Overclock.net › Forums › Components › Hard Drives & Storage › SSD › Samsung 840. From 2mb/s to 5000mb/s !!! Confusing.
New Posts  All Forums:Forum Nav:

Samsung 840. From 2mb/s to 5000mb/s !!! Confusing.

post #1 of 14
Thread Starter 
Hello. I recently got my first SSD (Samsung 840 evo 250GB) for my Toshiba P70-A-P11Q Notebook. Unfortunately there are my things that got me confused.
I will try to be as brief as possible:

I installed Windows 8.1 on the SSD as also the Samsung Magician Tool and enabled RAPID. Please note the Laptop supports AHCI and the drive runs on SATA-3.

After i did this configuration i run for the first time some benchmarks. Those are the numbers that i got.


Thinking that my speed was low. I installed the Intel Rapid Storage Technology Driver 12.8.2.1000. Rebooted and did some further testing. Please note here, that i am not 100% sure that those results are because of the Intel driver. As you see i got 2mb/s write on 4k.



I removed the Intel drivers and tried again a benchmark. This time i got the exactly opposite strange results. Suddenly i was seeing numbers like 5000mb/s on some benchmark apps. Furthermore if you see my first Benchmark from AS SSD Benchmark with the last one you will see a huge difference on almost all the sections. There is no relation anymore. I got a huge write increase but also a huge read decrease rate.




I am totally confused. Please let me know following:

1) Do i really need Intel Rapid driver? If yes is there any specific version i should download ?
2) Which benchmark should i trust? How do i know after all what are my true data rates
3) Is there something i could check, test to get some stable results ? Right now i get different numbers on every run.
4) Can you explain the above behavior ?

Thank you in advance.


PS. a small update. I just run again AS SSD and got now 1300mb/s on 4k-64trd. That more than double as much that i had a few minutes ago. There is surely something wrong with the numbers. Right ?
Edited by Alarm - 3/3/14 at 3:33pm
post #2 of 14
I think you would normally want the Intel RST installed but not 100% on that.

Definitely turn off RAPID if you're trying to get a good idea of drive performance. RAPID uses some of your system RAM for caching read and writes and it's the reason for the really high numbers.
post #3 of 14
Thread Starter 
Good morning.
First of all thank you for the reply.

Ok i could install the RST driver, the quesiton then is which version? The one i did installed seems to give really bad results.
Why should rapid reduce performance ?
post #4 of 14
Quote:
Originally Posted by Alarm View Post

Good morning.
First of all thank you for the reply.

Ok i could install the RST driver, the quesiton then is which version? The one i did installed seems to give really bad results.
Why should rapid reduce performance ?

I would usually go with the latest version but I haven't really looked into it.

RAPID increases performance but you want to disable RAPID for now because you won't be able to get accurate benchmarks of the drive's performance with it turned on. Once you're done troubleshooting the drive then you can turn RAPID back on.
post #5 of 14
Thread Starter 
Alright. Thank you for the tip smile.gif
post #6 of 14
Thread Starter 
So i have installed the latest Intel RST drivers, disabled RAPID (Magician) and rerun the test.

Does this look any better, assuming the drive runs on Sata-3 interface ?

post #7 of 14
Those speeds look like the drive is connected to a SATA 2.0 port. What was the model of the laptop again? Searching for P70-A-P11Q just brings up this thread.

Here are the results from my 500GB EVO with RAPID disabled:

post #8 of 14
First as stated above, the model number for your laptop in your first post is wrong, so we can't check its hardware.

The order of your benchmark screen shots is confusing, as well as the drivers you said you used. Your first AS SSD test shows you are using an Intel Rapid Storage Tech (IRST) driver, but you then said you installed an IRST driver, (12.8), or was that just a different version?

Next, in order to really check the true performance of your EVO, you must disable RAPID. As you have seen, RAPID can be wildly erratic, and your results are the most erratic I've ever seen. When you run benchmarks, you should always connect that PC to AC power, if you test with just the battery you will get different (and lower) results. The very different results are most likely seen because it is a mobile PC.

In your first post, you said your SSD "runs on SATA -3". It is true the EVO is a SATA III capable SSD, but your non-RAPID results are clearly SATA II, as Aibohphobia said.

We can't check your notebook since we don't know the model number, but even if it has a SATA III capable chipset, it is not uncommon for manufactures to only use the SATA II ports on the chipset. If your notebook has more than one connection for drives, then the other one might be a SATA III connection. But don't be shocked if all the SATA connections are SATA II, I've seen that many times with mobile PCs. That is done supposedly for "compatibility" reasons, or to use less power, but it is also easier (and cheaper) to validate the PC's functioning using the SATA II ports rather than SATA III. The manufacture might also not use the SATA III ports on a lower priced model, just to separate them from the more expensive models.

