See:
http://www.tomshardware.com/news/samsung-950-pro-ssd-nvme-3d-vnand,30135.html
They state they learned a lot from their previous M.2 models via their customers.
But they certainly didn't do anything with that information.
Same UBX controller (runs too HOT!!!).
7 Watts Burst power consumption (????).
512GB max capacity (1TB promised in 2016... yawn...).
$349 for 512GB seems reasonable on the surface. But if the sustained over time performance is still below SATA3 SSD's, then this is just another ripoff move from Samsung.
If I'm wrong on the heat/power/throttling issues I'll apologize with a hat in my mouth.
But if this is what it seems to me (Samsung trying to cash in on old/useless tech...), then I hope they don't make one cent from this new 'retail' line. And finally get it drilled into their heads that we want the whole package (balance, baby, balance), not some idealistic semblance of it.
I know I won't waste my time testing this product for them with the spec's as indicated.
Let's see what a (good/great) review shows us about this regurgitation of M.2 done wrong. Again?
-
tilleroftheearth Wisdom listens quietly...
-
tilleroftheearth Wisdom listens quietly...
Can't believe another SSD can be added so quickly and, it's not Samsung!
lol...
See:
http://www.tomshardware.com/news/avexir-s100-ssd,30143.html
What makes this SSD one to avoid? JMicron ring any warning bells for anyone?
And 'bling' on an SSD. Gawd... we're not 14 anymore... sigh... sigh..
and sigh...
If it sells in the $0.10 per GB range, it might be worth recommending to my clients' younger children. -
http://www.pcper.com/news/Storage/Samsung-Launches-950-PRO-300000-IOPS-and-25-GBsec-M2-V-NAND-SSD
"Equipped with Samsung’s 32-layer V-NAND and using the NVMe protocol enabled by a new UBX controller..."
And then the Author later stated in the comments:
"They stated during Q+A that this drive will only draw ~6 Watts and that they have 'changed their algorithm' to make the heat related throttling a non-issue. It's also worth stating that the SM951's are fine so long as you have a *minimum* of air flow over them or simply don't write to them at a continuous >1GB/sec."Last edited: Sep 23, 2015 -
tilleroftheearth Wisdom listens quietly...
So you think it's okay to expect this to be used at 40% of it's rated speed? What's the point?
And how exactly do you limit it to 1GB/s anyways?
But then again, it can only be used for mere seconds (at rated speed) before Samsung will declare it abused for warranty purposes.
These are not performance drives. They're 3's dressed up as a 9...
This is SF, take two... with all their duracrap marketing bs (no, limiting the performance is for your benefit... and theirs, to get them past warranty and you out of their hair)...
-
I feel like you're over reacting. Why not just wait until reviews on the unit comes out? Anandtech and PCPER usually do some good in depth reviews.
You're acting like Samsung is the Devil incarnate. -
tilleroftheearth Wisdom listens quietly...
It's not?
They game benchmarks on everything else including their ssd's. Just has to be a voice against them with their fail products.
Not all (the 2TB models are interesting, one more than the other...), but they don't compete on price or performance in real world use. But they do have one hell of a marketing campaign, right? (Yeah; I'm talking about the reviewers that don't push their 'testing' past idiotic bm's 'scores' - and that includes the majority of them out there).
-
-
tilleroftheearth likes this.
-
-
-
-
Micron/Crucial forever. No it's not the fastest, but it sure as hell is the most reliable and least prone to crap like this.
Jarhead and tilleroftheearth like this. -
-
-
http://forum.crucial.com/t5/Crucial-SSDs/MX100-will-not-boot-sometimes/td-p/158815
http://www.anandtech.com/show/5424/crucial-provides-a-firmware-update-for-m4-to-fix-the-bsod-issue
http://www.anandtech.com/show/3694/crucial-releases-realssd-c300-firmware-fix
http://ccm.net/faq/29514-crucial-m4-how-to-remove-the-5200-hours-bug
http://forums.macrumors.com/threads/do-not-update-crucial-m4s-firmware-to-000f.1357662/
https://bugzilla.kernel.org/show_bug.cgi?id=71371
All i'm saying is, there is no perfect manufacturer. Just some that are better than others.
