Surprised it took this long for AT to do a piece on this. I've been steering people away from the Kingstons for awhile unless they're on a serious fire sale.
Still faster than a conventional HDD for certain, but it's like buying a V8 and being given the V6.
Excepting DDR memory, do not purchase Kingston brand's any other else what.
Kingston is a Taiwanese Co, operation is by the Taiwanese, not a very good reputation at them, fully unlike US or EU or Japan major companies, also not as good as the Korean major companies
Firstly sells you several qualified samples, and then next is sneaky shoddy, some of the Asia companies accustom so.
Please, don't generalize an entire nation. A lot of US storage companies work with Taiwanese companies to get the job done, a lot of SSD engineers are Taiwanese. Source: I work for a US storage company that's acquired plenty of other storage companies, I've been working on SSDs for 4 years.
Timon, you are just stupid. Kingston is an American company based in California! So going by your logic, don't buy American products because they are obviously shoddy.
Kingston has also been known to switch SDRAM/DDR suppliers so the buyer couldn't assume that if they bought the same part # at a different time or from a different source, that they would receive an identical product. Granted it's of less concern since DDR either meets the spec'd timings and frequency as a primary and mandatory feature, or it doesn't.
Ach who cares. :D The point A.B. was trying to make was obvious enough. Last thing we need on a tech forum is an argument about car engines. ;)
Re the NAND change, one of these days I hope someone sues a vendor for doing this sort of thing. So the sequential read speed drops by more than half, almost down to SATA1 speed, and Kingston doesn't see a need to tell anyone? It's very bad practice. An old Vertex2 would leave the newer V300 in the dust (having typed that, I figured it sensible to do a quck test just ran AS-SSD on my ancient 3.5" Vertex2, it gives 254MB/sec seq. read, 113MB/sec seq. write, 18MB/sec 4K read and 77MB/sec 4K write, so I was right, better than the newer V300).
I find some modern SSDs very disappointing. They're often little better than the best SATA2 models from several years ago (do people actually buy the SanDisk Pulse? I'd rather use an old Vertex3), or they've cut corners in other ways, eg. the M500 (terrible write speed; it's only advantage is cost). More money for the vendors, but prices for a mainstream model haven't dropped for consumers for ages (avg. cost of a decent 240GB/256GB has only now matched the low they reached in Jan/13).
Well, they are using compression, so under certain scenarios (like highly compressable data) this drive will perform indentical to older one. So yeah, this is a dick move Kingston..
Exactly my point. They could stick the lowest grade flash in there and it wouldn't change the score in atto. Atleast they could use micron's sync flash.
Indeed, Atto is a poor test of an SSD. Using AS-SSD, a 60GB Vertex2E is half the speed of a 240GB Vertex2 for sequential write, about 43% slower for seq. read, etc., yet Atto gives almost identical results for both models. It's no wonder vendor marketing people use Atto for their PR.
In working for corporate America for 15 years I definitely agree with you. Any company that has decently savvy bloated management structures in place will definitely be able to handle "cooking the books". Once I handled a specific city/market where the sales results were the absolute worst in the entire company and we were a Fortune 30 company operating in all 50 states. Despite the massive capital outlay, the return on customers paying revenue was embarrasing. The amount of outside sales reps got doubled for this market while the quota got cut in half from the prior year. None to my surprise, I listened to my boss "proudly" present to our VP's our great trends and traction we were gaining.....He presented Year over Year growth in attainment to quota. Well, if the quota is 50 instead of 100 then I can sell 25 widgets this year versus 50 widgets last year and my year over year comparison would be 50%. No change on the cover, but in reality it is half the revenue being brought in from the previous year. There is a study, a graph, an expert, a Sean Spicer, and a template to make the most convincing argument in the present day. Kingston will suffer and Trump is being unmasked as well. It's only a matter of time
Imagine buying a new laptop, and instead of getting that Intel i5 processor you expected to be in it, the laptop manufacturer swapped it out with some cheapo AMD Opteron to save them money. This is pretty close to the same thing.
You can't knowingly release a product that is >50% slower than advertised speeds and not suffer severe consequences.
nice when you don't get what you see on the box and it ends up been better
had one where it was a dual core N2830 ATOM based spec on the box but it was a Quad core N2940 ATOM in the laptop (even the spec label that is typically stuck on the laptop stated it was a N2830 when it in fact had a factory installed N2940)
Disappointing that they are doing this without transparency, but at least this has resulted in some lower price points for the ssd market. Recent sales on Newegg had the 120GB for $66 and the 240GB for $125.
I'm willing to bet prices have been falling on their drives in recent weeks because of the negative publicity they've been getting around tech forums like Anand, [H] and others. I'm sure sales have fallen, hence the drastic discounts lately.
They have every right to change NAND, they just need to be up front about it and put accurate specs on their product accordingly. They basically made the switch and hoped no one would notice, which is stupid. How many enthusiasts buy SSDs and never do a performance test on them?
