We got the first glimpse of Transcend's SSD370 at Computex last year and now the drive has been in retail for quite some time. The interesting bit about the drive is its custom Transcend firmware, which is accompanied by a relabeled Silicon Motion SM2246EN controller (Transcend calls it TS6500). The SM2246EN has increasingly been gaining popularity among SSD vendors, but the SSD370 is the first time I've come across a custom firmware solution. Silicon Motion's business model is similar to SandForce's in the sense that it can deliver the whole package (controller hardware & firmware), but also allows custom firmwares (although I don't know if customers get full source code access, which at least SandForce doesn't allow).

Relabeling the controller is nothing unheard of because at least OCZ and Toshiba have done it in the past. Usually the reason why manufacturers do this is to ensure that their drive isn't mixed with the other drives that use the same controller because even with the same controller the drives can be totally different (excellent case in point is Marvell based SSDs). You would be surprised how often I still see people classifying drives based solely on the controller silicon, so I certainly understand the manufacturers' motivation as it's not exactly fair to judge a drive based on the controller alone. At the end of the day, it's the firmware that designates the drive's performance and reliability - the controller is just a SoC that does the number crunching. I can't say I'm a big fan of controller relabeling because it creates confusion and may be seen as a dubious marketing act, but as long as the manufacturer is open about the real source of the silicon, it's not something that deserves a big call out in my opinion. 

Transcend SSD370 Specifications
Capacity	128GB	256GB	512GB	1TB
Controller	Transcend TS6500 (rebranded Silicon Motion SM2246EN)
NAND	Micron 128Gbit 20nm MLC
DRAM (LPDDR2)	256MB	512MB		1GB
Sequential Read	550MB/s	560MB/s	560MB/s	560MB/s
Sequential Write	170MB/s	320MB/s	460MB/s	460MB/s
4KB Random Read	70K IOPS	70K IOPS	75K IOPS	75K IOPS
4KB Random Write	40K IOPS	70K IOPS	75K IOPS	75K IOPS
Idle Power Consumption	305mW	320mW	325mW	335mW
Read/Write Power Consumption	1.21W / 1.92W	1.28W / 3.11W	1.43W / 3.22W	1.76W / 3.46W
Endurance	150TB	280TB	550TB	1180TB
Warranty	Three years
Online Pricing	$60	$105	$200	$394

The SSD370 is available in capacities from 32GB to all the way to up to 1TB. I decided to leave out the 32GB and 64GB units from the specification table as I suspect these are mostly OEM-focused models because (to be honest) there isn't a significant retail market for drives smaller than 128GB anymore. In addition to the SSD itself, the retail package includes a 3.5" desktop adapter and for drive migration Transcend provides a cloning function through its own SSD Scope Utility.

By default, the SSD370 doesn't support AES and TCG Opal encryption, but Transcend has a customized firmware that enables encryption. I suspect the custom firmware is aimed towards PC OEMs because the major market for self-encrypting drives (SEDs) is still in business PCs. eDrive, however, is not supported at the moment, although Transcend's plan is to add support in the near future.

The SSD370 doesn't support slumber power modes (HIPM+DIPM) because Transcend decided to disable the feature due to some prior compatibility issues with hosts that didn't properly support the feature. DevSleep, however, is supported according to the data sheet, but there are no power figures to support that, so I'm thinking it's a feature that can potentially be enabled through a custom firmware if a customer requires it.

In the NAND department Transcend uses Micron's 128Gbit 20nm MLC NAND, which is fairly common in more value-oriented drives. Micron has also started shipping its 16nm NAND to customers and I think Mushkin's Reactor was the first SSD to adopt it, but we will likely see many SSD OEMs transitioning to Micron's 16nm during H1'15 as the shipping volumes increase.

Interestingly the SSD370 employs partial power loss protection as there are ceramic capacitors on the PCB to provide power in case of a sudden power loss. Ceramic capacitors are fairly low capacity and can't provide the necessary power to flush everything from the DRAM cache to NAND, so user data in the DRAM is still in jeopardy, but the capacitors ensure that data at rest (i.e. in lower pages) and the NAND mapping table are safe. That's similar to what Micron did with the M600 and I suggest you read the review if you are looking for a more in-depth explanation regarding client-level power loss protection.

Another intriguing feature is what Transcend calls StaticDataRefresh Technology. As the 840 EVO performance degradation bug taught us, the charge in cells degrades over time, which results in errors when the cell is read. ECC can fix a certain number of error bits, but if the limit is exceeded corrupted data will be sent to the host. The StaticDataRefresh technology monitors the error rates and when a preset threshold value is reached, the data will be rewritten to restore the correct cell charge level. I suspect all SSDs do this because it's vital to ensure the health of old data, but it's the first time I've seen it mentioned in a data sheet.

