You're not imagining things – USB memory sticks are getting worse

A German data recovery specialist has confirmed what many Reg readers will have suspected: USB memory sticks are getting less reliable. The cause, as you might have guessed, is inferior memory chips, while the move to storing multiple bits per flash cell also plays a part. CBL Data Recovery posted that the quality of newer …

[Grogan]

I got so sick and tired of all the crappy USB sticks available. I could go around to every store in the area and not one would have a decent one. I went through several and they all had such crappy, pulsating write speeds that they were pointless.

It seemed to me they were missing some SCSI logic and buffering wasn't working correctly, in addition to it being crappy memory.

So I ordered a big hundred dollar Corsair GTX that does about 400 mb/sec (according to iotop "actual write speed") for both reads and writes.

[Anonymous Coward]

Snap. I have one of those which has done very well for years, used mostly for pratting around with OS install images. I also keep a datashur pro around; which is not so much a performance thing as it is a security one.

[Chewi]

My SanDisk stick died last week. No, not some recent purchase, but one from 2008. I didn't write to it all that heavily, but I did read from it all the time. I was actually quite sad about it. I highly doubt its replacement will last that long, but I'll be sure to choose it very carefully!

[Anonymous Coward]

Depending on your use case, it might be worth looking at M.2 drives. They cost more because you need an enclosure but they'll last a hell of a lot longer, and they'll be faster.

[Lon24]

Size isn't everything

How big was your stick? most old ones were probably 8/16 or even 32GB. Since then we have found many alternatives to storing larger lumps of data. But if you just need to boot a computer or transfer some files then you may not need more today. But try and get a 8/16GB stick or SD card these days and even if you are lucky you won't find it cheaper than 32/64GB.

Paying for capacity you won't use which makes what you do use less reliable is not progress.

[KarMann]

Re: Size isn't everything

I've found that kind of problem with one of my sound recorders, where the time to boot and read the drive contents is proportional to the size of the micro SD card you have in it, and it's getting pretty slow around 8 GB. I was surprised how hard it was to get reasonably-likely decent-quality (i.e. a brand name I've actually heard of before) SD in sizes < 16 GB.

[BenDwire]

Re: Size isn't everything

Have you tried partitioning it? OK, so you won't have the full capacity, but you will have better speed. Possibly.

[CowHorseFrog]

Re: Size isn't everything

WHy would partitioning matter ?

SD Cards are block devices, the OS is simply reading diferent blocks from device, whether it reads a block from one partition, or a block from the 2nd partition, it doesnt care. A block is a block, partitioning and the file system are artificial layouts that only exist in software, on the hardware level its still the same, read a block.

[cyberdemon]

Re: Size isn't everything

He meant partitioning such there is a large area of unallocated space, i.e. the filesystem does not use the whole device. The device will only read blocks from the defined filesystem

[CowHorseFrog]

Re: Size isn't everything

The comment says the cards get slower when partitions getlarger than 8G, it never mentions efficient use of space.

[KarMann]

Re: Size isn't everything

I think it sounds like it's worth a shot. I don't even know why it feels the need to read all that despite the nearly-empty filesystem, but it might well be that if I partition e.g. a 32 GB SD with just a single 4 GB partition, it won't spend the time looking at the other 28 GB. I'll have to do some more extensive testing on that sometime.

[Terje]

Re: Size isn't everything

I suggest you look at electronics suppliers like mouser, they have quite a wide selection of microsd, and it's searchable by for example technology, so if you want slc you can filter for that and yes you can still get that if you want!

[the spectacularly refined chap]

In other news...

... the gamers paying a premium for memory with colourful "heat spreaders" glued on top are being ripped off left, right and centre. It's really quite effective at disguising the provenance of chips and you don't even need a packaging facility or laser etching facitilies to do it.

This is the kind of thing thats as old as the hills.

[sev.monster]

None of this is new, no-name fly-by-night manufacturers (including those crappy branded ones you can order in batch) have been pumping these things out for years. I guess the main issue is that they are becoming more prevalent as unknowledgeable buyers get the cheapest 1TB super ultra hyper fast USB thumbdrives on Amazon and are surprised when they try to write more than 256MB and it starts overwtiting old data. And by then, the seller has bolted.

[Zoopy]

Simple solution?

Make Amazon liable for defective products, and don't allow them to disclaim that liability via customer license agreements.

I know, I know, there's no political will to enact that. But a girl can dream.

[Malcolm Weir]

Re: Simple solution?

So your plan is to hold the reseller liable for wholesalers or manufacturers malicious or unintended failures or side-effects? That would certainly reduce Amazon's pervasiveness in the market, but with the added consequence of eliminating all resellers entirely!

[Dan 55]

Re: Simple solution?

I think they can be held responsible for mixing up real with fake goods in their own warehouses, for a start. Then after that they can be held responsible for supplying dangerous electrical goods or toys, crappy pwnable IoT shit, or TV sticks which try to hack everything on your LAN about 5 seconds after booting.

No reason why they can't meet the legal minimum that every other shop has to.

[Zebranky]

Re: Simple solution?

"So your plan is to hold the reseller liable for wholesalers or manufacturers malicious or unintended failures or side-effects?"

Why not? that's exactly how consumer law in may more civilized locations around the globe works, the retailer is responsible for dealing with the warranty and compling with consumer law.

The Retailer then claims back the costs of this from the wholesaler/manufacturer. If the Manufacturer refuses to play ball the retailer sues the manufacturer and will almost always win because the obligations to the consumer is enshrined in law.

Once strong consumer protection law in enacted there is increased pressure on the entire supply chain to only provide quality goods, because if you are a retailer who knows you will be on the hook for warranting that piece of junk the consumer had brought back there is no way in hell you are going to sign agreements with the Wholesalers/manufacturers that allow them to fob the cost of their junk failing onto you, you will simply find another good supplier.

I acknowledge that the rise of online clearing houses (ebay, amazon aliexpress, insert your favorite one here,...) has made this much more difficult to enforce than it was in the days of brick and mortar stores only, so if you need quality goods support your local brick and mortar stores and claim your warranties as necessary.

[ChoHag]

Re: Simple solution?

> if you need quality goods support your local brick and mortar stores and claim your warranties as necessary.

