The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.

    Why Are Larger SSDs Faster?

    Discussion in 'Hardware Components and Aftermarket Upgrades' started by Spartan@HIDevolution, Dec 16, 2017.

  1. Spartan@HIDevolution

    Spartan@HIDevolution Company Representative

    Reputations:
    39,579
    Messages:
    23,560
    Likes Received:
    36,854
    Trophy Points:
    931
  2. jaug1337

    jaug1337 de_dust2

    Reputations:
    2,135
    Messages:
    4,862
    Likes Received:
    1,031
    Trophy Points:
    231
    Hey LLT made a useful video for once :D
     
    Vasudev likes this.
  3. Papusan

    Papusan Jokebook's Sucks! Dont waste your $$$ on Filthy

    Reputations:
    42,706
    Messages:
    29,842
    Likes Received:
    59,619
    Trophy Points:
    931
    You won’t get slower speed with smaller sized Optane drives vs. the bigger one... New tech. See Intel’s new Optane drives.
     
    Vasudev and tilleroftheearth like this.
  4. 3Fees

    3Fees Notebook Deity

    Reputations:
    541
    Messages:
    970
    Likes Received:
    136
    Trophy Points:
    56
    They charge more for them so you have get something for the outlay. :D
     
  5. tilleroftheearth

    tilleroftheearth Wisdom listens quietly...

    Reputations:
    5,398
    Messages:
    12,692
    Likes Received:
    2,717
    Trophy Points:
    631
    The video points out a few of the reasons why they're faster.

    The rest of the reasons are that: (usually) better nand is used in larger drives, (usually) better controllers are used in larger drives, (usually) better firmware is used in larger drives and most importantly; I can OP a larger drive greater than 33% and still have more usable capacity and greater sustained performance (even if that greater performance exhibits greater and greater diminishing returns...).

    While supersizing your food is a bad idea for your health (no matter how much $$ you're 'saving', sigh...), buying the biggest SSD you can afford (and here, I would add 'continually') is directly related to how long the system will stay usable (not just to the original user; but to the second and third owners too).

    I have seen older systems (10 year old Precision series workstations) work almost identical and almost better than their current counterparts, when it comes to 'snap' (no question the newer platform is much more productive, of course...). By simply changing that old tech SSD's with anything current, of the biggest capacity possible (2TB SATA III SSD's are the current sweet spot). After maxing out the RAM and giving the system a good cleaning for hair/dust and a proper Windows 10 x64 Pro clean install, the owner of the older hardware kinda questioned why the old system was scrapped and the new system was bought (could have been easily used by a junior employee and still kept them productive for the next six months or more...).

    When any platform isn't optimized/maxed and/or balanced as well as it could be (O/S, Drivers, CPU, RAM, Storage and an optimal configuration of the preceding hardware/software...), the results are always dismal.

    Buying a cutting edge technology like an SSD and then getting a weak example of the same (wrong brand/controller/capacity) is a mistake I see too many make.
     
    Spartan@HIDevolution and Papusan like this.
  6. 3Fees

    3Fees Notebook Deity

    Reputations:
    541
    Messages:
    970
    Likes Received:
    136
    Trophy Points:
    56
    Okay, pun aside,actually they are not faster at least brand name ones!
    Case on Point
    Samsung 850 Pro SSD's
    256GB, 512GB, 1TB, and 2TB have the Mex Controller and all Read/Write 550/520 Mbps, IOPS 100K/90K, and MTBF 2,000,000 hours and 10 year warranty.

    Too many links to show this and all links show No Difference !
     
    Last edited: Dec 29, 2017
    Vasudev likes this.
  7. Papusan

    Papusan Jokebook's Sucks! Dont waste your $$$ on Filthy

    Reputations:
    42,706
    Messages:
    29,842
    Likes Received:
    59,619
    Trophy Points:
    931
    960 Evo
    upload_2017-12-30_6-5-23.png
    But you won't see the big difference in normal daily computer use :D
     
    Raiderman and Vasudev like this.
  8. tilleroftheearth

    tilleroftheearth Wisdom listens quietly...

    Reputations:
    5,398
    Messages:
    12,692
    Likes Received:
    2,717
    Trophy Points:
    631
    Spec's and 'scores' don't show a difference because they can't. I'm also not here to convince anyone in particular that a difference is there either.

    What I'm saying is that just using very many examples of SSD's from most of the available spectrum including capacity and brands I can see a difference from my very lightest use case to the very heaviest workloads I (and my clients) use them for.

    Is the price difference worth it for most? Maybe not (but unless they actually try it for themselves, they don't know what they're missing either...). But for those that are buying SSD's for their promised performance? They are cheating themselves out of that maximum possible performance if they buy on mere 'specs' to save mere $$...

    I've seen SSD's give me (sustained) effectively USB 2.0 speeds or they can give me something closer to up to ~55% to ~60% of the SATA III or PCIe x4 theoretical connection used... when they're properly and optimally setup.

    The biggest defining factor on what (sustained) real world speeds I can achieve? The size/capacity of the SSD in question and the correspondingly better controller/firmware used on them.

    And when we're comparing apples to apples (high end 'older' to high end 'current') newer and bigger is (almost) always better.

