A free tool to quickly detect counterfeit flash (2017) | Hacker News Reader

A free tool to quickly detect counterfeit flash (2017)

fight-flash-fraud.readthedocs.io

217 pointspopol12131 comments

I bought a (little too) cheap large SSD off the internet. It was surprisingly slow, but it seemed to work fine, so I assumed that's the reason for the low price - until I tried to backup my other SSD on it. After the first ~50 GiB, all the writes suddenly failed and I could only perform reads.

After re-formatting it and attempting the backup a few more times, I was frustrated, so I searched the internet for related problems and found out about these so-called "chinese scam drives" that announce size to the drivers that is much larger than actual, and just throw away any writes above some memory address.

I quickly found f3 and tested it - and sure enough, it was a chinese scam drive. I reported the seller to the local inspection and they confiscated all the other drives and gave them a huge fine. I feel pretty smug about it.

What do you mean by "the local inspection"? The police?

the SSDPD

I suppose the US equivalent would be the FTC (assuming they can confiscate and fine?).

lol you’ve clearly never interacted with the FTC.

Did the seller happen to be based in your country? It seems unlikely to me the FTC would do anything about an online purchase to begin with, especially not if it was from China.

The US has an International Trade Comission that does seize counterfeit imports and those that violate trademarks without being deceptive clones. They aren't going to do much for one unit, but if an importer is bringing in container loads, then they may exercise the authority.

> Did the seller happen to be based in your country?

Yes, they were running some internet re-sell business, probably buying cheap stuff off Amazon and re-selling it locally for a higher price.

From the wording, the GP didn't mean a federal organization.

Does the US have no local trade protection organizations? They are organized in a way that I would expect their states to insist on this.

Not op, but if someone is advertising X (eg. 10tb o space), and the reality is Y (not 10tb of space), you can call "trades inspection" (Tržna inšpekcija), and they can issue fines, etc.

Any hope of AliExpress policing this at some point?

Site that only offers 2 weeks for order complaints after receiving? Not a chance, they love money and don't care about anything after item has arrived.

Had a similar experience with a friend's "10 TB" SSD. After I tested it with f3 and confirmed it was fake, I opened the case and found a 64 GB microSD card and an adapter/faker board.

"10 TB"? Why did you even need to open it??

Well, at first it only showed up as 2 TB, which was at least possible though unlikely. But f3 indicated it was faking 10, at which point I realized it was presenting several additional partitions that were so corrupted the OS wasn’t even making them available to mount. After that I opened it just to see what was inside.

I suspect LorenPechtel's terse question implies that a 10 TB drive would weigh appreciably more with real hardware than a microSD card.

SSDs can be pretty lightweight. And also sometimes the fake ones have a piece of metal or even a rock glued to the inside of the thing to make it feel heavier.

Weight doesn't even have to be accounted for. There's simply no technological or marketing reason to manufacture a 10TB SSD. An odd size like that is already a massive red flag.

They do make 10 TB+ SSDs:

https://www.allhdd.com/kioxia-kcmyxrug15t3-15.36tb-ssd

Infact, Kioxia's enterprise line goes up to 30 TB/drive.

You linked a 15.3TB drive which doesn't dispute his point. 10TB has never been a manufactured SSD size that I'm aware of. In the enterprise we've got approximately 960GB, 1.92TB, 3.84TB, 7.6TB, 15.3TB, 32TB.

10 is skipped entirely, unless you can provide an example of any reputable manufacturer producing one.

Consumer drives don't follow that exactly, but consumer SSDs also don't hit 10TB+.

Exactly. I didn't say 10TB was too big, I said it was an odd size. There's no easy way to get close to 10TB when using components that are sized by powers of two, plus or minus varying amounts of overprovisioning depending on market segment, and GB vs GiB differences. ~8TB SSDs are common in consumer and enterprise markets. 16TB drives that expose 15.36TB usable space are common in enterprise, and 12.8TB usable space from ~16TB raw flash isn't unheard of. 10TB usable space isn't theoretically impossible, but it simply wouldn't make sense.