Unfortunately the brick and mortar stores have abandoned all but the most shiny and presentable of goods. They no longer stuff their warehouses and stock rooms with any of the add-ons and consumables that are essential but unsexy parts of owning modern consumer goods. Just try and pick up a hoover bag or food processor attachment without paying the Bezos tax. 100s of (incompatible!) food processors and hoovers though. With bluetooth.

It's been an uphill struggle against internet shopping for years for the very purpose of supporting brick and mortar stores but I'm getting tired of it. If I have to go to the internet for the essential bits and pieces you can't be bothered to keep in your warehouse, I'll probably buy all the shiny overpriced (but equally poorly made) stuff you *do* stock while I'm there.

And where are the cashiers?

[nobody who matters]

Re: Simple solution?

<........If I have to go to the internet for the essential bits and pieces you can't be bothered to keep in your warehouse, I'll probably buy all the shiny overpriced (but equally poorly made) stuff you *do* stock while I'm there...........>.

And the reason they don't stock the essential bits and pieces is because so many people buy them off the internet that there isn't sufficient demand in the physical shop to make having the store room space/warehouse for them economical to sustain.

Which then exposes you to the knock-off crap masquerading as the real thing on Amazon et all.

Viscious circle isn't it!

[BinkyTheMagicPaperclip]

Re: Simple solution?

Hoover bags - use Argos. I know Sainsburys have gutted them a bit, but for a lot of things they're pretty decent, and their reviews are honest.

Currys are actually pretty decent. Yes, sometimes the mark up over online is considerable, but at other times it is not and you do need to make *some* allowance for operating a physical store.

It does depend what you're buying though, the range of slightly unusual computer accessories that you could previously expect to find in a store or online has reduced to increase profit margin, and in those cases the only convenient options are admittedly Ebay and Amazon

[sten2012]

Re: Simple solution?

Having dealt with Currys in the past it's one high street retailer I'll actually celebrate going under.

Frankly, I'd rather support Amazon.

I don't know what they're like these days but they've screwed me enough times I don't intend to find out.

[Steve Davies 3]

Re: Currys

Have an upvote for wishing them to go bust. Cue looking for a bread maker just before Crimble. Website says that they have several different ones in stock at local store. Drive to local store. Zero in stock. Spotty faced assistant said 'we don't have any in stock and it has been like that for months.' All they had with a decent selection were Toasters and Air-fryers.

Sites like eSpares are also good places to go for things like cleaner bags, as is Ebay. No Baldy Bezos tax to pay there (until AMZN takes it over)

[BinkyTheMagicPaperclip]

Re: Currys

eSpares *used* to be good. Now at least for some products they will only push their third party alternative that is 'just as good'. After a faff trying to get strimmer wire that worked without hassle the only way of getting a genuine Flymo part was from Flymo themselves.

Their shipping is also extortionate.

[Malcolm Weir]

Re: Simple solution?

Sorry for the delay...

But your assertion seems a bit suspect: in the UK, for example, the Consumer Rights Act of 2015 imposes obligations on the retailer for a period of 6 months (Part I, Chapter 2, Subsection 14). Every other obligation is on the manufacturer, and it is unquestionable that a solid-state memory device failing sooner than expected (which is what we're talking about) is much more likely to be a warranty issue, not a trading issue.

So the situation where fake goods are delivered in place of genuine would indeed be a reseller problem (as long as the "fakeness" is real; I see a lot of cases where low-duty-cycle products are expected to behave like high-duty-cycle ones, which often results in a "caveat emptor" situation where a product was not fit for the purpose the purchaser intended, but the reseller had no way of knowing that). And indeed it is more than a little dishonest to pretend that Amazon et al don't address these sorts of problems (post-paid return shipping, refunds to original methods of payment as well as store credit, etc). I mean, Amazon does operate outside of the US, so to use your words, Amazon handles situations in precisely the same way as any other reseller where "consumer law in may more civilized locations around the globe works".

Or are you alleging that Amazon (et al.) in the UK, Germany, France, Ireland, Norway etc. do NOT adhere to local consumer law?

So it appears to me that you are unaware of how companies like Amazon operate, how consumer law operates, etc and are just eager to bash Amazon. And indeed there are plenty of things for which Amazon-bashing is appropriate, but those do not seem to have anything to do with the situation under discussion.

[Roopee]

Re: Simple solution?

Precisely.

When I used to do on-site technical support for home users I moved from varying sound a stock of bits and pieces that were useful for my clients and profitable for me (in terms of reasonable retail markup) to persuading/insisting that clients buy things such as PSUs and SSDs themselves, with my help if necessary - so that I didn’t have to carry the retailer risk of dealing with product problems.

[Anonymous Coward]

Re: Simple solution?

They are already liable under law. You don't have a contract with wholesalers or manufacturers do you. You have a contract with Amazon if you buy from them,

[Martin M]

Re: Simple solution?

Amazon beg to differ for Amazon Marketplace purchases. Point 11 of their Conditions of Use states that:

Amazon allows third party sellers to list and sell their products at Amazon.co.uk. In each such case this is indicated on the respective product detail page. While Amazon as a service provider helps facilitate transactions that are carried out on the Amazon website, Amazon is neither the buyer nor the seller of the seller's items. Amazon provides a service for sellers and buyers to negotiate and complete transactions. Accordingly, the contract formed at the completion of a sale for these third party products is solely between buyer and seller. Amazon is not a party to this contract nor assumes any responsibility arising out of or in connection with it nor is it the seller's agent. The seller is responsible for the sale of the products and for dealing with any buyer claims or any other issue arising out of or in connection with the contract between the buyer and seller.

[Jimmy2Cows]

Re: Simple solution?

Terms and conditions be damned. This might be a contract clause, but if it it trying to override local law it will not be valid.

Amazon may say they are simply the facilitator of the transaction, but if Amazon is taking your money and delivering the goods, how is that any different from say Currys PC World? If Amazon is taking their cut, that seems to make them the retailer, and therefore liable under consumer protection laws. Even if they don't want to be.

If they are just putting the buyer in contact with the seller, that is slightly different. But then Amazon should not be processing the transaction, and should not be storing the seller's wares. They should be passing the transaction through to the seller in full, and taking their facilitation fee at some later time. If the transaction goes through Amazon, if they are storing the seller's goods in their warehouses, they are the reseller. Whether they like it or not.

