Crucial m4 128GB vs OCZ Vertex 3 MaxIOPs 120GB

[Ao1]

I'd like to have the option to select my own "life-curve" on the SF drives, I'd rather use the drives for 2-3 years with "unlimited" writes than being hindered by "life- preserving" limits that go beyond the useful life of the drive.
I'd say 2-3 years is sufficient for me, by that time there would be better alternatives with better performance, larger capacity,..., well most things improve over a 2-3 year period.

Exactly. Who is going to care after three years anyway? Better to have an off switch. (Although that does not address why one is needed in the 1st place).

EDIT; maybe it is something to do with the amount of OP?

[Antioch]

Hmm, interesting discussion. I do agree with the comments about drive life and usefulness in it's prime.

I feel that I'm leaning heavily towards a single m4 128Gb right now, though.

[Anvil]

The 128GB m4 is just superb, I like it a lot.

Here's an AS SSD comparison of my m4's.

forgot: 256GB to the left and 128GB to the right.

as_ssd_mbps_128vs256.JPG
as_ssd_iops_128vs256.JPG

Cant speak for the 120GB SF drives but the V3 240GB is just marvellous

[vafan13]

curious that the M4 beats the vertex 3 in the "2010 storage bench" , yet anand switches to a new bench the "2011 storage bench" and the V3 wins handily. hard to tell these days. we need more of these drives in the hands of us enthusiasts to come to our own conclusions

The 2010 bench doesn't use SATA III, 2011 bench does. It was shown earlier in the review that the V3 is hit much more by dropping down to SATA II than the m4, probably because more of its performance comes from sequentials.

[bluestang]

I feel that I'm leaning heavily towards a single m4 128Gb right now, though.

The 128GB m4 is just superb, I like it a lot.

Yeah me too...but I can get 2x64GB M4's cheaper than a single 128GB M4. I'm torn Wonder how the 64's in R0 compare to the single 128?

[Antioch]

Yeah me too...but I can get 2x64GB M4's cheaper than a single 128GB M4. I'm torn Wonder how the 64's in R0 compare to the single 128?

I don't know but I can only keep pointing you back here:
http://www.xtremesystems.org/forums/...=268711&page=8
Where TheHardCase says that 2x m4 64GB in RAID0 is bad.

Maybe you should pester him about that - he may have had that setup for a while.

[Anvil]

In real-life, not much of a difference I'd expect, with the single drive you'd get TRIM and TRIM can do wonders, so, unless there are special needs for raid the single drive would be the better option overall imho.

I suspect that the m4 in raid *might* need more cleaning than the C300, just a hunch from a small number of tests, could be wrong though
*if* that was to be the case it could be corrected in the next firmware update.

[cherrypik]

Would I notice a difference between a single C300 128gb vs M4 128gb?

[Antioch]

Would I notice a difference between a single C300 128gb vs M4 128gb?

I don't think so. The C300s have higher random IO speeds but the m4s have higher sequential throughput.

If you don't have a Sandy Bridge system to put the m4 in (meaning a native Sata3 controller) then the C300 is going to be faster in general usage. With a Sata3 controller the higher sequential reads from the m4 begin to come into play -- but again it's a balancing act between random vs sequential.

[Computurd]

i was under the understanding, and i know over at OCZ forums this is common knowledge, that the previous generation V2s were hit by write throttling after each and every sector on the drive had been written once. that once you had hit each piece of nand in its entirety, that the drive would throttle itself, permanently. Then the only recourse is secure erase to reset the drive.

The 2010 bench doesn't use SATA III, 2011 bench does. It was shown earlier in the review that the V3 is hit much more by dropping down to SATA II than the m4, probably because more of its performance comes from sequentials.

yes, but the bench itself has changed radically. There are so many factors involved, compressible v incompressible, random v sequential, that you could effectively alter it to go either way tbh. curious of if he ran the new bench on sata 2, i would bet money the V3 comes out on top with the new bench regardless of what you run it on. im not saying that is intentional, just that is my impression. The M4 goes from a very solid showing in the old bench, to suddenly getting trounced by 2x in many areas....nah. doesnt add up. methodology may be flawed here a bit imo.

[cherrypik]

I don't think so. The C300s have higher random IO speeds but the m4s have higher sequential throughput.

If you don't have a Sandy Bridge system to put the m4 in (meaning a native Sata3 controller) then the C300 is going to be faster in general usage. With a Sata3 controller the higher sequential reads from the m4 begin to come into play -- but again it's a balancing act between random vs sequential.

Actually this would be for a sandy bridge desktop.

[TheHardCase]

I just picked up 2 x M4 128GB & set them up in Raid 0 128K
Now I can say these are 100% better then M4's 64GB in Raid 0 128k.
What a big difference almost double the speed.