Test Systems

For AnandTech Storage Benches, performance consistency, random and sequential performance, performance vs. transfer size, and load power consumption we use the following system:

CPU	Intel Core i5-2500K running at 3.3GHz (Turbo & EIST enabled)
Motherboard	ASRock Z68 Pro3
Chipset	Intel Z68
Chipset Drivers	Intel 9.1.1.1015 + Intel RST 10.2
Memory	G.Skill RipjawsX DDR3-1600 4 x 8GB (9-9-9-24)
Video Card	Palit GeForce GTX 770 JetStream 2GB GDDR5 (1150MHz core clock; 3505MHz GDDR5 effective)
Video Drivers	NVIDIA GeForce 332.21 WHQL
Desktop Resolution	1920 x 1080
OS	Windows 7 x64

Thanks to G.Skill for the RipjawsX 32GB DDR3 DRAM kit

For slumber power testing we used a different system:

CPU	Intel Core i7-4770K running at 3.3GHz (Turbo & EIST enabled, C-states disabled)
Motherboard	ASUS Z87 Deluxe (BIOS 1707)
Chipset	Intel Z87
Chipset Drivers	Intel 9.4.0.1026 + Intel RST 12.9
Memory	Corsair Vengeance DDR3-1866 2x8GB (9-10-9-27 2T)
Graphics	Intel HD Graphics 4600
Graphics Drivers	15.33.8.64.3345
Desktop Resolution	1920 x 1080
OS	Windows 7 x64

• Thanks to Intel for the Core i7-4770K CPU
• Thanks to ASUS for the Z87 Deluxe motherboard
• Thanks to Corsair for the Vengeance 16GB DDR3-1866 DRAM kit, RM750 power supply, Hydro H60 CPU cooler and Carbide 330R case

Performance Consistency

Performance consistency tells us a lot about the architecture of these SSDs and how they handle internal fragmentation. The reason we do not have consistent IO latency with SSDs is because inevitably all controllers have to do some amount of defragmentation or garbage collection in order to continue operating at high speeds. When and how an SSD decides to run its defrag or cleanup routines directly impacts the user experience as inconsistent performance results in application slowdowns.

To test IO consistency, we fill a secure erased SSD with sequential data to ensure that all user accessible LBAs (Logical Block Addresses) have data associated with them. Next we kick off a 4KB random write workload across all LBAs at a queue depth of 32 using incompressible data. The test is run for just over half an hour and we record instantaneous IOPS every second.

We are also testing drives with added over-provisioning by limiting the LBA range. This gives us a look into the drive’s behavior with varying levels of empty space, which is frankly a more realistic approach for client workloads.

Each of the three graphs has its own purpose. The first one is of the whole duration of the test in log scale. The second and third one zoom into the beginning of steady-state operation (t=1400s) but on different scales: the second one uses log scale for easy comparison whereas the third one uses linear scale for better visualization of differences between drives. Click the dropdown selections below each graph to switch the source data.

For more detailed description of the test and why performance consistency matters, read our original Intel SSD DC S3700 article.

Default	Transcend SSD370 256GBTranscend SSD370 512GBADATA Premier SP610 256GBADATA Premier SP610 512GBCorsair Force LS 240GBCorsair Neutron XT 240GBCorsair Neutron XT 480GBCorsair Neutron XT 960GBCrucial MX100 256GBCrucial MX100 512GBIntel SSD 730 480GBIntel SSD Pro 2500 240GBJMicron JMF667H 240GBMicron M600 256GBMicron M600 1TBOCZ ARC 100 240GBOCZ Vector 150 240GBOCZ Vertex 460 240GBOCZ RevoDrive 350 480GBPlextor M6S 256GBSamsung SSD 840 EVO 250GBSamsung SSD 840 EVO mSATA 1TBSamsung SSD 850 EVO 250GBSamsung SSD 850 EVO 500GBSamsung SSD 850 EVO 1TBSamsung SSD 850 Pro 256GBSamsung SSD 850 Pro 1TBSamsung XP941 256GBSamsung XP941 512GBSanDisk Ultra II 240GBSanDisk Extreme Pro 240GBSanDisk Extreme Pro 480GBSanDisk Extreme Pro 960GBSanDisk Extreme II 480GBSanDisk Ultra Plus 256GB
25% Over-Provisioning	Transcend SSD370 256GBTranscend SSD370 512GBADATA Premier SP610 256GBADATA Premier SP610 512GBCorsair Force LS 240GBCorsair Neutron XT 240GBCorsair Neutron XT 480GBCorsair Neutron XT 960GBCrucial MX100 256GBCrucial MX100 512GBIntel SSD 730 480GBIntel SSD Pro 2500 240GBJMicron JMF667H 240GBMicron M600 256GBMicron M600 1TBOCZ ARC 100 240GBOCZ Vector 150 240GBOCZ Vertex 460 240GBOCZ RevoDrive 350 480GBPlextor M6S 256GBSamsung SSD 840 EVO mSATA 1TBSamsung SSD 850 EVO 250GBSamsung SSD 850 EVO 500GBSamsung SSD 850 EVO 1TBSamsung SSD 850 Pro 256GBSamsung SSD 850 Pro 1TBSamsung XP941 256GBSamsung XP941 512GBSanDisk Ultra II 240GBSanDisk Extreme Pro 240GBSanDisk Extreme Pro 480GBSanDisk Extreme Pro 960GBSanDisk Extreme II 480GBSanDisk Ultra Plus 256GB

