Why did formatting win 7 computers to windows 10 double their CPU-Z bench score? - windows-7

I didn't want to ask this question here. I asked it on superuser but didn't get an answer.
https://superuser.com/questions/1420073/why-did-formatting-win-7-computers-to-windows-10-double-their-cpu-z-bench-score
So I work at a company as an IT guy while I am doing my computer engineering degree. Doing hardware and software maintenance of computers is part of my job. I have had a weird experience with two of the computers. These two computers(one desktop one laptop) were the slowest computers in the company. The laptop is Dell Inspiron N5010 with i3 370M(2 cores, 4 threads) processor. The desktop is HP 500B MT with E5800(2 cores 2 threads) processor.
At first, both of these computers had windows 7 running on them. CPU-Z(1.87.0) benchmark of the desktop was 113(single thread), 227(multithread). The laptop was 82, 267.
After I formatted these computers with windows 10 and ran the same CPU-Z version benchmark, I got exactly double performance with both computers. Both single threading and multithreading scores got doubled.
After formatting with windows 10, desktop got 270, 510. Laptop got 180, 520.
What is causing this? Physical core number stayed the same. Logical core number stayed the same. I am baffled.

Is it possible that you upgraded from 32 bit Windows 7 to 64 bit Windows 10?
According to this FAQ under the point What algorithm does the benchmark use... they state that
the 32-bit version keeps using the legacy x87 instructions, resulting
in almost half of the x64 performance
edit: please remove question here because it is not about code. I answered on superuser as well

If the difference in speed is noticeable, it might have been an issue with the drivers on WIndows 7, or it might have had something to do with huge pages (enabling huge pages could boost the CPU performance significantly).
If you can't notice the difference in speed/responsiveness, it might just be a bug in CPU-Z (Have you tried the newest version 1.88?).
Going from Windows7 to Windows10 should not on its own result in such drastic changes in performance, the CPU benchmarks should be pretty close. Windows versions are also important, I've seen tests between Win10 1803 and Win10 1809 which show approx 10% increase in FPS in favor of 1809 (but that's GPUs not CPUs).

Related

How to limit PC performance to test software

I am developing a .NET application, and have the luxury of doing this on a fairly powerful desktop PC. I want to ensure it runs okay on PCs with much lower spec, but I don't have spare machines kicking around and can't really afford to buy them. Is there any way to simulate a lower-spec PC on my current PC, to get a feel for how the software might run?
Any help or advice would be very much appreciated.
*My PC is Windows 7 Ultimate 64-bit with 8-core Intel i7 and 16GB RAM.
You could install VMWare and install any OS you want, with any hardware specs you want, provided that they don't exceed your current working hardware of course.
Keep in mind that VMWare is just a virtualization layer. It emulates an OS but you are still running your code on the same i7.
http://www.asp.net/mobile/device-simulators Here is an example of several Visual Studio plug-ins that emulate devices. You can also install Windows 8 and run hyper-v. It's great for this kind of thing.

Visual Studio 2010 and/or win7 64 bit limits

Does anyone have experience writing apps in VisualStudio 2010 C Premium that uses large amounts of ram and multiple cpus?
I am about to order a workstation with Dual hex-core Xeon 5690 processors (12 cores total, 24 hyperthreaded) and 48 gigabytes of RAM, but first would like to know if VS can handle that number of cores and RAM.
(Of course this is all 64bit) I can't seem to find a straight answer either from MS or the hardware vendor, or from the Web.
Thanks
Update: someone just sent me this link
I realize now that that my question was mis-directed. The real issue is whether the target OS can address that much RAM and run dual cpu.
So, unless I'm misreading it, the infomation in the link above means that if you want to write an application that will run on Windows 7, and even if you require the 64bit version, you are limited to 16 gb. The only way to get around that is require users to run Win 7 Professional, Enterprise or Ultimate 64 bit versions.
It's a little tricky to be sure quite what you mean. VS2010 can certainly produce output that takes full advantage of such hardware. And the IDE itself will run very nicely on such a roomy machine.
Your update discusses memory limits imposed by Windows itself. You say:
If you want to write an app that will run on Windows 7, and even if you require the 64bit version, you are limited to 16 gb. The only way to get around that is require users to run Win 7 Professional, Enterprise or Ultimate 64 bit versions.
The reality is that nobody will buy a machine with more than 16GB RAM and then install an OS edition which does not support that amount of RAM. That would just be a waste of money. If your app requires more RAM than that and your customers are prepared to get hold of such a machine, then they will be quite happy to put the Pro version of Windows on it.
Visual Studio is just an IDE. Limitations are imposed by the compiler and an OS, and are usually listed in help under Limitations or something alike. Unfortunatelly, I don't have C compiler installed, but try searching through help while waiting for other answers.

