Jump to content
IObit Forum
Top Free Driver Updater Tools Best 25 PC Optimization Software Best 22 Antimalware Best 22 Uninstaller Software IObit Coupons & Discount Offers PC Optimizer Mac Boost Advice IObit Coupons A Good Utility Program From IObit IObit Promo Codes IObit Coupon Codes IObit Coupons and Deals FAQs Driver Booster Pro Review

Crystal Disk Mark


Toppack

Recommended Posts

'Crystal Disk Mark' is a great little utility program, for testing SSD, Hybrid and Hard Drives.

Good way to compare drive speeds.

 

I'm getting surprising results from some of our older drives! (very Slow) :cry:

 

Also, I did not realize there was that much difference in the Read and Write Times, of a drive !

(even New drives that are not fragmented)

 

It's Time to Up-grade! 8-)

 

Download link:

http://www.majorgeeks.com/CrystalDiskMark_d5574.html

Link to comment
Share on other sites

  • 3 weeks later...

I've been trying to use this program to determine the differences in the three Defrag Options in the Smart-defrag program,

and how they effect drive-Read/write-speed.

Although I see differences, I've determined that it's Impossible to get Accurate test data,

since there is no way to get a consistant Base-line fragmentation (referance point). :cry:

 

Once you defrag using one option, the drive will Never have the exact same amount of fragmentation again,

so you can Not run accurate comparison tests. :-?

Link to comment
Share on other sites

hi Toppack

Seriously you are right!

Puts one in mind of the measuring problem in quantum mechanics - the act of measuring cannot be done without influenzing the item measured.

Cheers

solbjerg :-)

 

 

I've been trying to use this program to determine the differences in the three Defrag Options in the Smart-defrag program,

and how they effect drive-Read/write-speed.

Although I see differences, I've determined that it's Impossible to get Accurate test data,

since there is no way to get a consistant Base-line fragmentation (referance point). :cry:

 

Once you defrag using one option, the drive will Never have the exact same amount of fragmentation again,

so you can Not run accurate comparison tests. :-?

Link to comment
Share on other sites

I was using this program to test and compare some new USB-3.0 Flash-sticks to older 2.0 sticks.

I found that USB-3.0 is about 9 times faster in Write mode but only about 3 times faster in Read mode.

But That's Okay with me. It's Great actually! :-D

 

The NEED for SPEED ! :

Link to comment
Share on other sites

I heard the SSDs Slow-down after the drives get more than half-full of data. :shock:

I do not have any SSDs to test,

So I tested a couple 16.GB Flash-drives and found that they do the same thing. :shock:

I then tested a 73% full HD and a 22% full HD, (same model) and found that they were still both about the same speed.

 

Seems Very Strange that SSDs slow down as they get fuller ? :shock:

Reportedly there is enough difference to really notice the difference.

Link to comment
Share on other sites

Hi Toppack

 

You have certainly set yourself a difficult task. There are so any variables involved. Such as the controller architecture and wear levelling methodology.

It would seem that there is no great rush by those in the know, to change to SSD. The most optimum system seems to be a combination of SSD and HDD, where the SSD holds the OS and the HDD handles the data.

 

The following is a good link showing the various designs and comparisons of SSD's...HDD's and FD's.

Good luck with your testing.

 

http://en.wikipedia.org/wiki/Solid-state_drive

Link to comment
Share on other sites

At first SSDs seemed to the next great computer improvement

but I continue to find more reasons Not to like them. :cry:

We have had two of them Fail, with normal use, as Boot-drives.

It's almost like they are Designed to Fail after a pre-determined time of use ? :roll:

 

That's why I think our next Drive upgrade will be the WD VelociRaptor 10,000RPM Hard-drive,

if they continue to get Good Reviews.

Link to comment
Share on other sites

Hi Toppack

I think I will follow you there.

I have to change a couple of old harddisks (12 years) soon (Chinese 很快) (3-12 months) I think :-). They each hold about 36 bad sectors now and even though is is only a very small percentage of the disk it is a gentle reminder that they won't last forever.

I run a check disk fairly often (every month or so - or if I suspect the disk of bad behavior) and I have therefore not lost any data to bad sectors.

The SSD technology will probably win in the end, but like you I trust the proven technology better for now.

Cheers

solbjerg

 

 

At first SSDs seemed to the next great computer improvement

but I continue to find more reasons Not to like them. :cry:

We have had two of them Fail, with normal use, as Boot-drives.

It's almost like they are Designed to Fail after a pre-determined time of use ? :roll:

 

That's why I think our next Drive upgrade will be the WD VelociRaptor 10,000RPM Hard-drive,

if they continue to get Good Reviews.

Link to comment
Share on other sites

  • 4 weeks later...

I think I have figured out why those SSDs seemed to have Premature Failures.

I found out the older SSD designs did Not have 'Wear Leveling' in the SSD's Controller-firmware.

'Wear-leveling' prevents incoming data from being written to one area of the RAM, more than other areas.

So prevents Uneven writes, which shortens the time to failure.

Each time an SSD is written to, that is 'one more time' closer to when it will Fail.

 

Newer drives also have 'garbage-collecting' (Trim) capability, in the controller.

Leftover trash continues to slow down the drive, after repeated use, if not there.

Using 'Crytal disk mark' I could see that slow down very quickly, after repeated use of my new test SSD,

but it must be an older design without 'garbage-collecting' capability.

I'm sure that the Motherboard and OS both have 'Trim' capability.

(and I'm sure Trim is activated in W-7)

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...