Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


Benchmark and first impressions -> finalhost
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

Benchmark and first impressions -> finalhost

jsgjsg Member, Resident Benchmarker
edited May 2019 in Reviews

As they generously provide free test servers for 3 days, which is more than sufficient, I got one of those test VPS and benchmarked it. Here's the data, some remarks, and a summary:

Machine: amd64, Arch.: amd64, Model: QEMU Virtual CPU version 1.5.3
Mem.: 1.985 GB
CPU - Cores: 2, Family/Model/Stepping: 6/13/3
Cache: 32K/32K L1d/L1i, 2M L2, ? L3
Std. Flags: fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36
          cflsh mmx fxsr sse sse2 htt sse3 cx16 hypervisor
Ext. Flags: syscall nx lm lahf_lm

--- proc/mem/performance test single core ---
64 rounds~ 1.00 GB ->  125.40 MB/s
--- proc/mem/performance test multi-core ---
4 times 64 rounds ~ 4.00 GB ->  260.20 MB/s
--- disk test ---
Sequential writing 117.65 MB/s
Random writing     159.26 MB/s
Sequential reading 387.92 MB/s
Random reading     370.02 MB/s
--- network test ---
AU,MEL: 18.7 Mb/s
IN,CHE: 26.2 Mb/s
UK,LON: 46.0 Mb/s
DE,FRA: 46.4 Mb/s
IT,MIL: 45.4 Mb/s
FR,PAR: 46.6 Mb/s
RO,BUC: 38.4 Mb/s
GR,UNK: 42.4 Mb/s
US,DAL: 348.0 Kb/s
US,SJC: 24.7 Mb/s
US,WDC: 12.5 Mb/s
BR,SAO: 24.1 Mb/s
JP,TOK: 22.6 Mb/s
NO,OSL: 44.5 Mb/s

The CPU looks strange, probably a qemu thing. Several factors (like L2, the flag set, etc) point towards a stone age Xeon or Core something, the results, however, point more towards a modern proc in the 1.8 to 2.4 GHz range. As I said probably a qemu settings thing.

CPU performance is in the lower middle range. Really good VPS (with dedi cores) are 50 or more percent faster but the values here are OK and good enough for most typical VPS tasks like mid range dynamic web sites, etc.

The disks (which according to finalhost are SSDs) performance is clearly more in the average spindle range.

The connectivity - keep in mind that the test VPS is limited to 50 Mb/s - is quite nice. Really decent for a VPS in that price range. I've seen far worse. What I like is that even India, Japan and of course North-America are in the 2 digit Mb/s range (except for a strange slip with Dallas).

Summary so far: decent albeit nothing to write home about.

Where finalhost shines though is in another area. Their panel is well layed out, well usable and not confusing like some others. One point I'm always looking for is the panel console. Often enough some noVNC console is carelessly slapped in and with weird keyboard layouts. The finalhost panel however offers a properly, even nicely working panel console. That's a good sign; they seem to really care about their stuff and their clients. Nice.

The really big positive surprise however was something else: their support. It's just incredible. Not only do they respond quite quickly to tickets (well below 1 hour) but they also have one of those web chat thingies there and I waited less 1 min. when I had a small issue. To make it even better, their support also solved the issue right away. And that on a certainly busy day, mind you. Great!

To be frank, while their prices are OK, especially the starter "lure" packet, one can get very similar packages (technical aspects like performance) or even better ones (with a little effort and knowing where to look) for about the same price around here (at LET). As I don't need a lot of (their, again, excellent) support I'm not sure I would bite.

Considering however that they also threw in DDOS protection for free in their current special, it's a no brainer. I'll get one of their mid range VPSs.

I'll close with a somewhat personal/subjective remark: I like, really like their approach. Unlike most others who just rent some dedis and throw the usual software stack on them, finalhosting feel like a good mix of a solid business approach (again, with excellent support. that really deserves mentioning in bold letters) and a fresh techie approach. Of course, I'm a techie myself and many customers, say, business people won't care a lot. But still, it was quite often techies in a garage who brought major new impulses to the industry and market. I like taht in finalhost.

Recommended, in particular for people who need support.

(Should the VPS I buy be more than insignificantly different, I'll update this review).

Disclaimer: This review is based on finalhost's free test VPS. I did not ask for, nor was I offered any advantages or asked for any favours. The benchmark software I used is as always my own vpsbench (see my sig for binaries or source).

Thanked by 1jordynegen11
«1

Comments

  • jsgjsg Member, Resident Benchmarker
    edited May 2019

    Important major update!