This is appalling, when the swap is a 10 or 20% difference I'm annoyed but can at least understand the marketdroids claiming it's not a big deal because no-one will be able to tell the difference in real world use and that creating a new model for a "tiny" change would add confusion of it's own. This is nearly a 3x spread in performance, which is huge.
Kingston's earned itself an indefinite stay on my Do Not Buy list.
I have had a few Kingston SSDs - the KC100 and some V series. The latter were value SSDs. I had some performance issue with their V200 series (replaced with V+200 after complaints) but in general were still on the fence about Kingston in general.
I almost pulled the trigger on this last night, saw it was on sale for $66. Thanks for putting this out there, I will definitely spend $5 more to get something better.
I bought one of these a couple weeks ago. Sadly I knew of the issue but still bought it because it was the cheapest 128GB SSD at Microcenter. (Replacing a bad spindle laptop drive on my sisters HP so anything was better than that)
I've been a big Kingston fan so far. But my V200 is still out (for 6 weeks now) for its second replacement after some nasty sudden drive corruption after a few hundred hours of very slight use each.
This story is yet another reason never to buy Kingston again...
Sounds more like a V150 not V300 and far less than any improved V305. You can imagine if a car manufacturer swapped the V8 out of the sports car and put in a 4cyl and used the same model name and didn't replace the 400HP badge on the back.
The only thing I've ever bought from Kingston was server RAM, and it turned out to be sh@#. Doesn't surprise me that they're pulling these shenanigans.
Thanks for posting this. I purchased one 120GB model on sale recently for $70 (that went on sale again for $65). I went back and read old reviews before purchasing, including the one from Anandtech, and naturally it looked like a good drive for that price. Time to bug customer support for a return while it is still unopened. The only other Kingston SSDs I have are a trio of HyperX 3K drives for family members, and so far those had been solid (I only paid $65 on Black Friday two years ago). The good experience with those drives actually cemented the V300 purchase, and now Kingston is off the list indefinitely.
Props to Newegg for letting me return this junker. Just a note for anyone who has bought these on sale recently and hasn't opened the packaging, the firmware is listed on the bottom right corner of the Kingston sticker on the face of the drive. E.g. "521ABBF0" has the 521 firmware. So it's easy to tell which version you got without potentially killing your ability to return the thing.
I wonder if we're going to see another OEM go out of business. I'm sure even if they start producing the best SSDs out there, this scandal will keep sales way down. Just like OCZ.
The second this story broke the kingston v300 drives were on sale for $59.99 on Amazon and every other outlet. I got an inscreen ad on youtube.com showing me the drive advertised for dirt cheap on Amazon. It was not a targeted ad either. I never searched for SSD drives. The only thing I did was read this article. Unless , youtube ad generators searched my browser history. That could be it. But I didnt search for kingston V300 SSD. Either way, since they're outed....expect more price drops
They knew that if they changed the model number and offered subpar performance, any gain they got from using lower cost NAND would have been offset by the pricing they'd have had to use.
This way, they made tons of money off saps before they figured out what had happened.
Kingston just joined OCZ on my "No way, no how" list.
Ugh, I bought one of these just a week ago with the express aim of it actually speeding up my system over the spinning HD it previously had. Shame I've chucked the packaging away. I'll have to check the firmware to see which one I have.
Changing the performance characteristics of a product so drastically with a revision is just not on, it's really bad practice, it's dishonest and it is deceptive trading.
Maybe they've only done this in territories with poor customer protections in place. Would be interesting if people complained and they had to replace all the affected products. Retailers will be unhappy too - they didn't know Kingston have shat all over their own products, and now they'll be getting complaints and returns as awareness of the issue grows.
If they keep dropping in price I'll buy them. I always feel its not the raw MBps that counts but the access times. As long as they are still an order of magnitude higher in seek than a HDD thats fine with me. Plus they are cheap! More than handy for fast testing drives and swapouts.
Is there a reason that the benchmarking was done with incompressable data? My understanding is that "compression" is a main feature of the Sandforce Controller used on this SSD? Does using compressable data change the performance numbers?
Thank you for publishing this valuable information. I knew absolutely nothing about this NAND scandal and purchased Kingston SV300 at Amazon France last November. Luckily, the version I received (520A) was made with Toshiba NAND. I gathered more information on net and found out that those shipped with Firmware 505/520 are OK, but 506/521 are not. I checked the product page at Amazon France but there are only a few complains from buyers about speed of the product without much explanation. I posted the information I found to caution other Amazon buyers. What Kingston did may not be illegal but ethically very incorrect. I took a vow never to touch their product again, especially after I read this comment at Amazon USA: http://www.amazon.com/Kingston-Digital-Adapter-SV3... <quote> Kingston hid this fact by using solvent on the original labeling of the memory chips, then printed their own logo on top. Kingston does not manufacture any memory themselves. When SSD was shipped out for review it used quality Toshiba 19nm NAND but Kingston used solvent on chips labels and printed their own. This was odd, but the reason for it soon became obvious. After only a few months they switched to inferior quality chips with the same Kingston label. <unquote>
The only question people should be asking is 'does the drive perform to published specifications'?? That is all that matters. So, you got free cheese with your burger for a while...well, that cheese went away but its still the burger , as described. People are so fast to trash companies without using logic. If it performs to spec, that is all you should get.
