840 Evo question

[mojothehut]

Hey all.
I recently decided to run some benchmarks on my 840 Evo 512gb

I'm a little shocked at the results from Samsung Magician

So here's CrystalDiskMark results as well.

Should I be concerned about the low sequential read results? The drive has been running in my machine for about 6 months now, 250 of the 512gb used.

Running on a Z87, i7 4770k @ 4.4ghz rig, with 16mb 1600mhz ram and win7 pro

 

[Elixer]

No, those look normal.

 

[BonzaiDuck]

Hey all.
I recently decided to run some benchmarks on my 840 Evo 512gb

I'm a little shocked at the results from Samsung Magician


So here's CrystalDiskMark results as well.


Should I be concerned about the low sequential read results? The drive has been running in my machine for about 6 months now, 250 of the 512gb used.

Running on a Z87, i7 4770k @ 4.4ghz rig, with 16mb 1600mhz ram and win7 pro

have you tried using RAPID yet? It would be interesting to see what sort of benchies you get with that. You have 16GB of RAM -- might as well use it for something.

 

[mojothehut]

RAPID lead to crazy benchmarks...
245,835 read IOPS and 586 sequential read

Wiggy, well I'll run with this mode enabled for a while, as long as it won't "hurt" anything

Thanks =)

 

[BonzaiDuck]

RAPID lead to crazy benchmarks...
245,835 read IOPS and 586 sequential read

Wiggy, well I'll run with this mode enabled for a while, as long as it won't "hurt" anything

Thanks =)

Certainly the concern a veteran PC enthusiast would have: "Is the RAM-caching software totally reliable?"

So far, I've been using RAPID with my 840-Pro boot/system drive (512GB). Continuous use of RAPID hasn't caused any trouble at all. Further, it doesn't seem to make much difference in the usage of my 16GB RAM, although the most recent version of "Magician" uses more RAM than the previous limit of 1GB.

What I cannot understand in the OP's case, perhaps specific to his configuration, is the low sequential-read score in either situation. My 840-Pro score with the Magician benchmark is ~6,000 MB/s.

Also, in passing, I have tested the general software RAM-caching programs Primo-Cache and Super-Cache. They both use block-level caching approaches like RAPID, but you must choose the amount of RAM allocated in setting them up -- the caching software doesn't simply dynamically adjust the amount or RAM used. Especially, both programs seem to be as reliable as RAPID, and more especially -- PrimoCache gives more options in terms of allowing cached, delayed or "lazy" writes. This latter feature may increase the chance of data-loss from system crashes, but at least the software won't cause those crashes.

 

[skyhawk21]

Can you run hdtune please of the samsung ssd's and post a screenshot? thanks~!