Despite the custom Transcend firmware, performance consistency is an exact match with ADATA's SP610. I'm suspecting that the reason for low steady-state performance might be the hardware because the SM2246EN is a single-core design. Most controller designs today are multicore because today's NAND requires a lot of management and with multiple cores the NAND management can be dedicated to one or more cores, which leaves the rest of the cores available for host IO processing. In Silicon Motion's case, the one core has to take care of everything from host IOs to NAND management, which translates to lower overall performance as the controller can't keep up with everything that needs to be done.

Default	Transcend SSD370 256GBTranscend SSD370 512GBADATA Premier SP610 256GBADATA Premier SP610 512GBCorsair Force LS 240GBCorsair Neutron XT 240GBCorsair Neutron XT 480GBCorsair Neutron XT 960GBCrucial MX100 256GBCrucial MX100 512GBIntel SSD 730 480GBIntel SSD Pro 2500 240GBJMicron JMF667H 240GBMicron M600 256GBMicron M600 1TBOCZ ARC 100 240GBOCZ Vector 150 240GBOCZ Vertex 460 240GBOCZ RevoDrive 350 480GBPlextor M6S 256GBSamsung SSD 840 EVO 250GBSamsung SSD 840 EVO mSATA 1TBSamsung SSD 850 EVO 250GBSamsung SSD 850 EVO 500GBSamsung SSD 850 EVO 1TBSamsung SSD 850 Pro 256GBSamsung SSD 850 Pro 1TBSamsung XP941 256GBSamsung XP941 512GBSanDisk Ultra II 240GBSanDisk Extreme Pro 240GBSanDisk Extreme Pro 480GBSanDisk Extreme Pro 960GBSanDisk Extreme II 480GBSanDisk Ultra Plus 256GB
25% Over-Provisioning	Transcend SSD370 256GBTranscend SSD370 512GBADATA Premier SP610 256GBADATA Premier SP610 512GBCorsair Force LS 240GBCorsair Neutron XT 240GBCorsair Neutron XT 480GBCorsair Neutron XT 960GBCrucial MX100 256GBCrucial MX100 512GBIntel SSD 730 480GBIntel SSD Pro 2500 240GBJMicron JMF667H 240GBMicron M600 256GBMicron M600 1TBOCZ ARC 100 240GBOCZ Vector 150 240GBOCZ Vertex 460 240GBOCZ RevoDrive 350 480GBPlextor M6S 256GBSamsung SSD 840 EVO mSATA 1TBSamsung SSD 850 EVO 250GBSamsung SSD 850 EVO 500GBSamsung SSD 850 EVO 1TBSamsung SSD 850 Pro 256GBSamsung SSD 850 Pro 1TBSamsung XP941 256GBSamsung XP941 512GBSanDisk Ultra II 240GBSanDisk Extreme Pro 240GBSanDisk Extreme Pro 480GBSanDisk Extreme Pro 960GBSanDisk Extreme II 480GBSanDisk Ultra Plus 256GB

 