> There's no easy way to get close to 10TB when using components that are sized by powers of two...

I expect you can get any usable capacity you like by reserving some subset of the flash for onboard spare/scratch space. I think I remember long ago Anandtech doing some benchmarking the changing performance of some drives as they adjusted the size of this "housekeeping" section of the drive. No clue if it's adjustable on every drive, but it sure was on the ones they were testing.

Most drives don't have any special functionality for adjusting overprovisioning. You just don't touch a large chunk of the LBA space and you get more or less the same effect. Leaving part of the drive unpartitioned, or creating a partition but not putting a filesystem in it will accomplish that purpose.

Drive vendors can tweak this in firmware to make the drive appear to have lower accessible capacity (or higher, for fraudulent drives). But as I've said several times, doing so to make a 10TB product would not make sense. The drives that expose a 12.8TB usable capacity from 16TB of flash already have far more overprovisioning than almost anybody needs. Further reducing that to 10TB would be throwing away capacity for little or no performance gain and a useless improvement to write endurance. It's not a product any rational, non-fraudulent vendor would create, because there's no demand for such a strange configuration. The fact that it's theoretically possible to create such a product does not actually make a 10TB SSD less suspicious.

(Side note: you don't have to tell me about what Anandtech tested with SSDs. Been there, done that.)

There's no easy way to get close to 10TB when using components that are sized by powers of two

5 x 2TB? 16Tbit (2TB) NAND flash is available these days.

This old Intel SSD is 10 x 16GB: https://www.storagereview.com/wp-content/uploads/2010/04/int...

SSDs aren't sized in powers of 2 anymore. Even the flash itself isn't due to things like spare area (and TLC flash is internally actually a multiple of 3 times a power of 2 size.)

No, I'm saying that flash devices are pretty close to powers of two because that's how the chips come. I would presume any solid state device purporting to be 10TB would be fake unless I had clear evidence otherwise.

Just received 2 cheapo 64GB micro sd cards from aliexpress, they seemed legit, had tons of reviews with OK crystaldisk performance screenshots and... they're junk This tool quickly identified that they were counterfeit of type "Limbo", with 16GB of capacity instead of 64GB. Thanks to Michel Machado for writing this gem.

A part of me will see stuff like "3TB Flash Drive" on AliExpress for $3 and briefly consider buying it. Like, I know it's obviously not true, it's obviously a lie, but I also am curious about what I would actually get.

I never do it because I don't really see the point of paying money for something that will immediately go into a landfill, but it's always tempting.

Back when it was being done with external hard drives there were YouTube channels dedicated to breaking them open and experimenting with how they did it.

Fun stuff to watch on a boring afternoon.

The potpourri of a youtube channel, Atomic Shrimp, has done a couple interesting videos doing just that.

His channel is where I first heard of this scam.

https://youtube.com/watch?v=UsWx1iO-aeA

For consumer stuff like this, you'll obviously get something else. But sometimes you can get recycled components if the price obviously is not sane, where recycling is an option.

It's only fraud if the description mentions new and original. :)

My trick to get decently priced computer equipment is to buy used stuff from data centers; generally that stuff will sell for pennies because their goal is to "get rid of it", not "get rid of it for a profit". Desktop network cards and switches are almost always available for very little (sometimes literally just $1 + shipping).

Look at govdeals.com for stuff from universities, they seem to sell a lot of their surplus hardware there.

Govdeals is sort of addictive. I've only bought stuff on there a few times, but it's fun to look around and see stuff like "pallet of 400 laptops" selling for $600.

How do you find such things? In person or online somewhere?

Sibling comment mentioned Govdeals, which is a place I'll look sometimes if I can find something near me or they're willing to ship.

You can look on Craigslist as well, especially if you live in a large metro area (I live in NYC so it's easy to find lots of corporate surplus). Companies will often need to upgrade all their infra and liquidate their equipment. My printer is a giant nearly 20 year old HP office thing that I got for $30 + $100 for the Uber back home from an office that did a full IT renewal.

