | 4.5/5 |
| PNY Premier-X 128GB | 53.6 MB/s | 47.1 MB/s, V30 | 97.7% of label | 129 F | 4.5/5 |
| Silicon Power Superior Pro 128GB | 85.4 MB/s | 37.8 MB/s, V30 | 97.2% of label | 123.7 F | 4/5 |
| Metorage Industrial 64GB | 60.5 MB/s | 59.6 MB/s, V30 | 97.4% of label | 172.1 F | 4/5 |
The VIOFO 32GB's sustained floor was published as 14.9 MB/s from July to September 2026, from the card's first-ever run. Two full re-runs on two readers measured 34.3 and 34.2 MB/s, and the row now carries the primary reader's figure like every other card. The review carries the correction and what the July run was.
The Gigastone High Endurance Pro's row carries its used-state floor, 31.0 MB/s, measured after the card had been written end to end, which is the state a dash cam keeps a card in. Fresh from the pouch the same card floors at 59.6, and a full SD erase from a native host brings that figure back. The review has both runs and the erase; the explainer has what a quick format, a full format and an SD erase each do, and which cards never showed the drop.
Twenty-nine cards, one measurement chain. Read down the columns and the same shapes keep coming back, and not one of them is visible from a single review.
were not the size on their label. The SanDisk ImageMate holds 34.3 GB behind a 128 GB label while passing Class 10, U3, A1 and V30, and three no-brand cards sold as 128 GB hold 48.0 to 66.4 GB. Every other card delivered between 96.7% and 100.1%.
cards that print a video speed class sustain a higher one, every one of them by at least a full class. Seven beat theirs by two, holding V90 against a printed V30: the PRO Ultimate, the PRO Plus, the P9 Express, the SanDisk Extreme, the Lexar Silver Plus, the Lexar Blue and the Amazon Basics. None of them falls below its mark; the other nine hold exactly the class they print.
cards that state a write speed miss it by a margin that matters: the VIOFO 128GB by 29%, the Newegg TeamGroup by 20%, the ProGrade by 17%. It runs the other way more often. Twelve of the sixteen beat their claim, the SanDisk High Endurance most of all at 89.7 MB/s against a printed 40.
never write at one steady speed. The Botslab cycles between 34 and 82 MB/s every 9.5 seconds, the TeamGroup between 5 and 83 every 2.7, the PNY Premier-X between about 3 and 81 every 4.2, and the Gigastone on a nine-second saw-tooth. The FitcamX runs a 2.13 second rhythm that makes an honest card look broken, and the three no-brand fakes cycle faster than a one-second sample can draw. The four of them that print a V30 all hold it.
change their speed class depending on the reader. The P9 Express carries PCIe on its second row of pins: V90 and 596.7 MB/s read on an Express reader, V60 and 89.7 on an ordinary one. The PRO Ultimate holds V90 on a reader that speaks Samsung's fast mode, V60 on the plain bus a camera gives it. Read speed moves with the reader far more often, which the reader test measures; which reader to buy is its own question.
covers two different cards. The VIOFO High Endurance 32GB and 128GB share a brand, a package and a product line, and their identity registers name two different manufacturers seventeen months apart.
A review can tell you a card held 88.0 MB/s. It cannot tell you whether that is good. Every ranking below is computed at build time from the same results files the table above prints, across twenty-nine cards and thirty measured rows, and it is worth being blunt about what that makes it: these are percentiles of our bench, not of the market. We have tested these cards. We have not tested the ones we have never held, so n is printed beside every ranking on the page and you can weigh it yourself.