Default	Transcend SSD370 256GBTranscend SSD370 512GBADATA Premier SP610 256GBADATA Premier SP610 512GBCorsair Force LS 240GBCorsair Neutron XT 240GBCorsair Neutron XT 480GBCorsair Neutron XT 960GBCrucial MX100 256GBCrucial MX100 512GBIntel SSD 730 480GBIntel SSD Pro 2500 240GBJMicron JMF667H 240GBMicron M600 256GBMicron M600 1TBOCZ ARC 100 240GBOCZ Vector 150 240GBOCZ Vertex 460 240GBOCZ RevoDrive 350 480GBPlextor M6S 256GBSamsung SSD 840 EVO 250GBSamsung SSD 840 EVO mSATA 1TBSamsung SSD 850 EVO 250GBSamsung SSD 850 EVO 500GBSamsung SSD 850 EVO 1TBSamsung SSD 850 Pro 256GBSamsung SSD 850 Pro 1TBSamsung XP941 256GBSamsung XP941 512GBSanDisk Ultra II 240GBSanDisk Extreme Pro 240GBSanDisk Extreme Pro 480GBSanDisk Extreme Pro 960GBSanDisk Extreme II 480GBSanDisk Ultra Plus 256GB
25% Over-Provisioning	Transcend SSD370 256GBTranscend SSD370 512GBADATA Premier SP610 256GBADATA Premier SP610 512GBCorsair Force LS 240GBCorsair Neutron XT 240GBCorsair Neutron XT 480GBCorsair Neutron XT 960GBCrucial MX100 256GBCrucial MX100 512GBIntel SSD 730 480GBIntel SSD Pro 2500 240GBJMicron JMF667H 240GBMicron M600 256GBMicron M600 1TBOCZ ARC 100 240GBOCZ Vector 150 240GBOCZ Vertex 460 240GBOCZ RevoDrive 350 480GBPlextor M6S 256GBSamsung SSD 840 EVO mSATA 1TBSamsung SSD 850 EVO 250GBSamsung SSD 850 EVO 500GBSamsung SSD 850 EVO 1TBSamsung SSD 850 Pro 256GBSamsung SSD 850 Pro 1TBSamsung XP941 256GBSamsung XP941 512GBSanDisk Ultra II 240GBSanDisk Extreme Pro 240GBSanDisk Extreme Pro 480GBSanDisk Extreme Pro 960GBSanDisk Extreme II 480GBSanDisk Ultra Plus 256GB

TRIM Validation

To test TRIM, I filled a 128GB SSD370 with sequential 128KB data and proceeded with a 30-minute random 4KB write (QD32) workload to put the drive into steady-state. After that I TRIM'ed the drive by issuing a quick format in Windows and ran HD Tach to produce the graph below.

And TRIM works as expected.

AnandTech Storage Bench 2013

Our Storage Bench 2013 focuses on worst-case multitasking and IO consistency. Similar to our earlier Storage Benches, the test is still application trace based – we record all IO requests made to a test system and play them back on the drive we are testing and run statistical analysis on the drive's responses. There are 49.8 million IO operations in total with 1583.0GB of reads and 875.6GB of writes. I'm not including the full description of the test for better readability, so make sure to read our Storage Bench 2013 introduction for the full details.

AnandTech Storage Bench 2013 - The Destroyer
Workload	Description	Applications Used
Photo Sync/Editing	Import images, edit, export	Adobe Photoshop CS6, Adobe Lightroom 4, Dropbox
Gaming	Download/install games, play games	Steam, Deus Ex, Skyrim, Starcraft 2, BioShock Infinite
Virtualization	Run/manage VM, use general apps inside VM	VirtualBox
General Productivity	Browse the web, manage local email, copy files, encrypt/decrypt files, backup system, download content, virus/malware scan	Chrome, IE10, Outlook, Windows 8, AxCrypt, uTorrent, AdAware
Video Playback	Copy and watch movies	Windows 8
Application Development	Compile projects, check out code, download code samples	Visual Studio 2012

We are reporting two primary metrics with the Destroyer: average data rate in MB/s and average service time in microseconds. The former gives you an idea of the throughput of the drive during the time that it was running the test workload. This can be a very good indication of overall performance. What average data rate doesn't do a good job of is taking into account response time of very bursty (read: high queue depth) IO. By reporting average service time we heavily weigh latency for queued IOs. You'll note that this is a metric we have been reporting in our enterprise benchmarks for a while now. With the client tests maturing, the time was right for a little convergence.

What surprises me is that the SSD370 does so well in the 2013 Storage Bench even though its IO consistency is leaves a lot to be desired. Again the performance is a very close match with the SP610 and overall the SSD370 is a fairly capable middle-class performer.

AnandTech Storage Bench 2011

Back in 2011 (which seems like so long ago now!), we introduced our AnandTech Storage Bench, a suite of benchmarks that took traces of real OS/application usage and played them back in a repeatable manner. The MOASB, officially called AnandTech Storage Bench 2011 – Heavy Workload, mainly focuses on peak IO performance and basic garbage collection routines. There is a lot of downloading and application installing that happens during the course of this test. Our thinking was that it's during application installs, file copies, downloading and multitasking with all of this that you can really notice performance differences between drives. The full description of the Heavy test can be found here, while the Light workload details are here.

