OCZ Vertex Drives

[babalouj]

Here are my results on Vista with my single Vertex 120gb aligned to 128. I am currently running it as my OS drive and its very very snappy! It is leaps and bounds faster than my WD6400AAKS. Loading times for my games are ridiculously low and there are zero stuttering problems. I can install multiple games/programs at once and still surf the internet will no stalls.

[annihilus]

can you post a crystaldiskmark 1000Mb run?

[babalouj]

Here is my crystalmark. The difference between my ATTO and this crystalmark is that the ATTO was done before I was using the vertex as my OS drive. The crystalmark was done while using the vertex as my OS drive. The drive is currently at 30% capacity.

[annihilus]

vey very nice

I just noticed because I was looking at the 4k results, your sequential writing is lower than the 512k?

[babalouj]

That is probably an artifact of having the OS and program on the disk you are testing.

[Ao1]

Hi babalouj

Would you mind running another test?

This is a test that Marios created. and only takes a couple of minutes. I have attached the batch file as a zip file.

Try this under Vista or Seven
Make a batch flie bnch.bat
Edit
Copy paste the following

winsat disk -ran -read -drive %1 -count 1 -iocount 256 -ransize 16384
winsat disk -ran -read -drive %1 -count 1 -iocount 256 -ransize 32768
winsat disk -ran -read -drive %1 -count 1 -iocount 256 -ransize 65536
winsat disk -ran -read -drive %1 -count 1 -iocount 256 -ransize 131072
winsat disk -ran -read -drive %1 -count 1 -iocount 256 -ransize 262144
winsat disk -ran -read -drive %1 -count 1 -iocount 256 -ransize 524288
winsat disk -ran -read -drive %1 -count 1 -iocount 256 -ransize 1048576
winsat disk -ran -write -drive %1 -count 1 -iocount 256 -ransize 16384
winsat disk -ran -write -drive %1 -count 1 -iocount 256 -ransize 32768
winsat disk -ran -write -drive %1 -count 1 -iocount 256 -ransize 65536
winsat disk -ran -write -drive %1 -count 1 -iocount 256 -ransize 131072
winsat disk -ran -write -drive %1 -count 1 -iocount 256 -ransize 262144
winsat disk -ran -write -drive %1 -count 1 -iocount 256 -ransize 524288
winsat disk -ran -write -drive %1 -count 1 -iocount 256 -ransize 1048576
winsat disk -seq -read -drive %1 -count 1 -iocount 256 -seqsize 65536
winsat disk -seq -read -drive %1 -count 1 -iocount 256 -seqsize 131072
winsat disk -seq -read -drive %1 -count 1 -iocount 256 -seqsize 262144
winsat disk -seq -read -drive %1 -count 1 -iocount 256 -seqsize 524288
winsat disk -seq -read -drive %1 -count 1 -iocount 256 -seqsize 1048576
winsat disk -seq -write -drive %1 -count 1 -iocount 256 -seqsize 65536
winsat disk -seq -write -drive %1 -count 1 -iocount 256 -seqsize 131072
winsat disk -seq -write -drive %1 -count 1 -iocount 256 -seqsize 262144
winsat disk -seq -write -drive %1 -count 1 -iocount 256 -seqsize 524288
winsat disk -seq -write -drive %1 -count 1 -iocount 256 -seqsize 1048576

Save it under c:\user\yourname\
Start run cmd and type "bnch c"
You may change "c" with any partition you want to check.
In case you have a small partition you get better results.
In case you have formated the whole disk as one partition you get reliable results for the whole disk or raid.
Though we may change the parameter -drive %1 with -n %1 (not applied in raid) to check the whole disk or raid %1 no matter how many partitions we have.
In this case we type bnch 0 or bnch 1 to check the first or the second disk respectively
The numbers at the end are IO sizes. I have used the full IO size range, but we may add any IO size we want to check as follows
-seqsize <65536-1048576> bytes
-ransize <16384-1048576> bytes
The iocount set at 256 produces quick results.
-iocount <256-5000>

In a cmd window you might also copy past the following to get results similar to HDTach.
winsat disk -seq -read -drive c -count 1 -iocount 256 -seqsize 65536 -v

[NapalmV5]

^ by short stroking ssds performance drops doesnt increase as on hdds

ive tried it on my ssds.. best to have any ssd @ 100%

marios/anyone else saying smaller ssd partition increases performance.. mind posting some benches?

[Ao1]

^ by short stroking ssds performance drops doesnt increase as on hdds

ive tried it on my ssds.. best to have any ssd @ 100%

Hi NapalmV5, That reference is to HDD on another thread. Marios asked me to try a small partition on a SDD to confirm his suspicion that performance would drop. I tried to find out but could not get my SDD to create a smaller partition.