[DJO]

Re: Simple solution?

I think it would be that Amazon are acting as an intermediary agent for the seller and as the point of contact with the customer it's up to them rectify any issues and then take action against the seller,

[Lurko]

Re: Simple solution?

"Terms and conditions be damned. This might be a contract clause, but if it it trying to override local law it will not be valid."

Well, given this is a UK website, the prevailing local laws are those of the UK, and unfortunately masquerading as a "marketplace" is already well established as an entirely legal way of dodging retailer responsibilities. I think you'll find every citizen in the land would agree that if Amazon advertise it, process the payment, and facilitate the transaction, then they should be held to be retailers for all aspects of product safety and consumer rights. But what's the chances of the Morons of Westminster agreeing? The current shower have had nigh on fifteen years to close that loophole but have not even attempted to do anything, and the next lot aren't going to be much better.

[Steve Davies 3]

Re: Well, given this is a UK website

Sorry folks, since it moved to a .com URL, it is a US site. The IP address is allocated to Cloudflare, a US company and I might be wrong the owners of the domain are now a US company.

US laws might apply to cases that go to court. It all depends on the wording of the contract in the section that covers dispute resolution. I once saw the T's&C's of one company in the US that demands that all disputes are dealt with at a court in a drive by city in TX that meets for TWO days a year. Legal Yes. Ethical? Hell no.

[Anonymous Coward]

Re: Well, given this is a UK website

Hmmm. I entered "amazon.co.uk" into my browser and it loaded up the Amazon UK website, the URL remained as co.uk.

[Malcolm Weir]

Re: Well, given this is a UK website

Cloudfare is just a content delivery network. They'll provide delivery for any content, so where they're located is irrelevant.

The website is owned by Situation Publishing (a fact they cunning hide in the "About Us' section). The editor in chief is Chris Williams, who seems to be based in San Francisco...

[Jimmy2Cows]

Re: Amazon is not a party to this contract nor assumes any responsibility arising

Maybe not, but they should be. There's a clear loophole here where low-quality sellers can set up shop, hide behind Amazon's contract terms, scam a load of unwary buyers with shoddy crap, then shut down and disappear when the complaints start coming in.

Amazon as the facilitator must bear some responsibility, they are the enabler. They are using their presence to put otherwise unknown sellers in front of millions of eyeballs, sellers who otherwise wouldn't get a second glance. Amazon should be obliged to ensure goods sold in their name, which is essentially how this is seen, are of adequate quality and aren't cheap knock-offs or dangerous crap.

[I could be a dog really]

Re: Amazon is not a party to this contract nor assumes any responsibility arising

Yes, and similar issues regarding VAT - setup, trade, scarper, don't pay the VAT to the taxman. And reimbursement for some stuff where Amazon felt there could be a reputational issue with the "setup, sell crap, scarper with the cash before the returns come back" business model.

As a result, Amazon tried holding back money from sellers - which immediately led to complaints that it was hurting legitimate sellers as they were stuck with cashflow problems while Amazon held onto most of their money for months.

[Anonymous Coward]

Re: Simple solution?

Well then you are not buying from Amazon are you. You are only paying via Amazon. When this becomes troublesome, just buy a new one and send the broken one back as DOA. Then you can do it all via Amazon.

[CowHorseFrog]

Re: Simple solution?

I always thought if a company displayed goods or servies and then accepted payment for those goods or services they were selling said product.

I guess amazon is doing all that out of the kindness of their heart a bit like a neighbour giving away mangos for free outside their home.

[sev.monster]

Re: Simple solution?

For anyone reading this after, do note that this only applies to third-party sellers.

For items shipped & sold by Amazon, the terms are different, and Amazon takes a bigger role and bigger cut in the process. This is one of the perks of selling directly to Amazon, since they store the product in their warehouses, manage shipping in all aspects, handle customer service and returns, and offer Prime benefits and other deals to entice users without you having to do anything.

If it says sold by someone else under the listing, that's where the stated terms apply; and in such situations, Amazon is no different from eBay. You don't blame eBay when a seller lists a product using their services and lies about it do you? Both Amazon and eBay offer seller protection in such cases, though Amazon's is weaker. And while Amazon's cut is smaller than wholesaling to them when using third-party listings like this, it's still larger than eBay's—all of which makes me wonder what the draw is.

[mirachu]

Re: Simple solution?

We actually already have consumer protection laws that do that in Europe. The solution is to sell stuff from established brands and ignore random crap.

[heyrick]

Re: Simple solution?

"So your plan is to hold the reseller liable for wholesalers or manufacturers malicious or unintended failures or side-effects?"

Why not? Isn't that how it is supposed to work with actual shops selling actual things?

There seems to be rather a loophole in that the tat bazaars will take your money, take their commission, and then try very hard to have no responsibility if there's a problem. More often than not you have no idea who you're actually buying from (and if you look it's a bogus name like "Jenna Ortega distribution" and an address somewhere in China).

[Malcolm Weir]

Re: Simple solution?

Nope. Resellers are only on the hook for 6 months (at least in the UK), which is entirely reasonable if you're not trying to bash resellers...

USB memory sticks failing after more than 6 months are therefore the problem of the manufacturer. Which was the point.

[rafff]

Re: Simple solution?

" hold the reseller liable for wholesalers or manufacturers malicious or unintended failures "

Your contract is with Amazon or other merchant, not with the manufacturer. The entity you have the contract with is responsible for the the goods being of merchandisable quality. IANAL.

[Malcolm Weir]

Re: Simple solution?

Yup, for the delivery of the product. Sorry to burst the bubble, but resellers have nothing to do with warranties, only to do with delivery of what as promised.

You do, though, *also* have a contract with the manufacturer, which is where the warranty comes in. Product safety recalls happen, mysteriously without resellers getting involved. And resellers go out of business...

Incidentally, if Amazon delivers defective product, they'll pay for you to return it and credit or refund you or replace the item! Whooo-Hooo! Success! Just what people claim they don't do!

(There's a bizarre thread going on here that Amazon is simultaneously party to a contract and allegedly not honoring that contract, which is nuts because Amazon is just as subject to trading standards legislation and they know it, because they DO have L!!)

[katrinab]

Re: Simple solution?