    When a properly setup platform/ssd checks/updates my O/S and store apps in a few seconds vs. a few minutes; the difference (even in this extremely 'light use' example) is worth hundreds of $$'s to me. Because that is easily the time I'll gain over the next few years of usage out of them.


    I'll point this out again; warranty for any storage device that I've actually put mine or my client's data on is a moot point. I will never return/exchange a storage device with data on it - and neither will any client I represent either. Warranty is simply a marketing ploy that I don't factor into my purchase decision. Once I decide to keep a drive I've finished testing (with 'fake' data...); the warranty is effectively over.

    The reasons I'll keep it? SNAP! Sustained speeds. Sustained responsiveness. Keeps cool in the platform used. Very little ('felt') performance variance. NO STUTTERING/HICCUPS. Price (if there is an otherwise comparable product to choose from).

    Those that believe SSD's are all the same (if the specs are similar) are not familiar with the concept of the whole being more than the sum of it's parts. Some products have it; most don't and there are always a few that don't that fail spectacularly at it. To me; the differences are vast between those type of examples of any product category we can discuss in these forums.

    3Fees, we don't need to agree on this issue. I'm okay with that.

    But the differences my clients and I see are real and worth mere $$ to alleviate as much as is possible today. Even if the majority of the world is effectively blind to the existence of those differences and/or doesn't value the benefits gained from the best examples vs. the lessor ones (and again; not best/worst in 'specs' - rather; real world use).


    Wishing you and everyone here a Happy and Prosperous 2018.

    Take care.

     
  9. 3Fees

    3Fees Notebook Deity

    Reputations:
    541
    Messages:
    970
    Likes Received:
    136
    Trophy Points:
    56
    The Statement " Why are Larger SSD's Faster ? " Its a universal affirmative statement and what I have shown is one exception to it, which in terms of Logic negates and disproves the statement in entirety, thus, therefore ect in conclusion, The Statement is False !

    I have tested the two Samsung 850's Pro's I have on installed on two different notebooks-both saturate the Sata III band as published when running Sata III, both notebooks web surf, download, bring up pdf programs, office programs and such really fast as fast as larger Samsung 850 Pro's do, a few seconds.

    Have a Nice Day
     
    Last edited: Jan 7, 2018
  10. p0och

    p0och Notebook Enthusiast

    Reputations:
    4
    Messages:
    46
    Likes Received:
    9
    Trophy Points:
    16
    Guys you need to understand the basics of how SSD operate to understand the natural posotive co-relationship between size and speed.

    when SSD's "delete" stuff - they're not exactly deleted in the formal sense. (branching out a bit here but this makes file recovery a whole lot easier which can be viewed as a positive or negative issue)

    Therefore - drive encryption became somewhat a norm - bitlocker OPEL SED u name it - this effects speeds though but protects from anyone who wishes to "forensically" retrieve data you do not want them to have.

    ok any back to size and speed - the messages posted before me are accurate. there is an race to compress as much nand into a single wafer much like how every new gen intel cpu does but for different reasons.

    So the larger the SSD drive is (assuming you don't trigger the TRIM function before conducting the test) the better the perf will be on the benchmark given the tests will be conducted on free space.

    So forexample if you had 2 exact models but with different size like a 250gb samsung 850 evo and a 512gb - running a simple crystal disk mark test on both will display superior results in particular with sequential speeds but we all know IOPS is probably more relevant.

    There are of course other tests and tweaks in the benchmarks to ensure you get somewhat close to real world result outcomes but this is a lot of work for very little gain,

    key takeway here - size matters and while TRIM is a very good tool - if you value your SSD life over performance much like overclocking - stick with the default in windows 10 which is once a week.

    Other relevant notes:
    Size - majority of SSD devlopment is swayed by the direction of notebook design which are in my view mixed.

    The extreme being the Microsoft series which even the SSD is soldered - while we've all somewhat tolerated with soldered ram such as everyone's fave Lenovo X1 Carbon series.

    the point is - m2 or NGFF is the new norm and even the usual 2280 size is now being challenged to produce 2242 versions (think the size of your wifi card) '

    And now NVME or SSDs using the pcie bus are also changing towards the 2242 format. The Thinkpad T480 currently uses the Toshiba produced nvme drive which is in 2242 form factor.

    Similiarly Toshiba has recently announced their RC100 line which i think is the same as the one Lenovo uses. WD has folllowed suit while Liteon has theirs too.

    With this knowledge - it's going to be an interesting next 6 mths when the drives come into the market. For older laptops that had empty WWAN m2 slots in the pcie interface - this is an avenue to explore.

    I think currently the HP Zbook g1 series (haswell) - someone recently obtained a 2242 pcie nvme ssd and tested it as a boot drive with positive results.

    At the moment though - if you have a laptop wiith a m2 slot that supports both sata and nvme - pricewise go for the high quality m2 sata ssd. Budget NVME ssd currently sold under decent brands like ADATA or Kingston uses the pcie3x2 format (2 lanes instead of 4) which when compared to a Samsung msata ssd in perf terms are really equal while the energy consumption from using a pcie ssd is worse. I'd never use a nvme for these reasons.

    They're only really useful for datacentres due to their latency advantage.