If you're using sata drives why not use uas mode? Not every device has support for it, but it's something that can drastically improve desktop workloads.
The graphs are meant to be analyzed standalone - one by one - to identify whether there is any sudden drops in read or write speeds / whether that is correlated to temperature in any way.
The reason they are in a drop down selection is just because they are results from running the same workloads on the drive, and it is easier for readers to check up with that drop-down, rather than scroll down to view each of the graphs one by one.
For 5-7 graphs I would greatly prefer scrolling instead of using a dropdown. Maybe load dropdown as default with a button "show all graphs" for those who don't like it.
Any chance we'll see a follow-up with some more reasonably priced drives like the Sandisk Extreme? That drive is a beast, especially in 4-32k performance
Yea the sandisk extreme is supposed to be the fastest usb 3.0 according to reviews, why not do a review of the common usb drives which the majority of users would have access to.
It'll probably never happen because the OEMs are only interested in sending high end products for review. I'd really be interested in seeing if a $20-35 flash drive would be significantly faster than the $10-15 models I generally get.
"It'll probably never happen because the OEMs are only interested in sending high end products for review. "
Anandtech did a partial review before, but only on speed.
"I have one of these and yes when it hits its stride it really flies but sometimes it seems rather picky with what USB3.0 ports it will decide to play ball with"
Works great on intel chipset ports, never had the chance to test with others
I have one of these and yes when it hits its stride it really flies but sometimes it seems rather picky with what USB3.0 ports it will decide to play ball with.
For some reason I still feel that USB3.0 is at the stage USB2.0 was around 2002 when you had to be careful which chipset (VIA - Bad NEC - Good) you used. It's not fully sorted yet.
Is the Brand X flash drive available to purchase now? If it has fundamental flaws, shouldn't you be telling us about it now as opposed to when a nebulous firmware upgrade becomes available?
The 'fundamental flaw' in BrandX is the same as that in the Mushkin Ventura Ultra. It is just that the extent of throttling is different.
I can safely say that the drive we tested is $150+. If that is your budget, you should choose from the ones that have been named and reviewed in our list.
At AnandTech, we believe in giving manufacturers a chance to resolve the issues before either recommending or not recommending them. In its current state, I will not recommend BrandX, so you should just ignore it while considering possible purchase options. It doesn't matter who the manufacturer is :)
*sigh* So the shoddy product doesn't lose. The always positive review ad carasoul doesn't lose. Until we get to to the end user everyone wins! Is your audience your readers or the companies providing ads and samples? Its difficult to tell when you are protecting companies. Why not name them, let the reader know there is a fix (or not) planned and protect the people buying the products. You know... Like a review site. If you think its a bad sample then you can communicate that. This everything is Facebook positive thumbs up makes me skeptical of every review blurring into ads.
If a product is on the market that under-performs, you should be putting the readers first and telling them not to buy it, then re-evaluating the product when it is eventually fixed.
How can I ignore BrandX when you won't tell us what it is? Absolutely unacceptable.
What you are basically saying is that you are holding back a bad review because you don't want to upset the company concerned. Unbelievable... Simple question - who do you write your reviews for: your readers or your suppliers and advertisers?
That is horrible, absolutely horrible and I don't know when Anandtech "switched" to this policy. If the product is pre-production and hasn't come to market then fine. But if you are telling us you are testing a marketed product and won't release the name because of crap performance that is unacceptable. Answer this and then think about it....if BrandX had been awesome would you have kept the name a secret? I thought not...
FWIW : I have told the manufacturer that they are no different from Mushkin when it comes to throttling, and that the product should not be sold in the current state.
Btw, there are plenty of products that we evaluate, but review never gets done because we are waiting for the manufacturer to resolve some issue that we report. This is the exact reason why we haven't reviewed BrandX. The reason it is presented in this article is that is is just a sample point we gathered with our evaluation methodology.
This is a fine balance between writing the review for the readers and providing constructive feedback to the manufacturers.
Consider the scenario where I don't write about 'BrandX' in this article - The reader doesn't know that there is a 240GB USB3 SSD out there which throttles to 20 MBps. He MAY end up buying that without even knowing about the issue.
Now, consider someone who has read this article with BrandX. He knows that there is a USB3 SSD at the 240GB capacity point that throttles. He is definitely going to do his groundwork before purchasing the unit. There is a very HIGH probability that the unit he ends up buying might not be BrandX.