Um...no, just no. That's like saying that if Intel sampled 4770K processors, then silently replaced the cores with a version that could score the same in 1 benchmark but delivered 1/2 to 1/3 the performance in a bunch of other real world scenarios nobody should care. The asynchronous NAND versions of these drives have far lower sequential write performance for non-compressible data than a friggin' mechanical hard drive. The write speed is on par with an external 5400rpm mechanical drive over USB3. Kingston is taking this one on the chin, and they deserve every second of it. They can switch NAND all they want, just do the right thing and change the product name/SKU.
480GB 420ABBFO made in Taiwan okey. http://postimg.org/image/f5fs9efhn/ Slow 4k write but test wasn't even close to optimal. It is system disk win 8.1 and a lot of programs running, including utorrent. atm it looks like the problem is with 120 and 240GB v300 drives.
Kingston is making it's last grasp for money and trying to unload their SSD inventory and then file their SSD line for bankruptcy. Can't make any other sense on the fact that they are knowingly lying to their customers and thus losing a lot of future customers. It is easy to break the companys imago but it takes years to build it back.
On 3/22/2014 I purchased a Kingston SSDNow V300 Series SV300S37A/240G, and it benchmarks at 520 reads and 380 writes on a low cost A8-6600K system. Cost me $110 at newegg, so I'm very happy with it. Just old inventory? I'd love to hear what other owners are seeing.
I read this article, but i decided to buy a V300 240gb anyway. Mainly because of the price (150$ in my country) and i though i might get lucky with the firmware. I thought wrong. FW 521 has in ASssd and CrystalMark about 160-185 read and 190-240 write. In Atto is great, 400-550. But almost all ssds perform like that in atto so i dont take it as a real benchmark.
Now i wait to return the product because im the 10days waranty when i can return it no questions asked. Il go with the Samsung Evo 250gb or Adata's sx900 256gb. Still thinking which one suits me best (tlc vs mlc).
Kingston ssd, no more for me. Just sticking to ram. Even the Crucial m500 has more speed than That v300 new FW. An its cheaper!
So if you did get the 506 or 521 firmware version, is there a [relatively easy] way to downgrade to the 505 or 520? Or perhaps upgrade to a newer version?
5 months ago I got 120gb drive with 506 fw and I paid $130 for it so imagine how SAD I am. I have gigabit home connection and I get 30 MB/s at max with V300 and 110 MB/s at max with 2TB HDD, but wait, I got faster boot time!!! This crap ssd should not be sold.
Here is Kingston's Official response to my query: Kingston said: I can certainly understand your concern but it's important to understand why this is. Crystal Disk and AS SSD uses incompressed data during its testing. The data cannot be compressed further so much without the write speed being affected. Because both of these benchmarks use fully incompressible data they are nowhere near a real world workload for typical desktop-notebook computing. ATTO disk benchmark uses uncompressed data. So the Sandforce drives can perform its compression and thus the increased speed of this drive is reflected in the benchmark. It is also the program used by our engineering staff to document the speed ratings for the these drives during testing. Unfortunately CDM gives a different lower score with these Asynchronous NAND but does not mean the drives are subpar in any way, form or fashion. Just different testing methods used by these benchmarking software. If CDM tested the same as ATTO using uncompressed data there would be no problem.
I have seen this posted in other forums last year including one where a Kingston rep denied there were performance issues and denied they were screwing their customers.
I just got one of these drives before reading all about this. To add insult to injury, my drive appears to have been used. There are clear scratches around one of the screw holes. This despite it coming sealed in its package.
Kristian, is there any update? I trusted this darned brand and bought it a some days ago, haven't received the units, but I suspect it will be the slower type.
Hi i recently bought this SSD but a 480 GB variant, unfortunately read about this nand switch later, can u please tell me does this nand switch performance lag also present in the 480 GB variant of kingston ssdnow v300? please do reply as your reply will definitely help to request a return ASAP.
So I built my first custom gaming rig in April 2014 I believe. I chose to use this particular drive for Windows 10 as Windows 10 literally was just launching. I have another internal traditional 3TB drive for mass storage. I can state that the speed was incredibly fast the entire time I used the drive, but that may be due to it being my first time using a SSD. However, I was dissapointed when the drive suddenly died and barely booted up one morning after only owning it for 1.5 years at most. Once this occurred I researched the issue and found out that runtime/lifecycle kills SSD's and apparently I had no clue about Kingston's software and utilities that I promptly installed. It told me that I had 1% life left in the drive. My contact to Kingston was the typical response in that I am screwed because it is out of warranty, not a defect of their workmanship, and have a good day. Now, 3.5 years after building my rig I was updating www.pcpartpicker.com and deciding what to do with some of my old notes. This is how I ran across this article and lots more about Kingston and SSD's. Long post - short - I don't care if Kingston is an American company and I am an American, they will never get any of my business in my personal or professional life due to the consistently poor decisions I have experienced and read about. There are consequences for being a half-assed company and sometimes it takes longer than you would think for it to catch up to that company, but it will. I will withhold my comments on the list of companies...lol...(Sprint, Nokia, Blackberry)
What guts - here is the direct verbiage from a piece of collateral on Kingston's website regarding the V300. They are spinning the issue and stating that due to NAND suppliers and benchmark software that sometimes you will over-achieve their 450MB stated speeds. They got big balls.
