Samsung 870 QVO 1 TB

[Searing]

Exactly. It doesn't matter how fast your processor is if it's got nothing to do. It could be 3 GHz, 4 GHz, or 5 GHz... it doesn't matter. If it's got nothing to do then it's wasted clock cycles and most importantly, wasted electricity. But we're going off on a tangent here.

Basically, if you're working with huge media files like large images in Photoshop or huge video files in Adobe Premier, then yes... sequential reads and writes are going to be something you're going to want to worry about especially if you're having to scrub the timeline a lot. However, for the average user, CrystalDiskMark's RND4K Q1T1 is the number you're going to want to pay close attention to since that will give you a much better idea of how much faster (compared to the traditional HDD) your system will be.

In essence it's why I have no want to replace my current Samsung 970 EVO. It gives me the performance I want. What more could I want? Replacing it would be fixing something that isn't broken.

Except I replaced my 1TB 970 EVO (that I paid a whopping $700 CAD for on launch and felt regret over ever since) with 2 x 1TB ADATA SX8200 Pro for less. I do enjoy the extra storage space. I have a 2TB Intel 660p I picked up for cheap ($190 USD) also. It's fine but slows down in a way the Adata and Samsung don't.

 

[Gmr_Chick]

Lol, this has to be a first time I'm seeing "terrible, do not buy" right on the review itself on TPU. Don't recall seeing it even on things like bad, old day PSUs. Exactly true though, you can get one of those Silicon Motion NVMe 1TB drives for a bit less if you're budget limited I think...

Took the words right out of my mouth, my dude. Holy hell. Like, I consider TPU to be very neutral when it comes to product/hardware reviews, so when I read "Terrible, do not buy" right in the HEADLINE, I knew something was up!

Good work TPU!

 

[Caring1]

So it's not something a Firmware update can fix, it's just a shitty product?
Thanks for the warning.

 

[ironwolf]

Hope this doesn't break the spirit of all the hardcore Samsung SSD fans out there.

 

[X71200]

I've known that for a while now, I thought everyone knew that. CrystalDiskMark's RND4K Q1T1 number is the number you want to pay attention to since that number is what most closely mirrors that of real world performance.

If you really want to see some pathetic numbers, take a look at what most spinning rust drives give you with CrystalDiskMark's RND4K Q1T1 number. Most drives are less than 1 MB/s whereas a good SSD will be somewhere in the ballpark of 20 to 40 MB/s.

HDDs are for the colder data for the most part, their method of R / W is no good towards 4k. As for SSDs, a good, modern SSD doesn't really have 20 MB/s 4k, that's more like an SSD from 2010.



With firmware and scheduling from 2017, too.

And yet, look at the number of posts looking for or recommending NVMe drives because of the high sequential numbers.

High sequential of NVMe doesn't necessarily mean low 4k. In fact, those drives do the 4k better than a lot of SATA stuff as well, because of the age and NAND used in the SATA drives. I don't have doubts that the 980 Pro will have solid 4k, probably equal to that of a decent PCI-E stick / U.2. Even the Phison PCI-E 4.0 drives don't really have "bad" 4k.

 

[Totally]

Pro

• Low cost per GB (this translates to being inexpensive)

Con

• Way too expensive

Can someone enlighten me, I'm not following here

 

[semantics]

Can someone enlighten me, I'm not following here

Relative to even their own offerings the low price isn't low enough
Samsung 860 EVO 2TB can be had $279.99
Samsung 870 QVO 2TB can be had $249.99
That isn't cheap enough, there is a little to be said about the 8TB version have capacity that isn't offered else where but the price isn't low enough to justify the performance, the price is low though.

 

[Palladium]

Hope this doesn't break the spirit of all the hardcore Samsung SSD fans out there.

Well, they only have themselves to blame for their lack of critical thought, especially those who bought the previous garbo 860 QVO.

 

[Th3pwn3r]

And yet, look at the number of posts looking for or recommending NVMe drives because of the high sequential numbers. Look at TPU's stubornness in keeping QD1 numbers under wraps (much like other reviewers, this must be a manufacturer guideline or smth).

I want NVME ssd for the cleanliness of it. 2 less cables I have to route or try to hide.

 

[watzupken]

Generally when I see QVO or anything QLC, I skip. But I can't help but click on this because this is probably the harshest header I have ever seen. No beating around the bush and in your face message. I feel it is good for reviewers to be up front about a certain product to give the manufacturer the message. I've seen quite a few review sites that don't drive the message across by choosing to use more "friendly" conclusions or try to give inflated scores than what is worth.

Shocking for how long the SX8200 has been on top as the best drive. I really thought after it released the market would respond with other products but it is still here years later number one.

Its one of the better drives mainly because of its price to performance. It doesn't ask you for an arm and a leg, but yet provide good performance.

 

[okbuddy]

how come could say do not buy, they provided a sample for ya, for free bud

128 layer coming later this year

 

[Hotobu]

Soooo, you're saying it's good?

 

[trparky]

Soooo, you're saying it's good?

I read that and then I looked at your avatar. I laughed. Yep, sarcasm.

 

[Shou Miko]

I guess I have to bite the apple and just buy for an EVO SSD to make sure I get a great drive

I have tried Crucial, Intel, Gigabyte and Kingston drives but I keep going back to Samsung when talking about SATA based SSD's.

 

[Berfs1]

Would you actually recommend the 860 QVO over the 870 QVO (for the same capacities)? I know y'all didn't test it but considering 860 QVO should be cheaper, would that just be the better option?