    As I sometimes need FreeBSD for a project I did the above benchmark on FreeBSD. Now it turned out that FreeBSD on KVM has some quite significant performance issues - which made the tested VPS look worse than it is.

    Happily a free test server works for 3 days, so I could do the benchmark again on Linux (Debian 9).
    But there is a bit more to it. It seems the interest in test VMs was quite high which lead to the test node(s) being quite loaded.
    Finally yet another factor was significant: the test VPS's network is limited to 50 Mb/s but the VMs one can buy have 100 Mb/s so benchmarking a test VM lead to lower results than what one gets when buying. However @jordynegen11 was friendly enough to offer me a more realistic test VM that was upgraded to 100 Mb/s.

    Here are the results and a short discussion of what I found:

    Machine: amd64, Arch.: x86_64, Model: Intel(R) Xeon(R) CPU E5-2660 v2 @ 2.20GHz
    OS, version: Linux 4.9.0, Mem.: 1.980 GB
    CPU - Cores: 2, Family/Model/Stepping: 6/62/4
    Cache: 32K/32K L1d/L1i, 2M L2, 16M L3
    Std. Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat
              pse36 cflsh mmx fxsr sse sse2 ss sse3 pclmulqdq vmx ssse3 cx16 pcid
              sse4_1 sse4_2 x2apic popcnt tsc_deadline aes xsave osxsave avx f16c
              rdrnd hypervisor
    Ext. Flags: fsgsbase tsc_adjust smep erms umip syscall nx pdpe1gb rdtscp lm
              lahf_lm
    
    --- proc/mem/performance test single core ---
    64 rounds~ 1.00 GB ->  235.86 MB/s
    --- proc/mem/performance test multi-core ---
    4 times 64 rounds ~ 4.00 GB ->  523.29 MB/s
    --- disk test ---
    Sequential writing 926.25 MB/s
    Random writing     1.807 GB/s
    Sequential reading 4.449 GB/s
    Random reading     2.917 GB/s
    --- network test ---
    OK,LON:-> 94.9 Mb/s
    AU_MEL:-> 23.0 Mb/s
    IN,CHN: -> 44.5 Mb/s
    DE,FRA: -> 94.4 Mb/s
    IT,MIL: -> 93.6 Mb/s
    FR,PAR: -> 94.7 Mb/s
    BR,SAO: -> 31.0 Mb/s
    US,DAL: -> 50.2 Mb/s
    US,SJC: -> 34.8 Mb/s
    US,WDC: -> 48.5 Mb/s
    JP,TOK: -> 55.6 Mb/s
    GR,UNK: -> 88.0 Mb/s
    NO,OSL: -> 92.7 Mb/s
    

    A very different picture! This VPS (with linux) is a really nice one whose performance is in the netcup dedicated core VPS league.

    As for the disk, while I saw a very considerable increase under FreeBSD on a real customer VPS (as opposed to the test VM) bringing speed to a quite acceptable level (about 300 MB/s writing and about 600 MB/s reading) speeds really took off under linux. Those values are among the best I've seen so far on any VPS.

    Note: I always do multiple runs and the results naturally have some variation on VMs. In this case the shown processor/memory results are among the lower ones I've seen while the disk values are among the higher ones. Be reminded that it's generally wise to take benchmark results for virtual machines meaning "your VPS might be up to 20% slower or even faster, depending on your node and some other factors".

    Regarding connectivity the results haven't changed a lot, except for the fact that the current benchmark had the virtual adapter set to 100 Mb/s, which makes for a picture with some more contrast.
    Obviously the results for near targets, in particular the big 5 in Europe (FRA, AMS, PAR, Sweden/Danmark, and Milano) are close to the adapters limits at about 90 to 95 Mb/s.

    Way less boring and, in fact, really interesting is the fact that "exotic", far away and in some cases known to be difficult targets (like e.g. India or Brazil) show really good results with the finalhost network. Chennai, India and Sao Paolo, Brazil being in the range of tens of Mb/s and, in fact, not considerably worse than major North-American targets is plain astonishing. And even down under, Oz-land, which is condemned to slow network speed by physics (distance, too few and thin cables) achieves almost 25 Mb/s. Excellent, particularly for a rather low priced VPS (here with the LET special)!

    For those interested, I also ran some network tests on a node with DDOS-protection. It adds some latency ( +- 15 ms) but throughput is still very good (close to what you see here).

