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

Thumbdrive Restoration

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

TollhouseFrank

Senior Headphone Guru
​
Joined
Nov 29, 2004
Location
T3h Intr@tub3z!
This was written as a stream-of-consciousness piece while I was doing the tests. I present the notes as-is, no spellchecking or reformatting.

Tollhouse's Notes said:
I am testing to see if a well-used flash based usb drive can be sped up using a 0-write tool to 'clear' all the registers in the drive. Before doing this, however, i'm going to do a baseline test of the drive as is before doing both a clean format and then test, and then doing a zero-fill and then test.

The drive as is is a Kingston Datatraveler 16gb. It is a USB 2.0 standard drive. As is, windows reports it is 2.15gb used and about 12.8gb free out of a possible 15gb. Over the past few months, it has been written to, erase, written to, erased, hundreds of times. While the read speeds are fine, I have noticed the writes and re-writes have slowed down dramatically.

Before doing the test, I am going to format the drive so that it is seen as "clear", even though as many of us know, a flash-based storage method isn't really clear.. the information is there and thus affects write/rewrite speed.

The first tool I am using is called HD_Speed. It's tests have to be run seperately. The first thing I am doing is a Read Test and will let it run for a minute or so to make sure it is getting a good baseline reading.

The end results from this first read test are as follows (from the log file, a 2 minute run)

[Test Stop 2/1/2010 10:23:53 AM] errors: 0 average: 16.7 MB

I will now do a write test and the results are as follows (from the log file)

[Test Stop 2/1/2010 10:52:22 AM] errors: 0 average: 13.4 MB

So with this as a baseline, I am now going to 0-fill the drive and see if that affects either test in any significant manner. The program I am using to do the 0-fill is Active Killdisk Free. After having done the 0-fill, I had to reinitialize and format the Kingston Thumbdrive for windows to see it.

Now it is on to the read/write tests to see if the 0-fill had any affect on the drive.


[Test Stop 2/1/2010 12:06:55 PM] errors: 0 average: 21.3 MB

A very unexpected jump in speed! Gives me a lot of hope for the write/rewrite portion of the test!

[Test Stop 2/1/2010 1:04:05 PM] errors: 0 average: 13.4 MB

Unfortunately, it does not help the average write/rewrite time.

So all in all, the results of this test were surprising, just not in the ways I was hoping.
Click to expand...
 
will be running this test on at least 2 other drives, possibly a 3rd and 4th other drive in the next 2 days or so to see if there is a consistant set of test results for this.
 
You already wiped it? You didn't take my suggestion of running ATTO. That is going to show what file sizes that it is slowing down on. Just a straight through speed benchmark isn't going to show much as that isn't its strong point.
 
i alreayd know the file sizes it is slowing down on. It's anythign smaller than 1mb. THere is a huge dropoff below 1mb. It goes from 13.4mb/s write to about 6mb/s write. However, I decided to just do a pure read and write test to see if 0-fill had any affect whatsoever, and it appears it does (at least on the 1 drive i tested).

I'm going to test some others tomorrow... a usb1.1 drive (2gb), a usb 2.0 drive (2.0gb) and a usb 2.0 (8gb) all from different manufacturers so i'll hopefully get a better dataset than just the kingston drive.
 
Yes, but having an actual benchmark is going to give much more clear results than "yeah, it was slow". Not trying to be mean or anything, I just like having solid numbers. It also shows much MUCH of a difference formatting it will make. If it was only 5%, I wouldn't do it. If it was 50%, then I'd seriously think about it.
 
i already restored all the files that were on the drive before. I keep an image file around of each thumbdrive i use for just such a purpose (in case it gets corrupted by a virus on a customer pc, for instance).

I'll try to remember to run the ATTO test this time on the Kingston Drive before I run the others.
 
Hmm... I have a 4 gig PNY drive that's been my tech drive for a few years. Broken and battered, as I stuck it in a tight USB port, the casing cracked and I had the housing in my hand, and drive stayed in the USB port.. With about 30MB free, deleting and and putting files on it for about 3 years straight. Its still as fast as when I got it. I'm thinking that the flash RAM that is used on these drives are not prone to the short falls of SSD's... Unless it uses the same technology...
 

Attachments

  • PNY-XLR8-4gig.jpg
    PNY-XLR8-4gig.jpg
    95.9 KB · Views: 202
Last edited:
Joeteck said:
Hmm... I have a 4 gig PNY drive that's been my tech drive for a few years. Broken and battered, as I stuck it in a tight USB port, the casing cracked and I had the housing in my hand, and drive stayed in the USB port.. With about 30MB free, deleting and and putting files on it for about 3 years straight. Its still as fast as when I got it. I'm thinking that the flash RAM that is used on these drives are not prone to the short falls of SSD's... Unless it uses the same technology...
Click to expand...


I wanted to add that I have a HP T5720 thin client, that I have upgraded from a 512MB flash drive to an 8 gig. I noticed some major slow downs with this system, and would love to test your little program with that. This is what the drives look like.
 
downloaded atto this morning right after i got to work. gonna be lettin' it have a go at the 16gb Kingston drive to see what happens, then gonna give it a whack at the two 2 gig drives (usb1.1 drive and another usb2.0 drive) that are 90% full to see what happens
 
first test though is going to be my personal 8gb PNY Attache'. I've used it hard the past year, lots of large and small file swapping. Not been reformatted and not been zero-filled yet. Hopefully this will give a truer test of how well zero-filling will help after I run Atto Before and After.

