• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Silicon Power UD70 2 TB

W1zzard

Administrator
Staff member
Joined
May 14, 2004
Messages
28,955 (3.75/day)
Processor Ryzen 7 5700X
Memory 48 GB
Video Card(s) RTX 4080
Storage 2x HDD RAID 1, 3x M.2 NVMe
Display(s) 30" 2560x1600 + 19" 1280x1024
Software Windows 10 64-bit
Priced at only $210 for the tested 2 TB version, the Silicon Power UD70 is highly affordable, yet offers good performance thanks to the combination of Micron 96-layer QLC flash and Phison E12 controller. Unlike some competing value-oriented drives, a DRAM cache is included, too.

Show full review
 
Have been using SP's M2's for 3 years now and if they keep this low price per GB over time (cough cough Chia mining), next year I'll upgrade
 
2Tb with 530 TBW ? quite low.
 
Could the high temps have affected write speeds?
Over 80c seems quite high to me and most likely caused some throttling.
 
I'll still chose TLC over QLC every time.

QLC's horrible drawbacks were supposed to be offset by a 25% cost reduction. I ain't paying TLC prices for a QLC product, sorry.

With HMB now a viable thing that makes $0.10/GB TLC drives very capable jack-of-all-trades with no real downsides, why would you even consider a more expensive QLC drive with glaring raw-write deficiencies and obviously worse endurance?
 
Theres always one more clueless sucker you can peddle your shitty QLCs to unfortunately.
Obviously us who are informed and have a clue will stay clear of those because we can just grab the next TLC at the same price, but unfortunately a lot of people are ignorant and will just buy what is the cheapest, even if its only 1% which is really sad.
QLC certainly has potential as a more cost-efficient technology for SSDs but unfortunately, not like this. I want to see QLCs below $0.06/GB (realistically closer to $0.05 per) before I'd seriously consider one (with TLC being around $0.09-$0.1).
 
Could the high temps have affected write speeds?
Over 80c seems quite high to me and most likely caused some throttling.
the chart clearly shows that the write speed does not change?
blue line = write speed
red line = temperature

a lot of people
The majority of those, with their ultra-light workloads, wouldn't even notice the difference between QLC and TLC
 
The majority of those, with their ultra-light workloads, wouldn't even notice the difference between QLC and TLC
I think if this came installed in a laptop you'd be okay with it. It's QLC but the SLC cache is plenty good enough for use as an OS drive and light-duty applications. You wouldn't choose to buy it yourself over better and faster and higher-endurance and cheaper drives, but it's definitely fit for purpose and I'd be happy to have one in a prebuilt.

The consumer experience, whilst SSD-class, is relatively terrible compared to the other drives you've tested; The slowest windows boot drive in your dataset except the flawed and terrible BX500 and poor showings in the program installation (windows updates are implied here)/search/AV scan - arguably the four most important cornerstones of a typical consumer SSD experience.

Another typical use case for a large and cheap drive like this would be a capacity upgrade from a previous drive - if someone was migrating data from a full 1TB drive to this new 2TB drive, the very first experience of using the new SSD could run into the abysmal QLC raw write rate. I'm less concerned about this because it's not a common occurence - once or maybe twice ever in a drive's consumer lifespan; It would just suck as a first impression!

One thing that also isn't tested but an important feature for 15W ultraportables is power consumption and most QLC drives perform poorly here as the controller has to work twice as hard for the 33% capacity improvement over TLC. It has been a while since I saw a detailed power consumption test - probably because NVMe power consumption is much harder to test than SATA, but bad TLC SATA drives could use enough power that they'd significantly hurt battery life in a thin&light laptop over a more efficient MLC one.
 
[ ... ]
The majority of those, with their ultra-light workloads, wouldn't even notice the difference between QLC and TLC
Oh, I'm aware of that. It's just that as long as you can peddle QLC at like, 95% of the price of TLC to suckers we're not seeing QLC drop to much saner pricing compared to TLC (like, lets say between 65% and 70%), which is quite unfortunate because there's some serious potential in QLC's price advantage if only it manifested.
 
Oh, I'm aware of that. It's just that as long as you can peddle QLC at like, 95% of the price of TLC to suckers we're not seeing QLC drop to much saner pricing compared to TLC (like, lets say between 65% and 70%), which is quite unfortunate because there's some serious potential in QLC's price advantage if only it manifested.
Agree 100%
 
Back
Top