Pedros Posted December 17, 2010 Share Posted December 17, 2010 Hello to all. I've installed a Raid 0 using 2 corsair force 60gb ssd. Even though the performance is high... i'm not getting the same performance i see in so many benchmarks that, give a average read/write of +500Mb. Bench of my raid http://img.photobucket.com/albums/v319/Postmodum/read.png So, is there any catch to improve performance on this? It's not that i'm not happy or i will notice anything, i just want to know what variables must i be aware, performance wise, that can change how SSD's perform. An the graph is all messed up ... the picture bellow is very consistent. I have the OS installed on the SSD RAID Here's a bench of another force raid( 40Gb not 60 ) http://img80.imageshack.us/img80/6568/cf40i01.png Thank you Pedro Link to comment Share on other sites More sharing options...
Yellowbeard Posted December 17, 2010 Share Posted December 17, 2010 The original specifications were generated using ATTO. Try running ATTO and post a screenshot. Also, it's impossible to properly assist without your system specifications. Please fill them into the dropdown menu as requested at registration. Link to comment Share on other sites More sharing options...
Pedros Posted December 17, 2010 Author Share Posted December 17, 2010 Oh, sorry, going to do that asap. Link to comment Share on other sites More sharing options...
Pedros Posted December 17, 2010 Author Share Posted December 17, 2010 Ok system is filled and here's the screens Atto: http://img.photobucket.com/albums/v319/Postmodum/atto.png HDTune http://img.photobucket.com/albums/v319/Postmodum/read.png It's weird, writing higher values compared to reading... PS.1) I installed latest Intel Rapid Storage driver too. Link to comment Share on other sites More sharing options...
Corsair Employee RAM GUY Posted December 17, 2010 Corsair Employee Share Posted December 17, 2010 You drive seems to be fine compared with the reference ATTO results, so there is no problem that I see. Link to comment Share on other sites More sharing options...
Pedros Posted December 17, 2010 Author Share Posted December 17, 2010 So, the fact that's showing higher writing values, compared to reading... is it because i'm using the drive for the OS and the reference ATTO values from Corsair are "clean" drives? Link to comment Share on other sites More sharing options...
chorner Posted December 18, 2010 Share Posted December 18, 2010 They like you to test with ATTO because it provides easily compressible data. HD Tune pro might be reading/writing random uncompressed/uncompressible data. Kind of a 'gotcha' of Sandforce based SSD's if you will - highly dependent on the kind of data you're writing. You'll find even though Intel SSD's are rated "lower", actually usage patterns puts the Intel drives as high, if not higher performing on average. I'm currently running 2x120GB drives in RAID 0 configuration at the moment, while having tested the Intel drives and paying close attention to good reviews. It's the 4K read/write performance that generally will give you the feeling of snappiness on these drives. Maximum sustained throughput doesn't mean much for every day use, especially since if you're installing things from a DVD or CD, or from a storage HD... you're not likely to see even half of that maximum sustained transfer rate. Kind of misleading... but two things with SSD's are the extremely quick random access time, and the 4K read/write performance. Link to comment Share on other sites More sharing options...
Pedros Posted December 18, 2010 Author Share Posted December 18, 2010 Chroner, thanks for the reply. My only problem is the inconsistent graph i got with HD tune. I usually see the benches and they look almost like a flat line, without any major spikes ... Mine is all messed up. My question is, are the benches that i see, with almost flat line consistency and with higher read/write values and access time, clean drives, that are installed as a 2nd hard drive, and not running an OS? Link to comment Share on other sites More sharing options...
Wired Posted December 18, 2010 Share Posted December 18, 2010 Probably, but you'd have to ask the person who ran those benches to be sure. Link to comment Share on other sites More sharing options...
richardd43 Posted December 20, 2010 Share Posted December 20, 2010 Your numbers look good. Instead of worring yourself to death about a benchie, load up your games/apps and enjoy. I use bench checks to check my system against itself if I make a change. No 2 computers will bench the same so you never really get a truely accurate comparison. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.