‘Toshiba considering to buy OCZ consumer SSD business’

We’ve just posted the following news: ‘Toshiba considering to buy OCZ consumer SSD business’[newsimage]http://static.myce.com//images_posts/2013/10/myce-toshiba-ocz-95x75.png[/newsimage]
[quote][/quote]
Read the full article here: http://www.myce.com/news/toshiba-considering-to-buy-ocz-consumer-ssd-business-69183/

            Please note that the reactions from the complete site will be synched below.

Speculation aside.
If this deal was to go through, it would be good for both OCZ and Toshiba.

[QUOTE=Dee;2704485]Speculation aside.
If this deal was to go through, it would be good for both OCZ and Toshiba.[/QUOTE]
What internals and specifications make an SSD Consumer or Enterprise? Where is the line drawn? Would a PCI-E SSD be considered Consumer or Enterprise? Or is it just write cycles that make the difference?

Enterprise.
Endurance (high endurance NAND, and probably NAND overprovisioning).
Power failure protection.
Longer validation cycle for the hardware and firmware.
A performance profile geared for enterprise use.

[QUOTE=Dee;2704485]Speculation aside.
If this deal was to go through, it would be good for both OCZ and Toshiba.[/QUOTE]

I agree and I hope it is true.

I feel the result would enable Toshiba to give Samsung a run for their money, which would be good for the market.

Regds, JR

[QUOTE=JReynolds;2704515]I agree and I hope it is true.

I feel the result would enable Toshiba to give Samsung a run for their money, which would be good for the market.

Regds, JR[/QUOTE]

Toshiba has been the leader in this case. Toshiba had everything.

What OCZ neeeded was not a new partner. Not more capital.

Hope for Chinese to overwhelm Samsung. They are the only worthwhile competitors for now. Japanese manufacturers lost the edge not because lack of fine design or technology or capital but because of the same reasons other rich industrial nations gave up manufacturing.

This is what the “Senior Storage Editor” of the site says about the rumors:

“If the report is accurate this is a win-win for everybody. Toshiba will gain a controller to go with its flash and OCZ will escape from the low margin consumer market it created,” Ramseyer Said. “OCZ CEO Ralph Schmitt recently stated that OCZ had issues in 2013 securing flash in large volumes. Toshiba 19nm NAND flash has been proven superior to IMFT 20nm flash yet most of OCZ’s recent products shipped with IMFT 20nm flash due to Toshiba 19nm shortages.”

Does anyone think that means anything? You think Toshiba lacked the technology to compete against Samsung? Japanese needed technology to compete? That’s like saying Hitachi cannot make better IPS panels than LG. South Korean manufacturers thoroughly depend on Japanese technologies and equipment for production and that is true even in 2013 and will continue for at least one or two more decades, more likely forever.

OCZ bought Indilinx for one reason, and that was not money. It was Indilinx that needed money. Indilinx was a competitor to Mtron and other South Korean start-ups in the SSD sector, but it was soon obvious companies with one millionth capital and one thousandth number of researchers and developers could not produce SSD as cheaply as Samsung. It once seemed, about 10 years ago, by some inside South Korea and by others outside South Korea, Samsung would be disintegrated and employee salary was sometimes half to one third of what it is now. Mtron disappeared by about 2010.

Toshiba can make as good SSD products as Samsung. It did, and it still does. OCZ does have some fine niche products to complement Toshiba’s storage product offerings, but I doubt it will mean anything to Toshiba. Indilinx is inferior to Samsung, for some very obvious reasons. Samsung is vastly inferior to Toshiba, for even more obvious reasons.

What made Toshiba and other Japanese electronics and semiconductors especially in the making of DRAM and NAND business difficult to survive was not lack of technologies, or lack of the number of skilled technicians and engineers, and researchers and managers able to innovate and continue to innovate. They were less willing to reduce cost and prices to OEM. They gave up making something when competitors offer lower prices. What Samsung does to achieve that is often extreme, but so far foreigners have been interested only in the stock shares of Samsung, not how they work.

It would work if it was Intel and Micron to buy Toshiba, or Toshiba to buy Intel and Micron. An even better combination would be Toshiba + Seagate since both companies make both HDD and SSD while each has unique advantages in marketshare and technologies. Apple buying OCZ would work as well. Apple is not famous as a company of philanthrophists and it once switched from Motorola to Intel when Intel offered better performance for money, so while Apple buying OCZ could make OCZ employees become overnight millionaires, it wouldn’t mean all future Apple computers and tablets using proprietary OCZ design.

[QUOTE=Kenshin;2704529]Samsung is vastly inferior to Samsung, for even more obvious reasons.[/QUOTE] Reasons may not be as obvious as you think! :eek:

[QUOTE=Dee;2704510]Enterprise.
Endurance (high endurance NAND, and probably NAND overprovisioning).
Power failure protection.
Longer validation cycle for the hardware and firmware.
A performance profile geared for enterprise use.[/QUOTE]

I think PCI-E SSD was once considered enterprise, but it’s already changed. Both Apple and Samsung decided to make it mainstream, and both chipset and device manufacturers seem to agree popular standards widely used so far including USB 2.0 and SATA 300/600 are too bottlenecking.