Generally when I see QVO or anything QLC, I skip. But I can't help but click on this because this is probably the harshest header I have ever seen. No beating around the bush and in your face message. I feel it is good for reviewers to be up front about a certain product to give the manufacturer the message. I've seen quite a few review sites that don't drive the message across by choosing to use more "friendly" conclusions or try to give inflated scores than what is worth.

I agree, this is a really good and honest review.

 

[W1zzard]

random QD1 performance (TPU refuses to give us)?

TPU's stubornness in keeping QD1 numbers under wraps (much like other reviewers, this must be a manufacturer guideline or smth).

Are we looking at the same review? QD1 has always been there? Left-most point on the charts.

Pro

• Low cost per GB (this translates to being inexpensive)

Con

• Way too expensive

Can someone enlighten me, I'm not following here

The drive comes in at 13 cents per GB, which is a low price price per GB if you consider nothing but "price per GB". But if you take into account performance offered etc, the whole package, then it's a terrible deal and virtually every SSD on the market is either much faster, or substantially cheaper per GB

Would you actually recommend the 860 QVO over the 870 QVO (for the same capacities)? I know y'all didn't test it but considering 860 QVO should be cheaper, would that just be the better option?

No idea, don't have any data for 860 QVO. Just buy the MX500 or Seagate 120 for the same price, you even get +2 years warranty

 

[Shou Miko]

I agree, this is a really good and honest review.

@W1zzard always makes reviews that are worth the time reading and understand it's just a shame the products ain't always great but that's not the reviewers fault.

I am always excited for a review here on TechPowerUp when it's a product I am really interested in but in general I check a lot of the reviewers here because they are always good in-depth and the quality is always top notch. This is also why I keep being on this forum it's one of the best and people always know something.

 

[bug]

Are we looking at the same review? QD1 has always been there? Left-most point on the charts.Yeah, like anyone can

Yeah, like anyone can put a number on a point in a chart... When it comes to numbers, you're conveniently hiding it in "mixed usage", "average of".

 

[W1zzard]

Yeah, like anyone can put a number on a point in a chart... When it comes to numbers, you're conveniently hiding it in "mixed usage", "average of".

The world is not just QD1, or QD2, or QD4, or QD128 ... different people have different requirements, that's why I have the line chart, so everybody can pick the scenario he wants to look at.

For the bar charts my 80% QD1, 20% QD2, 5% QD4 mix is a good approximation of reality.

What mix do you propose? 85/15/5 ? 90/10/5? It's not gonna make a substantial difference

 

[bug]

High sequential of NVMe doesn't necessarily mean low 4k. In fact, those drives do the 4k better than a lot of SATA stuff as well, because of the age and NAND used in the SATA drives. I don't have doubts that the 980 Pro will have solid 4k, probably equal to that of a decent PCI-E stick / U.2. Even the Phison PCI-E 4.0 drives don't really have "bad" 4k.

Absolutely, but there's a far cry between the 20% better 4k performance NVMe offers, vs the 500-600% higher sequential numbers.

The world is not just QD1, or QD2, or QD4, or QD128 ... different people have different requirements, that's why I have the line chart, so everybody can pick the scenario he wants to look at.

For the bar charts my 80% QD1, 20% QD2, 5% QD4 mix is a good approximation of reality.

What mix do you propose? 85/15/5 ? 90/10/5? It's not gonna make a substantial difference

I propose a table: QD1 r/w, QD4 r/w, etc.

Edit: Also, highlight properly which are the numbers that are the most important. (Or not, if that puts you at odds with manufacturers )

 

[W1zzard]

the most important

my conclusion text, then the three charts on page 13, then write intensive, then thermals, then (for me, in that order) Steam Preload, WinRAR, ISO Copy, Photoshop, 50% mixed, and that's it, maybe local supply situation, which isn't part of my review

Or not, if that puts you at odds with manufacturers

Why do you keep suggesting that I'm trying to deceive my readers?

 

[bug]

my conclusion text, then the three charts on page 13, then write intensive, then thermals, then (for me, in that order) Steam Preload, WinRAR, ISO Copy, Photoshop, 50% mixed, and that's it, maybe local supply situation, which isn't part of my review


Why do you keep suggesting that I'm trying to deceive my readers?

I'm not suggesting you're deceiving anyone, I've been around for too long to think that. I actually appreciate that you actually take suggestions from readers when possible.
I was just saying most reviewers don't make it obvious enough which numbers will give you the real expectations. And I know there's a possibility the numbers are already obvious enough, it may only be about how I read these reviews. I was hoping others would chime in.

 

[W1zzard]

that you actually take suggestions from readers when possible.

Thanks

it may only be about how I read these reviews

I think it might be that, and that's fine. Everybody has different requirements and expectations, that's why I'm trying to include as much data as possible (without overloading the review). I'm not a fan of synthetic testing, years ago I used to have no synthetics at all, but I can see how they serve a purpose and should be included for completeness.

Working on new SSD testing rig right now (Ryzen, PCIe 4.0), biggest change will be that for real-life testing all drives are filled to 80% of their capacity, to simulate a "mostly full" environment.

I was hoping others would chime in

Yes please

 

[Shou Miko]

Working on new SSD testing rig right now (Ryzen, PCIe 4.0), biggest change will be that for real-life testing all drives are filled to 80% of their capacity, to simulate a "mostly full" environment.

Nice I cannot wait for this, I was thinking you have tested a lot of SSD's is it true that most users in general use of Windows 10 won't feel the difference between SATA vs NVME based SSD's?

I read some users experience that a SATA based SSD feels more snap and all-around more smooth when tested in a line up.

 

[Tsukiyomi91]

seems that the 860 EVO would be the better option for most of us after this review.