Can you _please_ sort the SSD's by worst-case ("disk full") numbers rather than best-case ones? I generally really like your reviews, but your sorting is simply wrong, and some horribly bad ssd's end up looking much better than they are because of it.
Particularly for things like the 99% latency numbers it is inane to sort by the best case, since the whole point is about near-worst-case latencies, and bad controller should simply not be given the benefit of the doubt.
Note that unless you actually trim the ssd, even an empty filesystem will act like a full one, since the ssd doesn't know which parts are used. So as far as the ssd is concerned, it's all full. So the argument that "most people have lots of room on their disk" is quite likely bogus to begin with, but possibly entirely irrelevant even if it were to be true.
Death666Angel: please learn to read. They are *not* "sorted by worst-case, just in reverse".
Look at the "ATSB - Heavy (99th Percentile Latency)" graph, just as an example.
In particular, look at the ADATA XPG SX8200 Pro 1TB one. Look at how absolutely *HORRIBLE* the latency is for that.
Yet the idiotic and incorrect sorting shows it as the second-best SSD on that list, because the *best-case* latency when the drive is empty is reasonable. But once it gets full, and $
Anybody who thinks that that drive should be second-best on that list is incompetent.
I still stand by "If you have an OS that uses GC as a valid algorithm, you desever all the crap you brought upon yourself." But have fun being a blast at parties! Learn to read fricking diagrams and stop bitching. Or start being the change you wanna see in the world!
the issue is only with the dramless drives when they are above 60-70% full witch you should avoid (the sandisk/WD blue recant controller is cida dramless but it has 10mb of ram on the controller it self witch seems to be enough to mitigate the lack of a full blown dram)
I see that you are still in denial about how it was you who couldn't read diagrams. I'd also like to point out that I'm actually trying to be the change I want to see exactly by asking for the sorting to be fixed.
Because the sorting clearly is wrong. I pointed out a very stark example of when a much worse drive sorts above the better ones.
Also, your lack of reading comprehension is showing in how you think this is only about TRIM. As I said, this is about disk full situations. And even with TRIM, the disk may simply be close to full. Not everybody buys an SSD that is twice as big as it needs to be.
I was also pointing out that even if your drive isn't full, it may well show the full behavior in reality.
Who uses their SSDs full to the brim and in sustained write mode? Honestly, that scenario is not even realistic for properly managed DB servers, let alone in client systems where the only wait time which actually happens is during system boot/application launch/data load on up to 80% full (in Anandtech-speak "empty" system). Client writes are all cached first and the write itself happens in background, the user does not have to wait anything.
AT does not even test this scenario properly, even their "Light" test is WAY too write-heavy for that.
A synthetic which would reflect that is something like "64kb random read" (runs are 16 clusters=64k on NTFS, and most DLLs are close to that size).
660P from Intel is $109 for 1 TB....; even though it is 'only' 2x PCI-e lanes capable, it is still more than 'snappy' for that sort of cost/capacity ratio....
"The write endurance ratings are still competitive with high-end drives that offer five year warranties"
The MP34 has over twice the endurance of any SSD utilizing the SM2262 with Micron NAND. I apologize, but I'm not understanding what you mean by "still competitive". Seems as if Phison is outclassing the competition in certain regards. A small sacrifice in performance for exceptionally more endurance.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
17 Comments
Back to Article
Samus - Wednesday, May 15, 2019 - link
Why is the WD Black missing from all the benchmarks - even the recently reviewed SN750 is missing?I’m at a loss here, you specifically mentioned it on the first page of the article, along with Samsung, yet included all the Samsung drives...
futrtrubl - Wednesday, May 15, 2019 - link
Last page. "..if we had the chance to test the 500GB WD Black SN750"kobblestown - Wednesday, May 15, 2019 - link
FWIW, I just bought a 480GB Corsair MP510 and the firmware is reported as ECFM12.2. I don't know if it's available for update of older devices though.ssd-user - Wednesday, May 15, 2019 - link
Can you _please_ sort the SSD's by worst-case ("disk full") numbers rather than best-case ones? I generally really like your reviews, but your sorting is simply wrong, and some horribly bad ssd's end up looking much better than they are because of it.Particularly for things like the 99% latency numbers it is inane to sort by the best case, since the whole point is about near-worst-case latencies, and bad controller should simply not be given the benefit of the doubt.