Out of interest how much of a performance drop did you see? Did you have problems creating a smaller partition on your SSD?

[JRW21]

Here are my results on Vista with my single Vertex 120gb aligned to 128. I am currently running it as my OS drive and its very very snappy! It is leaps and bounds faster than my WD6400AAKS. Loading times for my games are ridiculously low and there are zero stuttering problems. I can install multiple games/programs at once and still surf the internet will no stalls.

Why is the write speed so low?

[lowfat]

Why is the write speed so low?

They aren't bad

Only thing that is sort of weak is the 4k random writes. But the ATTO looks fine.

[Griff805]

Anyone else thinking ATTO isn't very accurate with the Vertex? Results i've seen are all over the place.

I'd really like to see some benchmarks other than ATTO- Anyone? Don't worry, I ain't gonna attack/make fun of any numbers or anything. Just want to see some real results- I'll keep my thoughts to myself....

[Rhys]

Anyone else thinking ATTO isn't very accurate with the Vertex? Results i've seen are all over the place.

I'd really like to see some benchmarks other than ATTO- Anyone? Don't worry, I ain't gonna attack/make fun of any numbers or anything. Just want to see some real results- I'll keep my thoughts to myself....

2x30gb - ICH10R 128K Stripe 128k Alignment, write back cache enabled, No tweaks other than that. Might get a couple more or though the fourth might be capped by the onboard raid, but 3 drives looks to perform great, so may just go with 3.
Attachment 96001

Attachment 96002

[Griff805]

Thanks for posting-

[XS Janus]

Did anyone measure if this "alignment" is beneficial in real world situations? If yes by how much?

The thing is... I feel that if it really shows to be a benefit of alignment then Vertex is once again an SSD product not really intended/suited for the masses as you must once again jump through hoops to run it as best as it can run...

I mean it isnt really an issue if the difference isn't noticeable. But than if we are already at that "its fast enough" point, than why all the Raids and future products... j/k

[Rhys]

Did anyone measure if this "alignment" is beneficial in real world situations? If yes by how much?

The thing is... I feel that if it really shows to be a benefit of alignment then Vertex is once again an SSD product not really intended/suited for the masses as you must once again jump through hoops to run it as best as it can run...

I mean it isnt really an issue if the difference isn't noticeable. But than if we are already at that "its fast enough" point, than why all the Raids and future products... j/k

It should only be mainly beneficial for Xp users as Xp does not align the drive. If you start with a completely fresh drive in vista, vista aligns the partition.

I run XP and tested a single 30gb drive without any alignment, and if I remember correctly atto showed that it had read speeds of 140-150 mb/s as opposed to 190-200 mb/s. I the time I did save the bench marks so i could compare them, but in my excitement to get them up to speed I forgot to transfer them to another disk before I re-partitioned.

I couldn't tell you If they ran a OS without any problems as I didn't install anything on it at that point.

[One_Hertz]

Did anyone measure if this "alignment" is beneficial in real world situations? If yes by how much?

Changes based on what SSD you have. Aligned versus unaligned makes no difference for my X25-Es in sequential rates or in random reads/writes. I haven't done any real world tests though.

[Ao1]

It should only be mainly beneficial for Xp users as Xp does not align the drive. If you start with a completely fresh drive in vista, vista aligns the partition.

Wrong. Tony (OCZ) is saying "you have to align" with Vista 64 as well as disable indexing and prefetch on the Vertex.

http://www.ocztechnologyforum.com/fo...ad.php?t=52462

Anyone else thinking ATTO isn't very accurate with the Vertex? Results I've seen are all over the place.

My guess is results are affected by the implementation, or otherwise, of alignment, OS cache and other tweaks recommended by OCZ.

[Ao1]

Where are all those review sites at?
All I see is lots of trusting people buying their drives blindly and testing ATTO, IOmeter... +one maybe two slim and early reviews...

Still no reviews, but at least now you can see Vertex customer feedback on the OCZ forum.

[Ao1]

I think that our definition of high end differs!
X25-e is bargain bin junk that degrades just as much as the x25-m!
I think they might be included in my favourite cereal pack soon!

The following link is what I think most of us would consider high end or at least high end mainstream! i.e. we might have to save up for it for a month or to!
http://www.gadgetquid.com/storage/te...torage-device/

Edit: Perhaps the X25-e could be considered high end, if a degradation fixing firmware was released to help some of their customers faltering X25's off their crutches

Rhys, I have owned a OCZ SDD. I now own 2 Intel X25-E SDD's. I have run them for two months as my primary OS with none of the OCZ recommended tweaks and have not bothered with alignment. I have not noticed any affects of the reported degradation of the X25-M on my X25-E's despite running synthetic benchmarks. If I had I would not be hiding the fact.