When you say data centers, what do you mean? If I was looking on Craigslist or whatever for this, how would I be able to tell which seller is a data center?

A few sellers on eBay are honest

  Description:
  This item is upgraded by 32 GB Pen Flash Drive to 2TB
  32GB-2TB, the actual capacity is 32G, the computer displays 2TB, the detection is also 2TB, more than 32G things can be stored, but not displayed.

  https://www.ebay.com/itm/234810575961

That's not bad. Most counterfeits I get are just 2GB so hardly useful. Still, the counterfeit firmware makes compatibility issues more likely, so better to buy one without the counterfeiting.

Is there any reason I would deliberately purchase one of these over a standard 32GB drive? Trying to think of if there’s a case where I would want to trick some hardware into thinking it has a 2TB drive when it really doesn’t?

> more than 32G things can be stored, but not displayed.

That's certainly an interesting way to phrase it. I use /dev/null for a similar purpose with the added benefit that it can store petabytes of data, but sadly we don't have the technology to display it yet.

Lol. That's up there with the [1-bit Bloom filter](https://www.xkcd.com/2934/)

But why would anyone want a fake 2TB though?

Surely they write this to cover their ass when people don't read the entire listing.

Makes backups cheap and easy. Also if you _know_ it won't restore, you stop worrying about whether it might or not.

what you are buying is a piece of paper with a picture of an XBox360 on it

The issue is that if you exceed the (invisible) limit, you start overwriting the existing data. That's Bad (TM), since there's no indication that it's happening, until you attempt to retrieve the data and discover it's corrupted.

The subject of this topic, f3, will quickly probe the drive, determine the true size, and create a partition on it of the actual usable size. Using that partition, you will never lose any data. If you try to create a partition in the rest of the drive, it will be immediately corrupted, no tools will show it as good, you won't ever be able to get around to putting data on it to lose it.

> will quickly probe the drive

Some scam SSDs are smaller drives with hacked controller firmware that is bright enough to avoid detection with quick scans - you have to write beyond the real capacity then re-read to confirm corruption.

Have you found cases where "f3probe --destructive" didn't work for you?

Not first hand, though I've used tools of similar intent and found the quick heuristics, while usually very good, not 100% reliable, and for storage I want to be sure not almost certain.

Note that I was warning about quick tests, I'll have to try “f3probe --destructive” on something next time I have some playtime to see whether I'd consider that quick! Of course, it could be much quicker than the full capacity write some tools use to find a defective or deceptive drive¹² but quicker and quick aren't the same.

If a quick test says the drive is fake then it is, and many fakes can be detected these ways, but it takes more to convincingly state a drive isn't fake.

[1] Only a little over true capacity should be needed for that, in the case where writes to anywhere up to capacity will work (which is common as creating a filesystem more complex than FAT/exFat often requires activity similar to what quick write/read heuristics do).

> Using that partition, you will never lose any data.

... unless/until the underlying flash fails. Which it probably will; there's every reason to expect these drives are made with the cheapest, worst flash available.

>A few sellers on eBay are honest

no, that's not honest, farthest thing from it, that's what makes it an actual scam. by having that in the fine print, you can't return it, and they won't be flagged.

You can certainly return an item if the title of the listing was inaccurate.

In my experience, eBay won't "flag" dishonest sellers of counterfeit storage products, anyhow. I tried to get a listing selling counterfeit Samsung micro SD drives taken down, but eBay did nothing. I was refunded, but the unsuspecting will continue to be duped.

I don't think buyers of these SSDs want to return it. The only reason to buy such an SSD is to resell it. Effectively they are enabling other scammers.

In what world is a title that says 2TB for a 32GB drive "honest"? Is your view that a lie becomes honest as long as you add some truths after it?

It's a fact of life that titles can't always be a fully accurate description of the whole subject. Otherwise we wouldn't need the part that comes after the title...

> It's a fact of life that titles can't always be a fully accurate description of the whole subject.