The rows are grouped by host bus and ranked only against their own group. Of the three buses represented here, UHS-I is the only one carrying enough cards (twenty-seven) to support a ranking at all. Any bus below that line is listed with its measurements and no percentile, and earns one once it reaches eight cards.
| Card | Label | Worst-10s sustained write | Committed sequential write | Peak temperature |
|---|---|---|---|---|
| Amazon Basics 128GB | 128 GB | 188.1 MB/sfastest of 27p98 | 189.4 MB/sfastest of 27p98 | 171 F23rd coolest of 27p17 |
| Lexar Silver Plus 128GB | 128 GB | 187.3 MB/s2nd fastest of 27p94 | 182.7 MB/s2nd fastest of 27p94 | 158.0 F18th coolest of 27p35 |
| Lexar Blue 128GB | 128 GB | 155.1 MB/s3rd fastest of 27p91 | 152.6 MB/s3rd fastest of 27p91 | 162.0 F20th coolest of 27p28 |
| Samsung PRO Plus 512GB | 512 GB | 141.6 MB/s4th fastest of 27p87 | 142.2 MB/s4th fastest of 27p87 | 169.6 F21st coolest of 27p24 |
| Samsung PRO Ultimate SD 512GB | 512 GB | 133.1 MB/s5th fastest of 27p83 | 137.3 MB/s5th fastest of 27p83 | 144.4 F13th coolest of 27p54 |
| SanDisk Extreme 512GB | 512 GB | 127.6 MB/s6th fastest of 27p80 | 135.8 MB/s6th fastest of 27p80 | 172.3 F25th coolest of 27p9 |
| SanDisk High Endurance 128GB | 128 GB | 88.0 MB/s7th fastest of 27p76 | 89.7 MB/s8th fastest of 27p72 | 151.4 F15th coolest of 27p46 |
| Samsung P9 Express 256GB (UHS-I fallback) | 256 GB | 87.0 MB/s8th fastest of 27p72 | 87.7 MB/s9th fastest of 27p69 | 140.9 F11th coolest of 27p61 |
| Samsung EVO Plus 128GB | 128 GB | 83.6 MB/s9th fastest of 27p69 | 82.0 MB/s12th fastest of 27p57 | 135.0 F9th coolest of 27p67 |
| Vantrue Endurance PRO 128GB | 128 GB | 83.4 MB/s10th fastest of 27p65 | 84.7 MB/s11th fastest of 27p61 | 170 F22nd coolest of 27p20 |
| SanDisk High Endurance 512GB | 512 GB | 68.0 MB/s11th fastest of 27p61 | 135.0 MB/s7th fastest of 27p76 | 174 F26th coolest of 27p6 |
| Kioxia Exceria G2 128GB | 128 GB | 62.6 MB/s12th fastest of 27p57 | 65.3 MB/s13th fastest of 27p54 | 143.3 F12th coolest of 27p57 |
| SanDisk Ultra 128GB | 128 GB | 59.7 MB/s13th fastest of 27p54 | 60.7 MB/s15th fastest of 27p46 | 155.0 F17th coolest of 27p39 |
| Metorage Industrial 64GB | 64 GB | 59.6 MB/s14th fastest of 27p50 | 60.5 MB/s16th fastest of 27p43 | 172.1 F24th coolest of 27p13 |
| VIOFO High Endurance 64GB | 64 GB | 59.5 MB/s15th fastest of 27p46 | 61.2 MB/s14th fastest of 27p50 | 174.4 F27th coolest of 27p2 |
| VIOFO High Endurance 128GB | 128 GB | 55.8 MB/s16th fastest of 27p43 | 57.0 MB/s19th fastest of 27p31 | 160.8 F19th coolest of 27p31 |
| Botslab 128GB (bundled) | 128 GB | 53.9 MB/s17th fastest of 27p39 | 59.0 MB/s17th fastest of 27p39 | 135 F9th coolest of 27p67 |
| PNY Premier-X 128GB | 128 GB | 47.1 MB/s18th fastest of 27p35 | 53.6 MB/s20th fastest of 27p28 | 129 F5th coolest of 27p83 |
| FitcamX 64GB (bundled) | 64 GB | 43.8 MB/s19th fastest of 27p31 | 47.1 MB/s21st fastest of 27p24 | 131 F6th coolest of 27p80 |
| Silicon Power Superior Pro 128GB | 128 GB | 37.8 MB/s20th fastest of 27p28 | 85.4 MB/s10th fastest of 27p65 | 123.7 F4th coolest of 27p87 |
