What's wrong with sandforce?

[groberts101]

I have seen with my own eyes shills for OCZ on amazon.com reviews, and judging by the shrillness of support for OCZ and vague disparagement (as well as ludicrous disparagement of the M4, which has quite acceptable random 4k speeds, which are what really matter, not some stupid compressible sequential speeds) here I suspect there are less-than-unbiased people here as well. Also, at least one person in this thread is an OCZ SSD tester and has openly stated as much.

I wouldn't even begin to dispute that first part.. but on the other hand.. if you look hard enough at any site(ehh hmm)?.. you'll always find a few of those types lurking around.

As to the second part which insinuates that I'm overly biased because I get drives to test for hours at a time(much of which turns into "actual work" because it involves tearing apart various systems and software installs to do it.. not to mention that I don't really have that time to spare in the first place)?.. you'd be dead wrong.

Now, of course I can't come in here and crap all over the company that I test for on a repeated basis just to lower guards.. or solidify some of the points being made in the dozens of "hate spun" threads like this one(some of which quite obviously have some merit behind them).

But that means squat when it comes to overviewing the Sandforce controller's as a separate entity all unto itself. The very fact that I've used and tested dozens of these particular drives from MANY manufacturers and studied most, if not all, of the other mfgrs offerings, along with their owners and issues along the way. Having purchased and returned more than $10,000 worth of SSD's(I actually have receipts) does in fact give me perspectives for this industry that most here will never understand.

Hell.. I've even gone so far to help troubleshoot issues with other mfgrs drives on other mfgrs forums for that matter! Think about that for one millisecond and ask what that says about someone who would go out of their way to do something like that? I'm certainly gracious for what opportunities I've been given and try to pay it forward whenever possible.. but I've earned that right and respect with far more time and energy into these things than many here combined.

And.. and.. and.. I'm certainly not as easily bought and paid for as some would like to believe. Far from it actually and I'm starting to get a tad bit more pissed off when people start to continually insinuate that I'm not being genuine with my input when I decide to take the time to share it. Sure.. add all the salt you want to.. I know I do since it's the internet and motives are tough to read through simple text.. but don't always assume that you've got me pegged and dismiss my credibility so easily. Although I may not be able to fill in all the blanks here and there in regards to this particular "flamebait".. I do know that firsthand knowledge trumps speculation every time. If I really wanted a job with them?.. I'd just go towards that end and it would become quite obvious where my loyalties LIE.. pun intended with that last word. lol

The biggest problem I see with many folks around here, aside from senselessly venting their life's frustrations while anonymously hiding behind a keyboard, is that they get too easily caught up in believing what they read in a few thousand words and quickly decide to buy into what they "LIKE to believe" without having enough facts to formulate viable opinions on a matter. Because of that.. they are often quite simply "products created by the internet". Welcome to the internet boys and girls. :hmm:

 

[Ao1]

And how many do you have ?

Because I have one. I also have 17 others. My experience is different than yours.

Vertex 4 and 3 and also Samsung 830 " feel" faster. My Toggle NAND Sandforce also feels faster. I also have dual sandforce in RAID0. My next drive will be an 840 pro or vector.

The crucial "feels" faster than only the crappy agility 3 I have in my 20TB WHS Flexraid server.

Yes I've benched them.

I have most brands of SSD and I spend a lot of time looking at real world I/O performance and comparing synthetic benchmarks. Sure the M4 does not bench as well as other SSD's, but for real world I/O, which is a world away from synthetic benchmarks I can assure you the M4 is an exceptional performer when it comes to read I/0's.

 

[groberts101]

I have most brands of SSD and I spend a lot of time looking at real world I/O performance and comparing synthetic benchmarks. Sure the M4 does not bench as well as other SSD's, but for real world I/O, which is a world away from synthetic benchmarks I can assure you the M4 is an exceptional performer when it comes to read I/0's.

Heck.. even if I did work for OCZ.. there's no doubt in my mind whatsoever!

I bought and returned one myself and also installed on in my nephews system.. and it would still be well seated in my "top 10 best drives" to own even by todays standards.

 

[Ao1]

I'd say its in the top three for read performance along with the 840Pro and M5Pro (256GB). I don't have a Vector so I can't comment on that drive, but benchmark speeds don't always reflect real world performance, simply because you are testing two different things.