**edit**

just finished running ATTO on my personal drive. has probably 4 gigs 'hidden' files from stuff i erased but due to the type of technology of flash memory, is still technically there. Doing 0-fill now then will re-run the ATTO Test, then will run the HD_Speed test just to line it up against the Kingston to see if there were similar improvements.
 
Last edited:
Cool testing man... looking forward to this... Also wondering why HDTach results were posted in here as it was well noted thats only for mechanical drives (unless there was an update) b/c of inconsistent results.

Anyhoo, I second ATTO used along with your app at its stock settings as well to show us where these slowdowns and improvements are. Its a quick test anyway. ;)
 
You are more than welcome to support your assertion some kind of links. You dont have to believe me though...:

http://benchmarkreviews.com/index.p...k=view&id=270&Itemid=38&limit=1&limitstart=11

Traditional tests such as IOMeter, HD Tach, HD Tune, and Passmark PerformaceTest all use buffered spot-samples for testing and are not reliable for SSD benchmarks. After testing with HD Tach RW for the better part of two years, I know that this tool is perfect for testing Hard Disk Drives, but SSD technology adds an element to all of these benchmarks which render them less accurate. Simpli Software doesn't promote their product for SSD testing, so I can waive the differences I've noticed between controllers, system memory, and drive cache mechanisms to the fact that they've never intended their 'HD' software for SSD products. Regardless, these tests are not suitable for SSD product benchmarks, and do not render an accurate performance profile.
Click to expand...

EDIT: I do realize these are not SSD's, but, someone correct me if I am wrong, these work off similar principles. The bottom line is that its not a MECHANICAL drive so I would believe, though admitedly I am not sure, that the applications listed for MECHANICAL drives wouldnt be as accurate as those for SSD's...If anyone can shed some light on the situation that would be appreciated. Getting correct, acurate information posted is KEY.
 
Last edited:
Here are some of the results from ATTO.

For the Kingston (this is after putting all the files back on it after the zero fill.)

results%20on%20post%200-fill%20kingston16gb.PNG

I'm going to post all the PNY results (my personal drive) after i get all the files back on it and run ATTO one more time to see how it fares with the 2.5gb's of data on there vs. when i ran ATTO after the zero fill (also ran ATTO pre-zero fill on the PNY... so i can have all 3 results)
 
TollhouseFrank said:
Here are some of the results from ATTO.

For the Kingston (this is after putting all the files back on it after the zero fill.)

results%20on%20post%200-fill%20kingston16gb.PNG

I'm going to post all the PNY results (my personal drive) after i get all the files back on it and run ATTO one more time to see how it fares with the 2.5gb's of data on there vs. when i ran ATTO after the zero fill (also ran ATTO pre-zero fill on the PNY... so i can have all 3 results)
Click to expand...

Do you have a before Screen shot?
 
I've come out of hiding and wanted to voice my opinion on the subject on HD benchmarking tools... And I've been hanging out and reading rather than posting on many sites including OC forums... (side note) The new design looks great! Good work guys! It has been awhile...lol

Anyway, in response to Earthdog and Joeteck about SSD vs thumb drives benchmarking tools... SSD's have a much higher rate of throughput than a thumb drive...USB 2.0 has a maximum of about 35MB a sec. SSD's have on board cache where as a thumb drive does not and will not suffer from inaccurate test using HD tach. No cache to read from to throw off the tests... So in Joeteck's defense, he's is correct.. I have tested many thumb drives in my day, using all sorts of tools. Thumb drives just don't have the gusto to keep up with SSD's... well not yet at least... And if you look carefully at his (Joetecks) tests, it clearly shows them to be darn close if you ask me... HD tach says 32.8 MB/s and ATTO says 32.1MB... I think we can say that HD tach can be used on thumb drives...
 
I stand corrected. All I wanted was supporting information instead of backlash regarding my "attitude", and I (finally)got it!!! Thank you VERY much Mr Fix it!!! :)
 
Unfortunately, no pre-0fill on the Kingston drive. Kinda screwed that one up.

Anyways, here are the PNY Attache' screenshots.

This first one is my drive (8gb) pre 0-fill. It also still has on it all my files, is totally fragmented, and still has all the 'ex-data' from the various other things i had on it... close to 4gigs....

pny%20attache%208gb%20pre%200fill.PNG


This next one is my drive post 0-fill. Totally Empty Drive.

pny%20attache%208gb%20post%200fill.PNG


Now for the coup-de-grace.... my drive post 0-fill plus all my 2.5gigs of data back on it.

pny%20attache%208gb%20post%200fill%20plus%20data.PNG
 
You must log in or register to reply here.

Similar threads

Overclockers.com
FRONTPAGE Corsair MP700 PRO 2TB NVMe PCIe 5.0 x4 SSD Review
Replies
3
Views
521
Woomack
W
BROOKLYN ZOO
coldboot (warmboot) issues @ R7-5800x | B550 Aorus Elite V.2 rev.1.0 | AGESA 1.2.0.5
2
Replies
31
Views
958
EarthDog
EarthDog
S
CMT32GX5M2X6200C36 *4 Sticks Z790-E GAMING WIFI i9-13900K XMP Unstable
Replies
4
Views
674
Woomack
W
S
Rebuild / upgrade looking for fine tuning advice
Replies
0
Views
707
Savlaka
S
W
Gigabyte B650 Aorus Elite AX v1.0 - unable to flash BIOS
Replies
14
Views
2K
Whity67
W
Back