OCZ May Be On Its Last Legs 292
itwbennett writes "OCZ, one of the first commercial solid-state drive (SSD) makers has been blaming a shortage of NAND for its woes for some time now, but things have taken a precipitous turn for the worse: 'For its second fiscal quarter ended August 31, 2013, revenue was $33.5 million, a huge drop compared to revenue of $55.3 million for the first quarter of 2013 and revenue of $88.6 million for the second quarter of 2012. The net loss for this quarter was massive, $26 million, a doubling of the $13.1 million loss in the same quarter last year.' The company has burned through cash, its stock collapsed, and now so have sales. Meanwhile, other SSD makers are doing well. So what is happening here?"
Tiniest violin (Score:5, Insightful)
They burned too many customers with "enterprise" devices that'd fail almost immediately, then treating the customers like shit when they did.
They bet too heavily on high performance, while not maintaining the kind of behavior that would bring back the customers who want devices like that.
The reason Dell and HP can get away with burning customers is simple: there's always another person who needs a cheap laptop.
Not many people need a new PCIe SSD.
Good riddance.
Of course (Score:5, Insightful)
Re:Easy. (Score:5, Insightful)
If you could provide a source(even if your numbers aren't completely accurate) you would make me very happy. I have been unable to find anything that discusses reliability of different manufacturers like you just described.
I have always sworn by Intel while friends have bought OCZ(because they were cheaper per GB) and several have had nothing but problems but others have sworn their OCZ was rock solid. On the other hand, I bought only Intels since the day the G2 series hit the market. Every single one is still in use and none of them have had any problems. In fact, I haven't had to reinstall windows as often as I've had to in the past. Not sure if its because Win7 is better than WinXP, the SSDs are more reliable than platter based disks, or both.
But even then, I still swear by Intel every time a friend makes a recommendation, regardless of the benchmarks and the (often) slightly higher price per GB.
Re:Easy. (Score:3, Insightful)
Oh my God! The money quote:
Re:Rebates (Score:4, Insightful)
Funny... Similar experience didn't convinced me to avoid company XYZ, but instead to completely avoid any and all mail-in rebates... The whole idea is a complete scam that is easily and frequently abused.
Re:Full of BS (Score:2, Insightful)
OCZ hurt the entire SSD industry (Score:5, Insightful)
There's an even better reason why nobody wants to sell flash to OCZ -- they've tainted the entire SSD industry so badly with their crap drives, no reputable manufacturer of flash wants to have its good name tarnished by association with them.
A lot of OCZ's problems were self-inflicted, with Sandforce's active complicity.
For example, Sandforce's engineers came up with an ugly, performance-killing hack that allowed the drive to avoid corruption if it were powered-down mid-write so they could officially claim that the ultracapacitor was "optional" in "cost-sensitive applications". OCZ built drives without the ultracap, then had Sandforce furnish them with firmware that DISABLED THAT SAFETY MEASURE to avoid killing their drives' write performance in benchmarks.
Mark my words. If OCZ doesn't go bankrupt on its own accord, they're eventually going to get put out of business by a class-action lawsuit like the one that nailed HP almost 20 years ago. I'm talking about the one where HP's management intentionally ignored their engineers, and sold CD burners that didn't have enough RAM to buffer a complete track & instead depended upon Windows to feed them a steady stream of data with a degree of lockstep precision that Windows could neither promise nor reliably sustain even though their own engineers told them it couldn't work reliably, and was GUARANTEED to turn at least 5-20% of discs burned into coasters (back when a blank CD cost SEVERAL DOLLARS).
HP's engineers DID have a way to allow the drives to be reliably used without the buffer... write the .iso file to a FAT16 volume, then boot directly into DOS from a floppy to do the burning. However, like OCZ's management (who wanted the performance of an ultracap-protected drive, without the cost of the ultracap itself), HP's management wanted a cheap drive that could burn CDs under Windows, even if it meant they had to knowingly LIE about its ability to actually DO it.
Re:OCZ hurt the entire SSD industry (Score:5, Insightful)
It didn't need 650mb, it just needed to be a lot bigger than the absurdly small buffers HP shipped with.
Think of an assembly line at a cookie factory with a badly-designed packing mechanism that blindly assumes (and depends upon) there being a cookie every 24 inches -- centered on a white dot printed onto the belt -- without fail, and shuts down the entire assembly line if it finds a gap without a cookie.
Now, assume the cookies get placed on the conveyor belt by one person who has a bucket of cookies in hand, seated in front of a 12-inch gap where the conveyor belt emerges from one slot, passes across an open area, and disappears into a second slot. The employee has exactly 5 seconds to grab a cookie from the bucket, and exactly 5 seconds to place the cookie on the dot on the conveyor belt before repeating. Now, suppose the employee is holding the cookie, ready to place it on the conveyor belt, and sneezes. To avoid spreading infection, he or she turns around to sneeze away from both the cookies and conveyor belt. Unfortunately, the sneeze takes 6 seconds to perform and recover from, so the dot disappears into the second slot without a cookie. If we're burning a metaphorical CD with those cookies, that sneeze has just caused a coaster.
THAT was the fundamental problem with HP's small buffer. It depended upon having the undivided attention of Windows for frequent, short intervals of time with ZERO tolerance for distraction.
In contrast, a larger buffer would be like an assembly line that shuffles cookies towards multiple bins. As soon as a bin is full, the flow of cookies into it gets temporarily halted (with enough room to buffer/queue a few cookies in the meantime), a new empty bag falls into place, and the queued-up cookies are allowed to fall into it immediately, then continue until the next bag is full.
In the real world, it's ALWAYS harder to guarantee data at some precise trickle than to allow it to just gush in spurts and be buffered at the same net data rate.A lot of people think "realtime" means "fast". It doesn't. It just means "deterministic" (often, deterministically-constant). A large buffer allows you to deliver a deterministic trickle of data transmitted in a bursty, non-deterministic manner.
Re:Tiniest violin (Score:2, Insightful)
Except for the people who don't own a single optical drive. Not trolling, Being an Apple user first I freaked out. Then now back to Linux and somehow don't miss them. So when I put a server and a desktop machine together, I didn't put an optical drive in them at all.
I think the right way to do it is to give users a bootable USB drive. Or offer the download with a utility to make a drive bootable.
I have probably 50 driver CDs in a shelf I never opened and this plastic nonsense has to end right now.
my 2c ...
Re:Avoid all brands with "4 GAMERZZZ!!11" marketin (Score:2, Insightful)
To be fair, I've only had one pair of Corsair DIMMs fail, ever. That one was because a power supply failed and shorted the motherboard and RAM as a result. less than 5 minutes on a phone call to Corsair and I had an RMA number. I even told them exactly what happened, I think they even sent a postage paid return shipping label. I now have 3 Corsaird Power Supplies in use (one for almost 3years, the other 2 less than 2 months) no issues with any of them yet.
Oh, and I'm 34. I'm more than happy to pay a premium if the Quality and Customer Service back up the price.
Re:Full of BS (Score:5, Insightful)
My experience with all drives, solid state and spinning, is a 100% failure rate... eventually.
Re:Full of BS (Score:3, Insightful)
If you couldn't find a citation then maybe you shouldn't have fucking said it.