| Newegg TeamGroup 128GB | 128 GB | 35.5 MB/s21st fastest of 27p24 | 39.8 MB/s22nd fastest of 27p20 | 133 F8th coolest of 27p72 |
| VIOFO High Endurance 32GB | 32 GB | 34.3 MB/s22nd fastest of 27p20 | 35.3 MB/s23rd fastest of 27p17 | 153.0 F16th coolest of 27p43 |
| SanDisk ImageMate 128GB | 128 GB | 31.8 MB/s23rd fastest of 27p17 | 33.2 MB/s24th fastest of 27p13 | 150.1 F14th coolest of 27p50 |
| Gigastone High Endurance Pro 128GB | 128 GB | 31.0 MB/s24th fastest of 27p13 | 58.5 MB/s18th fastest of 27p35 | 132 F7th coolest of 27p76 |
| Generic LD 128GB | 128 GB | 14.6 MB/s25th fastest of 27p9 | 15.5 MB/s25th fastest of 27p9 | 111 F2nd coolest of 27p94 |
| Generic MICRODRIVE 128GB | 128 GB | 12.8 MB/s26th fastest of 27p6 | 13.8 MB/s26th fastest of 27p6 | 118 F3rd coolest of 27p91 |
| Generic PRO PLUS 128GB | 128 GB | 11.9 MB/s27th fastest of 27p2 | 12.5 MB/s27th fastest of 27p2 | 110 Fcoolest of 27p98 |
Two rows are not a distribution, so nothing on this bus is ranked. These are the measured figures, for reference. This bus earns percentiles at eight cards.
| Card | Label | Worst-10s sustained write | Committed sequential write | Peak temperature |
|---|---|---|---|---|
| Lexar Professional 1667x 128GB | 128 GB | 109.5 MB/s | 119.3 MB/s | 142.7 F |
| ProGrade SDXC UHS-II V60 128GB | 128 GB | 102.8 MB/s | 108.4 MB/s | 145.1 F |
One row is not a distribution, so nothing on this bus is ranked. This is the measured figure, for reference. This bus earns percentiles at eight cards.
| Card | Label | Worst-10s sustained write | Committed sequential write | Peak temperature |
|---|---|---|---|---|
| Samsung P9 Express 256GB (SD Express) | 256 GB | 161.3 MB/s | 212.9 MB/s | 194.8 F |
These are the same figures the corpus table prints, so any caveat attached to a number up there travels with its ranking down here, including the duty-dependent floor footnoted beneath the table.
It came inside a Mercylion G401 dash cam and prints no speed at all. Its registers match the VIOFO High Endurance 64GB's apart from the serial and the build date, and on the same reader it lands within 1.2% of that card on every write figure: 60.5 MB/s committed, 59.6 through its worst ten seconds, V30 held on two readers, every byte verified, and 172.1 F under load.
Read the review
Silicon Power’s everyday card prints no speed at all, and its website promises 80 MB/s write. Measured, it writes 85.4 MB/s for the first 4.7 GB, then a flat 39 for as long as you keep going, which holds V30 with a quarter to spare. Every byte verified, 123.7 F under load, and an identity register carrying Silicon Power’s initials under an ID the Linux kernel names Kingston’s.
Read the review
The entry-tier card that prints a read speed and declines to print a write speed. Measured, the write is 53.6 MB/s committed in a four-second cache cycle; V30 holds at 47.4 MB/s on both readers, every byte verified twice, 129 F under load, and the capability registers put it on the TeamGroup and Gigastone platform under a PNY identity.
