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.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
In our hearts you will always be a king. No matter if dethroned and defamed.
Uhm, I already demanded them to change the quote (to their disadvantage) a while ago. So feel whatever you please but fact is that there is no reason to assume that I won't make the demand if it's justified.
Sorry, the world isn't black and white and I'm not interested in either blindly loving or blindly hating any provider. Hint: I really really dislike AnthonySmith yet when they had a really nice promo and their product turned out to be really nice and good I gave them a positive review.
I will not hunt down, punish or bash any provider based on emotions, let alone based on what I perceive as mud pit emotions, nor will I praise any provider without good reasons and hard data.
So, again: If they put a ridiculously low limit on NVMe VPS too, I will make that demand. If they don't I will not.
That was and is BS. I never asked for the "king" tag, I never felt like a king, and I'm not interested in such a title/tag/whatever. So, for the "dethroning" I'm actually grateful (and did expressly ask for it myself) and the "defaming" is but a wet dream of the mud pit.
Have a nice weekend.
I heard about contabo years ago for being good value because apparently you are allowed to slam the cpus 24/7 non-stop. Pretty sure it is still the case right now. I have never used contabo before, but I'm pretty sure there's still customers flocking to contabo solely for the "cheap" hardware and the ability to slam cpus 24/7. Due to these customers existing and contabo not caring about such behavior, the network and disk limitations are quite understandable.
I'm not sure how much a quote from a randomer on a forum that either a) you’ve not heard of or b) you know is a fucking shitfest is going to contribute to sales.
@jsg is a very opinionated person. He is a very very opinionated programmer. He is very opinionated with politics. And it seems he is also opinionated on the use case of vps. I see where he is coming from. I think his benchmarks are about consistency. If something runs well now, he expects it to run just as well every hour of the day.
If you ask me, if that consistency was really necessary then you would have went with a dedicated server instead of a vps. But that's just me, an unopinionated person without any important services. Maybe if you had a database server being slammed 24/7 or something like that then you would care for having a consistently good vps. Unfortunately most people are also unopinionated in this world.
(a) it seems your assumption of how deep people dig is wrong. (b) Not everybody, especially not intelligent people with a good education, give a lot (if anything) about "shitfest" actors views.
One may call that opinionated or not, but a performance difference of 5x and more is very significant and I bet that 95+% of users strongly prefer to see a reasonable spread. So if a VPS almost always stays within say a +-20% performance frame it's considered acceptable, but if it's +- 250% most people are very unlikely to consider that OK.
But hey, I'm certainly not blocking your way; feel free to buy such a VPS.
Jesus Christ, if they 'dig deep' it's going to be even worse.
I suspect most people will just see that blurb and think 'that's a bit weird' and carry on doing whatever they were doing on the site.
Depends, e.g. on whether they know LET or not.
I don't know what "people" think or not, I can only tell what I think when seeing "testimonials": I don't care; at all.
But hey, some users here - not me - took that quote to be totally important, important enough to be fought over.
If they do, they'll know it's a shitfest. If they're in WallOfText Club, they'll probably shoot their load and immediately buy a VPS, so I hope you'll take responsibility for that and process refund requests directly.
Well, 1 person did.
You're right, but all I can see from your results is the min and max. I believe your benchmarking runs over an entire month? Without values like the variance, a graph or the full data, there's no way to tell if your min and max are outliers.
I get that your benchmarking is supposed to show the full picture versus a quick glimpse like yab does. But it is a blurry picture. Especially since your formatting and script changes every so often and it is hard to keep up even if one reads your entire wall of text.
Until the day I see people whining on LET about being suspended by contabo for cpu abuse, I won't be buying anything from contabo.
You need a decent provider, where the cpu load is so low, that you can her the neighbours fuck at night.
Funny. When I present a full benchmark output there are complaints about, oh how sooo long it is and "yellow wall of text". When I shorten it and reduce it to the core, there are complaints, too.
What I usually provide is all the information including min and max of all benchmark runs both in values and in percentage.
That btw. is dimensionally more and better than any other review I've seen here so far.
The data I provide do not change and certainly not often. What occasionally "changes" is some additional info/result types. And no, my benchmarking is not so much about the "full picture" but rather about (a) comparability, and (b) lots of result sets gathered over multiple days (or even weeks) and during different times of day and night as opposed to the commonly seen single "snapshots".
And again, unlike most other reviews, with mine one can see whether a VM under test performs consistently or not. In fact I even have time stamps on each and every test run, so that, should it it be required, I can find out whether performance e.g. is different during day and night hours.