In the UK, retailers are liable for that. Amazon get round that by not being the retailer.

[jahill]

Re: Simple solution?

In the UK, the seller is liable for a faulty product, not the manufacturer. Seems to work here (much as Amazon would like to claim that it doesn't apply to them after three months, even when they're the seller).

[Richard 12]

Re: Simple solution?

It's already the law in civilised countries.

When I buy something my contract is with the seller, not the manufacturer.

[chris street]

Re: Simple solution?

Why not? Thats excatly how it works in the UK/EU if Amazon sell something dodgy, they are liable.

Not stopped them from being an enourmous retail force here either.

[Anonymous Coward]

Re: Simple solution?

They are responsible, in UK law anyway. The consumer has a contract with the retailer, not with the manufacturer, and the retailer is responsible for providing a quality product and for support and refund up to a reasonable period after purchase.

[CowHorseFrog]

Re: Simple solution?

Amazon became a seller when they added the Amazon tax to the purchasing experience (tm)...

[Anonymous Coward]

Re: Simple solution?

What do you mean? They are already liable for products that are sold by Amazon themselves. For up to six years under the law depending on what it was and the expected lifetime.

They don't make it easy to claim though. So to avoid the required bs, just buy a new one and sent the old one back in the box as DOA.

[Dave314159ggggdffsdds]

Re: Simple solution?

They make it incredibly easy to claim. You send them a message, they send you a refund. One and done, almost every time.

[Anonymous Coward]

Re: Simple solution?

Well no, only within the initial refund period. After that you have to contact customer services (which the options for they do their best to hide), waste time in an often lengthy conversation while "I will just check that for you" etc. and quite often you need to return the goods too depending on what it was as wait for them to receive and inspect them.

If you don't want the money but just want a working item, what I said is far easier and faster.

[Dave314159ggggdffsdds]

Re: Simple solution?

Rubbish. You tell them it's not working, and they issue a refund - send them a message instead of using chat, if you really find the five minutes on text chat to be too long. They very rarely want to pay to get things back so they can throw them away - unless, of course, you've repeatedly scammed them.

[Anonymous Coward]

Re: Simple solution?

<Different AC here>

Perhaps for you. I had something expensive (£400) that i bought directly from Amazon fail outside the return period. Their online support were busy trying to fob me off on the manufacturer (Sony, who generally make good products, but have a pretty poor name for post-purchase support), until I pointed out Amazon's obligations under the Consumer Rights Act 2015, and that as the retailer it was their job to resolve it, failing which I'd resort to enforcing my legal rights.

[Anonymous Coward]

Re: Simple solution?

Only if you are talking about £5 items. Anything of any value they want back first. And you have to wait for that return to be received and inspected to get your refund.

[Anonymous Coward]

Re: Simple solution?

I needed a CPU with integrated graphics to flash a BIOS on a motherboard so that it would support the CPU I wanted to install. So I bought one off Amazon, flashed the mobo, and got a refund from Amazon (after sending it back).

Never had any pushback from Amazon, they refunded me no questions asked.

[Dave314159ggggdffsdds]

Re: Simple solution?

Make them liable? They voluntarily accept liability. Conspiracy nuts will conspiracy, though.

[phuzz]

Re: Simple solution?

In a way Amazon are liable for defective products, because they're the ones giving you a refund.

[Roland6]

The troubling bit is the finding: “ there were also branded products among them”…

So is that Sandisksamsung/WD et al usb memory stick really worth the label and the money?

[Charlie Clark]

From what I've read recently, yes. Had a no-name stick die on me at some point recently: it wasn't in use and was hardly used, just handy for the out share of photos, etc. Got Sandisk as a replacement and I'll see how that goes. The premium for the brand isn't that much but data transfer rates can be significantly better. Ask me in 5 years whether it's still working!

With the cheap sticks, I think we're seeing what we know happens with tools, cleaning supplies, etc: at some point they are so cheap that they can only be crap. Not that you always have to buy the brands, some of which are guilty of the same tricks, but it needs consideration.

[ibmalone]

About five to ten-ish years ago when USB sticks had dropped down to about £1/GB I occasionally stopped by Ryman's and checked whether they had Sandisk devices on sale, occasionally they'd be half price. I ended up with a couple of fairly cheap 64GB and 128GB devices (was handy to have some extra storage for my laptop at the time), alongside quite a few 16/32GB ones (some also came from supermarkets on the same basis). Have not bought any other brand in a long time after I had a string of cheap ones fail on me. They haven't been heavily used, but all are still going strong. Whether that applies to more recently manufactured ones (particularly given the story) I've no idea.

The 128GB does get noticeably warm though. It's useful but I try to avoid trusting anything long term to these things.

[Roland6]

Currently, I would only buy selected Sandisk devices from the high street (PC world, Asda, Tescos et al), with data rates on the package, as most brands/models were all low performing, okay for the odd Word document but not for anything that would require larger individual data transfers. Hence the surprise that PC World actually stocked a decent stick and although you might have to ask (they tend to be kept under the counter/in the stock room, due them being easily pilfered, are available in 1~2 at most stores ie. Need one now quantities.

[Mark Zip]

Use ValiDrive to check if the stick you bought actually has the advertised capacity

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

Steve Gibson over at GRC has written a tiny little utility to check the actual (vs. advertised) capacity of USB sticks.

As is often the case with Steve's utilities, the name is not great, but it does the job well.

[steelpillow]

multiple bits per flash cell

So. Not actually binary any more. I seem to recall that Bletchley Park's first offerings were decimal - approx 3.2 bits' worth per vacuum-valve "cell".

[mark l 2]

My father bought a 64GB Integral USB memory stick from a Tesco at Christmas so I could copy some videos from my PC onto it for him, and it was terrible!

I was trying to copy about 60GB of videos to it and i was getting max of 6MB/s transfer speeds, sometimes down to 3MB/s so it took me hours.

For a while I was even wondering if it was one of those dodgy ones that would report an incorrect capacity and start overwriting data after it filled up a couple of GB, but all the 60GB of data was accessible once it eventually copied over

If i compare it to my few year old 32GB Kingston Datatraveller G3 on the same PC i can get in the region of 70MB/S write speeds on that.