Product Builds and Speed Differences – V300 In order to achieve a balance of price and performance, we must maintain the flexibility to source NAND Flash components from various Tier 1 NAND manufacturers. At times, this will mean that there is a difference in benchmarked performance, where certain builds outperform our advertised specification (450MB/s Read / Write) while other drives will meet the advertised specification. Regardless of revision, all V300 SSDs still demonstrate 10x performance when compared to HDDs.
To this day I won't ever buy another Kingston product due to how they handled this. Basically they did a bait and switch, and then when AnandTech called them on this Kingston basically told them to piss off.
AnandTech. The site that pretty much drove SSD makers to get their drives up to par. I think Anand even gave a reasonable suggestion that Kingston maybe rename the new revision V305 or something.
Basically Kingston knew very much what they were doing, and they didn't give a care.
I was really sad because I really liked Kingston and their Hyper-X memory line for years before that. But never again.
Dear Kingston, have you changed your ways? Land a 2TB NVMe drive at my doorstep and I'll stop reminding everyone about this URL whenever you launch a new drive. ;)
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
75 Comments
Back to Article
Anonymous Blowhard - Monday, March 3, 2014 - link
Surprised it took this long for AT to do a piece on this. I've been steering people away from the Kingstons for awhile unless they're on a serious fire sale.Still faster than a conventional HDD for certain, but it's like buying a V8 and being given the V6.
Kristian Vättö - Monday, March 3, 2014 - link
I wanted to get official word from Kingston instead of going by what is posted in the forums, hence it took longer.Anonymous Blowhard - Monday, March 3, 2014 - link
Fair enough. Don't want to expose yourself to potential legal issues by reporting "hearsay" ... even if it's highly confirmed hearsay.timon_comment - Thursday, March 13, 2014 - link
Excepting DDR memory, do not purchase Kingston brand's any other else what.Kingston is a Taiwanese Co, operation is by the Taiwanese, not a very good reputation at them, fully unlike US or EU or Japan major companies, also not as good as the Korean major companies
Firstly sells you several qualified samples, and then next is sneaky shoddy, some of the Asia companies accustom so.
apudapus - Thursday, March 13, 2014 - link
Please, don't generalize an entire nation. A lot of US storage companies work with Taiwanese companies to get the job done, a lot of SSD engineers are Taiwanese. Source: I work for a US storage company that's acquired plenty of other storage companies, I've been working on SSDs for 4 years.Ez8004 - Sunday, May 4, 2014 - link
Timon, you are just stupid. Kingston is an American company based in California! So going by your logic, don't buy American products because they are obviously shoddy.mindless1 - Tuesday, May 20, 2014 - link
Kingston has also been known to switch SDRAM/DDR suppliers so the buyer couldn't assume that if they bought the same part # at a different time or from a different source, that they would receive an identical product. Granted it's of less concern since DDR either meets the spec'd timings and frequency as a primary and mandatory feature, or it doesn't.it_progress - Wednesday, August 6, 2014 - link
timon, Kingston is an American company. I suggested you to wiki it if you know how to use internet.Daniel Egger - Monday, March 3, 2014 - link
That's sort of a dumb comparison... There're plenty V6es that are better than some V8es.mapesdhs - Monday, March 3, 2014 - link
Ach who cares. :D The point A.B. was trying to make was obvious enough. Last thing
we need on a tech forum is an argument about car engines. ;)
Re the NAND change, one of these days I hope someone sues a vendor for doing
this sort of thing. So the sequential read speed drops by more than half, almost down
to SATA1 speed, and Kingston doesn't see a need to tell anyone? It's very bad
practice. An old Vertex2 would leave the newer V300 in the dust (having typed
that, I figured it sensible to do a quck test just ran AS-SSD on my ancient 3.5"
Vertex2, it gives 254MB/sec seq. read, 113MB/sec seq. write, 18MB/sec 4K
read and 77MB/sec 4K write, so I was right, better than the newer V300).
I find some modern SSDs very disappointing. They're often little better than the best
SATA2 models from several years ago (do people actually buy the SanDisk Pulse?
I'd rather use an old Vertex3), or they've cut corners in other ways, eg. the M500
(terrible write speed; it's only advantage is cost). More money for the vendors, but
prices for a mainstream model haven't dropped for consumers for ages (avg. cost
of a decent 240GB/256GB has only now matched the low they reached in Jan/13).