It's also a fact of life that there is a gigantic amount of room between "not fully accurate" and "blatant lie"

> more than 32G things can be stored, but not displayed

"Honest"?

They’re suppliers to other scammers

> 2GB so hardly useful

I used to buy 2GB drivers all the time. The only reason I stopped is because it's hard to find anybody selling them.

It's plenty of space for a portable drive, it's enough for a Linux installation, it has lots and lots of uses.

But I wouldn't buy a 2GB driver that pretends to be larger. At least not willingly.

> It's plenty of space for a portable drive, it's enough for a Linux installation

RHEL9 installer is 10GB.

I did this on Amazon knowing it was a scam and immediately returned it just to drive up their numbers and hopefully get them shut down faster. I have no idea if that did any good, but what disappointed me was Amazon only gave me credit, not a full refund.

Yeah, aliexpress returns real money.

And there's honest stuff on there too, loads of it actually. I bought a 128GB SSD from there many years ago for $10. I expected nothing but like you say, I was curious about what I'd get and something faster than a shitty USB stick with decent storage capacity was all I was after. I verified it, and it works perfectly and it's 128GB as described, it just lacks cache so it bogs down on super large files. But for $10 I couldn't buy a better USB stick.

They may have true capacity but long-term retention is yet to be established: https://htwingnut.com/2022/09/11/ssd-as-long-term-storage-te...

There's even SLC USB drives being advertised, an otherwise extremely overpriced item these days, with sellers posting the exact part numbers and pictures of the insides as evidence. Those may be using recycled flash.

A few years ago Micro Center had a coupon for a free 120gb SSD, which I guess worked because I ended up going there and buying stuff in order to redeem it.

The SSD wasn't great, it was pretty slow, but it did work, and I have used it in conjunction with a USB SATA adapter as something where I can write lots and lots of tiny files for projects I work on [1], and where I don't particularly care if I break it. Surprisingly, it's still holding up...pretty good for free!

[1] One project I've done involves breaking up video files into a picture of every frame, then do some kind of image processing, write the new files, and then glue them back together, which means there can be a ton of writes on there that I don't want to be done on the expensive and irreplaceable internal SSD on my Mac.

https://www.trendforce.com/price/flash

TLC flash is currently less than $0.10/GB. $3 will probably get you a 32GB drive with flash of dubious quality.

This tool could save you from corrupting data but once you buy these counterfeit cards you're better off trashing them than requesting a refund, shipping will cost you more that the drive itself.

Chargeback - goods not as described. No need to return them.

Supposedly, card processors are meant to suspend accounts when they get too many chargebacks. How is that number determined? If it's a percentage of total charges, then places like Amazon/Ali* will have so many other charges these will pretty much round to zero. Also, these vendors are "too big to suspend", so chargebacks will do nothing like what they are meant to.

What do you mean they don’t do what they’re supposed to? They get you your money back. That’s what they’re supposed to do.

Dropping the merchant is a bonus.

The chargeback is meant as last resort though when a company is not cooperating. The company has already been refunding without having to do a chargeback. The chargeback is meant to let the processor know that their customer isn't holding up their end of the agreement.

You're correct that it's a last resort, but its primary role is to, literally, charge back to the original merchant. "[letting] the processor know that their customer isn't holding up to their end of the agreement" is a side benefit. As is eventually dropping the merchant when there are too many chargebacks to justify supporting them.

> Through a chargeback, your bank can try to get your money back from the seller on your behalf it isn’t a legal right, but your bank is committed to helping you, and will treat any claim fairly.

https://www.visa.co.uk/how-you-pay-matters/chargeback-purcha...

> When a customer disputes a debit or credit card transaction, the card issuer must determine whether to provide that cardholder with a refund for the transaction amount—also known as a chargeback.

https://b2b.mastercard.com/news-and-insights/blog/what-is-a-...

> A chargeback is a rules-based mechanism, with time-sensitive workflows, that enables the issuer and the acquirer to determine the financial liability of a disputed transaction.

https://www.mastercard.us/content/dam/public/mastercardcom/n...