I have been trying to find a reliable USB stick with quality NAND to use a portable SSD, as the M.2 enclosures are a bit to big for my needs but not found one yet.

[Roland6]

Recently had cause to want a performant usb stick, the only ones I could find were from Samsung, and surprisingly available from PC World

https://www.currys.co.uk/products/samsung-fit-plus-usb-3.1-memory-stick-256-gb-silver-10245087.html

Whilst not as fast as the SSD (only up to 400iMB/s) it was good enough to run a Windows VM from.

Trouble is what type of storage do they use and hence using it for a live VM might not be a good idea.

[Anonymous Coward]

I've always found the Samsung ones get REALLY hot within seconds though.

[gnasher729]

Quad level is slow. I have a two TB quad level SSD drive, and it goes down to 60MB per sec.

HOWEVER it can write to the same cells as one-bit cells at 1100 MB/sec. So you can write 500GB at full speed to an empty drive. If you copy more, you go down to 60MB/sec. If you stop copying, it copies from 1 bit to 4 big cells in the background. That’s 3.6GB per minute. So after one minute without writing you can write 3.6GB at full 1100MB/sec speed. Right now, that drive has 800 GB filled, 1200GB 4bit cells or 300GB 1 Bit cells unused so I can write 300GB at full speed.

Note that today tri-level SSDs are a lot cheaper, so there is much less price advantage. But I compared it to a 2TB spinning drive, which isn’t much cheaper, which is a lot slower, and a lot heavier - that 2TB SSD is just 45 Grams.

[anthonyhegedus]

How do you persuade the cells to be written to as 1-bit cells when they're 4-bit cells? Do you have some special tooL?

[druck]

It isn't the same cells. It's a slow 4-cell drive with a faster 1-cell cache, as long as you don't write more than the cache at any one time it appears to be quick.

[Robert Carnegie]

Some time ago, I got some displeasing USB sticks devices carrying the Integral brand. They were sold in a multiple-blister pack i.e. two or three USB sticks packaged on one card, they were more boxy and less sleek than my usual black rectangular Integral sticks, and their speed, specifically writing data, was slow and I think stop-and-go. Whether they were genuine Integral or not, I think it was a bad deal. Perhaps the message is to not buy USB sticks in a multi-pack if high quality of product is important to you.

[Dave314159ggggdffsdds]

I bought a 128gb Kingston one recently, well aware it was incredibly slow. Wrote to it once, leaving it going in the background for however many hours it took to fill it up; reading is a bit slow, but once it's indexed (by my car's audio player) it's fine for the purpose.

Thing is, it cost about 6 quid. Really, nothing to complain about at that price. If you want fast ones, they cost much more - but they are available for those who so choose. For what I'm using it for, I'd rather have a super-cheap slow one, and I suspect most people are happy with that trade-off.

[John Brown (no body)]

Yep, for that sort of situation, it's basically a "write once" device, so even the cheap crap is good enough. I'm using a couple of fairly old 8GB sticks for audiobooks. I'm on the road a lot so a full 8GB stick does me a couple of weeks. When it's getting near the end, I fill up the other one ready to swap them over, so they each get filled every 2-3 weeks and I've head years out of them so far. Even the quad level cells with 600 cycles reported in the article should least me a good few years in my low write usage pattern.

[Dickie Mosfet]

The good old Kingston DataTraveler

Upvoted for giving the Kingston DataTraveler a mention.

I have a silver 512MB DataTraveler II that I bought in 2006 (Image) and a 2GB DataTraveler 100 (Image), and they're still going strong.

(I think I also have a 2GB DT G2 somewhere that's still sealed in its original blister pack)

You can still find DataTraveler sticks on eBay with the separate plastic end-cap. I imagine that they're old enough to be using the original SLC chips.

[Marty McFly]

ValiDrive

Run, don't walk, over to Steve Gibson's website and get ValiDrive.

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

You need this to detect deliberately misleading USB storage.

[Charlie Clark]

Re: ValiDrive

Thanks very much for the link!

[Mike_R]

Re: ValiDrive

For Linux try f3 (lowercase)

F3 (Fight Flash Fraud or Fight Fake Flash) tests the full capacity of a

flash card (flash drive, flash disk, pendrive).

F3 writes to the card and then checks if can read it. It will assure you

have not been bought a card with a smaller capacity than stated. Note that

the main goal of F3 is not to fix your removable media. However, there

are resources to mark the invalid areas.

[robinsonb5]

Re: ValiDrive

But wouldn't running a tool like this immedately take most of the drive's blocks out of circulation for wear-levelling?

Or can you do a "trim" operation on USB flash drives these days?

[BinkyTheMagicPaperclip]

Re: ValiDrive

According to a Google search, if the stick supports UASP then it *should* also support TRIM.

I would have thought a tool like F3 without a subsequent TRIM would impact on performance, but the performance should still be 'good enough'. You can't claim a '60Mb/s stick' is 60Mb only when unused, it needs to handle it once the stick is full, space freed, and data written again.

It's still far better to have reliable but slow performance over partial or complete data loss due to a controller that lies about the stick capacity

[robinsonb5]

Re: ValiDrive

Interesting - I didn't know about UASP - but it appears to be a pretty new development?

It's not performance I'm concerned about - it's lifespan.

The way wear levelling works (or at least did on CompactFlash 20 years ago, when I learned about it after killing a CF card in an IDE adapter*) is that the device does logical-to-physical block mapping, and keeps track of how many times each physical block has been written to. There's a small pool of "spare" blocks (not counted in the nominal capacity) which it swaps out when a heavily-used physical block is re-written.

When the device is brand new, all the blocks on the device are unwritten and thus in the "spare" pool - so if you re-write one small file 1000 times, those writes will go to 1000 physical blocks instead of hammering one block with 1000 writes. If you do a test that involves writing to the entire device and then don't trim it, the pool of spare blocks is now depleted to just the small number of over-provisioned blocks, so a bad write pattern can now wear out the device much faster.

[* use dd to write a disk image to the card, then mount it without the noatime flag. It won't last long!]

[BinkyTheMagicPaperclip]

Re: ValiDrive

I don't know - I don't actually use USB sticks heavily enough to worry about their lifespan. Now SD cards on the other hand, those I now refuse to buy anything but the endurance products, after too many failures or the cards suddenly going 'read only', and generally I try to use SSDs instead.