The same story continues in our 2011 Workloads. The performance isn't 850 Pro level, but the SSD370 is competitive among other value-oriented drives. The performance of the 128GB model is a bit of a letdown, but that was expected since 128Gbit NAND has a severe impact on parallelism at 128GB and lower.

Random Read/Write Speed

The four corners of SSD performance are as follows: random read, random write, sequential read and sequential write speed. Random accesses are generally small in size, while sequential accesses tend to be larger and thus we have the four Iometer tests we use in all of our reviews.

Our first test writes 4KB in a completely random pattern over an 8GB space of the drive to simulate the sort of random access that you'd see on an OS drive (even this is more stressful than a normal desktop user would see). We perform three concurrent IOs and run the test for 3 minutes. The results reported are in average MB/s over the entire time.

Random performance is a match with the SP610 too. The SSD370 isn't the fastest drive on earth, but its performance is decent given its value focus.

Sequential Read/Write Speed

To measure sequential performance we run a 1 minute long 128KB sequential test over the entire span of the drive at a queue depth of 1. The results reported are in average MB/s over the entire test length.

No surprises in sequential performance either. At 128GB and 256GB the write speeds are limited due to lesser parallelism, but in both cases the performance is fairly normal to drives with 128Gbit NAND.

AS-SSD Incompressible Sequential Read/Write Performance

The AS-SSD sequential benchmark uses incompressible data for all of its transfers. The result is a pretty big reduction in sequential write speed on SandForce based controllers, but most other controllers are unaffected.

Performance vs. Transfer Size

ATTO is a useful tool for quickly benchmarking performance across various transfer sizes. All capacities of the SSD370 for fairly well at all IO sizes and there's nothing concerning that catches my eye.

Click for full size

Power Consumption

As Transcend disabled device initiated power magament (i.e. DIPM), the slumber power consumption is significantly higher compared to drives that have DIPM enabled. For desktops that isn't a big deal because there is no battery life concern, but laptop users are better off with a drive that properly supports DIPM. 

Power consumption under load is also a bit high given that many modern drives are able to stay below 3W.

Final Words

From the performance perspective, the SSD370 is a very competitive value drive. It doesn't top the charts, but it provides a very good balance of consistency and peak performance. In most workloads, particularly our real-world testing, the SSD370 performs better than the MX100 and Ultra II, which have been my go-to value drives. Under very intensive workloads, the ARC 100 is slightly faster thanks to its great consistency, but most people who are looking for value drives won't have such harsh usage anyway. 

The lack of DIPM support and the high slumber power consumption is a bit disappointing, though. Given how similar modern SSDs are in terms of performance, the power consumption really matters because additional battery life is easier to notice and more concrete than a few percent increase in performance. Obviously that doesn't apply to desktop users, but the majority of PCs are laptops now, so it just seems illogical to disable DIPM. Hopefully that's something Transcend can enable through a firmware update.

NewEgg Price Comparison (1/23/2015)
	120/128GB	240/250/256GB	480/500/512GB	960GB/1TB
Transcend SSD370	$73	$111	$190	$404
Transcend SSD340	$74	$105	-	-
Samsung SSD 850 EVO	$86	$140	$235	$476
Samsung SSD 850 Pro	$118	$170	$367	$630
SanDisk Extreme Pro	-	$150	$260	$508
SanDisk Ultra II	$70	$110	$215	$405
Crucial MX100	$70	$109	$214	-
Plextor M6S	$76	$130	$270	-
Intel SSD 730	-	$160	$318	-
Intel SSD 530	$80	$128	$240	-
OCZ ARC 100	$70	$99	$190	-

The pricing of the SSD370 is extremely competitive. It's practically undercutting the MX100 and Ultra II, which makes it one of the cheapest value drives on the market, and the SSD370 is also listed at even lower prices on Amazon Prime right now. Only the ARC 100 is cheaper, but on the other hand it also lacks support for low power states and doesn't come in 1TB capacity either.

I don't hand out the "Recommended by AnandTech" award very often, but I think it's justified in this case. It's not an overstatement to say that the SSD370 is overall the highest performing value drive and on top of that the pricing is very alluring. I would still, however, recommend the MX100 and Ultra II for users that are concerned about battery life because of SSD370's high idle power consumption. If Transcend SSD370 was able to fix that through a firmware update, it would be safe to say that the SSD370 would be the best value SSD on the market.