No offense intended here.. but most people who start threads like these.. or even those who respond to them for that matter.. don't understand enough about how these things work sometimes. Not all is as it seems here to the casual forum goer or even those who have had issue with these drives.
First of all. Sandforce pushes almost the entire validation process onto its vendors which is contrary to other controller mfgrs in the industry.
So with that in mind it seems pretty simple to figure out that those who go to market first and sell the most drives would appear to be most problematic from the casual users perspective. Did OCZ jump the gun and shoot themselves in the foot by releasing without sufficient testing? Yes they did, IMHO. But others were not without issue either though even as they waited to release their versions of these drives. Was it worth the gain in the end even at the expense of lost credibility(no one likes to be pushed into the beta-testing process) and higher RMA numbers? Their stock(along with others using this controller) says YES IT WAS! Speed sells like sex in this industry and many just want more and more.
Hence much of the reason why OCZ has the bad rep that it does since Sandforce has fueled their SSD portfolio(from the very first SF-1xxx series controller released) more than any other controller to date. They were first to market.. they sold more.. they fixed more. Simple stuff when boiled right down.
The Everest and Everest 2 controllers will work to change that negative mentality as there is obviously much more to an SSD than just synthetic benchmark speeds. I can tell you from firsthand experience that the Octane is much better in actual usage than the benchmarks will show.
This new controller also finally gives OCZ complete in-house control to validate in a much more timely manner(no long drawn out ping-pong match to get things sorted out). Which unbeknownst to others.. is much of the issue with OCZ(and other vendors as well) and Sandforce's partnership since the process is a constant overly long ping-pong match as the firmware is tested and tweaked back and forth between the two. This of course leaves the end-user who would see and advise of issues to be stuck in the middle and ultimately at the end of the line for actual fixes.
The funniest thing I see out of all this is the interpretation from others(and even those using other vendors models) is that they think firmware code is entirely developed on each individual vendors dime. Not true at all as the trickledown effect applies in a very big way across the spectrum of vendors using these controllers.
Should also be obvious that OCZ is watched and followed more than any other SF vendor due to the transparency of their forum environment and the sheer volume of products sold(and having issue). They release firmware before all others(usually by several weeks) and the others watch to see where it goes before implementing them into their code as well. So, if one vendor finds an underlying issue in usage.. applies workarounds or tries a quickly implemented(weeks in "Sandforce time") firmware patch? ALL vendors are paying VERY close attention to what is being proven to work across that spectrum.
So in a way, OCZ actually leads the way in testing of these particular controllers(although it appears to be at the end users expense) as they are constantly in the habit of sticking their necks out further than others to get higher sales volume for these products. This of course leaves them open for ridicule and the assumption that others are doing much more in-house validation before releasing their fixes. Everyone is using everyone and ultimately Sandforce uses them all to further their product, is all this crap amounts to.
And it is very much fact that all issues with the SF-2281 controller was not squarely on OCZ's shoulders. I witnessed firsthand(through beta-testing these controllers), and all over the net(including other vendors), a multitude of Intel driver fixes(IME was HUGE in this regard).. improperly defaulted power mgmt setting workarounds.. along with board CMOS issues(hotplugging ON by default was a biggie here too as it told Intels drivers that the drives were external and removable) that were able to take unstable SF controlled drives and make them COMPLETELY stable. Many cases were remedied before the very first firmware fix was ever released with many of these workarounds. That should say something in itself as to the potential for other underlying issues being at play.
It was only after all the underlying issues were completely understood that OCZ(along with others) was able to feed Sandforce enough info to replicate things on their end to be able to actually apply workarounds from the firmware end of the picture. And don't be so niave to think that everyone wasn't watching everyone throughout the troubleshooting process. I and a few others with higher vantage points to these issues also saw the obvious feet dragging going on as Intel was entering the picture with plans of using this controller as well. Money talks of course and that's just life in general though and it's neither here nor there at this point. Just sayin, is all.
And FYI here.. when it takes the SSD controller mfgr, several vendors using them, mobo mfgrs, tens of thousands of users, and the world’s largest chipset mfgr to build a subsystem and resultant firmware that works seamlessly across various hardware configurations to figure all these things out?.. you can bet your ass that the controllers firmware is much more complex than many give it credit for. Intel jumping into the game this late is simply to gain a bit of the very popular Sandforce market and further position themselves for the next gens release. Even lower latency.. higher incompressible write speeds.. what's not to like about such a potentially good controller? They'd be stupid NOT to get a headstart into getting an even bigger slice of that pie.
Welcome to the Sandforce follies.