    I'd like to close with a final remark re. finalhost's support. It's really incredible. Once I became a customer I started noticing some small issues with their home grown (and actually quite nice) user interface/panel (which, however still could have a bit more features). Not only did they react within minutes but they even repaired/changed some of the small issues right away. I'm really impressed by their support.

    As for value vs. price I think that finalhost is not the cheapest. Quite low and very reasonably priced, yes, but not the cheapest. On the other hand you get excellent performance (it seems their vCPUs are virtually dedicated) and extremely good and quick support.
    Hence my recommendation is to have them in ones top go-to providers list if your priority is a good but not the cheapest price, good performance and quick good support. Then the ratio of what you get vs what you pay is indeed excellent.

    Disclaimer/Disclosure: Unlike when doing the first benchmark, I'm meanwhile a customer of finalhost. But - an important but - I did not receive any bonus nor did finalhost ask me for any favours other than to please update the review, which I did (a) for the sake of fairness, and (b) because testing under FreeBSD (then not known to me) did indeed make their VPS look much slower than it actually is under linux. Plus, that problem is KVM/FreeBSD related and concerns every KVM provider.

    Thanked by 1jordynegen11
  • angstromangstrom Moderator

    Ah, so this was what was behind that other thread (on FreeBSD on KVM).

    In general, though, in any review that you do, it would be a good idea to state at the outset which OS you're using to run the benchmarks.

  • jsgjsg Member, Resident Benchmarker

    @angstrom said:
    Ah, so this was what was behind that other thread (on FreeBSD on KVM).

    In general, though, in any review that you do, it would be a good idea to state at the outset which OS you're using to run the benchmarks.

    Yes, I've learned that now. Will do in future reviews.

  • angstromangstrom Moderator

    Re the support: I don't doubt your experience with their support, and your experience may even be characteristic of the support experience that every customer gets (I have no reason to doubt this), but they did know that you wanted to test/benchmark their VPS, didn't they? Or did they have no idea that you wanted to test/benchmark their VPS?

  • jsgjsg Member, Resident Benchmarker

    @angstrom said:
    Re the support: I don't doubt your experience with their support, and your experience may even be characteristic of the support experience that every customer gets (I have no reason to doubt this), but they did know that you wanted to test/benchmark their VPS, didn't they? Or did they have no idea that you wanted to test/benchmark their VPS?

    As far as I know they did not know that in beginning. If there is one factor that might indeed be relevant, I guess it's more that I'm a techie and them being techies, too, it might be possible that my experience with their support is somewhat different from a non-tech savvy client. But even then their support would be good because they seem to be (a) really friendly and (b) really interested in seeing their customer happy and well served.

    But clearly: I'm a human being and no matter how hard I try to be fair, unbiased, and objective (which I seriously try my very best) there will always slip some subjective perception through the net.

    To this day, however, there has not been a single provider, nor a single customer claiming that I was anything but trying hard to be fair and unbiased.

  • KuJoeKuJoe Member, Host Rep

    It's refreshing to see something positive about new hosts on LET. Thanks @jsg for taking to time to put together this review out in the sea of negativity. :smile:

    Thanked by 1jsg
  • angstromangstrom Moderator

    @jsg said: But clearly: I'm a human being and no matter how hard I try to be fair, unbiased, and objective (which I seriously try my very best) there will always slip some subjective perception through the net.

    To this day, however, there has not been a single provider, nor a single customer claiming that I was anything but trying hard to be fair and unbiased.

    No need to take anything personally! As I said, I don't doubt what you said about your experience.

    It's just that you seemed to emphasize the friendliness of the customer support a lot, which made me wonder whether they knew that you wanted to review their VPS. (By the way, I suspect that they did, but my suspicion doesn't matter!)

    Consider this: ideally, a restaurant critic does a restaurant review (both of food and service) unannounced. If it's announced beforehand, it doesn't necessarily invalidate the review, but it does lead a reader to wonder whether the fact of the announcement was a influencing factor in some way. For example, because of the announcement, they had their best chef there that day, and the nicest waiters, all who tried extra hard.

    (You see what I mean.)

    Thanked by 1vimalware
  • jsgjsg Member, Resident Benchmarker
    edited May 2019

    @angstrom said:

    @jsg said: But clearly: I'm a human being and no matter how hard I try to be fair, unbiased, and objective (which I seriously try my very best) there will always slip some subjective perception through the net.

    To this day, however, there has not been a single provider, nor a single customer claiming that I was anything but trying hard to be fair and unbiased.

    No need to take anything personally! As I said, I don't doubt what you said about your experience.

