[2008] Graphics Test. Looking for results from Vista users. Use Post # 39 attachment.

I recently upgraded to Vista Business from XP Home. I found that my Visual Basic 2008 graphics applications is running much much slower in Vista than it was in XP. I took a small piece of my application and created a test program. I've set my computer up for dual boot so I now have XP Home and Vista Business on it. I've run my test program from both OSs and the results are tremendously different.

I have a custom built computer that has a Microstar Int. KT6 - V mother board. It has an AMD Athlon XP 3200 + 2.2 GHz with 2 GB RAM. It's a 32 bit processor. I do have the 32 bit version of Vista installed. I have only PCI slots and one AGP slot on my mainboard. I'm using a NVIDIA GeForce2 Graphics card that I bought back in 2005 with 32 MB RAM. My Vista Experience Index subscores are as follows:
Processor : Calculations per second - subscore 3.6
Memory(RAM) : Memory operations per second - subscore 4.2
Graphics : Desktop performance for Windows Aero - subscore 1.9
Gaming Graphics: 3D business & gaming graphics performance - subscore 1.0
Primary hard disk : disk data transfer rate - subscore 5.3

The test program that I created runs 3 screens that are the same except for the pen thickness. Screen 1 uses a pen thickness of 1. Screen 2 uses a pen thickness of 2 and screen 3 uses a pen thickness of 8. When I run the Graphics Test program my results from XP Home on my computer are as follows.
Screen 1 duration = 7,781 milliseconds or 7.781 seconds
Screen 2 duration = 7,796 milliseconds or 7.796 seconds
Screen 3 duration = 7,796 milliseconds or 7.796 seconds

All pretty much the same.

From Vista Business my results are as follows:
Screen 1 duration = 119,062 milliseconds or 119 seconds - Pen Thickness = 1
Screen 2 duration = 20,234 milliseconds or 20.23 seconds - Pen thickness = 2
Screen 3 duration = 21,609 milliseconds or 21.6 seconds = Pen thickness = 8

Usually when the pen thickness is lower the drawing is quicker but my Vista results show something very unusual happening. When the pen thickness is 1 the drawing is 6 times slower than when it is 2. The drawing is about 15 times slower in Vista than in XP when the pen thickness is 1.

I'm going to attach the program I created and someone can let me know if they were able to install it and run it from the attachment. If you run it let me know your results if you run it from Vista and let me know your computer set up. What's all the goods on your computer. What kind of graphics card do you have. Is it PCIexpress, PCI or AGP.

Next post will have attachment.

Are you running Aero Glass? With such a low Graphics score that could be a problem. Try turning down the visual effects in the OS and see if that makes a difference. If it does then you can assume that anyone running Vista on reasonably recent graphics hardware should have no issue.

Remeber I'm looking for Vista users. If you have XP and you want to run it and give me the results that's ok but I'm primarily looking for Vista users. After you install the Graphics Test program it won't place an icon on the desktop for you but if you look in programs it should be there as Graphics Test. Go to post # 39 for attachment.

Aero Glass is off. I tried adjusting the visual effects before and no improvement. I tried it again and still no improvement. I went to the visual effects screen and clicked adjust for best performance. I ran another test and there was zero improvement. I tried putting an ATI Radeon HD 2400 PRO AGP graphics card in my computer and the subscore for Graphics went from 1.9 down to 1.0. There's a whole other story on trying the ATI card.
http://www.vbforums.com/showthread.php?t=531015

When I get the money I'll upgrade my computer but in the mean time I'm wondering how my application will run on computers that are more capable than my own.

I tried downloading and installing from the attachment but it doesn't work. I've created a setup program that will install Test Graphics fine but when I zip it something gets left out.

Here's another attachment. I just tested it and was able to download, unzip and install successfully. Graphics Test will be in list of programs after installation.

If you run the program and give me the results let me know the essentials on your computer as I said before and if you could also give me the Windows Experience Index Score and subscores. Start > Welcome Center > Show More Details (top right of screen) > click Performance in lower left and then you'll have the screen that gives you the Windows Experience Index and the subscores for different areas of performance. If nothing else give me the Graphics subscore and Gaming Graphics subscore.

I'm sorry but I had to remove the attachemnts as they were executables. Its against the Forums olicy to allow them. Please upload your source code so members can compile their own versions for testing for you.

