• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

What are the most trustworthy web sites to see USB 3 Flash Drive speed rankings

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

c627627

c(n*199780) Senior Member
Joined
Feb 18, 2002
There are so many flash drives out there and some USB 3 Flash Drives are so much faster than others...

Also, when they rank them, is it safe to say that a 32GB ranking is the same as a 64GB version of exact same name/model flash drive?
 
I've done quite a few on our front page, working to get more as well.

No, it isn't the same.
 
Generally the larger flash drives are faster of the same series. Same as SSDs
 
So for example, if you want to know the ranking of a say, SanDisk UltraFit 64GB drive, you cannot just look at the write/read numbers of a SanDisk UltraFit 32GB drive, correct?

The question still remains about the data base of real life speeds. I know that if I looked I could find some but thought forum members have some suggestions from personal past experience of where they looked?

It is not the same transferring a single file. It is not the same transferring relatively small amount of data. I just transferred a 17GB folder filled with a mixture of small and huge files and the writing process took a while but the reading process, when 17GB folder was being transfered FROM the flash drive, that was extremely fast.

I wonder if overheating issues in these ultra fit drives contribute to these writing slow downs... Or maybe other drives would take as long or longer to write.
 
Last edited:
Just saw your post Janus67, so the numbers of a 64GB of the series would be the same or more often, faster than 32GB, but never slower?
 
So for example, if you want to know the ranking of a say, SanDisk UltraFit 64GB drive, you cannot just look at the write/read numbers of a SanDisk UltraFit 32GB drive, correct?

The question still remains about the data base of real life speeds. I know that if I looked I could find some but thought forum members have some suggestions from personal past experience of where they looked?

It is not the same transferring a single file. It is not the same transferring relatively small amount of data. I just transferred a 17GB folder filled with a mixture of small and huge files and the writing process took a while but the reading process, when 17GB folder was being transfered FROM the flash drive, that was extremely fast.

I wonder if overheating issues in these ultra fit drives contribute to these writing slow downs... Or maybe other drives would take as long or longer to write.

I have included real world testing in my results, they're very representative of my "seat of the pants" impressions as well.
99% of the time it's writing where flash drives struggle, hence my testing of writes only in the real world portion.

Just saw your post Janus67, so the numbers of a 64GB of the series would be the same or more often, faster than 32GB, but never slower?

The vast majority of the time this is a correct assumption.
 
Click on Effective Speed if needed - larger the drive the faster it seems to go according to the websites

http://usb.userbenchmark.com

So they only test Sequential, Random 4k, and Deep queue 4k?

In my reviewing I've found it best to look at an actual real world test.
Benchmarks on flash drives, especially when it's compressible data, can be vastly different from actual usage.
 
This turns out to be the thread topic... Which rankings represent the writing speed for a large folder filled with a mixture of large, huge, and small files.
I have no complaints about READ speeds. Those are plenty fast... But guess what, I have plenty of time to wait for reading - it is when you are in a hurry to take off, that is when speed is of importance... Therefore, looking for real life Write speeds...
 
Well its relative there, i have never had a file copy faster then ~80MB/s (from HDD or SSD) even though my pen does ~96MB/s, so either my drivers arent working 100% or USB3 has a low limit ?

Sandisk Extreme 32gb.jpg
 
The thing that gets me is that there is no standard to measure that figure you just posted.
Our drives can only be measured relative to one another using exact same program to measure them.
If you use a different program, you appear to get different results.

So to that end, it would be nice not to see a figure - but a ranking system, so you can tell relatively, where the drive you want to get stands.
Surely there are charts out there, surely we can kind of tell which ones are reliable because given the same model/capacity drives those sites should be similar when it comes to identical drives.
 
C6, this dilemma is exactly why I have my folder test right before the conclusion. You simply can't replace testing of a real world folder.
 
Do you have a WRITE speed list of all 3.0 drives you tested? What were some of the fastest?
Why are 3.1 drives on the market currently not faster than 3.0 drives?
 
TBF the only real world testing i ever done apart from ATTO was daily using Teracopy on multi gigabyte files :p as to 3.0 vs 3.1 it really depends on the drive itself doesn't it ?
 
There are only a few 3.1's anyway and their actual listed specs are no faster than 3.0.
What WRITE rate would you expect to see in a good USB 3.1 Flash drive?


I agree that a lack of standard of measuring drive speeds is definitely problematic. Same program - same test folders/files - is the only way to truly measure them.
But who's to say what program and what test files should be used...
 
I have received a 128GB Samsung USB 3.0 Flash Drive fit USB 3.0 drive to replace an identical looking SanDIsk UltraFit 3.0.
Both Read * Write figures fluctuate, depending on what tests you use, but both fluctuate in the range of +/- 40 MB/Sec Write and +/- 140 MB/Sec Read.

Since I have both now, this was the starting point for me really thinking about this issue.

I came to the conclusion that this is an extremely important topic, about which few people have real knowledge about.


Before doing a file transfer test, I did research and quickly came across articles like this, listing reasons why copy-paste is NOT a measurement of read/write performance:

https://blogs.technet.microsoft.com...-good-idea-and-what-you-should-do-instead/+/-


Both drives when measured by a benchmark program do +/- 40 MB/Sec Writes, but are they both the same?


First of all, a side note, the +/- 40 MB/Sec is a really important figure. Reads are three to four times faster and they are okay on many USB 3 flash drives.
It is the WRITES that are a problem and it is the figure of +/- 40 MB/Sec that needs to be surpassed if we are to speak of a "better" or rather faster USN 3.0 drive.
Would be interested in hearing if anyone has come across drives faster than +/- 40 MB/Sec Writes and what they are, and what you used to get a figure significantly larger than +/- 40 MB/Sec
Writes.


Second, here's the big picture.
I replaced by SanDisk because I was in a hurry and had to wait and wait and wait for it to complete because as you know these things get hot but when they do,
they actually copy, pause to cool down, copy, pause to cool down.

Well it is alleged at various Hot Deals forums that Samsung version does not run as hot as SanDisk's version, and since it doesn't - there is less pausing in between of large file transfers, and consequently, one would imagine, the drive is faster...


But is it really?


They are both identical when programs used to measure speed are concerned, but how do you measure which one would transfer a large amount of files, if you can't use Windows copy-paste to test for reasons outlined in that article. But did the article refer to measuring a NUMBER or did the article also outline why same case scenario copy-paste is not the same?
 
So the question stands on the validity of using Windows copy-paste as a benchmark....

In the meantime, I nevertheless took a laptop and disconnected its internet connection.
I used the same USB 3.0 port and used the same folder located on a slow mechanical laptop drive.

This was a 6.26 GB folder. But it had almost 18,000 files in its subfolders, making the USB 3.0 transfer process take longer because of the shear number of files and perhaps because of the length of transfer activity, making the drives hotter and hotter....

• A brand new just purchased Samsung Fit 128GB USB 3.0 drive took 16 minutes and 16 seconds whereas
• An old SanDisk Ultra Fit 128GB USB 3.0 drive indeed did take a lot longer, in fact it appeared to be heading towards being twice as slow when at 95% it appeared to overheat so much that it it completely stopped the transfer and remained at 95% for 10 minutes and then finally resumed to finish at 36 minutes 28 seconds.

Perhaps it is true that even though these drives score near identical at short bursts of a speed test, the SanDisk perhaps handles heat much worse and that is why it took so much longer.
I hope to acquire a brand new SanDisk identical drive and to then repeat this test.

I saw the old SanDisk pause continue pause frequently, especially toward the end of the transfer.
 
Back