Did you run the first few AS SSD tests something like this; Run AS SSD with RAPID enabled. When complete, immediately disable RAPID and run AS SSD again. If so, that could explain the poor write performance in the second AS SSD test you posted. I can explain that later if what I wrote is the case.
post #9 of 14
Thread Starter 
Toshiba P70-A-Q11 is the model as i mentioned already thumb.gif
post #10 of 14
Thread Starter 
parsec, first of all thank you very much for taking the time to give such a complete answer.
Quote:
Originally Posted by parsec View Post

First as stated above, the model number for your laptop in your first post is wrong, so we can't check its hardware.

The model of the laptop is correct. Here is the link to the specs. http://www.toshiba.ch/de/laptops/satellite/satellite-p70-a/satellite-p70-a-11q/ . From what i know from Toshia, the entier P70-A share the same chipset. Unfortunately Toshiba doesn't give any info about SATA. Neither on the Swiss page nor on any other international page.
Quote:
Originally Posted by parsec View Post

The order of your benchmark screen shots is confusing, as well as the drivers you said you used. Your first AS SSD test shows you are using an Intel Rapid Storage Tech (IRST) driver, but you then said you installed an IRST driver, (12.8), or was that just a different version?

As for the tests made, i hope the following list below helps a bit.
1st test: Rapid enabled + RST drivers that windows 8 installed (if there were any. i am not sure about this) (Screen 1)
2nd test: Rapid enabled + RST version 12.8.2.1000 (Screen 2)
3rd test: Rapid enabed + No rst drivers (Screen 3)
4th test Rapid disabled + Latest RST drivers (Screen 4) .
Quote:
Originally Posted by parsec View Post

Next, in order to really check the true performance of your EVO, you must disable RAPID. As you have seen, RAPID can be wildly erratic, and your results are the most erratic I've ever seen. When you run benchmarks, you should always connect that PC to AC power, if you test with just the battery you will get different (and lower) results. The very different results are most likely seen because it is a mobile PC.


On the 4th run rapid was disabled , latest RST drivers and yes the laptop was on the AC thumb.gif
Quote:
Originally Posted by parsec View Post

In your first post, you said your SSD "runs on SATA -3". It is true the EVO is a SATA III capable SSD, but your non-RAPID results are clearly SATA II, as Aibohphobia said.

On this i am really confused. I am not 100% sure if the drive is on a SATA 2 or SATA 3 interface.
I have downloaded several apps, but all of them show different results.
For example: Samsung magician shows that the drive runs on SATA3. Intel RST shows that the drive runs on SATA2.
I am just giving the links of the images so that i won't overflood the topic with images. On each of them you will see a different SATA speed.

http://i1252.photobucket.com/albums/hh565/rockreactor/Clipboard04_zps62d0e576.jpg
http://i1252.photobucket.com/albums/hh565/rockreactor/Clipboard02_zps2b6a381b.jpg
http://i1252.photobucket.com/albums/hh565/rockreactor/Clipboard02_zps2b6a381b.jpg
http://i1252.photobucket.com/albums/hh565/rockreactor/Clipboard04_zps5d3d2f1d.jpg
http://i1252.photobucket.com/albums/hh565/rockreactor/Clipboard02_zps59625bc3.jpg

My thought was to put my SSD on caddy 1 on the laptop and the HDD on caddy 2 (swapping them)
Quote:
Originally Posted by parsec View Post

We can't check your notebook since we don't know the model number, but even if it has a SATA III capable chipset, it is not uncommon for manufactures to only use the SATA II ports on the chipset. If your notebook has more than one connection for drives, then the other one might be a SATA III connection. But don't be shocked if all the SATA connections are SATA II, I've seen that many times with mobile PCs. That is done supposedly for "compatibility" reasons, or to use less power, but it is also easier (and cheaper) to validate the PC's functioning using the SATA II ports rather than SATA III. The manufacture might also not use the SATA III ports on a lower priced model, just to separate them from the more expensive models.

Did you run the first few AS SSD tests something like this; Run AS SSD with RAPID enabled. When complete, immediately disable RAPID and run AS SSD again. If so, that could explain the poor write performance in the second AS SSD test you posted. I can explain that later if what I wrote is the case.

I am sorry i did not understand your question very well about the RAPID and AS SSD.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: SSD
Overclock.net › Forums › Components › Hard Drives & Storage › SSD › Samsung 840. From 2mb/s to 5000mb/s !!! Confusing.