Only company i have't really seen complaints about is Sandisk in relation to SSDs, and that's because they're relatively new to the SSD market and they also copy what everyone has already done (along with avoiding their mistakes), but unfortunately for Sandisk, they've been losing money lately. -
Yeah...C300 and M4 are ancient SSDs by today's standards, though I've never had problems updating firmware on either. I'm sure other forum members would chime in to say the Crucial/Micron drives are among the most reliable on the market. After deploying dozens and dozens of M4s, M500s, M550s, and M600s, zero have returned to me for problems of any sort, whereas this Samsung problems are fresh on modern drives, which is rather inexcusable.
Obviously, your mileage may vary.Jarhead, TomJGX and tilleroftheearth like this. -
i'm pretty sure the BX100 and the MX100 had issues where they refused to be seen in the BIOS because of initial FW.
There aren't really many Samsung issues out there. The ones that are have been fixed.
The read speed bug has been fixed on both the 840 EVO 2.5" and M-sata.
From the looks of it, Samsung might be doing something for the regular 840 as well thanks to Allyn's pushing.( http://www.overclock.net/t/1574166/ssd-advice/10#post_24437468)
Queued TRIM in Linux. They haven't fixed this one yet, but its not vital. Just disable Queued TRIM or update your Kernal so it has the blacklist and regular TRIM will run on the drive. (Crucial drives had the same issue)
Data Loss bug with TRIM. It was found out that the issue was not cause by the Samsung SSDs, but by the Linux kernal. ( http://techreport.com/news/28724/samsung-docs-detail-linux-trim-bug-and-fix)
Heat with the SM951. I really don't think its that big a deal. I was thinking about this earlier today and the only ppl i've seen complaining about this is regular consumers. This is an OEM/Enterprise level drive, which would mean it is meant to be sold to Businesses and OEM manufacturers that would modify the drive anyway (heatsinks, cooling setups, etc.), its not meant to be sold to regular ppl, even enthusiasts. So i think the 950 PRO is not going to have the same issues the SM951 does. -
tilleroftheearth Wisdom listens quietly...
The heat/throttling and low QD1 performance vs. 2.5" SATA3 based SSD's is real and it is a big deal if you want to extract the maximum performance offered/marketed for more than a few seconds a day.
I notice you don't comment on my responses to your questions? Samsung doesn't have a real world workload leg to stand on, correct? Synthetic BM 'scores' are meaningless in the real world.
I'm positive the 950 Pro will not have the same issues too. Following in the steps of the 1TB 850 EVO mSATA model, I'm sure it will have its own unique issues all its own.
-
Starlight5 Yes, I'm a cat. What else is there to say, really?
tilleroftheearth, if one's main concern is capacity, there pretty much ain't alternatives.
-
Which questions? the ones on the other thread? I've never heard of Tweaktown, but reading the rest of the review, they still recommend the Samsung as the only SSD you should ever need to get. If the Sandisk is much better, why do they still recommend Samsung over it? and i mean that question seriously, not sarcastically or a "hah, i win".
I didn't know the 1TB M-SATAs had that issue. has anyone else reported it? -
i thought some of the thing is pretty obvious, coming from samsung/korea in general are about looks from the outside. which is why i go for intel SSD. -
-
The best answer to a defective feature support on hardware is not to tell the customer to simply not use that feature. The best answer is to fix your hardware's problems.ajkula66, tilleroftheearth and TomJGX like this.
One more SSD to be wary of... Yeah, it's Samsung again...
Discussion in 'Hardware Components and Aftermarket Upgrades' started by tilleroftheearth, Sep 22, 2015.