XNA Windows Phone Simulator on 'Mac-Windows' vs. regular Windows

I installed my .net stuff recently on a Mac, i.e. preinstalled a version of Windows 7 before using Bootcamp. Does anyone have an explanation, why the Windows Phone 7 Simulator is so slow, compared to a Simulator installed on a 'regular' Windows system when deploying a target onto it?
Performance overhead might be caused because of the non-nativity of the hardware platform you are running it on. The emulator itself has a set of requirements - if these aren't met then you should expect serious drops in performance and stability (in case it starts).
When you're comparing the performance "Regular Windows", do you mean on another machine (PC)? If so, it could be down to hardware differences - (Graphics card, processor speed, less RAM, slower hard disk).
It could also be down to drivers - I don't know much about the Mac hardware, but it's possible Windows drivers aren't as good for hardware that's more commonly used by Macs.

Can GPU capabilities impact virtual machine performance? [closed]

Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 9 years ago.
Improve this question
While this many not seem like a programming question directly, it impacts my development activities and so it seems like it belongs here.
It seems that more and more developers are turning to virtual environments for development activities on their computers, SharePoint development being a prime example. Also, as a trainer, I have virtual training environments for all of the classes that I teach.
I recently purchased a new Dell E6510 to travel around with. It has the i7 620M (Dual core, HyperThreaded cpu running at 2.66GHz) and 8 GB of memory. Reading the spec sheet, it sounded like it would be a great laptop to carry around and run virtual machines on.
Getting the laptop though, I've been pretty disappointed with the user experience of developing in a virtual machine. Giving the Virtual Machine 4 GB of memory, it was slow and I could type complete sentences and watch the VM "catchup".
My company has training laptops that we provide for our classes. They are Dell Precision M6400 Intel Core 2 Duo P8700 running at 2.54Ghz with 8 GB of memory and the experience on these laptops is night and day compared to the E6510. They are crisp and you are barely aware that you are running in a virtual environment.
Since the E6510 should be faster in all categories than the M6400, I couldn't understand why the new laptop was slower, so I did a component by component comparison and the only place where the E6510 is less performant than the M6400 is the graphics department. The M6400 is running a nVidia FX 2700m GPU and the E6510 is running a nVidia 3100M GPU. Looking at benchmarks of the two GPUs suggest that the FX 2700M is twice as fast as the 3100M.
http://www.notebookcheck.net/Mobile-Graphics-Cards-Benchmark-List.844.0.html
3100M = 111th (E6510)
FX 2700m = 47th (Precision M6400)
Radeon HD 5870 = 8th (Alienware)
The host OS is Windows 7 64bit as is the guest OS, running in Virtual Box 3.1.8 with Guest Additions installed on the guest. The IDE being used in the virtual environment is VS 2010 Premium.
So after that long setup, my question is:
Is the GPU significantly impacting the virtual machine's performance or
are there other factors that I'm not
looking at that I can use to boost the
vm's performance? Do we now have to
consider GPU performance when
purchasing laptops where we expect to
use virtualized development
environments?
Thanks in advance.
Cheers,
Dave
EDIT:
The HDDs in the two systems are 7200 RPM, the E6510 having 500GB vs. the M6400 have 2x 250GB in a non-RAID configuration.
Also, when I turn off some of the graphics features of Windows 7 (host and guest) by going to non-Aero themes, VM performance visibly increases.
Just to closer off this question with my findings, what we have discovered was that driver performance was limiting the perceived performance of that virtual machine. With the default Dell drivers, which are built for "stabilty" the virtual machines would be visibly impacted in "visual" applications like IDEs (Visual Studio 2010) such that VS 2010 could not keep up with my typing. When we had some nVidia reference drivers installed, the IDEs were crisp and you couldn't really tell that you were in a VM anymore, which was my experience with the M6400s.
Thanks to everyone who threw out some ideas on the subject.
I am running two VMs on my development system simultaneously, one for development, and one for TeamCity. My graphics card on my Dell Optiplex is an ATI 2450, which is, quite honestly, complete crap. Personally, I have found RAM and CPU to make the most significant impact on my desktop. But since you are on a laptop, have you thought about the disk? Our M6400 has an SSD, and perhaps that is the biggest difference for your two laptops. I would not expect GPU to affect anything, unless of course you are trying to use the experimental Direct3D features in VirtualBox.
You guys are looking in the wrong places. Go to bios look for virturalization extensions AMD-v or VT-X. Off by default on most system. if it dosent have that option take a look at Sun Virtual box runs good on my older laptop with out virt support.
A GPU can significantly impact performance of any system. Visual Studio, for example, has a huge performance difference between on board video vs dedicated graphics.
That said, I would expect there are other differences. First, how do the two hard drives compare? notebook manufacturers love putting slow disks in machines in order to beef up their battery longevity numbers; and other the other side, sometimes they put in the faster drives to boost performance numbers. It really depends on what the new machine was marketed towards. Along these lines some hard drives also have configuration settings to determine their power / performance / noise levels. Depending on the drive you might be able to tweak this.
Another expected difference is the quality of memory. Nearly every dell I've used has had second or third tier ram installed. Sure they might both be DDR3 of a certain Ghz, but the quality of the chips is going to determine how they really perform. Sometimes by 200% of more.
Beyond those you start getting into chipset differences, mainly in the hard drive controllers. You can't do anything about this though.
The next thing I can think of is drivers. Make sure your up to date on everything you can. Also, test both Dell and nvidia supplied drivers. Sometimes nvidia has better drivers, sometimes the original ones from dell are better. That part is a crap shoot.
And, finally, consider blowing away the new machine and doing a complete reinstall from the bare metal up. Before installing any anti-virus or CPU sucking software, test your VM performace.

