OCZ Technology was the first to do the switcheroo when swapping from 3Xnm to 25nm NAND which caused the storage capacity to shrink. That didn’t go well.
But it seems that other companies want to learn it the hard way.
Two SSD makers were caught switching to slower NAND after series of review sites posting encouraging reviews. It first started with Anandtech who reported about Kingston V300 SSD. It was reported in their review that its first-generation V300 uses Toshiba 19nm toggle-mode 2.0 NAND, but silently made a switch to Micron 20nm asynchronous NAND. This resulted to a different of performance and Bandwidth.
It was found that its Toshiba version has an NAND interface of 200 MB/s, whereas the Micron counterpart has bandwidth of 50MB/s. According to what the data with Anandtech’s reviews and what their readers have provided, Kingston V300 with Toshiba NAND showed AS-SSD incompressible sequential read of up to 475MB/s, but the Micron version reported ~170MB/s. The incompressible Sequential Write also reported the same, where the Toshiba version performs ~150MB/s, but Micron performs ~85MB/s.
It was pointed out that Kingston was considering to switch the model to V305, but decided to go against it and stick the Micron version as V300.
While Anandtech had reported issues with Kingston V300, Tweaktown reported the Bait-and-Switch NAND with PNY Optima 240GB. This SSD uses new Silicon Motion controller SMI SM2246EN controller.
A user sent a message to TT that he found a SandForce-type firmware version listing with his PNY Optima 240GB SSD with 5.4.1 firmware. What also confirmed the suspicion is that PNY listed 5.6.0 firmware update in its website, followed by SandForce flash utility. PNY did confirm this finding by the end-user, and justified it by saying the following:
Yes we did ship some Optima SSD’s with SandForce controllers, but only if they meet the minimum advertised performance levels (in most of the benchmark tests, LSI controllers outperform SMI controllers). The readers assumption that PNY has abandoned SMI controllers is wrong as we have been shipping mostly SMI controllers, but also utilizing LSI to fill in the gaps.
All SSD makers should not make this as a habit. In fact, Kingston and PNY shouldn’t have done that unless they wanted to risk denting its reputation for providing SSDs. While Kingston is being silent about it, PNY tried to justify it. Both companies however decided to use existing model versions to push slower performing NANDs as quietly as possible, especially after strings of review pointing out that a particular SSD model is good, if not best.
If a company follows such a trend, irrespective of what any review sites says about an SSD made by a company that swapped NANDs in the past for whatever reason with the existing model, users are going to be very skeptical in choosing it. After all, its their money and they trust that they get what they pay for and what they see in reviews. Preventing this entirely is lot better than having a dented reputation and then attempting to repair it. We hope that SSD makers do not follow this trend.