 

[groberts101]

I'd say its in the top three for read performance along with the 840Pro and M5Pro (256GB). I don't have a Vector so I can't comment on that drive, but benchmark speeds don't always reflect real world performance, simply because you are testing two different things.

true.. true.

Personally speaking.. I would kill to have an SSD with the reads of those drives combined with the writes of the Vector. But.. I tend to write the hell out of my drives while I multitask so I can only imagine(not having personally tested either of those other 2) that the Vector "maybe/might/could" fit my particular workflow better than the others you mentioned there.

TBH.. at that level of performance?.. it's all about brand, price, and/or availability while splitting extremely thin hairs. I wouldn't kick any of them out of my systems.. that's for sure.

 

[taltamir]

Ok. I'm going to try to remove your OCZ bias and hatred for sake of making my point. Lets assume OCZ doesn't exist anymore.

What OCZ bias? He posted a chart from an independent source showing the so called "slow as a pig" drive getting 516MB/s compared to the fastest drive getting a whopping 547MB/s and the fastest OCZ 529MB/s

Posting a benchmark from a third party is the very definition of not biased.

 

[jwilliams4200]

What OCZ bias?

I think he means that anyone who does not agree with him is biased.

 

[Concillian]

Early life-cycle results will have significant long term effect, and Sandforce early life-cycle results were ABYSMAL, with some early OCZ implementations seeing 20+% return rates.

Even if issues are resolved, the damage this does to reputation is irrevocable.

For an example of this, look no further than Windows Vista. At launch it had issues... primarily poor 64 bit driver support, but also was a memory hog, among other things. This resulted in a pretty damaging launch. At Service Pack 1, MS fixed most of the inherent issues, and hardware companies had largely fixed 64 bit driver issues, resulting in an OS that is pretty much the same thing as Win7 under the hood... but it was too late word was out that Vista was junk and reversing that public opinion would be an uphill battle.

Hence the release of Vista 2.0, or Win 7.

Regardless of current Sandforce reliability, public opinion has been formed. You will hear bad things about Sandforce at least until they release their next controller.

I have no knowledge of current reliability, but I'd venture to guess that the issues have mostly been resolved.

 

[Coup27]

I think he means that anyone who does not agree with him is biased.

Unfortunately the general tone of this thread is that anybody who doesn't like OCZ or doesn't rate OCZ's products is branded as biased. This is even after they have linked to independent sources confirming failure rates among OCZ products or independent sources who have carried out speed tests.

I have no doubt that there is enough information out there to write a 10,000 word thesis on the subject of "SSDs and OCZ" but brushing past all of that including the monumental blame game which goes with it, the blunt reality is OCZ's SSDs fail a lot more than their competitors.

OCZ's first generation of SSDs using JMicron were not fit for sale and it wasn't until huge public weight had amassed that they finally accepted and took action. Intel's X25-M was solid.

Both generations of SandForce SSD's have been hugely problematic while in comparison Samsung's SSD's have remained virtually flawless. The m4 has had bugs, all of which have been promptly corrected. Intel have blotted their copy book with the unresolved 8MB bug.

Then we had the Petrol and Octane, more terribly unreliable drives according to information in this thread. Personally I was not aware of this but it does not surprise either.

This very brief history doesn't even mention the switching of NAND process node without warning or publication and only doing anything about it after pressure from the media or the blatant misleading over the origins of the first Barefoot controller. I cannot remember which SSD it was but they also released an SSD which could only be updated when running as a slave drive in a system without Intel's RST installed after it was released long before the toolbox and OS independent firmware update ISO's were ready. Er, what?

I am sure there is far more as well but my memory isn't what it used to be but I genuinely cannot understand how anybody can defend OCZ's reputation, product quality or business ethics to this day.

 

[Zxian]

Both generations of SandForce SSD's have been hugely problematic while in comparison Samsung's SSD's have remained virtually flawless. The m4 has had bugs, all of which have been promptly corrected. Intel have blotted their copy book with the unresolved 8MB bug.

This is the only statement from your entire post that is actually on topic. It's just proof in the pudding that this thread has degenerated so far off from the original topic of discussion.