Read the review
The question people keep asking, answered on two readers. Longsys silicon under Amazon's name: 156 MB/s sustained on the ProGrade, 188 on a SanDisk QuickFlow, the 200 read claim beaten without Amazon's own reader, every byte verified, and 171 F against an 80 C rating.
Read the review
The first card on the bench to print TLC on its face, in a pouch that prints nothing else. It holds V30 in both of its states: 59.6 MB/s through its worst ten seconds fresh from the pouch, 31.0 once the card had been written end to end, and back to 54.5 after a native-host SD erase. Every byte verified, and among the coolest cards here.
Read the review
The middle capacity of a line tested three times over, supplied by VIOFO with an A119M Pro. It prints 40 MB/s and holds 59.5 through its worst ten seconds on two readers, verifies every byte, and runs at 174 F doing it, the hottest of the three VIOFO capacities.
Read the review
A used card with real dash cam service history. It reads 180 MB/s, holds V60 against a printed V30 and verified all 511.8 GB twice, then stopped writing for 3.6 seconds mid-stream, on two readers, thirteen times in three hours. No other card here has ever done that.
Read the deep dive
A dash cam card that prints no read speed, no write speed and no endurance rating, then beats the one class it does print by 2.8x and verifies every byte. Its one measured weakness is heat: the highest peak of any 128GB card on this bench, in an air-conditioned room.
Read the deep dive
Three no-brand cards bought in one order, identical packaging, three names. All three report 134.2 GB and hold 48 to 66 of it, and not one raised a read error while losing the rest. Only writing the whole card finds it.
Read the test
Lexar's cheapest tier prints one speed number and no write figure, then beats the claim it does print and holds 155.1 MB/s, level with the pricier Silver Plus on the same reader. On a faster reader the Silver Plus pulls ahead; the Blue does not.
Read the deep dive
The card most people already own, and the one that promises least: a 10 MB/s floor, no published write speed anywhere in the line, and no video class. It beat that floor six times over and verified every byte after nearly eight years.
Read the deep dive
The only used card on the bench and the only one that prints no speed at all. Its own CID register dated it to June 2019, and then all 128 GB wrote and read back with zero bad bytes…
Read the deep dive
It came free in a dash cam box and then outwrote almost every card here, including 512 GB cards that cost real money. 157.0 MB/s committed on the ProGrade, 182.7 on a SanDisk QuickFlow, no cache cliff anywhere, and a camera that cannot use a byte of it…
Read the deep dive
Beats its printed 190 MB/s read at 193.3 on a host that can show it, holds V90 against a printed V30, and returns every byte of 512 GB. It is also the slowest small-file writer of its class.
Read the deep dive
The fastest sustained writer on this bench that is not on a PCIe bus, and it beat several UHS-II cards costing more. It exceeds its own printed 130 MB/s write at 142.15, holds its speed flat across all 512 GB with no cache cliff anywhere…
Read the deep dive
The first microSD Express card on the bench, and the first tested twice on one day because it has two personalities. Its second row of pins carries PCIe, not a faster SD bus…
Read the deep dive
The cheap card that turned out to be completely real, and completely slow. Every byte of its capacity is there and verified, and it holds every class mark it prints…
Read the deep dive
The first 512GB card on the bench, and the first whose printed speed the standard interface cannot reach at all. It beat the 130 MB/s write on its own face, handed over 100.13% of half a terabyte with zero mismatched bytes…
Read the deep dive
The first full-size SD cards on the bench, and the first UHS-II. Two cards that agree on every number a shopper can read: 128GB, UHS-II, V60, 250 MB/s…
Read the deep dive
The card in the FitcamX box wears a look-alike SD logo where the licensed one belongs, and then holds every byte of its 64GB anyway. Its speed graph appears to fall apart twice a second…
Read the deep dive
The card taped inside a dash cam box, and the strangest write curve on this bench. It cycles between 34 and 82 MB/s every nine and a half seconds, for as long as you care to watch, and still never drops below the V30 floor it prints…
Read the deep dive
The first pair we tested as a matched set, and they turned out to be two unrelated cards wearing one name. Their identity registers name two different manufacturers, seventeen months apart…
Read the deep dive
A counterfeit, and the most interesting one we have seen: it holds 34.3 GB against a 128 GB label while passing Class 10, U3, A1 and V30. Fast flash, fraudulent number. It is the proof that you cannot catch a fake card by benchmarking it.