Ian.
jabber - Tuesday, March 4, 2014 - link
But when 99% of folks are just used to that old 2005 4200rpm 120GB HDD that pushes 50MBps there are plenty who will think that Pulse SSD heavenly.psiboy - Sunday, August 2, 2015 - link
That's called Fraud! Criminal charges would be laid here in Australia!Subyman - Monday, March 3, 2014 - link
Did they update the packaging? The previous V300 shows "up to 450MB/s" on the box. Obviously, that is no longer possible.hojnikb - Monday, March 3, 2014 - link
Well, they are using compression, so under certain scenarios (like highly compressable data) this drive will perform indentical to older one.So yeah, this is a dick move Kingston..
Flunk - Monday, March 3, 2014 - link
Based on the wording you could write "up to 1,000,000MB/s" and still be legal. "up to" is the weasellyest wording in the industry.Kristian Vättö - Monday, March 3, 2014 - link
Kingston uses ATTO for the marketing benchmarks and there is no slowdown since it's all compressible data.hojnikb - Monday, March 3, 2014 - link
Exactly my point. They could stick the lowest grade flash in there and it wouldn't change the score in atto.Atleast they could use micron's sync flash.
mapesdhs - Monday, March 3, 2014 - link
Indeed, Atto is a poor test of an SSD. Using AS-SSD, a 60GB Vertex2E
is half the speed of a 240GB Vertex2 for sequential write, about 43%
slower for seq. read, etc., yet Atto gives almost identical results for both
models. It's no wonder vendor marketing people use Atto for their PR.
Ian.
dpjtpa - Sunday, October 1, 2017 - link
In working for corporate America for 15 years I definitely agree with you. Any company that has decently savvy bloated management structures in place will definitely be able to handle "cooking the books".Once I handled a specific city/market where the sales results were the absolute worst in the entire company and we were a Fortune 30 company operating in all 50 states. Despite the massive capital outlay, the return on customers paying revenue was embarrasing. The amount of outside sales reps got doubled for this market while the quota got cut in half from the prior year. None to my surprise, I listened to my boss "proudly" present to our VP's our great trends and traction we were gaining.....He presented Year over Year growth in attainment to quota. Well, if the quota is 50 instead of 100 then I can sell 25 widgets this year versus 50 widgets last year and my year over year comparison would be 50%. No change on the cover, but in reality it is half the revenue being brought in from the previous year. There is a study, a graph, an expert, a Sean Spicer, and a template to make the most convincing argument in the present day. Kingston will suffer and Trump is being unmasked as well. It's only a matter of time
wicketr - Monday, March 3, 2014 - link
Imagine buying a new laptop, and instead of getting that Intel i5 processor you expected to be in it, the laptop manufacturer swapped it out with some cheapo AMD Opteron to save them money. This is pretty close to the same thing.You can't knowingly release a product that is >50% slower than advertised speeds and not suffer severe consequences.
whitehat2k9 - Monday, March 3, 2014 - link
I'd like to see a laptop with an AMD Opteron in it...leexgx - Thursday, April 2, 2015 - link
nice when you don't get what you see on the box and it ends up been betterhad one where it was a dual core N2830 ATOM based spec on the box but it was a Quad core N2940 ATOM in the laptop (even the spec label that is typically stuck on the laptop stated it was a N2830 when it in fact had a factory installed N2940)
smunter6 - Monday, March 3, 2014 - link
Disappointing that they are doing this without transparency, but at least this has resulted in some lower price points for the ssd market. Recent sales on Newegg had the 120GB for $66 and the 240GB for $125.RaistlinZ - Monday, March 3, 2014 - link
I'm willing to bet prices have been falling on their drives in recent weeks because of the negative publicity they've been getting around tech forums like Anand, [H] and others. I'm sure sales have fallen, hence the drastic discounts lately.They have every right to change NAND, they just need to be up front about it and put accurate specs on their product accordingly. They basically made the switch and hoped no one would notice, which is stupid. How many enthusiasts buy SSDs and never do a performance test on them?
Dirty pool, Kingston. Dirty pool indeed.
hojnikb - Monday, March 3, 2014 - link
Will anandtech cover the new sandforce updete ? (5.2.5)Apperently TRIM issue is finally fixed (atleast acording to adata website).
DanNeely - Monday, March 3, 2014 - link
This is appalling, when the swap is a 10 or 20% difference I'm annoyed but can at least understand the marketdroids claiming it's not a big deal because no-one will be able to tell the difference in real world use and that creating a new model for a "tiny" change would add confusion of it's own. This is nearly a 3x spread in performance, which is huge.Kingston's earned itself an indefinite stay on my Do Not Buy list.
MartinT - Monday, March 3, 2014 - link
I've just bought a V300 last week, it's the other very slow version with firmware 521A.I'll return it, Kingston should have known better than to sneek this by. Shameful.
gamoniac - Monday, March 3, 2014 - link
I have had a few Kingston SSDs - the KC100 and some V series. The latter were value SSDs. I had some performance issue with their V200 series (replaced with V+200 after complaints) but in general were still on the fence about Kingston in general.Not after this shady incident though. Shame.
