Windows 10 Creators Update

[mpack]

ISTM that in order to solve a problem you first have to demonstate that a problem exists. A couple of anecdotes is not proof, and it certainly is not proof of a problem with VirtualBox. And as far as proving that something totally irrelevant is a magic cure for a likely non existing problem... well, there are certain realms of ridiculousness where I can't bring myself to go.

Windows slow? I'll give you the short version: Windows Update. Ask Microsoft about a fix.

I tried to post the link for some less "academic" people

Does academic mean "people who know what they're talking about"? Yes, I can see why you would want to avoid those.

[socratis]

It seems that is easier to ask for proofs or to complain about the source of that info than to try to solve a problem and maybe to try o help other people.

I wouldn't even know where to start addressing your reply. The fact that you're implying that we're not helping people? Really? Take a look at mpack's and my posts. Add the number of posts, do some math and you'll find more than your lifetime's worth of helping people. Oh, I forgot, that's called evidence, proof. And some people are allergic to that.

It's as simple as that: I need to use 3-4 virtual machines ...Windows 7 ...After installing the Creators Update I have noticed that all these machines were very slow, unusable.

The fact that you didn't even come here with a problem and you're just bringing it up? "Slow VB" is not a problem, it's a symptom. And you got to ask yourself why that happened. mpack already gave you a possible answer to that. Another issue might be that you run out of resources when running 3-4 VMs. But no, of course not. That would mean doing some research. Instead you get the first hocus-pocus unsubstantiated advice you found on random_blogger.com and follow it. Not only that, but you come as an authoritative figure to spread misinformation. Seriously now? And you expect not to be met with skepticism?

If you think that I have to prove or to explain anything, I think that you'd better consume your energy asking Oracle to release an update to solve this problem, II really think that they can explain a lot better.

Of course you don't have to prove anything, why would you. Like I don't have to prove that a lot of people are really beyond stupid and beyond help. They are proving it themselves.

And of course I shouldn't forget to mention that we don't have to lobby Oracle for a fix, because they will fix your imaginary problem in their next imaginary update.

Or or stop whinning and revert that Creators Update.

Do you even know at which forums you are posting? You are not in the Microsoft forums, they released the Creators Update.

I don't have the time and/or intention to do any benchmarks or explanations.

Pity that you have the time and energy to register and post, just so that you can spread misinformation based on half-baked misunderstandings. You could have channeled all that energy into understanding the problem. And maybe helping others. Really helping...

[massula]

Hi, all!

My intention with this isn't put more gasoline on the fire, but only report the results of the tests I said I would perform in my previous post on this thread

TL;DR: This method didn't work on my system.

My host is a Acer Aspire E15, 4 GB RAM,
Intel Core i5-6200U 2.3 Ghz CPU, dual boot with
Windows 10 Home Single Language with Creators Update (1703 version) and Ubuntu 16.04.

VirtualBox in Windows is 5.1.22 r115126. In Ubuntu is 5.1.4 r110228 (more on this later).

My guest is a Windows XP machine, 1 GB RAM, and 1 CPU assigned, without any advanced option enabled.

This guest's vhd (virtual hard disk) is an .vmdk file, dinamically allocated, with 15,00 GB total size, and 14,23 actual size.

I've made four clones:

a) 1 converting the vhd to .vdi format, with dinamically allocated size.

b) 1 converting the vhd to .vdi format, with fixed size.

c) 1 keeping .vmdk format, with dinamically allocated size, and the last one.

d) 1 keeping .vmdk format, with fixed size.

I've made not so scientific benchmarks, measuring the time between the click on start button at VirtualBox manager, and the first functioning click in the guest's start menu.

The measurement of the original one is 3m48sec

The startup times from the clones float arount 3m50sec and 4min10sec. I've took care to measure the second boot, since the first one will be a little bit bigger, because Windows recognize the vhd as new hardware.

To give an idea, the same original guest, that in Windows now spends almost 4 minutes to start, boots in 28 seconds in Ubuntu. This was the performance I had before on Windows 10 (maybe a little bit bigger, but booting within 1 minute), and this is what I'm expecting now.

When I did my first post on this topic, I didn't had the Creators Update yet. After I've downloaded it, VirtualBox stopped working completely. Even the virtual connections disapeared. The solution from my problem was download and install the newest version (5.1.22).

I've noticed this same slowness on other machines I have around, one with more robust specs. As a side note, I'm experiencing the same problem with VWware Player as well, so I'm convinced this is a Windows 10, not VirtualBox problem.

Before someone suggest it, I can't switch full time to Ubuntu now . Although the idea amuses me, and I had used this setup until last year, now I need some "unvirtualizable" software, so I spend more time on Windows than in Ubuntu.

Last, I would like to thank MariusFilo, socratis, mpack and everybody who post something on this forum.

Everyone is trying to help, in a way or another.

[massula]

So, digging this forum, I've found the problem in my system, and it turned out to be the almost the same explained on this thread

viewtopic.php?f=6&t=82516

I've also had Avast on host.

At first, I've tried only disable Avast for a certain period of time, and it didn't worked.

After, I've tried to include VirtualBox.exe in Avast Game Mode, and it didn't worked also.

Third, I've uninstalled Avast completely, rebooted, and guest looks a little bit faster, but not as in Ubuntu.

So, I've also uninstalled Malware Bytes, that was also present on this host, rebooted computer, and voilá!

Now, the same guest that was spending almost four minutes to boot earlier today now starts with 31 seconds.

So, for now, I will need to rely only in Windows Defender.

Wish me luck!

[socratis]

There's no "good luck" because Windows Defender is a very decent antivirus. And it's actually the only one that doesn't choke your host. In fact you should never have more than one antivirus, they're conflicting.

Glad you got it going and thanks for the feedback. Antivirus are becoming more and more intrusive, we'll keep the advice handy.

[mpack]

There's no "good luck" because Windows Defender is a very decent antivirus.

I see that Microsoft is being accused of antitrust issues concerning the bundling of Defender in Win10, in the EU courts I believe, and after a complaint by Kaspersky. I find myself in the curious position of actually supporting Microsoft on this one. ISTM that all of the other AV vendors have a vested interest in getting you to pay for bloatware protection you don't actually need, whereas Microsoft does not because they are not charging. Also you could consider that all AV does is patch up gaping security holes in the OS, in which case it might be argued that MS is perfectly entitled to remove the holes in its own product itself. Should they be required to leave defects in the OS unaddressed just because a third party market has built up around those defects? Should I have demanded that the VBox devs stay out of the VM cloning business just because CloneVDI previously had that market? I think I may have got short shrift on that one...

ISTM this is clearly different from the browser wars case, when Microsoft's motive was clearly to coopt the Internet into the MS revenue stream for its own benefit alone and was therefore contrary to consumer interests.