 

[groberts101]

I think he means that anyone who does not agree with him is biased.

stones.. glass houses.. and all that.

statistical facts.. or not... like them.. or not.. OCZ helped this industry push forward at faster pace just from the sheer volume of sales, competitive strategies, and overall availability more quickly created from pushing to hit shelves first.

Of course, some might say that OCZ has stunted this industries growth and we've all paid the price as we're stuck back peddling to fix the early release issues created from shoving them out the door so quickly to gain those early leads in market share too. Now, there might actually be some merit to that point too.. but I do know that people still stood in line in front of me and bought multiple drives from them regardless of all that's been said around here over the years.

Fortunately, there were other companies that also shoved similarly controlled products out their doors and had those same issues to easily dismiss those claims to some extent.

So at the end of the day.. where would we all be right now with this tech's offering if we were all waiting around for Intel to validate a controller that was over a year old and already outdated by the time "they fixed all the bugs"?

Then let's wonder where our drives current level of firmware evolution would be?

And.. how much would you guys be paying for these so called perfect drives had OCZ not lowered their bottom lines and forced others to do the same to stay competitive along the way?

While I'm certainly no statistical engineer like some here.. I'm guessing that you'd not be looking at 3rd.. or even 4th gen controllers that achieve excellent steady state performance and paying .50 per GB right now. That's where.

 

[jwilliams4200]

.. I'm guessing...

That is the main source of the problem right there.

Some people prefer to base their conclusions on real data rather than guesses.

 

[groberts101]

That is the main source of the problem right there.

Some people prefer to base their conclusions on real data rather than guesses.

and there's proof right there that the "other side" shows their ass around here too.

real data from France, the internet, and XS?

So, get off that big ass horse of yours and give us a break already! You just study slices of the pie, read forums, and come up WAGS just the same as everyone else does. I have no doubt that you're a sharp guy.. but the sheer level of condescension you often exude proceeds you and ultimately undermines you credibility and many of the points you try to get across. :thumbsdown:

 

[jwilliams4200]

You just study slices of the pie, read forums, and come up WAGS just the same as everyone else does. :thumbsdown:

Just because you do not understand basic probability and statistics does not mean that others in the discussion do not. You can claim that conclusions of a statistical study with valid methodology and sufficient sample sizes are just guesses, but without pointing out specific logical or methodological flaws in the study, such a claim is worthless.

 

[cytg111]

The biggest problem I see with many folks around here, aside from senselessly venting their life's frustrations while anonymously hiding behind a keyboard, is that they get too easily caught up in believing what they read in a few thousand words and quickly decide to buy into what they "LIKE to believe" without having enough facts to formulate viable opinions on a matter. Because of that.. they are often quite simply "products created by the internet". Welcome to the internet boys and girls.

..

.. buy into what they "LIKE to believe" without having enough facts to formulate viable opinions on a matter.

- Like Anand for example? Put all that, other reviewsites and personal experiences from the forums together, now add that to the vertex2 "lets change the nand to this newer cheaper shite, change nothing else and hope noone will notice" and tell me what it equals.
I want to put a spoiler here, but .. resisting ...

 

[Coup27]

This is the only statement from your entire post that is actually on topic.

Are you having a laugh? I joined this thread on page 4, by page 1 it had already descended away from the actual topic and onto an OCZ discussion thread. If you are unhappy with the thread deviation, you need to look at the posters long before I joined in.

 

[Hellhammer]

So at the end of the day.. where would we all be right now with this tech's offering if we were all waiting around for Intel to validate a controller that was over a year old and already outdated by the time "they fixed all the bugs"?

Intel isn't the only SandForce OEM... While OCZ had an exclusivity with the SF-2281 at first, many drives based on the same controller were released within months after the Vertex 3. Intel also released a Marvell SSD 510 in the meantime, which is partially the reason why SSD 520 was released so late.

And.. how much would you guys be paying for these so called perfect drives had OCZ not lowered their bottom lines and forced others to do the same to stay competitive along the way?

At the ground level, OCZ does not control the prices. They buy controllers, NAND and DRAM from thirds parties, which means their pricing is subject to the industry prices (unless they are ready to make loss). One of the biggest reasons why SSD prices have fallen so much is the overproduction of NAND, which has forced manufacturers to lower the prices in order to sell their inventory.