    It's just that you seemed to emphasize the friendliness of the customer support a lot, which made me wonder whether they knew that you wanted to review their VPS. (By the way, I suspect that they did, but my suspicion doesn't matter!)

    Consider this: ideally, a restaurant critic does a restaurant review (both of food and service) unannounced. If it's announced beforehand, it doesn't necessarily invalidate the review, but it does lead a reader to wonder whether the fact of the announcement was a influencing factor in some way. For example, because of the announcement, they had their best chef there that day, and the nicest waiters, all who tried extra hard.

    (You see what I mean.)

    (a) I didn't take it personally. I * understood* your concerns and responded honestly.

    (b) I'm not a payed, professional restaurant tester. Plus, your argument fails because unlike in a restaurant where one might be able to quickly change some things, ahoster, in particular a smaller one, can't adapt/change that much on short notice. His hardware is what it is and so is his network.

    (c) Your argument fails again for another reason - even with a restaurant: To know that a test is coming up does not mean that the tester is known. So unless the tester has a big "I'm the tester" label on his forehead there is no or just a very minor problem.

    I wasn't taking your first remarks personally. But now I'm starting to get pi__ed off.

    I made considerable efforts. I've written a VPS specific benchmark. So it seem reasonable to assume that I'm interested in and serious about testing VPSs.

    On the other hand you had no comment with the first version. Now, however, that the review got considerably better, you are all over it. So much for assumptions of bias. Grab your own nose.

    Also note that I frankly and honestly introduced the revised review with a "mea culpa" plus I was utterly open and frank about the circumstances. Yet you insistingly continue to follow a non existing issue.

    So, what are the facts? Do you really think that finalhost created a web chat system for customer support just to look good for me? Do you really think that usually they they don't care and are unfriendly but for me they magically turned into friendly and seriously interested in their customers people?

    You might want to look at the question whether you are simply too mistrusting. Because of another fact I'd like to bring to your attention: So far not a single provider has ever tried to influence me or to ask favours. Activating your capability to reason you'll quickly find out why: because at the end of the day I'm just one guy here. One who wrote a benchmark, one who does tests and reviews - but still just one guy. If I did lie there would soon be plenty of others who'd tell the truth about the providers whom I allegedly painted too nice.

    Didn't happen so far, though. For a reason I guess.

    (And YES, I hate it when someone does actually constructive things for a community and invests time and work - and then someone comes up and only has to bicker on small details).

  • angstromangstrom Moderator
    edited May 2019

    Dude, get a grip.

    People do benchmarks around here all of the time. But it's nice that you tried to interpret your results for us. And for pointing out that there is a big difference between FreeBSD and Debian on KVM. And I don't question your numbers or your comments on them.

    The only reason that I reacted was the many superlatives in your review about their support, which just made me wonder whether they knew that you were reviewing their VPS -- this was all. You could just have answered "Yeah, they knew, but I was so impressed with their support that I wanted to mention it anyway, and I feel that everyone will receive the support that I did." But no: instead you decide to feel attacked and doubted for your nothing-but-generous contribution to the community. (Just like some other guy from the past who I won't mention.) Give me a break.

  • psb777psb777 Member

    jsg said: ... Now it turned out that FreeBSD on KVM has some quite significant performance issues - which made the tested VPS look worse than it is.

    Isn't it that FreeBSD on bare metals have major performance issues as well (compared with Linux of course; maybe less significant than those on KVM but still very noticeable)? And OpenBSD is significantly worse than FreeBSD as far as I'm concerned.

  • deankdeank Member, Troll
    edited May 2019

    @angstrom said:
    The only reason that I reacted was the many superlatives in your review about their support, which just made me wonder whether they knew that you were reviewing their VPS.

    Which is the prime reason not to take any reviews seriously when a product was given out for a purpose of reviewing. They'd make sure that his account is created on a fast node.

    It's a form of shilling in exchange of free goods.

  • jsgjsg Member, Resident Benchmarker

    @angstrom

    I got excellent support. Simple as that. What's your problem?

    I've seen really good support elsewhere, too, e.g. with netcup. But this, with finalhost, is a new level of it's own. It's way beyond anything I've seen so far.

    Why I mention it so clearly? Because unlike probably you or myself, there are actually many, many people out there for whom support is a, maybe even the criterion that makes having a VPS a failure or a success.

  • jsgjsg Member, Resident Benchmarker
    edited May 2019

    @psb777 said:

    jsg said: ... Now it turned out that FreeBSD on KVM has some quite significant performance issues - which made the tested VPS look worse than it is.