> A chargeback occurs when, after investigation of the dispute, we debit your account for the amount of the disputed transaction and credit the Card Member with this amount.

https://www.americanexpress.com/au/merchant/chargebacks-and-...

Notice none of them talk about it being a feedback mechanism. Because it's not. It's a refund mechanism used as a last resort.

> It's a refund mechanism used as a last resort.

you've just skipped over then entire part where the websites are freely refunding and not denying refund claims making the last resort of a chargeback entirely unnecessary

You are commenting in a thread about chargebacks as an option. You are the one that claimed chargebacks aren’t doing what they’re meant to do. They are, you just don’t know what they’re meant to do.

Whether or not a website does refunds on their own does not change the definition of a chargeback in any way shape or form.

Yes, you're so wrong. I've had merchant accounts before, and there is clear wording about the negative impacts of chargebacks. I've even done a chargeback as a consumer, and they asked me if I had already been in contact with the seller before making a claim.

So the blind advice of making a chargeback claim before making a refund/RMA type of request with the seller is really out of order in the steps to take.

Clearly the definitions from all three major processors, both from consumer and merchant documents, don’t mean anything. Your anecdotes are far more convincing, for sure.

> So the blind advice of making a chargeback claim before making a refund/RMA type of request with the seller is really out of order in the steps to take.

That wasn’t even your point in the first comment. It was that chargebacks don’t do what they say they do. Way to move the goalposts when your initial point was clearly proven wrong.

You don't need to ship them back. Aliexpress will fully refund fakes, if you send something resembling proof. I got refunds for even obscure stuff like opamps, transistors, etc., with just a quick video of a oscilloscope output.

Sometimes it's trivial to prove, like CMOS opamp with +6V absolute max Vcc supply happily working at +40V.

I did not know that. Thanks.

> Sometimes it's trivial to prove, like CMOS opamp with +6V absolute max Vcc supply happily working at +40V.

Double take

What was going on with that one?

Almost all cheap opamps on aliexpress are LM358 or similar supercheap bipolar design, no matter what's marked on the package.

> had tons of reviews with OK crystaldisk performance screenshots

Select the 1-star reviews. Usually someone helpfully posts proof of the scam.

Until a new scammer runs a protection racket offering to withhold the scam accusation for a price.

I've bought a couple things from Aliexpress, and payment processing is such a hassle[1], I don't know why you would use it to buy things that are easily found from domestic sources? Especially SD cards which are widely counterfeited.

[1] This was a couple years ago, maybe things got streamlined? Of my cards that don't have a foreign transaction fee, Aliexpress wants the phone number off the back of one, which is sketchy; no thanks. The second one, charges don't go through, and the issuer customer service can't even see the attempts; I have to ask them to disable security on my card for ~ 30 minutes, and then the charges go through. Billing showed from England, IIRC. Doesn't (edit: Didn't! thanks) support any intermediates I do (paypal/amazon pay) which is usually my goto for low trust transactions.

They do support PayPal nowadays. Paying on AliExpress is as easy as on Amazon, and a lot of things do have a 5 day delivery guarantee and it does actually work. I've been using it quite a lot personally(I'm in the UK).

That sounds like issues with your bank, not AliExpress.

I entered my card number with them years ago and it just works.

What I hate is their login and search - both horribly broken.

But as to why, it's a better selection than Amazon for many things (for example components) and usually cheaper.

Wow, they scam for that little of a price difference? 64GB costs almost nothing...

Yeah, really... I picked up a 128GiB card from the store on something of a whim a while back, and I'm not even sure I paid more than about $18USD (with tax) for it. I bet I could have gotten it cheaper, but I was impatient.

A helpful tool I recommend for buying from common outlets online is Fakespot [1]. It scans the reviews and looks for suspicious/fake reviews & other telltale signs of deceit. Mozilla recently acquired it, but you can scan Amazon URLs on their website, and they have a very helpful browser addon. There was a similar tool I used to double check called ReviewMeta, but they seem to be offline.

