A Benchmark For Just SSDs - AS SSD Benchmark

[Apoptosis]

A new benchmark is in the workds for Solid State Drives - AS SSD Benchmark

The benefits of the benchmark are:
* Good precision of results
* Simplicity of use
* Optimized for SSD
* Screenshot-function
* It's freeware

Min. sys req:
* Windows XP/Vista/7
* .NET Framework 3.0
* An SSD to test with 3gb free space on it

Known bugs:
+ You can only abort the test between the measurements
+ If you close the program without ending or aborting the measurement it leaves some temp-files on the partition you have tested

Download Link: http://alex-is.de/PHP/fusion/downloads.php?cat_id=4

Some Screen Shots that I took using it on my OCZ 120GB Vertex EX SLC SSD tonight using version 0.8.3451.39214 on a Vertex EX drive that has been used for a number of days:

[nsa666]

Thank you for using my benchmark on your reviews.

There was some updates and it reached final stage now.
It looks like this:

[FZ1]

Welcome to the forums NSA

[nsa666]

To implement some web-functions(update-notifications, database of results are planed) i ordered new webspace.

New link for download: http://www.alex-is.de/PHP/fusion/downloads.php?cat_id=4

[Apoptosis]

Thank you for using my benchmark on your reviews.

There was some updates and it reached final stage now.
It looks like this:

Nice, thanks for the update... I'll download this now as I'm starting a new SSD review.

[nsa666]

Update:

1.0.3462.36297
- Filename bugfix at screenshot-function of copy-benchmark

[nsa666]

Update:
1.1.3466.29641
* Update checker
* New score system
+ Small design changes

[DMB2000uk]

Wow, you're really churning out the versions, keep up the good work!

Dan

[nsa666]

Update:
1.2.3487.20916
* IOPS-View

[Apoptosis]

Update:
1.2.3487.20916
* IOPS-View

Just downloaded the latest build and tried it out yesterday.. How exactly does the point system work? Also, with so many different versions can the numbers between two versions be compared to each other?

intel_34nm_asssd.jpg (59.73 KiB) Viewed 186800 times

[nsa666]

There was some small changes on score system from 1.0 to 1.1. Score System gives lot more points for random-values than for sequentual.
It is Score=seq_write * 0.15 + seq_read * 0.1 + 4k_read * 2 + 4k_write + 4_64thrd_write + 4_64thrd_read * 1.5

P.S.: You should enable AHCI on your intel SSD that makes it A LOT faster.

[Apoptosis]

P.S.: You should enable AHCI on your intel SSD that makes it A LOT faster.

If only you could just enable it...

[FZ1]

P.S.: You should enable AHCI on your intel SSD that makes it A LOT faster.

If only you could just enable it...

Haven't tried it myself...
http://support.microsoft.com/kb/922976

To resolve this issue, enable the AHCI driver in the registry before you change the SATA mode of the boot drive. To do this, follow these steps:

1. Exit all Windows-based programs.
2. Click Start, type regedit in the Start Search box, and then press ENTER.
3. If you receive the User Account Control dialog box, click Continue.
4. Locate and then click the following registry subkey:
HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\Msahci
5. In the right pane, right-click Start in the Name column, and then click Modify.
6. In the Value data box, type 0, and then click OK.
7. On the File menu, click Exit to close Registry Editor.

[darren]

Hi nsa666,

First of all thanks for this app.

Now does this work with SLC ssd's also? When I run it on my 16GB SLC ssd with 8GB free space on XP SP3 it takes over an hour, the 4K writes are way slower than I know they actually are.

[nsa666]

Hi darren,

Yes it works with SLC SSDs and it shows exactly the performance MTRON specifys in the datasheet (about 130iops at 4k random write).

See: http://www.ssdisk.eu/downloads/datashee ... rev0.4.pdf page 9 (3.3.1).

4k*130iops=520kb/s=0,5MB/s

Unfortunatly CDM is wrong with this SSD, and SLC not automaticaly means higher performance than MLC, you also need a good Controller.

[darren]

Hi nsa666,

I see, thanks. You certainly seem to know your stuff. Is it possible to shorten the time it takes for the benchmark to run on slower ssd's or is it nessasary to get accurate results.

Would you consider adding HDD support aswell?

Thanks,
darren

[nsa666]

The benchmark realy writes 1gb of data(every test) on the SSD to get accurate result. if i reduce ammount of data that the benchmark writes it will lead to reducing of accuracy(see CDM that not realy wirtes the ammount of data even it you set it up to 1gb). Since every SSD generation is faster(on a cheap Indilinix-based MLC-SSD the benchmark takes about 5 minutes), it makes no sense to support slow SSDs. It is not planed to support HDDs either(there is a lot of HDD Benchmarks). You acutally can run this test on a HDD but it will take very long.

[darren]

OK thanks again nsa666. Keep up the good work

[darren]

One more thing nsa666, does AS bypass the SSD's onboard cache?

[nsa666]

No it does not. Its only bypasses windows/controller-cache. But SSD-Cache is just to small to make huge difference.