Why you can trust Tom's Hardware Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test.

Page 10 of 19:

Sequential 80 Percent Read Mixed Workload

SATA is a half-duplex interface, so when a hard drive or SSD is asked to read and write at the same time, the drive needs to prioritize and schedule the transaction. Some products handle this task better than others. The 80-percent read marker is an industry standard measuring point for client workloads. This test better represents daily use over the isolated test that uses 100-percent read or 100-percent write loads.

The larger-capacity SSD370 delivers higher performance than the 256GB model in the sequential mixed workload test. Again, this doesn't surprise us because the sequential write performance is lower. Compared to other products on the market, the Transcend SSD370 512GB delivers just middle-of-the-road performance and fails to stand out from the crowd and, more importantly, the Samsung 850 EVO 500GB.

Current page: Sequential 80 Percent Read Mixed Workload

Prev Page Random Write Next Page Random 80 Percent Read Mixed Workload

See more SSDs Reviews

TOPICS

Chris Ramseyer is a Contributing Editor for Tom's Hardware US. He tests and reviews consumer storage.

Sabrent's USB Type-C M.2 SSD Enclosure reduced to just $22 at Amazon

'World's first' AMD GPU driven via USB3 — Tiny Corp tests eGPUs on Apple Silicon, Linux and Windows also supported

Sandisk teases 'dynamite' new SSD controller dubbed Stargate

See more latest

16 Comments Comment from the forums

synphul

Good info to have. More problematic than the confusing product model numbers correlating to plastic vs aluminum bodies I think are the differences in the internals. I'm not sure how this is considered proper practice since changing the controller and the nand flash is basically selling a different quality drive even if the specs are loosely similar. Makes people like myself wonder how they can get away with this and yet nvidia had to answer to a class action lawsuit over the layout of their vram when in fact they did have the full 4gb of vram on the card. It just reeks of bait and switch. Especially when as mentioned reviewers are often given cherry picked units that don't necessarily reflect the common resale market most consumers will be met with.

It would be nice to see manufacturers give review sites a voucher instead, rather than sending a unit directly for testing. Good for 1 purchase of the part slated for review from the retailer of their choice to ensure it's random and more indicative of the lottery a consumer would face. That is so long as they were confident that a random off the shelf unit would fairly represent their products rather than one specially set aside for review.

It may not seem like such a big deal but with the guts swapped out for lower cost and potentially lower quality alternatives that's an issue. I wouldn't be happy purchasing a top tier tool set only to find out it included a number of budget store brand replacements in lieu of what I paid for just because it's 'basically' the same. In just about any other scenario this would create great backlash. Why is it just accepted as common practice in the form of ssd drives? Not that I'm sue happy or think lawsuits are the solution in an attempt for gains but some sort of reassurance that the customer is getting what they're told they're buying rather than sold on a song and given something entirely different.
Reply
crabdog

+1 for this review. It's nice to see someone report the truth and their real concerns without sugar coating.
Reply
Bannereus

good to see that not every product is top of the line, gives meaning to those that earn top recommendations

but please fix Section 5. "Test PERIMETERS ..."
Reply
Wisecracker

""Paralysis by Over-Analysis""
(leads to ridiculous conclusions)

The real-world tests show there is generally one percent or less difference between the drives tested, so the bottom line is ...

buy whatever SSD (with a 3-yr warranty) that is on sale.

Reply
JPNpower

Hey, Kingston did it, and they got off well enough
Reply
fudgecakes99

i bought one for 160 from amazon 512gb's, on sale. Have yet to boot it as skylate 6700k has yet to be released in america, from the charts i dont' see a major difference save about 70-90 mb's write difference, as someone moving up from a 5200rpm hdd i expect great things.
Reply
CRamseyer

Kingston's V300 went from the top ten selling SSDs on Amazon to off of the list when the info was picked up by review site news teams.
Reply
shrapnel_indie

Disappointing to see price-point win over quality... yet again.

Manufacturers have been sending cherry-picked "review" units of products for decades, nothing new there... sadly it's to get the best reviews possible while hoping consumers won't be able to tell the difference in the actual product.

The "product specification may change" disclaimer has been around for a while. Sometimes it is used due to certain components going away, becoming too expensive (for the price point the bean counters and marketing decide on), or (less seen as it means they care about the product and reputation) quality goes south. Sadly it is more often abused just to make sure tier-1/top tier executives get their (large as they can get away with) bonus checks.
Reply
kalmquist

The Crucial BX100 line uses the same controller as the Transcend SSD370. I've heard no reports of Crucial using cut-rate NAND, and I wouldn't be expect them to since Cruical (aka Micron) makes its own NAND. Recently, prices have been lower for the BX100 than the SSD370. It seems like the BX100 would be a better option for anyone who might consider an SSD370.

Reply
Dark Falz

I got a 256 and 512 about a year ago, main concern was price. I wanted the MX100 instead but couldn't find any in stock at the time. Have served me fine but definitely weren't worth the money. Has received no Firwmare updates/improvements (which I suppose is not always necessarily a bad thing, look at the 840). The plastic housing doesn't bother me except for the fact there's no temperature sensor in the SMART data, so I can't tell how warm it gets. Will probably upgrade to a 1 or 2 TB when they become more affordable and put this on my laptop instead. It performs much like any other SSD but disappointing to see it fare so badly in the theoreticals (when I bought it, there were not many reviews).
Reply