It's not a 100% foolproof way to determine if a vendor or product is fake, but it is helpful. There are some other things you can do to double check things as well. [2]

[1] https://www.fakespot.com/

[2] https://www.wired.com/story/how-to-spot-fake-reviews-amazon/

Every single product on Amazon has a fake spot of like D these days. They flooded the field.

For anyone looking for tools that do this, it seems like a good opportunity to mention Steve Gibson’s Validrive tool [0] if anyone out there is trying to help family and friends who might be scared off by a CLI tool, and I believe it’s non-destructive.

I’m glad to see more awareness of this issue and entrants into the space.

- https://www.grc.com/validrive.htm

It warms my heart to see that Steve's website appears frozen in time (and works just fine). I bought Spinrite nearly two decades ago and it saved my bacon more than once. Also loved to listen to him and Leo on Security Now in the very early days of podcasts ("netcasts" lol).

Security Now is still going strong and I listen weekly! And Steve recently committed to continuing past 1000 episodes (he was previously planning to wind things down).

He’s continuing to do awesome work and I deeply appreciate him for it.

Steve is an interesting guy and while his website and software development methods are downright ancient, I would trust his programs over almost any other vendor. SpinRite has saved me also. I'd encourage anyone who isn't familiar with him to check out the Security Now podcast. I have been listening since 2005.

I was surprised that when he bought 20 cheap "1 or 2 terrabyte" thumb drives, all of them were frauds. This was on Amazon just this past September.

Really? You were surprised? Something too good to be true actually turned out not to be true? From Amazon no less? I'm guessing you forgot the /s at the end of your comment

Well yeah, if you take a look at the "group photo" (https://www.grc.com/validrive/drives.jpg), you can see that all of them are either no-name, have "brands" like "Blanbok+" and "Dianww", and one of them is even a borderline counterfeit SanDisk product (the SD card). I suspect that if he had bought a (non-counterfeit) product from brands such as Kingston or SanDisk, he would have got the actual advertised capacity (although probably not as cheaply).

Does anyone have any experience with mass testing flash drives as part of an assembly line?

Another tool for testing flash drives that was recommended to me was H2testW.

The tool brings a huge problem. You need to buy one to know it is fake, thus making the counterfeiter efforts successful.

Bad product reviews on the sales site won't work as they can be easily circumvented if not removed.

Bad product reviews on 3rd party web site won't be effective as well.

I think the right tool is a website to show updated buck/TB prices. So we can avoid buying fake devices. The proem is how/who would keep those data up to date. And how to make that site a popular choice for buyers.

Once got a cheap "Sandisk" SD card from an Australian ebay seller, which had much less capacity than what was displayed when it was plugged in.

After I tried to store more and ran into issues, I did some investigation and found out that it was a counterfeit Sandisk.

I emailed a screenshot with proof to the seller. Because of the very good consumer protection laws we have here, I got my money back.

> Because of the very good consumer protection laws we have here, I got my money back.

Sometimes a scammer will give money back easily just to avoid a negative review, irrespective of local consumer protections or lack thereof, if they have currently managed to maintain a good rep (via exchanging good reviews with buyers who haven't tested the item yet, or have passed it on as a present and never will test it, from fake reviews, managing to get negative reviews cancelled the harder way, and other trickery).

I believe this was so long ago that reviews themselves weren't very widespread (or at least not for individual products. The seller had good ratings).

There's no need for this.

Buy name-brand storage from reputable sellers.

Of course the fantastically cheap stuff on Alibaba is fake. You don't even have to check.

I don't have a link, but I thought I saw that this was happening with name brand drives from third party sellers on Amazon as well. And given that Amazon co-mingles product, it's a crapshoot.

Dont even think about buying flash on Amazon. Use a reputable store like B&H, Digikey, Mouser.

Yes, this. Or just buy it in person at someplace like Target or WalMart (and not at a "dollar" store). Amazon cannot be trusted.