Anyways, any person who reads online reviews will find the problems that I have reported here in other sites, albeit with not as many comparison points that we have here. [ Eg: http://geardiary.com/category/reviews/ ] It doesn't ta
"Btw, there are plenty of products that we evaluate, but review never gets done because we are waiting for the manufacturer to resolve some issue that we report."
Then the fact remains that you are putting the interests of your advertisers and suppliers over and above the interests of your readers. Let's be honest here, as the guy above says, if the data had been awesome you *would* have run the review.
In essence you are essentially admitting that you are filtering out bad reviews and only presenting good ones. You're even asking us to read other sites to get the data and recommendations that for some reason you are deliberating holding back. That's... breathtaking - for all the wrong reasons.
A review is written only when there is a good balance of plus and minus points. Sometimes, the minus points may be more than the plus ones, making the product a tough recommendation. If there is nothing good in the product and it is no different from what is already in the market, what is the point of presenting the data point to the reader? Why should I waste time on writing a review if all I have to say is: "Don't buy this product" ; ?
Running a review process should benefit everyone - the author, the reader and the manufacturer. Put yourself in either of the non-reader shoes and ask for yourself.
FWIW, there is no product that I have reviewed where the 'product is awesome' and has no minus points.
It's not a waste if it prevents someone from buying a crap product. That's the entire point of reading reviews before buying ... to know which ones are crap (avoid) and which ones are good (buy).
If all you present are "good" reviews, how are we to know which products to avoid as lemons?
A well-written review that shows all the good and bad of a product is never a waste of time. Even if the end result is "don't buy this piece of crap".
How about a transparent review that puts your readers' interests ahead of the manufacturers'? Oh by all means don't run a standalone review if you don't want to. But at the same time, deliberately obscuring the identity of a poorly performing product only reflects badly on you and your review policy.
In my view, the better solution for an external USB 3.0 drive is: to use a USB 3.0->SATA adapter cable + a regular 2 1/2" SSD. For anything larger than about 32 Gb, it's also usable (without the adapter cable) as a bootable internal SSD drive.A 10" long USB 2.0 adapter cable was included with my Samsung EVO 840 120 Gb SSD. Or similar USB 3.0 versions can be found at Newegg or Amazon for ~$8.
Do all of those (other than the lacie) get treated by windows as a removable drive (rather than a 'fixed disk') - note that this is not the same as showing a 'remove hardware' option.. typically, even usb drives using a usb->ata controller get shown as 'fixed disks' (which enables such things as native partitioning, instead of treating them as a large single partition, in windows). More importantly, it makes them unusable for deploying OS's using Microsoft's deployment toolkit or SCCM.
What would be good is if you threw in a wild card entry to some of your reviews. In this case something like a consumer SSD such as Crucial's 128GB MX100 in a cheap USB 3 caddy and see what you can get for around the £60/$80 or even 256GB for around £90/$110, yes it is not the same form factor but would be an interesting comparision if price or capacity where bigger considerations than size.
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.
29 Comments
Back to Article
tuxRoller - Saturday, November 15, 2014 - link
If you're using sata drives why not use uas mode?Not every device has support for it, but it's something that can drastically improve desktop workloads.
ganeshts - Saturday, November 15, 2014 - link
Other than the Atom drive (which is not SATA-based), all others were tested with UASP enabled.cosmotic - Saturday, November 15, 2014 - link
It's very confusing when the comparison graphs' axis domains aren't consistent.mjcutri - Saturday, November 15, 2014 - link
+1ganeshts - Saturday, November 15, 2014 - link
Those are not specified as 'comparison' graphs. For the bandwidth comparisons, you can look at the dedicated review of the Mushkin Ventura Ultra.http://anandtech.com/show/8709/mushkin-ventura-ult...
The graphs are meant to be analyzed standalone - one by one - to identify whether there is any sudden drops in read or write speeds / whether that is correlated to temperature in any way.
The reason they are in a drop down selection is just because they are results from running the same workloads on the drive, and it is easier for readers to check up with that drop-down, rather than scroll down to view each of the graphs one by one.