Read the autopsy
The card that came with your dash cam, and the standard advice to bin it is wrong. It beat its published write spec by 31% and did not fade at all under loop-recording churn…
Read the deep dive
The card that under-sells itself. It committed writes at 89.7 MB/s against a printed 40, held V60 against a printed V30, delivered all 128.0 GB with zero mismatched bytes…
Read the deep diveCKBench is not only for review samples. Before a piece of storage earns a job on this bench, it gets the same suite as everything else, and the first one through was a USB flash drive about to become the boot device for our card-identity rig. It is not a memory card, so it is scored on its own terms and stays out of the card table above. It also produced the strangest finding on this page: a drive whose read speed depends on how the data on it was written.
The thumb drive that beat its only printed number and hid everything else. 148.8 MB/s read against a claimed 130, full capacity verified with zero bad bytes…
Read the bench notesShould I replace the card that came with my dash cam? Probably not, and we have now tested five of them. The bundled Kioxia beat its own spec and handled loop recording without flinching, and the Lexar Silver Plus that replaced it in the next camera outwrites every card on this bench, retail ones included.
How do I tell if my card is fake? Not with a benchmark. Check the reported size first, then fill the whole card with verifiable data and read it all back. The counterfeit we tested passes every speed test there is.
Why does my 128GB card only show 119 GB? That is normal. Your computer counts in binary, so 128,000,000,000 bytes displays as about 119 GB. Nothing was taken. A real shortfall looks very different, and the Kioxia review explains where the missing gigabytes on that card actually went.
Do I need a special card for a Nintendo Switch 2? Yes, it needs microSD Express, and the marking to look for is EX. An ordinary microSD will not run games on it, and a UHS-II card will not either: both rows of pins look alike but carry different signals. We tested one on two readers to show what the format actually delivers.
Which of these should I actually buy? Answer three questions in the card finder and it names one or two. If you would rather see the working, the full card picker takes a specific camera and lays every measurement out side by side against it.
How much footage fits on a card? Work it out for your own camera's bitrate in the dash cam storage calculator.
What does the endurance rating on a card mean? Less than it looks. Every hours ladder SanDisk and Samsung publish is exactly linear in capacity, so the figure is the capacity times a constant, and the bitrate that gives it meaning is printed on the box rather than in the datasheet. The endurance-rating explainer shows the arithmetic and what the bench measures instead.
Do you only test memory cards? No. Anything storage can go on the bench, and the gear we rely on ourselves goes on first. The SanDisk Ultra Fit 32GB got the full suite before we trusted it to boot our card-identity rig.
Every figure on this page comes from one results file per card, written by a bench built for this: writes are committed and confirmed before the timer stops, a cache-leak canary aborts the run if the operating system serves a cached read, and every test byte is seeded so it can be verified on the way back. Cards are scored on five dimensions against the job they are sold for, and a card that misrepresents its capacity is capped at 1 out of 5 no matter how fast it is.
The full method, the rubric, the unit conventions and an honest list of what we deliberately do not test are all on one page. Alongside it sits the reference we built while identifying these cards: every SD manufacturer ID we have decoded, and an honest account of what that number can and cannot prove.
The bench, the five-dimension rubric, the conventions, and the gaps we will not paper over.
Read the methodLook up the CID register your own card reports, and see why a matching manufacturer ID does not make a card genuine.
Read the register