Sure OCZ can have some control in prices if they buy in huge volumes (= drives the price down) but the real control is still in the hands of actual NAND manufacturers. There are also plenty of other SSD OEMs that compete in price, some which offer lower prices than OCZ (e.g. Mushkin's SSDs have often been cheaper).

While I'm certainly no statistical engineer like some here.. I'm guessing that you'd not be looking at 3rd.. or even 4th gen controllers that achieve excellent steady state performance and paying .50 per GB right now. That's where.

OCZ didn't really do anything to the controller industry until Vertex 4. Before the V4 all their SSDs were based on third party controllers and firmware. I don't doubt that OCZ has had their say about the SF-2281 and its firmware for instance, but again OCZ hasn't been the one in control.

In my defense, I'm not trying to badmouth OCZ, not at all. I just don't get the point about OCZ being magical in some way when it comes to SSDs. They have done plenty of good to the consumer SSD industry but they are just one among the many factors.

 

[groberts101]

In my defense, I'm not trying to badmouth OCZ, not at all. I just don't get the point about OCZ being magical in some way when it comes to SSDs. They have done plenty of good to the consumer SSD industry but they are just one among the many factors.

Surely no need for any defense as you seemingly have the ability to not tear others down while you work to get your points across.

My main point was that OCZ "HELPED" to prod the industry into a supply and demand price war or sorts. Pure nand availability alone will not do that since a company still has to package it up with a processor/firmware and put their sticker on it. Which aside from the self inflicted scars and haters created from it.. had an overall positive impact across the board and helped push others to develop faster/better models through the extra competition they created by pushing 12 different models onto shelves.

Most of us know full well that the market is not open ended for this tech just yet( I still meet many who've never even heard of an SSD) and most mfgrs still have to via for their share of the available market to continue keeping the stockholders happy. Some just aren't as multifaceted as the big boys are across the various industries and have more eggs in this particular basket, is all.

 

[blastingcap]

I'm not saying you are intentionally or unintentionally biased, sorry for lack of clarity in that regard. I am noting that there is a conflict of interest if you were to slam OCZ, which you agree is true. Personally I find your posts interesting but it needed to be mentioned, sort of like how IntelEnthusiast and Keysplayer publicly state their affiliations in their sigs. Nobody is saying that they are intentionally biased, but the possibility exists hence the public statement.

I wouldn't even begin to dispute that first part.. but on the other hand.. if you look hard enough at any site(ehh hmm)?.. you'll always find a few of those types lurking around.

As to the second part which insinuates that I'm overly biased because I get drives to test for hours at a time(much of which turns into "actual work" because it involves tearing apart various systems and software installs to do it.. not to mention that I don't really have that time to spare in the first place)?.. you'd be dead wrong.

Now, of course I can't come in here and crap all over the company that I test for on a repeated basis just to lower guards.. or solidify some of the points being made in the dozens of "hate spun" threads like this one(some of which quite obviously have some merit behind them).

But that means squat when it comes to overviewing the Sandforce controller's as a separate entity all unto itself. The very fact that I've used and tested dozens of these particular drives from MANY manufacturers and studied most, if not all, of the other mfgrs offerings, along with their owners and issues along the way. Having purchased and returned more than $10,000 worth of SSD's(I actually have receipts) does in fact give me perspectives for this industry that most here will never understand.

Hell.. I've even gone so far to help troubleshoot issues with other mfgrs drives on other mfgrs forums for that matter! Think about that for one millisecond and ask what that says about someone who would go out of their way to do something like that? I'm certainly gracious for what opportunities I've been given and try to pay it forward whenever possible.. but I've earned that right and respect with far more time and energy into these things than many here combined.

And.. and.. and.. I'm certainly not as easily bought and paid for as some would like to believe. Far from it actually and I'm starting to get a tad bit more pissed off when people start to continually insinuate that I'm not being genuine with my input when I decide to take the time to share it. Sure.. add all the salt you want to.. I know I do since it's the internet and motives are tough to read through simple text.. but don't always assume that you've got me pegged and dismiss my credibility so easily. Although I may not be able to fill in all the blanks here and there in regards to this particular "flamebait".. I do know that firsthand knowledge trumps speculation every time. If I really wanted a job with them?.. I'd just go towards that end and it would become quite obvious where my loyalties LIE.. pun intended with that last word. lol

The biggest problem I see with many folks around here, aside from senselessly venting their life's frustrations while anonymously hiding behind a keyboard, is that they get too easily caught up in believing what they read in a few thousand words and quickly decide to buy into what they "LIKE to believe" without having enough facts to formulate viable opinions on a matter. Because of that.. they are often quite simply "products created by the internet". Welcome to the internet boys and girls. :hmm:

 

[groberts101]

That is the main source of the problem right there.