[Antioch]

I just picked up 2 x M4 128GB & set them up in Raid 0 128K
Now I can say these are 100% better then M4's 64GB in Raid 0 128k.
What a big difference almost double the speed.

And do you have the benchmarks for the 64GB RAID0 setup? I'd curious to see the actual numbers.

[Mech0z]

Meh still no solid info about the MAX IOPS

[Ao1]

I found a bit of info on throttling in a SF patent. A range of measures in detecting overuse and then implementing throttling are described. Commands that impact lifetime are delayed to guarantee a required minimum life. Not sure if GC is one of those commands but if GC is deactivated it would explain why some drives appear to be unrecoverable.

Looks like the 2000 series controllers "have more options in the firmware for partners to set that allow throttling...."

There are also non-standard SATA commands that can set the drive to go into lower power modes.

SMART attribute 230 indicates if “lifetime throttling was active for last write” .

Seems that not all SF partners set the drive to throttle. (Don't know which ones don't)

[Andrea deluxe]

any test with newest vertex3 firmware?

[Nizzen]

Meh still no solid info about the MAX IOPS

So WHAT info do you want?

[Ao1]

I found a better diagram for the life curve. "Credits" = short term unsustainable write speeds that if sustained result in throttling.

www.storagesearch.com/sandforce-sf2000-1.pdf

[Mech0z]

So WHAT info do you want?

I want to see the 128GB M4 and the 120 IO in a same test scenarios not just random peoples ATTO and crystalmark as those are normally not accuarate to compare as some have more data on drives and setup might be different as well. (Unless its done on the same machine, I just dont want to compare 2 different peoples benchmarks where one person has 5% free space while the other has 80% as its not a fair comparision)

[bluestang]

So WHAT info do you want?

AS SSD MB/s and IOPs and PCMark HDD Suite...Thanks!

[Tony]

post made on the 19th Feb 2011

http://www.ocztechnologyforum.com/fo...l=1#post604816

you guys need to read more

Life curve throttle is set with the mptool, I told the world this 12 months ago (although you did have to read between the lines)

SF drives do NOT throttle for settled state, that is the normal running state and is unthrottled. The fact that an unmapped drive is faster is irrelevant to me as that lasts just a few days....marketing unfortunately has to focus on this though as it would only take 1 competitor to go against the grain and people start to think X is faster than Y etc etc.

All drives will now have active throttle, all controllers can and will do it as there are just not enough PE/c in nand to allow it to be not be set.

If you want to write more than an average desktop users daily amount to a drive you either have to set more OP, raid and add more OP to spread the load or buy enterprise drives or at the least SLC drives.

its that simple.

What life curve throttle is set to all depends on warranty period.. PE/c and how the drive is used...

sit their benching your drives and you will wear them out

Use them like normal, save downloads and games to a spinner and your ssd OS drive will last for years.

Throttle is here to stay, find ways to bypass it and you will only kill the drive MUCH FASTER. Drive logs catch full drive erases, data written etc etc, all controllers will do this. You would be amazed what the logs can tell about users habits with the drives.

The short of it is buy the drives that fit your needs, do not abuse the drives...just use them.

[Computurd]

raid and add more OP to spread the load

good point.

.marketing unfortunately has to focus on this though as it would only take 1 competitor to go against the grain and people start to think X is faster than Y etc etc.

unfortunate but true. they need a industry standard for testing, but doubt we will ever see it. i dont think there is an industry standard for testing of anything, from monitors to hdd to ssd to lawnmowers

buy the drives that fit your needs, do not abuse the drives...just use them

that should be written on the front of all SSD packages

[johnw]

Is it so difficult for OCZ to make a clear statement?

1) We do support disabling throttling on these Sandforce models: ....

2) We do NOT support disabling throttling on our Sandforce SSDs. However, it can be done with this tool (link) although you will lose warranty support (or whatever).

3) We do NOT support disabling throttling and will not provide a link to a tool to do it.

[Ao1]

Or 4)
We will sell you an un-throttled SSD, but the warranty is based on "X" amount of writes and not a period of time.

[Tony]

is it so difficult for ocz to state what?

The drives have to throttle, simple fact, without it you would have nand burn in weeks if a few months...then you try and RMA??? LOL

Except the product is as it is, Duraclass and life write throttle is needed. If you want faster lobby the nand manufacturers to get 100K PE/c on cheaper MLC and we will build drives just how you want them.

The nand dictates how the drives work remember

Guys...you want faster here is how to do it.

Raid 0

lose 50% of the array to OP

live with the day to day speed...the drive will last for years

dead simple solution