Dentons - Monday, March 3, 2014 - link
It's underhanded and devious. From Kingston, it isn't surprising.With Samsung or Micron offering quality products at equivalent prices, there's no reason to go with such dishonest vendors.
optimus1883 - Monday, March 3, 2014 - link
I almost pulled the trigger on this last night, saw it was on sale for $66. Thanks for putting this out there, I will definitely spend $5 more to get something better.ruthan - Monday, March 3, 2014 - link
I will not ever buy their product.Gunbuster - Monday, March 3, 2014 - link
I bought one of these a couple weeks ago. Sadly I knew of the issue but still bought it because it was the cheapest 128GB SSD at Microcenter. (Replacing a bad spindle laptop drive on my sisters HP so anything was better than that)Still super shady of Kingston.
Daniel Egger - Monday, March 3, 2014 - link
I've been a big Kingston fan so far. But my V200 is still out (for 6 weeks now) for its second replacement after some nasty sudden drive corruption after a few hundred hours of very slight use each.This story is yet another reason never to buy Kingston again...
yannigr - Monday, March 3, 2014 - link
I just removed Kingston from my future options. Any other companies doing that?hojnikb - Monday, March 3, 2014 - link
OCZ used to do that in the Vertex2 days..And there were reports of some lesser known ssds to be using fake flash that underperformed.
Sunburn74 - Monday, March 3, 2014 - link
You can't sell an inferior product at the same price as your superior product whilst using the same names for both.ThortonBe - Monday, March 3, 2014 - link
I was considering getting a Kingston, but thank god I didn't. I'll definitely be sure to avoid the brand in the future.tygrus - Monday, March 3, 2014 - link
Sounds more like a V150 not V300 and far less than any improved V305. You can imagine if a car manufacturer swapped the V8 out of the sports car and put in a 4cyl and used the same model name and didn't replace the 400HP badge on the back.DigitalFreak - Monday, March 3, 2014 - link
The only thing I've ever bought from Kingston was server RAM, and it turned out to be sh@#. Doesn't surprise me that they're pulling these shenanigans.sausagefingers - Monday, March 3, 2014 - link
Thanks for this article I was one of the numerous emails and I really appreciate it.Jeffrey Bosboom - Monday, March 3, 2014 - link
So, are you going to update your review with a disclaimer explaining that what's being sold now is different from what you tested?sausagefingers - Monday, March 3, 2014 - link
I think updating the original article with a link to this one is a good plan.Kristian Vättö - Tuesday, March 4, 2014 - link
Done!Communism - Monday, March 3, 2014 - link
Never Purchase an SSD that isn't from a NAND manufacturer.Never Purchase DRAM that isn't from a DRAM manufacturer.
Bob Todd - Monday, March 3, 2014 - link
Your RAM all has SK Hynix, Micron/Elpida, or Samsung on the packaging?DanNeely - Monday, March 3, 2014 - link
I assume he's using all Crucial ram (Crucial is Micron's house brand).Bob Todd - Monday, March 3, 2014 - link
Thanks for posting this. I purchased one 120GB model on sale recently for $70 (that went on sale again for $65). I went back and read old reviews before purchasing, including the one from Anandtech, and naturally it looked like a good drive for that price. Time to bug customer support for a return while it is still unopened. The only other Kingston SSDs I have are a trio of HyperX 3K drives for family members, and so far those had been solid (I only paid $65 on Black Friday two years ago). The good experience with those drives actually cemented the V300 purchase, and now Kingston is off the list indefinitely.Bob Todd - Thursday, March 6, 2014 - link
Props to Newegg for letting me return this junker. Just a note for anyone who has bought these on sale recently and hasn't opened the packaging, the firmware is listed on the bottom right corner of the Kingston sticker on the face of the drive. E.g. "521ABBF0" has the 521 firmware. So it's easy to tell which version you got without potentially killing your ability to return the thing.alaba - Tuesday, March 4, 2014 - link
I wonder if we're going to see another OEM go out of business. I'm sure even if they start producing the best SSDs out there, this scandal will keep sales way down. Just like OCZ.alaba - Tuesday, March 4, 2014 - link
Of course, I'm not saying OCZ's later SSDs were the best but they definitely improved but it was too little, too late.axiommods - Tuesday, March 4, 2014 - link
The second this story broke the kingston v300 drives were on sale for $59.99 on Amazon and every other outlet. I got an inscreen ad on youtube.com showing me the drive advertised for dirt cheap on Amazon. It was not a targeted ad either. I never searched for SSD drives. The only thing I did was read this article. Unless , youtube ad generators searched my browser history. That could be it. But I didnt search for kingston V300 SSD. Either way, since they're outed....expect more price dropslokhor - Tuesday, March 4, 2014 - link
I have one of these drives and was pretty shocked to hear about. Thank god mine has the 505 firmware. Never going to buy Kingston again though.HisDivineOrder - Tuesday, March 4, 2014 - link
They knew that if they changed the model number and offered subpar performance, any gain they got from using lower cost NAND would have been offset by the pricing they'd have had to use.This way, they made tons of money off saps before they figured out what had happened.
Kingston just joined OCZ on my "No way, no how" list.
Congrats!
psychobriggsy - Tuesday, March 4, 2014 - link
Ugh, I bought one of these just a week ago with the express aim of it actually speeding up my system over the spinning HD it previously had. Shame I've chucked the packaging away. I'll have to check the firmware to see which one I have.Changing the performance characteristics of a product so drastically with a revision is just not on, it's really bad practice, it's dishonest and it is deceptive trading.
Maybe they've only done this in territories with poor customer protections in place. Would be interesting if people complained and they had to replace all the affected products. Retailers will be unhappy too - they didn't know Kingston have shat all over their own products, and now they'll be getting complaints and returns as awareness of the issue grows.
jabber - Tuesday, March 4, 2014 - link
If they keep dropping in price I'll buy them. I always feel its not the raw MBps that counts but the access times. As long as they are still an order of magnitude higher in seek than a HDD thats fine with me. Plus they are cheap! More than handy for fast testing drives and swapouts.Storage Gal - Wednesday, March 5, 2014 - link
Is there a reason that the benchmarking was done with incompressable data? My understanding is that "compression" is a main feature of the Sandforce Controller used on this SSD? Does using compressable data change the performance numbers?Paris Kitty - Wednesday, March 5, 2014 - link
Thank you for publishing this valuable information.I knew absolutely nothing about this NAND scandal and purchased Kingston SV300 at Amazon France last November. Luckily, the version I received (520A) was made with Toshiba NAND.
I gathered more information on net and found out that those shipped with Firmware 505/520 are OK, but 506/521 are not. I checked the product page at Amazon France but there are only a few complains from buyers about speed of the product without much explanation. I posted the information I found to caution other Amazon buyers.
What Kingston did may not be illegal but ethically very incorrect. I took a vow never to touch their product again, especially after I read this comment at Amazon USA: http://www.amazon.com/Kingston-Digital-Adapter-SV3...
<quote>
Kingston hid this fact by using solvent on the original labeling of the memory chips, then printed their own logo on top. Kingston does not manufacture any memory themselves. When SSD was shipped out for review it used quality Toshiba 19nm NAND but Kingston used solvent on chips labels and printed their own. This was odd, but the reason for it soon became obvious. After only a few months they switched to inferior quality chips with the same Kingston label. <unquote>
reality4321 - Thursday, March 6, 2014 - link
The only question people should be asking is 'does the drive perform to published specifications'?? That is all that matters. So, you got free cheese with your burger for a while...well, that cheese went away but its still the burger , as described.People are so fast to trash companies without using logic. If it performs to spec, that is all you should get.
Bob Todd - Thursday, March 6, 2014 - link
Um...no, just no. That's like saying that if Intel sampled 4770K processors, then silently replaced the cores with a version that could score the same in 1 benchmark but delivered 1/2 to 1/3 the performance in a bunch of other real world scenarios nobody should care. The asynchronous NAND versions of these drives have far lower sequential write performance for non-compressible data than a friggin' mechanical hard drive. The write speed is on par with an external 5400rpm mechanical drive over USB3. Kingston is taking this one on the chin, and they deserve every second of it. They can switch NAND all they want, just do the right thing and change the product name/SKU.Korsi - Friday, March 14, 2014 - link
480GB 420ABBFO made in Taiwan okey.http://postimg.org/image/f5fs9efhn/
Slow 4k write but test wasn't even close to optimal. It is system disk win 8.1 and a lot of programs running, including utorrent.
atm it looks like the problem is with 120 and 240GB v300 drives.
Kingston is making it's last grasp for money and trying to unload their SSD inventory and then file their SSD line for bankruptcy. Can't make any other sense on the fact that they are knowingly lying to their customers and thus losing a lot of future customers.
It is easy to break the companys imago but it takes years to build it back.
compdoc - Wednesday, April 2, 2014 - link
On 3/22/2014 I purchased a Kingston SSDNow V300 Series SV300S37A/240G, and it benchmarks at 520 reads and 380 writes on a low cost A8-6600K system. Cost me $110 at newegg, so I'm very happy with it. Just old inventory? I'd love to hear what other owners are seeing.hojnikb - Friday, April 4, 2014 - link
Try as-ssd or crystaldiskmark random data and report back :)apothec - Friday, May 9, 2014 - link
I read this article, but i decided to buy a V300 240gb anyway. Mainly because of the price (150$ in my country) and i though i might get lucky with the firmware. I thought wrong. FW 521 has in ASssd and CrystalMark about 160-185 read and 190-240 write. In Atto is great, 400-550. But almost all ssds perform like that in atto so i dont take it as a real benchmark.Now i wait to return the product because im the 10days waranty when i can return it no questions asked. Il go with the Samsung Evo 250gb or Adata's sx900 256gb. Still thinking which one suits me best (tlc vs mlc).
