The Samsung 840 Pro was reviewed earlier, and it made a very good impression. That SSD flies!!! I won’t blame enthusiasts recommending these who people who can’t afford to spend a lot on such SSDs, but you get addicted to the performance it has. Something that one rarely feels, even at the time when there are a lot of companies rolling out SSD drives.
- Packaging and Specifications
- Internal Impressions
It should be noted that though 840 and 840 Pro use the same screws, I wasn’t able to open 840 as I did with 840 Pro. I wasn’t the only one, as Comp Kraft’s Shatul Durlabjhi wasn’t able to do the same. We actually had to go to Lamington at 7.30 in the evening and finally got the right tool for the right job.
- Disk Info, Endurance and TRIM Testing
- Test Bench and MethodologyIt gives me great pleasure to say that I get hardware support from the manufacturers to review not only their own products, but others on behalf of the readers. Its some companies such as Asus, Gigabyte, Kingston, Western Digital and Coolermaster who give me hardware support by updating my test rig no strings attached and that help me to help you! Thumbs up to these guys! I would like to thank
- Gigabyte India for providing Gigabyte 890GPA UD3H Rev 1.0 motherboard
- Asus India for providing Asus 990FX Sabertooth motherboard
- Kingston Taiwan for providing hardware support with memory kits and SSD drive.
- WD India for providing WD 300GB HLFS Velociraptor Hard Drive.
- Coolermaster India for providing Coolermaster GX450 RS-450-ACAA-D3 Power Supply
- AS SSD (Pass 3) Read and Write (Pass 3)
- ATTO (Pass 3) Test File Size= 0.5 to 8MB- Read and Write Pass 3
- Aida64 Access Test Write and Read Access time File Size 64KB Pass 3
- Boot Load Test (Windows 7 clean installation with AMD 12.6 drivers+ AHCI drivers pre-installed with Utorrent, Avira Security Suite, Asus Xonar DX+ 184.108.40.2064 Drivers, Netgear WG111 Wireless LAN USB drive software as start-up items) Pass 1-5 (Pass 1= System start from Power Off)
- TRIM Tests: Tested by comparing PCMark Vantage HDD scores with SSDs with Clean, 50% Fill, 75% Fill, idle for 1 hour for TRIM testing.
- CrystalDisk Mark Benchmark with Pass 3, each with 5 re-runs 1000MB File Size
- HDtune Pro Benchmark Sequential Write and Read Pass 3
- PCMark 7 HDD test
- PCMark Vantage HDD Test
- Anvil Benchmark: 4/16/32GB File Size in both 0% and 100% Fill test
- AS SSD BenchmarkThis synthetic test in gives a good idea about Sequential, 4K Aligned, 4K Aligned with 64Bit thread test and the Access time of the hard drive. It uses Compressed Data (like CrystalDisk Mark). In AS SSD read tests, there’s not much different between this and the Pro counterpart….but… In sequential, Samsung 840 is slower than Sandisk Ultra Plus II 256GB, let alone 840 Pro.
- ATTO Benchmark
Again, in read both drives are fighting each other but when it comes to write, it’s a whole different ball game for 840.
- AIDA64 Average Access Benchmark
840 simply has 0.01 bump ahead of 840 pro, when it comes to right…
- Boot Load Test
840 Pro takes 0.07 ms lesser time than 840 for access write test on an average basis.
- CrystalDisk Mark Benchmark
- HDTune Pro Benchmark
- PCMark HDD Benchmarks
Futuremark’s PCMark tests help end users to relate benchmarks with their everyday tasks and better understand the actual performance that would affect them with particular tasks.
- Anvil SSD Benchmark
It was obvious that the only way to reflect such potential of such SSD drives is by doing series of tests. Not just synthetic tests, but also simulate real world tests, endurance testing to ensure that even with a massive torture of 25TB+ host writes that runs at least 36- 48 hours in a row. And you can’t just stop there, you have to continue- like updating the benchmark database whenever a new firmware comes out (the world wide known reviewers do this provided they have the drive with them or gets purged from the graph database). It’s a very time consuming review even though SSD drives are pretty quick. Just how many reviews in India bother to test TRIM, endurance, performance level with different fill, compressible and incompressible data? For such people, all drives will perform practically similar. It needs time, it needs patience- and you can’t be stuck with the same benchmarks over and over again.