I don't see lifespan being an issue, a program such as f3probe should only need to write to the device once. After it's been established that the 16GB stick really has at least 16GB of concurrently usable flash, it's safe to use.

Surely the spare blocks are a buffer against an effective controller initiated trim. When the device is using the spare blocks to achieve fast write speeds, once idle it should be clearing some of the other blocks to replenish the spares

[Mark Hahn]

Re: ValiDrive

UASP is very not new. It's found on pretty much any usb-to-sata or -nvme adapter.

But the usb-stick market isn't sophisticated enough to care, so you won't find it there. Or at least no small/cheap/mass market ones support it, and it's surprisingly hard to find any that do.

Yes, if it supports UASP, it's almost certainly also got trim and smart. There's no such thing for usb storage that only supports BOT transfers.

[Alan J. Wylie]

I discovered the hard way that doing an initial backup of lots of files to a mounted USB stick with LUKS encryption caused so many writes that it wore it out.

Now, I create a temporary encrypted filesystem no bigger than the stick, back up to that and then "dd" the image onto the raw device.

[bazza]

>With quad level cells (QLC), for example, four bits are stored per cell, which means that 16 states have to be distinguished.

I was preparing to lambast such a sentence and ridicule it to pieces; 4 "levels" of voltage would be able to store 2 bits, not 4. However, it seems that by "level" the industry does indeed mean "bits", according to the wikipedia page on the matter. I fear I have done El Reg a temporary disservice.

Now, why can't they just call it as it is, either 4 bit cells or sexdeclevel cells, or hexadecalevel cells?

[Bebu]

QLC wears out after 1600 cycles...

Explains a mystery of mine.

For reasons lost in the mists of before my advent there were a bevy of vmware (esix) hosts whose storage were identical small (thumb nail sized ) global branded usb memory sticks inconspicuously stuck in the back of these servers.

One day soon after my advent most of these boxes stopped vmware~ing (possibly after a power outage) and wouldn't reboot except one solitary server which had remained up.

Once the existence of the usb sticks was established they were removed and checking them they were all missing great chunks of storage - more holes than a swiss cheese.

The PFY was dispatched by manglement to purchase more usb sticks with the promise to replace them with internal nvme ssd drives later (history repeating I suspect.)

A few days everything was restored and the one host that had survive the onslaught predictably had a different brand usb stick which puzzled me as to why all of one brand died after a wee tickle from the mains power the other didn't.

I am now guessing that these sticks all hit their 1600 r/w cycles about the same time.

Wasn't too long after this that I dis-advented myself from that circus.

Some things aren't worth defenestrating.

Today I would trust a 1.44Mb 3.5" floppy a whole lot more than these devices.

[Shalghar]

Re: QLC wears out after 1600 cycles...

"Today I would trust a 1.44Mb 3.5" floppy a whole lot more than these devices."

And your trust would be well deserved. Once every while i take out one of the C64s, one of the Amigas and the twice repaired RiscPC (CMOS battery dislikes prolongued periods of non use). Neither the C64 diskettes (5,25") nor the 3.5" from Amiga (880kByte/720 in DOS) or the RiscPC HD (1,6MB ADFS) have ever failed.

Concerning the C64 media, thats 40 years of storage reliability, even the few tapes i still have are still readable.

For Amiga and RiscPC thats still 30 years and the mechanical HDD in the RiscPC also still boots up without issues.

If you want it a bit older, the chip modules for the VIC20 also still work, but i believe ROM doesnt count here.

[bazza]

Re: QLC wears out after 1600 cycles...

I used to think that the weakness of floppy disks would be that the plastic film would eventually degrade. This has happened in old magnetic tape - it either goes stick or brittle. But floppies seems impressively far more resilient. I guess the fact that they're coated on both sides and the disk is a lot thicker than tape all helps.

IBM briefly talked about a storage medium based on just polythene. Microscopic dents melted into the polythene for 1's, unmelted for 0's. It's lifetime was thought to be near infinite (given the stability of polythene, so long as it doesn't get hot!).

[Shalghar]

Re: QLC wears out after 1600 cycles...

I do remember some concept of a roll of clear sticky tape as "holographic storage", using LASER and pretty interesting optics. Thats a few too many years away, though, might be in the late 80ies or early 90ies in one of the then pretty good computer magazines before they degraded from "hey lets build a coprocessor card for the amiga 500 with the new 68020" to "did you know that apple has a new browser ?" wastepaper of nowadays. :P

[Securitymoose]

I wonder what part of the world the main source of these devices is located…

Don’t tell me. I can make a good guess, especially if they come from a well known internet souk.

[khjohansen]

Re: I wonder what part of the world the main source of these devices is located…

*That* part of the world supplies ~80% of ALL devices (shrugs)

[Rich 2]

Stoprot

I don't know if it's of any use to anyone, but I wrote a program that generates Reed Solomon error correction data for any file, plus the means of using that data to recover a corrupted file.

I have used this for several years now and apply it to almost everything I back up to my NAS system

You can find it here - http://knockknock.org.uk/stoprot.tar.bz2

[Graham 32]

Re: Stoprot

There is also par2

[Anonymous Coward]

Quality, schmality

The 'high quality' Kingston USB flash drives we buy in my office (they're genuine, I've dissected some and even sent them back to Kingston) from 32GB to 128GB have pretty much all failed on me within a year and claiming warranty on them is slightly less pleasant than nailing your eyelids to a table.

Counterintuitively the most reliable have been the dirt cheap 1GB and 2GB trade show specials we used to have branded with logos, links etc. for various in house 'initiatives', I've got a few of those which are daily drivers.

[Rich 2]

Re: Quality, schmality

That would seem to confirm what is being discussed - the "dirt cheap" ones will probably have 'old' technology chips in them

[Anonymous Coward]

Re: Quality, schmality

Yeah, I've got several of the real cheap throwaway ones which are at least 6 years old and still rocking on nicely, I've not been able to break any of the Sandisk, PNY or Transcend flash drives I have either.

The Kingston metal cased ones got hot enough to be uncomfortable to touch and failed, unrecognised or 'last USB device you inserted' regularly, the plactic cased ones just stop being recognised or become unformattable.

Out of about 20 Kingston, I've got 3 which still work and I'm not convinced one of those wasn't acting up last time I tried to use it.