    Isn't it that FreeBSD on bare metals have major performance issues as well (compared with Linux of course; maybe less significant than those on KVM but still very noticeable)? And OpenBSD is significantly worse than FreeBSD as far as I'm concerned.

    No, not that I noticed. Now, I'm by no means an expert with FreeBSD; I merely use it sometimes if a project (or, more precisely, a client) requires it. But based on my observations I'd say that FreeBSD and linux (on bare metal) are about the same. Usually one does something a bit faster while it does somethings else a bit slower, but the difference is usually nothing significant.

    With OpenBSD it's a different story because many things done only single threaded and some peculiarities of OpenBSD and its project leader. That must not be negative; I guess it depends on ones priorities.

    @deank said:

    @angstrom said:
    The only reason that I reacted was the many superlatives in your review about their support, which just made me wonder whether they knew that you were reviewing their VPS.

    Which is the prime reason not to take any reviews seriously when a product was given out for a purpose of reviewing. They'd make sure that his account is created on a fast node.

    It's a form of shilling in exchange of free goods.

    That's pure Bullshit.

    The FACT is that their free test VPS is worse than the payed ones - and I based my benchmark on their test VPS. I changed that only in my update/2nd. version because of the FreeBSD problem and also because the difference between the considerably poorer test VPS and the actual (bought) ones became so significant that I felt that basing my review on the test VPS wouldn't reflect what one gets when buying a VPS.

    The FACT is that the node with the test VPSs was slower, possibly due to many test VPSs running.

    The FACT is that I got NOTHING for my review. Or, wait, no, I got something: A "thank you. Nice review" message.

    I paid the normal and full price for my personal VPS. And it's really normal. No extra vCore, no extra memory or bandwidth, no extra disk space.

  • deankdeank Member, Troll

    Dude, get a grip.

    Thanked by 2psb777 TimboJones
  • jsgjsg Member, Resident Benchmarker

    @deank said:
    Dude, get a grip.

    Actually I got a grip. Unlike you I invested some time and effort in this test/review. Unlike you I actually know what I'm talking about.

    Being at that: I try to actually contribute to this community. How about you? (Don't worry, I don't expect an answer and certainly not a sensible one).

    Thanked by 1TimboJones
  • angstromangstrom Moderator
    edited May 2019

    @jsg said: I got excellent support. Simple as that. What's your problem?

    Oh, man, wow.

    You insist on taking as an attack something that wasn't. Yes, there was a little skepticism on my part, for which I asked a question, but it wasn't an attack.

    Come to think of it, why in the hell world did you need so much support anyway?! Just install the bloody OS, run your bloody benchmarks, then write your bloody review for us ingrates here on LET.

  • jsgjsg Member, Resident Benchmarker

    @angstrom said:
    Come to think of it, why in the hell world did you need so much support anyway?! Just install the bloody OS, run your bloody benchmarks, then write your bloody review for us ingrates here on LET.

    2 main reasons: (a) the FreeBSD problem (on KVM) and (b) the fact that their home grown panel still has some rough edges. I noticed some of them and told them and always found them very friendly, constructive, and quick. In some cases they fixed a small thing basically right away.

    Now, I understand that one might differentiate between e.g. "I have a problem" and "they (their panel) have a problem" but I decided to lump it all together. Reasons: (a) in the end their problems become mine too (because, e.g. something in the panel is confusing or not working) and (b) my measuring scale for support is "I have a question or see a problem - how and how fast and how constructive does the provider react?".

    Thanked by 2angstrom uptime
  • deankdeank Member, Troll

    PMS overflow detected.

    Thanked by 1uptime
  • jsgjsg Member, Resident Benchmarker
    edited May 2019

    @deank said:
    PMS overflow detected.

    Then get some tampons or pills or see a doctor or psychiatrist.

  • angstromangstrom Moderator

    @jsg said:

    @angstrom said:
    Come to think of it, why in the hell world did you need so much support anyway?! Just install the bloody OS, run your bloody benchmarks, then write your bloody review for us ingrates here on LET.

    2 main reasons: (a) the FreeBSD problem (on KVM) and (b) the fact that their home grown panel still has some rough edges. I noticed some of them and told them and always found them very friendly, constructive, and quick. In some cases they fixed a small thing basically right away.

    Now, I understand that one might differentiate between e.g. "I have a problem" and "they (their panel) have a problem" but I decided to lump it all together. Reasons: (a) in the end their problems become mine too (because, e.g. something in the panel is confusing or not working) and (b) my measuring scale for support is "I have a question or see a problem - how and how fast and how constructive does the provider react?".

