Flash memory such as a USB stick consists or 4 parts. These include the:
Product from an unidentified supplier can have a mixture of controllers and wafer sizes.
There are lots of different types of flash technologies, the two most popular are multi-level cell (MLC) and single-level cell (SLC). MLC is cheaper than SLC but it has certain technical disadvantages. However, these issues only really apply to mission critical applications and high end industrial applications.
It is unlikely that Flash cards not sold by a well known manufacturer such as Sandisk are of an A grade quality, you get what you pay for. That being said, in most cases where the memory sticks are used as "promotional give aways" this does not matter. However low grade flash can be over or undersize, meaning that the master can hold more data than most of the supplied flash, causing problems when duplicating. The very small number of Flash can pass all write and read tests when duplicating, but when removed from the computer it may not retain the data. If the data is critical, specify only Grade A cards. An alternate saving can be to use grade A-cards, and re-verify each card on a different day. This will ensure that the card is capable of retaining data.
'Good flash' that has passed the original manufacturers test is commonly known as grade A. The 'rejects' are commonly known as grade B β I have even heard some companies refer to grade B minus, but there is no industry wide 'test' that is used to determine an A from a B
Flash chips are traded on the world market every day, just like every other electronic component. Price reflects supply and demand.
With Flash sales increasing all the time, most of the A stock is allocated to cards or for embedded application, hence there isn't a great deal on the spot market. If you are a manufacturer and you want to use A grade, then you will normally have an allocation and have multiple sources of product β i.e. supply chains with most manufacturer.
The big manufacturers don't use the B stock, instead they are released for sale on the spot market.
Plenty of companies trade in these products and just because they are B grade, does not necessarily mean they are poor quality and not fit for purpose as it ultimately depends on the purpose! Some flash might be good for average use if it is built with a good controller and sold properly β i.e. the trading company doesn't imply it's good for mission critical apps or that it will last for 10 years !
The reality though is that top class manufacturers chose not to use components that they deem to be sub-standard, but there are companies who do not think this is a problem.
Too many companies are competing with each other for sales hence they buy cheap, build cheap and do whatever they can to lower costs. This can be using B flash (sometimes up to 60% cheaper, but is does change every day), using a cheaper controller and generally doing the things that top class manufacturers don't subscribe to.
Most USB FlashDrive manufacturers have several bills of materials so they can use different components depending on what's available. Some are much better than others and the vast majority all purchase the same housings from the same manufacturers, hence products from lots of different companies can all look the same, but technically they can be very different.
There is a market price for flash, whether it is an SD card or a USB FlashDrive. If the price is too cheap, then ask why and probably stay away from it for critical applications.
SanDisk is the biggest card manufacturer in the world, and they are unique in that they are probably the only company who control all production elements β i.e. manufacturer of flash, design and production of controllers, assembly. SanDisk are strategically partnered with Toshiba and they share fabs (fabrication plants) - this is where the chips are made.
The above information has been supplied by Paul Norbury, MD of Cardwave Services Ltd, world leaders in Flash technology.