Doroga - Sunday, November 16, 2014 - link
For 5-7 graphs I would greatly prefer scrolling instead of using a dropdown.Maybe load dropdown as default with a button "show all graphs" for those who don't like it.
basroil - Saturday, November 15, 2014 - link
Any chance we'll see a follow-up with some more reasonably priced drives like the Sandisk Extreme? That drive is a beast, especially in 4-32k performancepoohbear - Saturday, November 15, 2014 - link
Yea the sandisk extreme is supposed to be the fastest usb 3.0 according to reviews, why not do a review of the common usb drives which the majority of users would have access to.DanNeely - Sunday, November 16, 2014 - link
It'll probably never happen because the OEMs are only interested in sending high end products for review. I'd really be interested in seeing if a $20-35 flash drive would be significantly faster than the $10-15 models I generally get.basroil - Sunday, November 16, 2014 - link
"It'll probably never happen because the OEMs are only interested in sending high end products for review. "Anandtech did a partial review before, but only on speed.
"I have one of these and yes when it hits its stride it really flies but sometimes it seems rather picky with what USB3.0 ports it will decide to play ball with"
Works great on intel chipset ports, never had the chance to test with others
jabber - Sunday, November 16, 2014 - link
I have one of these and yes when it hits its stride it really flies but sometimes it seems rather picky with what USB3.0 ports it will decide to play ball with.For some reason I still feel that USB3.0 is at the stage USB2.0 was around 2002 when you had to be careful which chipset (VIA - Bad NEC - Good) you used. It's not fully sorted yet.
Samus - Sunday, November 16, 2014 - link
Which controller/chipset was the USB host?ganeshts - Sunday, November 16, 2014 - link
http://anandtech.com/show/8241/lacies-2big-thunder...Intel Z97 ; USB3_1 in the Pro Wi-Fi ac board depicted in the link above. It is directly off the PCH.
Obviously, the same port is used to test all drives.
Samus - Sunday, November 16, 2014 - link
Thanks!OrphanageExplosion - Sunday, November 16, 2014 - link
Is the Brand X flash drive available to purchase now? If it has fundamental flaws, shouldn't you be telling us about it now as opposed to when a nebulous firmware upgrade becomes available?I assume that the drive is not yet available?
ganeshts - Sunday, November 16, 2014 - link
The 'fundamental flaw' in BrandX is the same as that in the Mushkin Ventura Ultra. It is just that the extent of throttling is different.I can safely say that the drive we tested is $150+. If that is your budget, you should choose from the ones that have been named and reviewed in our list.
At AnandTech, we believe in giving manufacturers a chance to resolve the issues before either recommending or not recommending them. In its current state, I will not recommend BrandX, so you should just ignore it while considering possible purchase options. It doesn't matter who the manufacturer is :)
hrrmph - Sunday, November 16, 2014 - link
"At AnandTech, we believe in giving manufacturers a chance to resolve the issues before either recommending or not recommending them."-1
hero4hire - Tuesday, November 18, 2014 - link
*sigh* So the shoddy product doesn't lose. The always positive review ad carasoul doesn't lose. Until we get to to the end user everyone wins! Is your audience your readers or the companies providing ads and samples? Its difficult to tell when you are protecting companies. Why not name them, let the reader know there is a fix (or not) planned and protect the people buying the products. You know... Like a review site. If you think its a bad sample then you can communicate that. This everything is Facebook positive thumbs up makes me skeptical of every review blurring into ads.OrphanageExplosion - Tuesday, November 18, 2014 - link
If a product is on the market that under-performs, you should be putting the readers first and telling them not to buy it, then re-evaluating the product when it is eventually fixed.How can I ignore BrandX when you won't tell us what it is? Absolutely unacceptable.
What you are basically saying is that you are holding back a bad review because you don't want to upset the company concerned. Unbelievable... Simple question - who do you write your reviews for: your readers or your suppliers and advertisers?
7Enigma - Tuesday, November 18, 2014 - link
That is horrible, absolutely horrible and I don't know when Anandtech "switched" to this policy. If the product is pre-production and hasn't come to market then fine. But if you are telling us you are testing a marketed product and won't release the name because of crap performance that is unacceptable. Answer this and then think about it....if BrandX had been awesome would you have kept the name a secret? I thought not...ganeshts - Wednesday, November 19, 2014 - link
FWIW : I have told the manufacturer that they are no different from Mushkin when it comes to throttling, and that the product should not be sold in the current state.Btw, there are plenty of products that we evaluate, but review never gets done because we are waiting for the manufacturer to resolve some issue that we report. This is the exact reason why we haven't reviewed BrandX. The reason it is presented in this article is that is is just a sample point we gathered with our evaluation methodology.