    Now, you see, this is a clarifying reply. (Yes, I agree, it was nice that they were so reactive.)

  • jordynegen11jordynegen11 Member
    edited May 2019

    Hi @jsg

    Thanks for your review. We really appreciate the time you took to review our services and the extra tips your gave us about our control panel, via the livechat.

    If you have further questions or more tips, you can always trow those in the livechat or create a support ticket.

    I can guarantee everyone will get the same kind of performance. And if for some magical reason not, just contact us and we figure it out for you!

    Thanked by 1jsg
  • CrelingCreling Member

    Thank you for your contribution. Hoping to see more in the bench if possible. For example, something like dd if=/dev/zero of=test bs=64k count=4k oflag=dsync it seems can judge if the vps is suitable for hosting a website with mysql more accurately.
    Btw, hoping to see the speedtest nodes in China in the bench as well. As we know, China is very special in various senses.(。^▽^)

  • jsgjsg Member, Resident Benchmarker

    @Creling said:
    Thank you for your contribution. Hoping to see more in the bench if possible. For example, something like dd if=/dev/zero of=test bs=64k count=4k oflag=dsync it seems can judge if the vps is suitable for hosting a website with mysql more accurately.
    Btw, hoping to see the speedtest nodes in China in the bench as well. As we know, China is very special in various senses.(。^▽^)

    Won't happen. Not needing to run nonsense tests like dd was the major reason why I wrote a proper benchmark software.

    As for other nodes or providers, that depends on them. If they want me to review their VPS I'll usually do.

  • SirFoxySirFoxy Member

    Man ppl love to argue on LET lol

    Thanked by 1uptime
  • donlidonli Member

    @SirFoxy said:
    Man ppl love to argue on LET lol

    No we don't.

    Thanked by 1uptime
  • SirFoxySirFoxy Member

    @donli said:

    @SirFoxy said:
    Man ppl love to argue on LET lol

    No we don't.

    boiiiiiiiii

  • @jsg said:

    @Creling said:
    Thank you for your contribution. Hoping to see more in the bench if possible. For example, something like dd if=/dev/zero of=test bs=64k count=4k oflag=dsync it seems can judge if the vps is suitable for hosting a website with mysql more accurately.
    Btw, hoping to see the speedtest nodes in China in the bench as well. As we know, China is very special in various senses.(。^▽^)

    Won't happen. Not needing to run nonsense tests like dd was the major reason why I wrote a proper benchmark software.

    As for other nodes or providers, that depends on them. If they want me to review their VPS I'll usually do.

    Sorry, where is this proper benchmark? I don't see one in the OP. It looks like a single run of a reduced set of the usual, non real world tests without a final ranking or quantifiable comparable result to produce usefulness.

    Thanked by 1uptime
  • jsgjsg Member, Resident Benchmarker
    edited May 2019

    @TimboJones said:
    Sorry, where is this proper benchmark? I don't see one in the OP. It looks like a single run of a reduced set of the usual, non real world tests without a final ranking or quantifiable comparable result to produce usefulness.

    It's not my fault when you miss whole threads and don't look at signatures.

    Regarding your other remark: This was quick test and I said so. If a provider wants more, I'm willing to run series of tests - and I actually did so, e.g. over multiple weeks. The result/review was published here at LET.
    In those cases the results are averaged but significant deviations are mentioned, too.

    Have a nice weekend

  • Leave Khaleesi alone.

    Thanked by 1uptime
  • @jsg said:

    @TimboJones said:
    Sorry, where is this proper benchmark? I don't see one in the OP. It looks like a single run of a reduced set of the usual, non real world tests without a final ranking or quantifiable comparable result to produce usefulness.

    It's not my fault when you miss whole threads and don't look at signatures.

    Regarding your other remark: This was quick test and I said so. If a provider wants more, I'm willing to run series of tests - and I actually did so, e.g. over multiple weeks. The result/review was published here at LET.
    In those cases the results are averaged but significant deviations are mentioned, too.

    Have a nice weekend

    Your signature that goes to a download without any information? That signature? I need to read your source to figure out whatever benefit you think you added?

    Regarding your test results, having a substantially lower sequential write than random either means something is wrong with the server or your test method. Either way, I have no real takeaway from this review. I mostly care about the specifics of why you needed support and the specific response, and that was really glossed over.

Sign In or Register to comment.