[khjohansen]

Re: Quality, schmality

YMMV - I choose to dodge the "SANSUMG" (sic) brand when perusing the Silk Road in the Sky - the "deny" (sic), QUEEDNS and Netac brands have been good to me ;)

[Dave314159ggggdffsdds]

Re: Quality, schmality

"claiming warranty on them is slightly less pleasant than nailing your eyelids to a table."

If you buy them via Kingston's Amazon store, the process is painless, IME. You send them a message (through Amazon) describing the problem and they refund your money.

[Anonymous Coward]

Re: Quality, schmality

We don't buy from Amazon because it's rife with fakes and easy for people to get caught out by not realising they aren't on a manufacturer's 'store' (and even sometimes when everything indicates you are, you still get burned).

Regardless of how easy your return/refund process was, I prefer stuff to work and save me the hassle of having to replace it.

We don't buy Kingston flash drives, cards or RAM any more.

[Dave314159ggggdffsdds]

Re: Quality, schmality

If you buy from the Kingston store on Amazon, you're buying from Kingston, but with Amazon's CS and delivery mechanisms. If you can't tell the difference between Kingston's store and other sellers, I don't think any of the advice here will help.

Not buying Kingston at all is defensible.

[Plest]

The only two USBs I've ever had that were shit hot where a Buffalo 16gb that I used for about 9 years as a boot device countless times for Linux, OSX and Windows, I still have it and at almost 10 years old now it's still working perfectly despite it's smashed case now wrapped in black leccy tape! The other device I have is a Rocket 512gb USB by Sabrient, absolutely superb! 3 years old now and still lightning fast and never had a single failure.

I have a drawer full of old microsSDs and USB sticks, almost like you have a drawer full of old biros. Just have to find time to test them and throw out the busted ones!

[Ozzard]

Oh look, it's tape

Mount and unmount physical media, read and rewrite regularly, avoid too many write cycles... just treat USB sticks as if they were tapes.

(Nope, still no "old fart" icon)

[MrReynolds2U]

SanDisk FTW IMVHO YMMV RTFM ISKWYDLS

I've been buying and using SanDisk USB sticks for years and I've only had one fail(†). But like any storage, if they contain anything that holds value, it exists in several other places too.

I wasn't unaware of the multi-level cell design until reading this, so I have questions:

* If I only use 20% of the capacity of a multi-level stick, will it just write to 1 bit/level of the cell, thus avoiding the slow downs and lifetime issues mentioned above?

* Is there a way to tell if they are multi-level when buying them?

† = it went read-only, so I haven't lost the data

[david 12]

Re: SanDisk FTW IMVHO YMMV RTFM ISKWYDLS

* If I only use 20% of the capacity of a multi-level stick, will it just write to 1 bit/level of the cell

It's been suggested that if you just leave such a device plugged in for a while, it will re-write as a background task, converting all the single-bit cells to quad-bit storage. So maybe not giving a resilience advantage?

If the first 25% of storage is fast, and then transfer slows down, then you've hit a re-write wall.

[BenMyers]

Can't be as reliable as SSDs

Unlike SSDs, flash sticks generally do not have controllers with firmware to replace failed blocks with spares. As I have found out the hard way, once you delete and rewrite memory some number of times, the flash stick becomes inoperative. Ditto with SD and microSD cards.

[Stuart Castle]

I've got a Sandisk Extreme Pro 512 gig USB. I might have the spec wrong, because I lent it to a family member a year ago, and haven't got the damn thing back. It is bloody fast, unlike a lot of cheaper USB sticks I've used. It's the third one I've had though. The first one failed just out of warranty, so I bought another. That failed, but was in warranty.

But, I generally use external SSDs, as they do tend to be faster, and I have to have a bag for my work laptop anyway, so don't really have a requirement for pocket size storage. I use the Samsung T range of SSDS. A couple of T5s and a T7.

The T5s are good, but I bought the T7 because it was faster, had 2TB of storage and could optionally be encrypted, with a finger print reader for access control. I also have an iodd st400, which is a 2.5inch drive case that can take an SSD or HDD, but offers the advantage that it can emulate a USB CD/DVD/Blu Ray drive with ISOs, multiple hard drives (using VHDs) and apparently a USB floppy (using IMA files). The emulation happens on the device, so it appears as those devices to the PC (including in the firmware).

Anyhow, the T7 offered incredibly fast access in the gigs per second range on a USB C gen 3.2 connection, and finger print security.

One day, I was backing my T7 up, so I could reformat it. I put my hand near it, and could feel heat radiating from it, so I touched it, and burned the palm of my hand. There were no smart warnings. I unplugged the drive, and let it cool down. It didn't start up I sent it back to Samsung, who did replace it.

I got a new T7 back, but it's noticably slower than the old one. It still averages hundreds of meg per second, but not gigs..

[StrangerHereMyself]

Magnetic storage

For my high-value data which needs to be stored for potentially a long time I still rely on magnetic (HDD) storage. Flash memory stored in a barn or shed somewhere will become unreadable if left unpowered for a couple of years (a decade or more) and subject to extreme temperature differences outside.

I'm pretty sure magnetic storage will last much longer under the same conditions and will still be (at least partially) readable in decades if not centuries.

[Dave314159ggggdffsdds]

Re: Magnetic storage

A hard disk that hasn't been spun up in a few years is likely to be toast for mechanical reasons like bearings seizing. The platters may still be readable with specialist equipment. However, bit-rot is also significant even over relatively short timescales unless the data is very robustly stored. Things like jpegs are _extremely_ flaky in that regard, where a single bit flipping can significantly alter the image; other formats, like ascii text, are much more robust, since a single bit flipping will usually be obvious, and only lead to corruption of a single character. But, you can't rely on the data stored that way to be absolutely correct.

So, if you want to store data on magnetic hard disks, extensive error correction and multiple copies are necessary, and so is powering the disks up fairly often.

[StrangerHereMyself]

Re: Magnetic storage

Very unlikely that bearings will seize if you let the drive acclimatize for a while (several days) before powering it up.

In any case: magnetic storage is MUCH longer lasting than Flash.

[Dave314159ggggdffsdds]

Re: Magnetic storage

I'm not sure what possible good letting seized bearings 'acclimatize' [sic] could do. They are already seized from sitting in the same place for too long.