Some people prefer to base their conclusions on real data rather than guesses.

ooh.. ooh.. this one's better.

I could surely say the same thing about your posts here and elesewhere.

The simple fact that you continually pull fragments out of others posts and spin them to your liking really does show that there's a "perfect tool" for every job.

And.. the statistical engineer side of you must really thank god that France's vendors keep track of mfgr specific returns, eh? Because without it.. you'd have to go back to digging/linking OCZ's support site. :sneaky:

 

[Coup27]

Personally I find your posts interesting but it needed to be mentioned, sort of like how IntelEnthusiast and Keysplayer publicly state their affiliations in their sigs.

This is a good point and also something I have thought myself. Please don't take this the wrong way as I certainly have nothing against you personally but I do think that as you are such an active contributor to the SSD forum that you should indicate in your signature that you are an OCZ beta tester. It at least avoids future issues where somebody finds out later down the line and retrospectively thinks the advice you have given may not have been down the middle. At least if you make this clear to all, there can be no animosity further down the line.

 

[groberts101]

I'm not saying you are intentionally or unintentionally biased, sorry for lack of clarity in that regard. I am noting that there is a conflict of interest if you were to slam OCZ, which you agree is true. Personally I find your posts interesting but it needed to be mentioned, sort of like how IntelEnthusiast and Keysplayer publicly state their affiliations in their sigs. Nobody is saying that they are intentionally biased, but the possibility exists hence the public statement.

okey dokey, then. No offense taken from you.. or intended to you.. if it's not intentionally used to somehow discredit my opinions on the matter. I just figure that I earned my Sandforce stripes the hard way and I have no issues or shortage of pride to flash them in people's face when I get a hair up there.

 

[groberts101]

This is a good point and also something I have thought myself. Please don't take this the wrong way as I certainly have nothing against you personally but I do think that as you are such an active contributor to the SSD forum that you should indicate in your signature that you are an OCZ beta tester. It at least avoids future issues where somebody finds out later down the line and retrospectively thinks the advice you have given may not have been down the middle. At least if you make this clear to all, there can be no animosity further down the line.

Nope.. not really too quick to misinterpret things based on the mood I'm in. Though I do admit to being human. And married with teenage sons. LOL

I look at it like this Coup. If I should be required to have sig's indicating my affiliation with OCZ as a tester to put guards down and get people to swallow what I'm giving/sharing in my spare moments while in the office?.. then there are plenty of others who should be doing the same around here. If anything.. I'm less rigid in my "I only recommend this brand" type posts than many others around here are.

So, when they start off with post disclaimers.. or include sig's that basically say.. "Please take everything I say with a grain of salt since I have a real beef with this company and often show extreme tendencies to sway people in one way or the other in a heavily biased fashion"?.. I will be glad to do the same.

I make no secret of my beta-testing but am simply not willing to go out of my way to oddly discredit my own posts by doing such a thing. It's the internet for christ's sake and IMHO, EVERYONE should be keeping a mental tally in the back of their minds as they study up and receive input from others. I say just add salt to it ALL and then see what falls off later on when you're trying to see what fits where.

Me personally?.. I put MUCH more stock in people.. and their input.. that go far far out of their way to help me through firsthand experience with anything I may have problems or questions with. People that study statistical probability and practice/preach/recite only what they read on the internet and forums?.. not so much.

 

[josephjpeters]

I'm sure glad I came to this thread to learn about Sandforce. /sarcasm

In all seriousness I think there were 2 or 3 constructive points made and I thank those who tried to stay on topic.

 

[WhoBeDaPlaya]

The fact that this thread has spun quite far from the original question of "What's wrong with sandforce?" is a little sad. So many people pointing the finger at OCZ in particular and not directly answering the OP.

It's a little better than the idiots constantly spouting "ZOMG Hitach DeathStar!".