As for "graphs" and the like: that translates to yet more work for me to do; I might even be willing to do that work but definitely not when the reward is just complaining, demanding ever more, and bashing or even trying to insult me. Before asking someone to do more free work look up "motivation".
I don't care at all whether anyone does or does not buy from Contabo.
Surely not having everyone mug you off and to have you hailed as the true King of Benchmark is all the motivation you need? C’mon, prove all those doubting motherfuckers wrong!
Anyone bringing up the "king" tag IMO is just an idiot with ill intentions. I never asked for that tag and in fact asked for it to be changed for something more modest - but hey since when do facts get in the way of idiots with ill will ...
But actually today I'm in a (granted kind of weird) way grateful for @jbiloh meaning well and giving me that tag, because it helped a lot to distinguish idiots and mud dwellers from homo sapiens.
This paragraph. So good.
Also this. I love it. Bravo!
I feel like someone sneezed at the table, and went on a 20 minute apology.
Any one in LET should know early results can be vendor manipulated during massive sales for a myriad of reasons. Anyone could be tricked into posting positive results. I’d hope most LET users understand that but … I suppose if that was true this 5 page post wouldn’t have been created. Sorry to see this bubbled up this way.
In this case, it's probably because contabo's servers were new? I think @jsg did benchmarks when contabo launched some new epyc servers and when they launched at new locations so their servers were probably somewhat vacant back then.
I’ve decided to substitute this reality and chosen to believe this involved a massive drug ring, ninjas, and some kind of eye-scanning AI identity stealing software.
Yes, the NVMe VPS benchmark was pre-launch and I said that clearly.
The other ones like the test of their us-american locations were on production nodes. Btw, in the cases where I benchmarked pre-lauch (or just launched) nodes I always did follow up tests and put the results on LET if the later results in production turned out to be significantly worse.
In fact the OP is based on benchmarks run on my own private VPSs to a large degree.
And I don't do that only with Contabo, I also do it e.g. with my NexusBytes VPS but that one didn't and doesn't show changes of any significance, neither does btw. the InceptionHosting VPS.
Because it doesn't make sense for a review to request to remove his result unless admitting the data/testing was faulty. It's a past tense quote about specific testing he performed. That's different than a present tense or future tense quote implying performance where the message is no longer true and requesting it removed would be appropriate.
In other words, their reputations are tied together for better or for worse. They want to leave a quote up by the Elizabeth Holmes of VPS benchmarks, that's on them.
The quote has no date on it and no numbers shown to back the statement.
Best case scenario, it is very much like an opinion.
Another point, “best tested so far” is subject to change as time progresses.
His statement keeps on changing
You can see the trend.
Here is a challenge from me for @jsg
Share what do you think NVME speed for Contabo and I will get VPS from Contabo.
If the speed is not same as your number (or within 10 percent range) then ask Contabo to remove your quote.
If speed is >= your mentioned speed (within 10 percent range), I will publicly apologize
Roll em back further, you might finally see (or not) a brain in there.
You are not in a position (anymore) to challenge me.
I'll do what I said. I'll wait what, if any, changes monday brings and if my NVMe VPS is slower or there are any other negative changes. If the result of their changes is a slower NVMe/ significantly lower IOPS I'll ask them to take the jsg quote out; if not then I won't.
Keep in mind that it was just a vague guess that the change is about limiting NVMe IOPS. It might as well just be changing memory stick or some maintenance. So, of bloody course my reaction is conditional and depends on what exactly they change and how.
Whether you'll apologize or not for this whole speculation dance and assumptions and interpretations you made is entirely up to you and to your quality as a human, I don't care.
Thank you, @jsg
Would you please be kind enough to share the mean NVME speed (that you find better than any other VPS) ?
You will dance around it, but I doubt it that you will share the number.
No matter how bad performance becomes, the quote will stay.
I already did back then.
Putting your subjective speculation in bold doesn't make it true. We don't need speculation, it's just a few days of patience and we'll have facts and data.
Edit: In case you meant the current result before any potential changes on monday, here you go:
That benchmark runs were done end of October, beginning of November (End of test Nov. 4 2021).
who is included in “we” ??
Well, this is why I distrust quotes, reviews, recommendations, testimonials etc. - unless I know the author(s) of the listed.
"Well, guys, your service is grea... oh shit, why your NVM is 3% slower than yesterday?! No, your service is no longer great, remove my testimonial".
I definitely need more popcorn in this case.