[druck]

Re: Magnetic storage

My primary storage on my Raspberry Pi based NAS is a Samsung T7 flash drive, but I have several layers of HDD backup drives on other Pi's both in the house and in other places, that replicate daily, weekly and monthly. I haven't had any problems with the HDD that spins up monthly despite it being the oldest at about 8 years old.

The single bit flipping issues affects JPEGs but also compressed files. I used to have lots of compressed files of various formats, but after finding some on an old drive were corrupted (luckily I had copies), I decompressed everything on the NAS. Storage is cheap enough not to have to zip everything like we did when discs were only MB or GB.

[Mage]

Resold subspec chips

Sinclair started that way with transistors and then had the IC10 (a Plessey) and IC12 (Texas) with higher powers in his ads, but they were actually sub-spec but working amps.

[Eclectic Man]

So, what USB sticks are reliable? And does this also apply to CF and SD cards used in cameras?

Image at: https://www.flickr.com/photos/139030274@N06/25864318996/in/dateposted-public/

[webstaff]

SLC high endurance ones.

If it's important, buy industrial ones from a proper supplier.

Single layer ones are still made for this purpose.

https://www.mouser.co.uk/ProductDetail/Swissbit/SFSD2048N1BN1WI-E-QF-111-STD?qs=CiayqK2gdcK%252BEJtkAdLXbQ%3D%3D

[webstaff]

High endurance SLC chips can be bought easily.

Plenty of industrial SLC SD cards and USB sticks exist.

Plenty of my customers moan about dash cameras not having footage after crashes due to worn out cards.

We.sent an email around with links for high endurance cards.

Magnitudes more right cycles.

10 times the cost for a quarter of the size of other cards

Wow we got hostile responses and we explained why they should consider it and had all the long term cost savings shown.

People are dumb and short sighted.

https://www.mouser.co.uk/ProductDetail/Swissbit/SFSD2048N1BN1WI-E-QF-111-STD?qs=CiayqK2gdcK%252BEJtkAdLXbQ%3D%3D

Doubly so for.most business owners.

I just don't give them sympathy now when claims they could have won go against them or systems die that could have been still up.

[razorfishsl]

Actually they are NOT getting worse.

What is getting worse is the amount being bought from China.

As regards the Changing of the Density, this has been on going for 20 years.....

It does not shorten the life of the cells, the issue is the range comparators, the level of the charge on the cell has to be detected, if the Vref for the comparators drifts , the WHOLE content will be misrepresented or will dissapear. (actually it is still there, but not detectable correctly)

They are actually more reliable is they are used now & again, since the cells get re-written based on the aging of the comparators.

If you write the contents out and the comparators Vref moves , then its game over.

also if you get hold of some very well known chip brand datasheets, after signing the NDA, you will see that they spec all the way down to 1 or 2 writes, with multiple reads.

they even spec specific pages missing... all this shite ends up in the market, since it may have a valid use in writing hardware that only needs to be written once at manufacture, but with changing parameter data.

all you are seeing is teh bottom of the barrel criminals going below the bottom of the barrel to make a profit., since now everyone else is doing the same.

it's all the big retailers, that are involved.

[SuperGeek]

This sounds familiar...

"many of the USB sticks actually contained microSD cards that had been mounted onto the circuit board and were being managed by an external controller chip."

This sounds like the way those scammy drives are built on WISH and AliExpress that YouTuber SMOOREZ regularly buys to alert people!

[wolfetone]

Re: This sounds familiar...

Who the hell is buying items from Wish and AliExpress and expecting it to be a quality item?

[anthonyhegedus]

Main storage

It amazes me how many of our customers still use USB sticks, not for backup or anything even vaguely functional, but for the main storage of important data. Or the only backup. I've heard people say that they keep important company documents on this "backup drive" (= a USB stick) because it's more reliable than a computer's own storage. Or even the server. People sort of feel an attachment to a system whereby the file is physically in their hand and can be placed in a jeans pocket.

For me, the only use for a USB stick is installing an OS or running a utility at boot.

[frankyunderwood123]

Memory "sticks"? - stopped using them almost a decade back

My goto is simply SD Cards - decent quality high speed ones.

And the only thing I use them for, unless they are in a device like a camera or smartphone? - writing an ISO to them.

I'm struggling to recall the last time someone gave me a memory stick with something on it, or I gave one to someone - probably the same as my title, easily a decade ago.

I do have a few knocking about in drawers, but always reach for an SD Card instead.

[Luiz Abdala]

Portable NVME enclosures.

I am looking at portable NVME enclosures with USB 3.0 interfaces and finding them more appealing by the day.

You sacrifice a lot of speed, but NVME is designed as high speed with the matching endurance, while skirting all those issues.

[Max Pyat]

They are already responsible:

When you buy something, your contract is with the retailer. Not with the manufacturer. So you're entitled to redress from the retailer

https://www.which.co.uk/consumer-rights/advice/what-are-my-statutory-rights-and-when-do-they-apply-acif94x8QOoH#what-are-your-statutory-rights

The retailer in turn can take action against its suppliers.

This is not in any way new and has a good logical and practical basis

[Conundrum1885]

Cheap USB

Has been an ongoing issue. I was astonished to find a 128GB on sale locally for £9 ! Tested and it did actually store 128GB but was slow as treacle so highly likely the low write garbage.

Seems that manufacturers often use multiple small microSDs in parallel sometimes soldered to the board as an alternative, the controller relies on each card having its own buffer so as long as all the cards work, everything runs. As soon as one goes bad the entire drive stops working abruptly.

Less reputable manufacturers use a 4GB card and a clever controller that does lossy compression using the host processor and fills in the gaps with random data while editing the file headers so that the checksums match which is what the OS detects. I only found this out when trying to copy data from one such drive and though the files 'looked' fine only segments of the data were there.

Interesting about the read amplifiers, that would track with the failed drive(s) here working when cooled as it obviously lowered the analogue threshold on a damaged amplifier somewhere in the chip stack permitting a full recovery. Wiping then rewriting the recovered data got it going again.

The old methods it seems work again, despite in this case the storage being solid state the laws of physics haven't changed so cooling a chip down can recover unreadable data in some cases.