This is generally good advice, but the sophistication over the years of some of the fake Sandisk stuff as one example has meant at times even seemingly sound storefronts have become polluted with fakes.

I imagine all sorts of random commingling type activities have happened in Amazon warehouses over the years, there is so many avenues for convincing fake crap to accidentally get sold in mainstream sales channels.

reputable does not mean amazon. they hide random sellers on some products and lie they are from the oem on the product page.

half of premium kingston models I got from the did not fall under the obvious and amateur scam this tool detects... but were lower quality chips or outright QA failed units. that half always fails around the 3x capacity writes. so infuriating.

I don't think anyone is suggesting to intentionally buy scam drives.

IMO there are no trustable SSD and SD card brands. They are all bad.

How so? I've never had any issues with any of the big name brands in the past, without any issue.

I got a few new USB drives at work for testing data centre hardwares. I normally would run f3 on new flash drives but this time the deadline is so rushed so I skipped that. Then I wasted an hour diagnosing a mysterious problem, and eventually I found out the usb drive is faulty after testing it using f3.

I then tested all of them and found out 4 out of 8 of them aren’t faulty, some of them died and disappeared.

So test your hardwares, test your hardwares that’s used to test hardwares. You will never know you can trust them unless proven.

Edit: badblocks, SMART test, memtest86 and memtest86+, prime95, Intel burn test, OCCT, iperf3, etc are equally useful.

I can't be the only one who read the title and thought this was about detecting a counterfeit Flash player.

It was actually silverlight in an adobe branded trenchcoat

Silverlight, now that's a name I haven't heard in a while. Next you're going to say Macromedia.

I misread the title and I thought it was some sort of hardware to identify vegan meat

This tool has a function to "correct" the capacity. I can't understand why that would be useful, I would not trust a device like this at all for any purpose.

I picked up a handful of "580gb" microSD cards off AliExpress recently, with the goal of using them in mp3 players around the house. I figured they'd be reject 64gb sdcards, and I was right - they averaged 50-58gb of usable storage. Perfect for writing once, reading many times, and cheaper than even 16gb legit cards. They can be useful, just know they may also die at any moment.

Has someone made a flash drive tester as a standalone hand-held device? That would be useful for buyers and incoming inspection. Haven't found one yet.

That's not a bad idea! It wouldn't be too hard to set this up with a raspberry pi as a prototype

I sort of wish there were places that sold new 2Gb drives as honest 2Gb drives rather than pretending they're 512TB. There are plenty of use cases (i. e. for a Gotek/Flashfloppy setup where 2Gb would hold more software than ever existed for the host machine, or as a cheap item you don't need to ask for back when you loan it to someone).

It's great that I can spend 10 bucks and get a competent 32Gb drive, but if we could get around $1, we could treat them as semi-disposable as floppies were.

2011, I bought a counterfeit Kingston card over Amazon. At the time this was not cheap

https://i.imgur.com/4XeaX.jpeg https://i.imgur.com/FZEYA.jpeg

I tried to fill in a warranty claim when I ran into problems. No dice. I always make sure the seller is quasi official now.

How do you check the sellers validity?

Steve Gibson of GRC (SpinRite, Security Now podcast, etc) made a free Windows tool specifically for this called ValiDrive:

https://www.grc.com/validrive.htm

If you are a Windows user, ValiDrive seems like the much easier choice.

Many comments assume (quite naively) that Big Marketplaces have no idea that it happens, and are fooled by shrewd counterfeit sellers who are always one step ahead. I suppose it is a complete fantasy, and those marketplaces have precise analytics on how big is the market of selling crap to illiterate consumers. If they voluntary abstain from that, and ban swiftly, they'd simply let competitors feast on that crowd, which would be bad for business™. Therefore, the handling of negative reviews, refunds, and fraud detection will always leave just enough margin for cheap counterfeits to be sold. As long as they both profit, it's a controlled dumpster fire.

This website traps me on mobile, can't use the back button. Fennec on android

Works perfectly fine for me. v125.3.0 with uBlock.

Reply on news.ycombinator.com