This is a fine balance between writing the review for the readers and providing constructive feedback to the manufacturers.
Consider the scenario where I don't write about 'BrandX' in this article - The reader doesn't know that there is a 240GB USB3 SSD out there which throttles to 20 MBps. He MAY end up buying that without even knowing about the issue.
Now, consider someone who has read this article with BrandX. He knows that there is a USB3 SSD at the 240GB capacity point that throttles. He is definitely going to do his groundwork before purchasing the unit. There is a very HIGH probability that the unit he ends up buying might not be BrandX.
Anyways, any person who reads online reviews will find the problems that I have reported here in other sites, albeit with not as many comparison points that we have here. [ Eg: http://geardiary.com/category/reviews/ ]
It doesn't ta
OrphanageExplosion - Wednesday, November 19, 2014 - link
"Btw, there are plenty of products that we evaluate, but review never gets done because we are waiting for the manufacturer to resolve some issue that we report."Then the fact remains that you are putting the interests of your advertisers and suppliers over and above the interests of your readers. Let's be honest here, as the guy above says, if the data had been awesome you *would* have run the review.
In essence you are essentially admitting that you are filtering out bad reviews and only presenting good ones. You're even asking us to read other sites to get the data and recommendations that for some reason you are deliberating holding back. That's... breathtaking - for all the wrong reasons.
ganeshts - Wednesday, November 19, 2014 - link
A review is written only when there is a good balance of plus and minus points. Sometimes, the minus points may be more than the plus ones, making the product a tough recommendation. If there is nothing good in the product and it is no different from what is already in the market, what is the point of presenting the data point to the reader? Why should I waste time on writing a review if all I have to say is: "Don't buy this product" ; ?Running a review process should benefit everyone - the author, the reader and the manufacturer. Put yourself in either of the non-reader shoes and ask for yourself.
FWIW, there is no product that I have reviewed where the 'product is awesome' and has no minus points.
phoenix_rizzen - Thursday, November 20, 2014 - link
It's not a waste if it prevents someone from buying a crap product. That's the entire point of reading reviews before buying ... to know which ones are crap (avoid) and which ones are good (buy).If all you present are "good" reviews, how are we to know which products to avoid as lemons?
A well-written review that shows all the good and bad of a product is never a waste of time. Even if the end result is "don't buy this piece of crap".
ganeshts - Wednesday, November 19, 2014 - link
Given a choice between ' no review ' or ' misleading review : http://goo.gl/giqJ19 ' , what would you prefer?OrphanageExplosion - Thursday, November 20, 2014 - link
How about a transparent review that puts your readers' interests ahead of the manufacturers'? Oh by all means don't run a standalone review if you don't want to. But at the same time, deliberately obscuring the identity of a poorly performing product only reflects badly on you and your review policy.vailr - Sunday, November 16, 2014 - link
In my view, the better solution for an external USB 3.0 drive is: to use a USB 3.0->SATA adapter cable + a regular 2 1/2" SSD. For anything larger than about 32 Gb, it's also usable (without the adapter cable) as a bootable internal SSD drive.A 10" long USB 2.0 adapter cable was included with my Samsung EVO 840 120 Gb SSD. Or similar USB 3.0 versions can be found at Newegg or Amazon for ~$8.Araemo - Sunday, November 16, 2014 - link
Do all of those (other than the lacie) get treated by windows as a removable drive (rather than a 'fixed disk') - note that this is not the same as showing a 'remove hardware' option.. typically, even usb drives using a usb->ata controller get shown as 'fixed disks' (which enables such things as native partitioning, instead of treating them as a large single partition, in windows). More importantly, it makes them unusable for deploying OS's using Microsoft's deployment toolkit or SCCM.lorribot - Monday, November 17, 2014 - link
What would be good is if you threw in a wild card entry to some of your reviews.In this case something like a consumer SSD such as Crucial's 128GB MX100 in a cheap USB 3 caddy and see what you can get for around the £60/$80 or even 256GB for around £90/$110, yes it is not the same form factor but would be an interesting comparision if price or capacity where bigger considerations than size.