Kingston ssd, no more for me. Just sticking to ram. Even the Crucial m500 has more speed than That v300 new FW. An its cheaper!
kugo2006 - Tuesday, May 13, 2014 - link
So if you did get the 506 or 521 firmware version, is there a [relatively easy] way to downgrade to the 505 or 520? Or perhaps upgrade to a newer version?GraveNoX - Tuesday, May 20, 2014 - link
5 months ago I got 120gb drive with 506 fw and I paid $130 for it so imagine how SAD I am. I have gigabit home connection and I get 30 MB/s at max with V300 and 110 MB/s at max with 2TB HDD, but wait, I got faster boot time!!! This crap ssd should not be sold.Colorado_Al - Tuesday, May 27, 2014 - link
Here is Kingston's Official response to my query:Kingston said:
I can certainly understand your concern but it's important to understand why this is. Crystal Disk and AS SSD uses incompressed data during its testing. The data cannot be compressed further so much without the write speed being affected. Because both of these benchmarks use fully incompressible data they are nowhere near a real world workload for typical desktop-notebook computing. ATTO disk benchmark uses uncompressed data. So the Sandforce drives can perform its compression and thus the increased speed of this drive is reflected in the benchmark. It is also the program used by our engineering staff to document the speed ratings for the these drives during testing. Unfortunately CDM gives a different lower score with these Asynchronous NAND but does not mean the drives are subpar in any way, form or fashion. Just different testing methods used by these benchmarking software. If CDM tested the same as ATTO using uncompressed data there would be no problem.
GarthBock - Thursday, May 29, 2014 - link
I have seen this posted in other forums last year including one where a Kingston rep denied there were performance issues and denied they were screwing their customers.the_law - Wednesday, October 8, 2014 - link
I just got one of these drives before reading all about this. To add insult to injury, my drive appears to have been used. There are clear scratches around one of the screw holes. This despite it coming sealed in its package.Encho - Saturday, February 14, 2015 - link
Kristian, is there any update? I trusted this darned brand and bought it a some days ago, haven't received the units, but I suspect it will be the slower type.JohnnyFingers - Monday, March 30, 2015 - link
I think maybe synchronous VS asynchronous NAND longevity is a more serious issue too??ahmadhanif - Sunday, April 26, 2015 - link
Hi i recently bought this SSD but a 480 GB variant, unfortunately read about this nand switch later, can u please tell me does this nand switch performance lag also present in the 480 GB variant of kingston ssdnow v300? please do reply as your reply will definitely help to request a return ASAP.dpjtpa - Sunday, October 1, 2017 - link
So I built my first custom gaming rig in April 2014 I believe. I chose to use this particular drive for Windows 10 as Windows 10 literally was just launching. I have another internal traditional 3TB drive for mass storage. I can state that the speed was incredibly fast the entire time I used the drive, but that may be due to it being my first time using a SSD. However, I was dissapointed when the drive suddenly died and barely booted up one morning after only owning it for 1.5 years at most. Once this occurred I researched the issue and found out that runtime/lifecycle kills SSD's and apparently I had no clue about Kingston's software and utilities that I promptly installed. It told me that I had 1% life left in the drive. My contact to Kingston was the typical response in that I am screwed because it is out of warranty, not a defect of their workmanship, and have a good day. Now, 3.5 years after building my rig I was updating www.pcpartpicker.com and deciding what to do with some of my old notes. This is how I ran across this article and lots more about Kingston and SSD's. Long post - short - I don't care if Kingston is an American company and I am an American, they will never get any of my business in my personal or professional life due to the consistently poor decisions I have experienced and read about. There are consequences for being a half-assed company and sometimes it takes longer than you would think for it to catch up to that company, but it will. I will withhold my comments on the list of companies...lol...(Sprint, Nokia, Blackberry)dpjtpa - Sunday, October 1, 2017 - link
What guts - here is the direct verbiage from a piece of collateral on Kingston's website regarding the V300. They are spinning the issue and stating that due to NAND suppliers and benchmark software that sometimes you will over-achieve their 450MB stated speeds. They got big balls.Product Builds and Speed Differences – V300
In order to achieve a balance of price and performance, we must maintain the flexibility to source NAND
Flash components from various Tier 1 NAND manufacturers. At times, this will mean that there is a
difference in benchmarked performance, where certain builds outperform our advertised specification
(450MB/s Read / Write) while other drives will meet the advertised specification. Regardless of revision, all
V300 SSDs still demonstrate 10x performance when compared to HDDs.
xrror - Sunday, February 25, 2018 - link
To this day I won't ever buy another Kingston product due to how they handled this. Basically they did a bait and switch, and then when AnandTech called them on this Kingston basically told them to piss off.AnandTech. The site that pretty much drove SSD makers to get their drives up to par. I think Anand even gave a reasonable suggestion that Kingston maybe rename the new revision V305 or something.
Basically Kingston knew very much what they were doing, and they didn't give a care.
I was really sad because I really liked Kingston and their Hyper-X memory line for years before that. But never again.
Gunbuster - Tuesday, May 21, 2019 - link
Dear Kingston, have you changed your ways? Land a 2TB NVMe drive at my doorstep and I'll stop reminding everyone about this URL whenever you launch a new drive. ;)