Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
More TLC NAND Trouble
According to the comments, it's the rebranded Phison controller that's to blame.
Quote:When I started reviewing the Trion 100, it exhibited very large performance swings between each of our real-application benchmark runs. After some testing I realized that this is due to the way the TLC memory is organized. The drive has a small 9 GB portion of its capacity working in SLC mode, which is very fast to read and write, but once its capacity is reached, the drive will either have to shuffle data from SLC into TLC cells, or write to TLC directly, which is much slower. As a result, during prolonged periods of disk activity, the drive will completely stop accepting new data, causing all writes to stall, which usually also means that task switching in Windows will hang, because almost all programs have a little bit of disk activity, which cannot be completed now.
For our benchmarks, I've improved the situation by adding a 180-second delay between each of our tests, which is not an unreasonable assumption for consumer oriented workloads. The only test that still shows huge issues is our data compression test, which unpacks an archive with lots of small files. In this scenario the drive will hang roughly every four seconds, for around 10 seconds, causing the whole OS to slow down or freeze. It gets even worse when you start deleting the unpacked files, in which case the whole OS freezes for around two minutes. I understand the argument that most consumers won't do intensive disk operations, but the whole system freezing from starved disk activity, suggests some kind of firmware issue to me, maybe the read/write operations aren't prioritized properly. Especially with consumers, the appearance of a hanging = broken computer could be an issue as these are less tech-savvy people and will either end up cursing SSDs in general, or making RMA claims with OCZ.
People are already catching on that TLC NAND is crap.
Quote:The downside of being the slowest modern drive we have tested is that OCZ can only compete in price, but unfortunately the pricing isn't aggressive enough for the Trion to be competitive at all. Currently the ARC 100 is even cheaper than the Trion 100, which just doesn't make any sense because even OCZ is positioning the ARC 100 higher on paper and it's undoubtedly a better drive all way around. There is usually some level of decline from the initial street price soon after the launch, but in all honesty OCZ needs to cut the prices by 15-20% for the Trion to have a place on the market. OCZ told us that it will be running promotions with discounts, but it remains to be seen if those can bring the prices down to a level where performance and price meet. It has the lowest performance, thus it needs to be priced lower than the competition to provide the value to the user because right now I wouldn't have to think twice about buying the BX100 or 850 EVO over the Trion 100.

All in all, getting TLC done right is far from an easy job as we have witnessed with here. I believe that with the first big wave of TLC SSDs coming this year we are going to see sub-par performance compared to MLC drives. Fundamentally I have no problem with that because even a "slow" modern TLC SSD is a significant upgrade from a hard drive, but it is time for the manufacturers to realize that the price should reflect performance. It's just silly to take up to 50% hit in performance and only offer a few dollar savings because any educated buyer will gladly pay the extra few dollars for a substantially better drive. Once other NAND vendors start to ship 3D NAND in volume next year, we will likely see the majority of client SSDs move to TLC because as Samsung showed with the 850 EVO, 3D TLC NAND can enable planar MLC-like performance, but in the meantime it seems like MLC SSDs will still provide better overall value.
Now the 850 Pro 2 TB has developed a performance issue. F*** Samsung.
What are you talking about?

The link showed pretty solid results
Here's what I am talking about:
Quote:We were surprised to see the 850 Evo 2TB outperform the 850 Pro 2TB in this series of tests. Unlike the previous benchmark results from PCMark 8, these run under more demanding conditions, with less time between each metric. They aren't enterprise-oriented, but come close to that level of load.

Normally we see the SanDisk Extreme Pro 1TB dominate this discipline, with the 850 Pro 1TB close behind. Under the lighter workloads, we expect the 850 Pro 2TB to outperform the smaller 1TB model. However, it just couldn't recover fast enough to climb back.

We would normally suspect a TRIM issue. But the 850 Evo recovered well and led the light workload portion of the test.
The latency numbers show us where the 850 Pro 2TB falls short and where the 850 Evo 2TB turns into a beast. Under heavy load, Samsung's 850 Pro 2TB took nearly twice as long to complete the tests as the 850 Pro 1TB. When the drives are given time to recover, the 850 Pro 2TB fails to clean the NAND or direct the data writes to clean flash. This is something Samsung will need to look into and fix in a future firmware update.

The 850 Evo 2TB performs better than we expected under demanding conditions, and then brings the service times down to class-leading levels when it's given time to recover.
We spent a lot of time exploring the 850 Pro 2TB performance abnormality we found, but not enough to isolate the problem. This is a bad time for Samsung to have firmware issues. Some users are still reporting slow-downs on the 840 Evo, even after the company attempted multiple fixes. The 840 Evo is the best-selling client SSD of all time. But the transition to three-bit-per-cell flash may have come too soon or without powerful-enough ECC to tame it.
I don't understand your post:

(07-15-2015, 12:34 AM)SteelCrysis Wrote:,4205.html
Now the 850 Pro 2 TB has developed a performance issue.  F*** Samsung.


At first, I thought oh no, now the 850 pro has an issue that shows up over time. Degenerates over time kind of issue.
But that is not at all what is going on. These are new and different drives, they perform different in different scenarios. Saying "now" kind of implies that there is an issue that came to be that wasn't there. The truth is, this is how it is lunching. It may be a trade off or consequence in the design. It is not unusual, every brand and every SSD has its great points and not so great points. Most drives that are on top, are not on top in every last test nor every last benchmark.

This "issue" is there for all to see before hand. It may be improved with firmware or it may never improve in that area. It's there for all to see, so I don't see the big deal about it.

Forum Jump:

Users browsing this thread: 1 Guest(s)