Using the laser guided temperature reader to read out the temperature from the surface of the casing with an open test rig and temperature read out on an air conditioner? Awww, that’s cute!!!!
Reason why manufacturers should make sure that reviewers know about solid state drives, such as the incident of a review which only had 2 benchmarks that would take 1 hour to complete and don’t maintain a graph database to properly will not find any potential flaws and flush down consumer’s money down the drain leaving a bad taste and questioning one’s credibility. Unsurprisingly, I got the same drive that the previous reviewer tested (same serial number). End result was that in reality the same drive has multiple write issue, unable to load up as a primary drive, complete and/or partial freeze. Most likely this rev version of the drive was one of the defects where the company made a recall few months ago. Thanks to this incident, I know which rev version not to buy. You’ll never know if these units are lying around for retail, because they sure did for reviewers!
If you think endurance testing is a waste, you need to be aware of an issue of a particular drive which got awards from certain review sites, but close analysis by some sites who bothered to do endurance testing found out that the drive’s flash NAND degrades eventually. It was also found out by others that many review sites got a different flash NAND chip compared to others, and it had different marketing. It was then a sour taste developed that flash NAND chips were not of the same in the same model. Another source also pointed out an issue of relabeling FLASH NAND over the existing chips, something that the manufacturer themselves never even knew about it until a reviewer highlighted it.
Sometimes there are a new rev version of controllers and manufacturers of the SSD do not follow a rev version system on their packaging. So user will not sure if it’s a new or improved one of not. Such was the case with another SSD. The newer ones were better, but the manufacturer didn’t release the newer firmware that fixes TRIM issue, despite the fact that the rival brands using the same controllers fixed it a long time ago. In the middle of it, the manufacturer used the new version of the controller which was said to consume lower power. I unfortunately got the one with the older rev version controller, but interestingly there was no rev version written on the packaging or on the SSD drive.
Hardware BBQ tries to keep up, finds, adopts and uses benchmarks and testing that will ensure its readers that they will get the good stuff in their hands! If I can do as much as possible with the limited resources and even time in hand, shame on those who can’t, but considered as ‘Tier 1’ reviewers.
Same packaging as 840 Pro, except the warranty period is of 3 years on this unit.
Same contents as on 840 Pro series SSD, so there’s nothing to add.
840 Pro used Samsung S4N021X01-8030 MDX controller which features 3x ARM Cortex R4s, with Samsung K4P4G324EB-FGC2 512MB LPDDR2 SDRAM Cache and 8x Samsung K9HFGY8U5A-CCK0 21nm MLC NANDs.
It’s the same on 250 GB 840 SSD Drive.
The drive was not used at all and came with DXT06B0Q firmware. The newest firmware available at the time of writing was DXT08B0Q and endurance test was done with the newer firmware.
so as a usual run I have stress tested the SSD drive with 26.13 TB of host write data on it to know that if the endurance is keeping up on that level.
The drive’s TRIM works as well and retains much of its performance as its on clean/secure erase state.
|Test Setup for:||Samsung 840 250GB SSD with DXM08BOQ firmware|
|Motherboard+ Processor||Gigabyte 890GPAUD3H Rev 1.0+ AMD 965BE|
|Memory||Kingston KHX1600C9D3P1K28G HyperX Genesis 8GB 1600MHz DDRIII|
|Primary OS drive||WD 3000HLFS Velociraptor 300GB/WD 320GB BLUE 320AAJS|
|Power Supply||Corsair TX750|
|Chassis||Open test bench|
The Benchmarks and tests that I am using are as follows:
The drive pretty much stays at 15 seconds boot, except on Pass 3. I wouldn’t be surprised if the newer firmware on 840 Pro would improve the boot speed and keep it relatively close to the 840 non-pro.
It’s the same scene throughout synthetic benchmarks: Write rules on 840 Pro.
In both cases, 840 Pro scores ahead of 840, with an exception of few tests with either or no noticeable difference.
If you’re someone who is shifting from mechanical hard drive, 840 250GB SSD will be a bigger step than LSI Sandforce powered drives. It is faster than the usual Sandforce drives and Futuremark tests highlights that nicely!
But if you’re someone who has a workload that WILL benefit from write, 840 Pro is your drive. For such drives, having 5 years warranty makes a lot of sense. Both are equally reliable drives. Other than that, it’s the difference of the warranty period and costs.
Samsung 840 250GB SSD Review