Apple’s Mac Pro is surely for the rich, or those who can spend a lot on workstations, so their adopting 1.2GB/s PCIe SSD cannot impact consumer SSD markets seriously. But Apple’s existing MacBook Air also has PCIe SSD. More and more Samsung Ativ computers, both in desktop and laptop and tablet categories, will come with PCIe SSD.

Alan’s question is valid.

  1. There were clear divisions between IDE and SCSI. SCSI was better programmed for servers and web servers needed to meet the requirements of exploding numbers of concurrent users. SCSI was always several times more expensive than IDE.

  2. SSD based on RAM and NAND was for enterprise. Some of the early-generation SSDs made of SLC NAND were enterprise, but some others were consumer. Differences between SLC-made enterprise SSD and SLC-made consumer SSD? 10% perhaps, but sometimes 3x faster IOPS and 2x faster sequential write speed.

  3. As MLC increasingly replaced SLC, the widely published specification indicating to the effect that SLC can last 10 times longer than MLC in the early years seemed ancient and it was soon forgotten. In 2011, Tom’s Hardware Guide posted some interesting articles on whether SSD can really outlive HDD. The article contained some case studies as well though evidences and statistics were not highly detailed.

But in discussions with ZT Systems, the company Intel cited, we discovered that the 0.26% AFR figure doesn’t actually take age into account and only covers validated errors. Frankly, if you’re an IT manager, you care about unvalidated errors, too.

According to the company’s CTO, Casey Cerretani, the final numbers are still being tabulated, but early estimates peg the unvalidated AFR for the first year closer to 0.7%.

By then, people who around 2007 preferred the most expensive “Pro” category of enterprise-oriented Mtron SLC drives were buying MLC drives made by Intel and Samsung. 10x more expensive SSD did not prove to be 10x as reliable and sometimes 10x more expensive SSD seemed to offer 10x slower speeds.

  1. Similar things happened with MLC vs TLC. TLC has not replaced MLC. SLC NAND was not 10x more expensive than MLC NAND. It was merely 10% or 50% more expensive. It was the RETAIL SSD makers like OCZ and Intel that made SLC SSD prices 10x more expensive than MLC SSD prices. Some people insisted TLC would last one year when MLC would last ten years. Today’s TLC drives write very fast, a lot faster than SLC drives of 2007 did, because they allow multiple TLC chips write at once. It works like striped SCSI drives, but far more efficiently. If TLC were really 100x more unreliable than SLC, everyone of the millions of 840 and 840 EVO drives sold two weeks ago should have died. If SLC is for enterprise and MLC is for consumer, where does TLC belong? Add to it that it was SCSI that monopolized the enterprise market. SCSI for enterprise, SLC for much poorer enterprise, MLC for risk-taking consumer, TLC for the incurable?

  2. Interface standards for data transmission have evolved as well. SCSI wasn’t enterprise. It was merely an interface standard. SCSI couldn’t replace IDE. IDE replaced SCSI for most because the cost of US$1 for IDE controller and IDE cable was so little compared to US$1,000 for SCSI controller and SCSI cable while most of the newest devices were available for IDE. Even lower-end web servers had Adaptec SCSI controllers and Cheetah drives. Some were bold enough to replace the latter with Intel’s 80GB/160GB MLC SSDs by hundreds of thousands.

  3. I regard what OCZ has said about enterprise success and NAND supply difficulties as lies.

[QUOTE=DrageMester;2704540]Reasons may not be as obvious as you think! :eek:[/QUOTE]

Sorry for that, I have to fix that and post once more since I cannot edit it, but it will do for now since everyone will know the obvious mistake.

[QUOTE=Albert;2704558]You meant Toshiba, then Samsung, right? Would you like someone to fix that typo?[/QUOTE]

Sorry for the trouble.

What I wanted to say was this:

Samsung is vastly inferior to Toshiba, for even more obvious reasons.

By which I meant Samsung learned everything they needed to do NAND and SSD from Toshiba just as LG learned everything they needed to produce LCD and promote IPS from Philips and Hitachi.

Whole South Korean electronics and semiconductor industries depend heavily on Japanese equipment manufacturers, which explains why many South Koreans complain it’s the Japanese that earn more profits when South Korea exports more.

[QUOTE=Kenshin;2704561]Sorry for the trouble.

What I wanted to say was this:

By which I meant Samsung learned everything they needed to do NAND and SSD from Toshiba just as LG learned everything they needed to produce LCD and promote IPS from Philips and Hitachi.

Whole South Korean electronics and semiconductor industries depend heavily on Japanese equipment manufacturers, which explains why many South Koreans complain it’s the Japanese that earn more profits when South Korea exports more.[/QUOTE]

No trouble at all. Typo fixed. :slight_smile: