Well I did what ive read is a no no that cuases issues with stability and also performance of the ssd.... I imaged my factory hdd as soon as I got with Acronis and then placed it on my new Samsung SSD when I got it and DID NOT do a clean install. I also changed the BIOS to ACHI which is another thing ive read causes performance loss..... well......
Ive never had even the slightest issue with stability and i finally benchmarked the SSD via the software that came with it and my results seem pretty good.
Test range: 1GB
I/O size
Seq:128kb
Ran:512kb
RESULTS
seq read:434
seq write:377
Ran read:1028
Ran write:755
These results seem to be on par with the advertised max speeds of the drive. Just thought Id share in case anyone else was wanting to do the same thing without fear of losing speed or stability. Maybe I was just a lucky one I dunno.
Anyone else have benchmarks with this same drive?
-
Seq. Read - 513
Seq. Write - 374
Ran. Read - 1030
Ran. Write - 825
At the same test setting.
This was after a Temporary File Cleanup and Performance Optimization. Also I did a clean install about 3 months ago. The drive is 68/119 GB used. -
Just did file clean up and optimization and got
seq read:506
seq write:372
ran read:1028
ran write: 756
makes improvements but I think Ive busted the thinking of imaging to the ssd and setting AHCI in bios causes significant performance loss.
FYI to see the difference of ssd and hdd I did the benchmark on my storage drive, the stock 7200rpm hdd that came with it and got this
seq read: 105
seq write: 105
ran read: 82
ran write: 100
HUGE difference. If you dont have a ssd GET ONE! -
Tinderbox (UK) BAKED BEAN KING
Sorry not an Alien
830 256gb
seq read:517
seq write:428
Ran read:1036
Ran write:850
EDIT : Random changed to 512KB
EDIT2 : Cloned from original hdd, but deleted all unnecessary partitions.
John. -
Tinderbox. You need to adjust the settings of the bechmark to the ones we used
-
Itd also be useful to know if you ckean installed or imaged. Main reason I made this thread was to verify or debunk the myth the imaging makes an ssd run slower than a clean install. I like to image to keep all my files settings and up to date drivers. Much easier than redoing EVERYTHING.
-
I am seeing about 1000 MB/s read and 800 MB/s write with my SSD's in RAID 0 configuration (clean install). I had similar results with the factory image, except the write was slightly lower. It is really such a miniscule difference that I believe it really isn't worth the time of reinstalling everything again. Now, I'm not saying that fresh installs are a waste of time, because they aren't. But if someone asked me if they should do a fresh install or keep factory state, I would suggest properly utilizing Alienware Respawn, and if they encounter problems - do a fresh install. -
there are couple different things that takes place:
- clean install is only needed to get rid of the unnecessary software that comes bloatloaded into newly purchased systems. It has nothing to deal with HDD or SSD being in use.
- one can clone OS installation from HDD to SSD just fine, however attention has to be given to the alignment of the SSD after the clone is completed. There is free software that will clone and align the partition at the same time (called AOMEI Partition Assistant), however Acronis will produce misaligned cloned image thus further action is needed;
- using misaligned partition shortens the SSD life significantly (twice is the best case), while also produces slower write speeds especially in the 4K writes (the most important for OS partition).
here's screenshots of performance difference between aligned and misaligned cloned partition (via SATA2 interface only):
forum.thinkpads.com • SSD performance on cloned XP Pro, Samsung 830 256GB -
I tried to align my ssd with that program. It went through all the reboot and now my computer wont even get to the windows splash screen..... beware using it. Off to clean install now. Might as well do it now since ive never done it before!
-
tilleroftheearth Wisdom listens quietly...
I am one who will always recommend a clean install.
Not because the 'clean' way is faster, right 'now'. But because the installation stays faster over the lifetime of the system (without needing a 'fresh' cloning or a future 'clean' install' anyways).
I also don't benchmark my system, nor decide which way to configure it based on 'scores'.
This is simply using many systems as-is and seeing that a clean install was less trouble over the lifecycle of the system than the few minutes saved over cloning.
The 'scores' won't tell you if your system is setup optimally.
Only time will tell.
Hope this helps.
Happy Holidays to one and all. -
Ok, well just finished up with the clean install and got rid of the partition for the back up. Updated everything. And got some intresting scores. Maybe a clean install does give a significant boots in ssd performance??
seq read: 517
seq write: 424
ran read: 916
ran write 817
Significant difference to me. -
what is significant is that you cant have random read or write that high, or in other words carefully pick your benchmark software.
use CrystalDiskMark next time. -
The point I was making is there are gains.... regaurdless of what software I used as long as I used the same for before and after. There are gains shown after a clean install over a acronis image. The point of the thread.
-
This is what I am seeing with my SSD's on a factory image restore from HDD to RAID SSD's.
Worse than a clean install? Perhaps slightly. Would I recommend a clean install over a factory restore? Not necessarily, it depends. They are both very legitimate options, it's up to you. -
Why my score is low in Cristal ? And fine Samsung magic software.
-
Wow dre.... does raid make the seq read and write double than just with a single drive? Your scores are about double of the many posted.
-
-
Didnt know that.... so two 128gb n raid0 will outperform a single 256gb of the same ssd?
-
Maybe, but RAID 0 doubles the chance of the drives failing, and when one of them does, you lose everything. Keep that in mind.
samsung 830 SSD benchmark
Discussion in 'Hardware Components and Aftermarket Upgrades' started by tyxpx, Dec 23, 2012.