Note that unless you actually trim the ssd, even an empty filesystem will act like a full one, since the ssd doesn't know which parts are used. So as far as the ssd is concerned, it's all full. So the argument that "most people have lots of room on their disk" is quite likely bogus to begin with, but possibly entirely irrelevant even if it were to be true.
Death666Angel - Thursday, May 16, 2019 - link
They are storted by worst-case, just in reverse. And if you use an SSD without an automatic trim OS, it's kinda on you, isn't it?ssd-user - Thursday, May 16, 2019 - link
Death666Angel: please learn to read. They are *not* "sorted by worst-case, just in reverse".Look at the "ATSB - Heavy (99th Percentile Latency)" graph, just as an example.
In particular, look at the ADATA XPG SX8200 Pro 1TB one. Look at how absolutely *HORRIBLE* the latency is for that.
Yet the idiotic and incorrect sorting shows it as the second-best SSD on that list, because the *best-case* latency when the drive is empty is reasonable. But once it gets full, and $
Anybody who thinks that that drive should be second-best on that list is incompetent.
ssd-user - Thursday, May 16, 2019 - link
Fat-fingered the response. The "and $" should be "and garbage collection happens, latency becomes horrid".Death666Angel - Saturday, May 18, 2019 - link
I still stand by "If you have an OS that uses GC as a valid algorithm, you desever all the crap you brought upon yourself." But have fun being a blast at parties! Learn to read fricking diagrams and stop bitching. Or start being the change you wanna see in the world!leexgx - Saturday, May 18, 2019 - link
the issue is only with the dramless drives when they are above 60-70% full witch you should avoid (the sandisk/WD blue recant controller is cida dramless but it has 10mb of ram on the controller it self witch seems to be enough to mitigate the lack of a full blown dram)ssd-user - Sunday, May 19, 2019 - link
Exactly. The point is that you should avoid those drives.Which is why they shouldn't show up at the top of the charts. They are not top drives, they are the dregs, and they should show up as such.
ssd-user - Sunday, May 19, 2019 - link
I see that you are still in denial about how it was you who couldn't read diagrams. I'd also like to point out that I'm actually trying to be the change I want to see exactly by asking for the sorting to be fixed.Because the sorting clearly is wrong. I pointed out a very stark example of when a much worse drive sorts above the better ones.
Also, your lack of reading comprehension is showing in how you think this is only about TRIM. As I said, this is about disk full situations. And even with TRIM, the disk may simply be close to full. Not everybody buys an SSD that is twice as big as it needs to be.
I was also pointing out that even if your drive isn't full, it may well show the full behavior in reality.
Sorry for not being your ideal party buddy.
peevee - Monday, May 20, 2019 - link
Who uses their SSDs full to the brim and in sustained write mode? Honestly, that scenario is not even realistic for properly managed DB servers, let alone in client systems where the only wait time which actually happens is during system boot/application launch/data load on up to 80% full (in Anandtech-speak "empty" system).Client writes are all cached first and the write itself happens in background, the user does not have to wait anything.
AT does not even test this scenario properly, even their "Light" test is WAY too write-heavy for that.
A synthetic which would reflect that is something like "64kb random read" (runs are 16 clusters=64k on NTFS, and most DLLs are close to that size).
MDD1963 - Wednesday, May 15, 2019 - link
660P from Intel is $109 for 1 TB....; even though it is 'only' 2x PCI-e lanes capable, it is still more than 'snappy' for that sort of cost/capacity ratio....peevee - Monday, May 20, 2019 - link
Why do they even use x4 PCIe when they cannot even saturate x2? Really, peak read of 1.4GB/s is pathetic.DyneCorp - Monday, June 17, 2019 - link
"The write endurance ratings are still competitive with high-end drives that offer five year warranties"The MP34 has over twice the endurance of any SSD utilizing the SM2262 with Micron NAND. I apologize, but I'm not understanding what you mean by "still competitive". Seems as if Phison is outclassing the competition in certain regards. A small sacrifice in performance for exceptionally more endurance.
DyneCorp - Monday, June 17, 2019 - link
Metrics*, not regards ha.crimson117 - Friday, April 24, 2020 - link
Looks like the new MP34's offer a 5-year warranty:256GB - TM8FP4256G0C101
512GB - TM8FP4512G0C101
1TB - TM8FP4001T0C101
https://www.teamgroupinc.com/en/product/mp34
https://www.teamgroupinc.com/en/catalog/act.php?ac...