Are you basing your statement on the PC Perspective review? Is so there are a few issues that I have with that review, which is specifically on the X25-M and NOT the X25-E by the way.

First, it is unclear from the review what exactly they mean by a drop in performance in terms of what that actually equates to outside of synthetic benchmarking. If it is a drop in benchmark performance that is not exactly a revelation. If it is a drop in performance in real world scenarios they failed to quantify what that meant in real term use.

It is unclear from the review if the reported drop in performance occurred after benchmarking the drive and continued use afterwards in normal use. Intel SDD's adapt to usage patterns so if they were synthetic benchmarking and then using the drive afterwards that would not be reflective of users experience.

Where did the statement "We will attempt these [fixes] by following Intel’s own advice on the matter, exploring two of their suggested methods." come from? When Intel issued the drives for review they asked reviewers to format the drives after each synthetic benchmark test. Why? Because the drives adapt to user usage patterns that would affect the validity of the tests. Intel have stated that they have not be able to replicate the reported findings in the PC Perspective review, so I somehow doubt they gave this advice to PC Perspective in response to the "problem" they "discovered".

Again I own Intel SDD drives and I have not seen the slightest degradation in real term use, but if I did would it be the end of the world? Not really. I use Windows image back up. It takes me 10 minutes to wipe the drive and restore the image. Defragging a HDD takes ages and admittedly whist you can do other tasks at the same time spending 10 minutes every couple of months to wipe the drive and reinstall the image would hardly seem to be the end of the world.

I can only comment on the X25-E, but perhaps owners of the X25-M could comment on their experience degradation?

[ixtapalapaquetl]

Regarding the 800 MB/s reads on the previous page...

That was my post over at OCZ, and the speed is clearly just a product of the 512k cache on the 5805. I posted the result in order to dispel some early rumors that began surfacing there that Vertex have issues with Adaptec cards - they obviously do not. Here you can see that the Atto result should not be taken at face value:

[WonderSausage]

Wrong. Tony (OCZ) is saying "you have to align" with Vista 64

This means that either A) Tony doesn't know what the fudge he's talking about, or B) The Vertex reports incorrect geometry to Windows Vista.

Vista (GUI or DiskPart) automatically aligns to the drive's reported geometry.

While it's true that XP does not do alignment, that doesn't mean you can't create an aligned partition for XP using Vista WinPE (e.g. download the AIK to create a WinPE bootable solution on optical or USB memory stick).

[One_Hertz]

wow performance degradation seems MUCH heavier on vertex drives than even Intel drives.

http://www.ocztechnologyforum.com/fo...ad.php?t=52526

Two X25-E degraded to the maximum (nearly a month of usage + a bunch of iometer runs):



The sequential writes are between 150-250MB/s, the 512kb writes are pretty stable and the 4kb writes are between 90-120MB/s. The reads are all dead stable.

[Ao1]

wow performance degradation seems MUCH heavier on vertex drives than even Intel drives.

http://www.ocztechnologyforum.com/fo...ad.php?t=52526

Rhys, I've attached HDDErase. Looks like you are going to need it as adding more drives to your array ain't going to help.

[Rhys]

Rhys, I've attached HDDErase. Looks like you are going to need it as adding more drives to your array ain't going to help.

Lol Thanks!
What is special about HDDErase? could I not just delete the raid partition in 'Disk Management' so that it is a RAW drive again, and then use diskpar to rebuild the partition and then image back over the OS (XP).

As you said "Defragging a HDD takes ages and admittedly whist you can do other tasks at the same time spending 10 minutes every couple of months to wipe the drive and reinstall the image would hardly seem to be the end of the world."

My raid array is still running ok so far and it is just over 75% full (so might add another drive anyway even if I'm not after additional performance), ATTO & HD Tach show only a relatively small difference from when the drive only had an OS on it, so I will see how it goes.

I think there is talk of supporting the trim function (will work in windows and Linux) in upcoming firmware releases, so this should help out allot, and the drives performance should see an upgrade also!

Has anyone heard anymore regarding Intel's solution for fragmentation?

[lowfat]

wow performance degradation seems MUCH heavier on vertex drives than even Intel drives.

http://www.ocztechnologyforum.com/fo...ad.php?t=52526

Two X25-E degraded to the maximum (nearly a month of usage + a bunch of iometer runs):



The sequential writes are between 150-250MB/s, the 512kb writes are pretty stable and the 4kb writes are between 90-120MB/s. The reads are all dead stable.

Here is my X25-M after one run of IOmeter. Less than 2 hours after doing a clean install on it too.



Happens everytime I do a large Iometer run. It took me a while to figure out what was causing it.