Jump to content
Corsair Community

Benchmark fiasco


Recommended Posts

I have been doing some research on the net and i finally understand why ATTO is the only bench recommended by corsair.


its because its the only benchmark that uses uncompressed data which gives the full reading of the Sandforce drives.


i have heard many times on here that others such as AS SSD and Crystalmark were made for HDD this is UNTRUE.


They are all SSD benchmarks and they are all fairly new in their approach.


They also use non compressible data which is why you see the lower scores.


these benchmarks ARE indicative of a drives performance in every aspect but only when used in conjunction with other drives using the same test.


i think it would be easier for people to compare their drives to otheres if there was some kind of thread / database which people could use as a reference for different drive models.


using ATTO as the benchmark is pretty poor since its not very user-friendly in its approach to displaying data. which is why people prefer the AS SSD and Crystal mark.

Link to comment
Share on other sites

I believe ATTO is harder to understand. It has many different settings that you can mess with before the test and if a person is unsure what all of those settings do, it can be a little overwhelming.


Everyone has different levels of technical understanding.


Many other tools give you a nice graph with one line for write speed and one line for read speed, with MB/s on the Y-axis and drive location on the X-axis, with a min and max for both clearly stated.


Many of those tests also don't test with varying block sides as ATTO appears to do, but is seems that ATTO doesn't to a read test to the entire drive, just a portion of it, which is fine for SSD since the throughput should be mainatined throughout the length of the drive. - This may also be why people say the alternative benchmarks are not designed for SSDs.




Link to comment
Share on other sites



Please see my thread that relates to this topic. http://forum.corsair.com/v3/showthread.php?t=89060

It pretty much concludes the same and explains why it is like this.

ATTO might be indicative of a drives performance, but not real world performance since you'll never use only 0fill in a real environment.


It's like a car that can run twice as fast as any other car, but only if you run it on special fuel which you can't buy anywhere. Pointless..


/ Aloha

Link to comment
Share on other sites

i have heard many times on here that others such as AS SSD and Crystalmark were made for HDD this is UNTRUE.



This is very version dependent. When I see users posting results with multiple versions, it's almost impossible to track the change log on some of these free utilities. The point is that utilities have to be updated as new controllers are introduced. And, I have yet to determine if these utilities began life as an HD bench.


Both of these utilities mentioned above have had frequent updates and there have been some inconsistencies with their results as we have expanded our SSD lineup. This is not to say that they are not good tools nor are we saying that they they cannot be useful. However, you cannot compare their results to ATTO results directly.


On our product page we clearly state that the results were obtained using ATTO. At the time the SandForce controller was introduced, ATTO was the only utility that would yeild both consistent results and results that were consistent with the manufacturer's stated specifications. We have used ATTO and only ATTO to publish results since our first SSD line. It's an industry standard test and we have to be able to publish results that our customers can use.

Link to comment
Share on other sites


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

  • Create New...