S-128 perceived better then P-128 In actual use

[matthewj]

Hello, I am running a SSD pilot. During the early stages of my pilot I used S-128 as the HD of choice and we had great performance. People felt their machines were quicker and better. However part way through we switched to P-128. Users said the performance was better but not great. I ran some a couple of tests and saw some strange differences. Not so much in how quick the drives were, but in their minimum write speed.

 

The S-128 did well having a minimum write speed of a 59.7 MB/sec on a Lenovo T61 and 72.1 MB/sec on a T400. However the P-128 did much worse having a minimum write speed of 2.9 MB/sec on the Lenovo T61 and 31.9 MB/sec on the T400. We feel this change in minimum write speed is why users feel the P-128 is not as fast as the S-128.

 

I have attached a PDF with results for you to evaluate. We have seen this across multiple P-128 drives. All machines had the same windows XP image. We did our best to control the variables we could. We are positive the change in results is because of the drive and not the software. Is there a fix or planned fix for this?

 

Thank you,

Matthew J.

SSD Eval.pdf

[leexgx]

the Free HD tune is an Read benchmark not Write

 

try this one CrystalDiskMark

http://crystalmark.info/software/CrystalDiskMark/index-e.html (press the ALL button and it should give you some results post the on here)

http://rd1ugg.bay.livefilestore.com/y1pK_eJ4k0yup4tF1iJf8u4TLIDHoUjLpeuN5AziqwXqEPdl-hUjAoxhVV4j9FPjKbtbo9V-sRkg7byoTaKZ9wB_Q/crystaldiskmark%20HDD%20vers%20S128.png

there is atto disk benchmark as well

 

before i fully comment both of them drives look like they are operating at SATA 150 speeds as the laptop/motherboard is only set to 150 in most cases as your busted rate does not go past 90mb/s on the T61 that seem like an chipset limit and the T400 looks like its going at 150 Sata

 

from tests i have done on my S128 when HD tune reads none erased space it reads at the Sata speed limit as to why the read speeds are around 75MB/s as thats what the drive is stated to operate at and mine still does (2 months old) but towards the end of the test at 75-100% it goes to 160mb/s (this be interesting what happens when i run out of erased space, but should not be an problem once firmware is out for TRIM support for windows7/linux)

 

what image software was used

 

but with SSDs access times is whats makes them fast not the data rate as proven in CrystalDiskMark hdds tank once very small reads come into play SSD can access 4k files are 20MB/s, personly you Not notice any diferance in speed between an S128 an P128 (more so on an laptop) as most programs on an laptop are going to be below 100mb any way most will CPU limited when an SSD is installed and a lot laptops are SATA 150 limited, even on desktops the 200MB/s data rate SSDs or higher does not mean the random read and Writes be much better (i would like to see yours)

[matthewj]

I am not worried about write speed. What i am worried about is how it how the P-128 takes a drastic dip in speed, where the s-128 does not. I would ignore the max speeds. Yes, Crystal gives better speeds, however it only shows the max, it doesn't show the range. I feel the HD Tune is a more accurate test as it shows the read over time, which is what matters during long operations.

 

I think the P-128 is a real miss or Corsair's part. Personally i don't think it out performs the S-128, even thought it is faster.

[leexgx]

have you tested them in an desktops,

 

what cloneing software did you use

 

HD tune is showing max data rate as its only reading the disk directly from one end to the other, the results you are getting in HD tune do not seem right for the P128, CrystalDiskMark is an file benchmark test bit more real world then hd tune

[RAM GUY]

I would agree you may want to try HD Bench or ATTO to test the drive performance.