How much space should I leave for my Windows 7 partition?

I'm setting up a complete .NET development environment on my Macbook Pro.
I'm using Visual Studio 08 team suite, SQL server 2008, MS Office and other tools (like FinalBuilder, RegexBuddy, Beyond Compare).
How big should my windows 7 (beta currently) partition be? Will 100GB be enough?
NOTE: I wasn't sure if this was programming related enough for SO, so I'll just let the community decide if this question is relevant.
100 GB should be more than enough for all those apps. I've installed win7 in a virtual machine and the virtual HDD ended up with a size of 7GB (that's only the OS of course). Trying the same with Vista, for example, uses about 25GB. It seems they're making it lighter.
You described my laptop. 100 GB would leave approx. 40GB for Users directory.
100 GB will be plenty. You'll have OS, apps, but no music, pics, videos. 100GB is probably overkill, especially if you can resize it if needed.
I have Windows 7 installed on a laptop with 2 100 Gb hard drives.
Currently I'm using 18 Gb and that's with most of the primary stuff installed, but not Visual Studio or SQL, but those probably won't use more than 10 Gb (I reckon). I do have Virtual XP Mode installed which is probably quite large too.
The Windows folder is about 9.3 Gb
The User folder is 3.2 Gb (but I have some large files on my desktop)
The Program Files is 3.0 Gb
The rest of the files on the OS-drive are mostly driver files which you don't have to leave on the drive itself.
So 100 gb would probably even be an overkill, but does give you some headroom!
Windows7 is going to be a little bit smaller than Windows Vista. So if you create partition big enough for Windows Vista, it will be perfect for Windows7.
See Engineering7 blog for more information about disk space in Windows7.
I would give as much as you could to Windows 7, since it will probably become your primary OS. I find that I rarely use my OSX partition, except for cracking WEP.
100 GB is barely enough. You can install Windows 7 and the mentioned programs along with lot of other stuff, but once you get to have some lots of trash there and there plus you happen to download movies and such it gets cumbersome.
Unless you're relying on some other device for things other than those tools, I recommend a larger space allocation, of at least 150 GB

Resources