You can not compare XP environments against Vista. Vista uses alot more memory and DirectX. Its like comparing Apples to Oranges.

Didn't know about that policy but now I do. I seriously doubt anyone would be interested in doing the work to put it all together that's why I created an install program but if no one does that's ok. If I don't get any feedback it's no big deal. Go to Post # 39 for attachments for Graphics Test.

I dont doubt that your app is safe but we have to apply the rules fairly to all. Its basically to help protect members from compiles that may be harmful to their systems since there is no source code they have no way to know if its safe. Now that you have uploaded your source I'm sure you actually may get more testers participating. :)

What if I were to provide a link to where they could download my app. Would that be ok? If they don't want to take the risk of downloading it they don't have to.

Nope, its the same thing and could get into "advertising" areas. Sorry

That's ok but it seems that I've seen many times on VBForums that people put up links for downloading of programs but maybe that was ok because they weren't programs written by VBForums memebers.

I've rewritten my program some so it would be easier to recreate.

HTML Code:

---

Option Strict On
Imports System.EventArgs
Imports System.Math

Public Class Form1
    Sub Form1_Load(ByVal sender As Object, ByVal e As System.EventArgs) Handles Me.Load
        XLimit = System.Windows.Forms.Screen.PrimaryScreen.Bounds.Width
        YLimit = System.Windows.Forms.Screen.PrimaryScreen.Bounds.Height
        Label2.Text = My.Computer.Info.OSFullName & R & Int(My.Computer.Info.TotalPhysicalMemory / 10000000) / 100 & " GB"
    End Sub

    Sub Timer1_Tick(ByVal sender As System.Object, ByVal e As System.EventArgs) Handles Timer1.Tick
        Dim i As Integer
        VisualVarietyFactor = 0 ' Adjust this value 0 to 9 to get different visual effect
        If IC = 0 Then
            StartTime = DateTime.Now
            H = XLimit / 2 : K = YLimit / 2
            Dph20Toggle = 1
            Ra(0) = 0.49 * YLimit
            RIC(0) = TP / 100
            For i = 0 To 99
                i2 = i + 1
                SRL(i2) = i * RIC(0)
                PX(i2) = Sin(SRL(i2)) * Ra(0) + H
                PY(i2) = Cos(SRL(i2)) * Ra(0) + K
            Next i
            Select Case Screen
                Case 2
                    PenThickness = 2
                Case 3
                    PenThickness = 8
            End Select
            If Screen > 1 Then
                Label2.Text = Label2.Text & R & "Screen " & Screen - 1 & " duration = " & TotalTimeForScreen(Screen - 1) & " milliseconds"
            End If
        End If
        Select Case VisualVarietyFactor
            Case 0
                istep1 = 2 : istep2 = 2
            Case 1
                istep1 = 1 : istep2 = 1
            Case 2
                istep1 = 1 : istep2 = 2
            Case 3
                istep1 = 1 : istep2 = 3
            Case 4
                istep1 = 2 : istep2 = 3
            Case 5
                istep1 = 2 : istep2 = 2
            Case 6
                istep1 = 1 : istep2 = 1
            Case 7
                istep1 = 1 : istep2 = 2
            Case 8
                istep1 = 1 : istep2 = 3
            Case 9
                istep1 = 1 : istep2 = 4
        End Select
        If VisualVarietyFactor < 5 Then
            Ra(0) = SelectValue2(IC, 100) * 0.006 * YLimit
        Else
            Ra(0) = 0.49 * YLimit
        End If
        Label4.Text = "Pen Thickness = " & PenThickness & " "
        IC = IC + ICincrement
        If Screen = 4 Then ICincrement = 0
        If IC = 500 Then
            TimeAfterStart = DateTime.Now
            TotalTimeForScreen(Screen) = TimeAfterStart.Subtract(StartTime).TotalMilliseconds
            IC = 0 : Screen = Screen + 1
        End If
        Label1.Text = "Screen " & Screen & "  IC = " & IC
        GraphicsTest = Me.CreateGraphics
        ' You can change SelectValue3 to SelectValue or SelectValue2 in below line or change SelectValue to one of the other 2 to get different results for color
 
        GraphicsPen = New Pen(Color.FromArgb(SelectValue3(IC * 2, 255), 160 - SelectValue3(IC * 2, 150), SelectValue(IC * 2, 50) * 4), PenThickness)

        If Screen < 4 Then
            i2 = 1
            For z0 = 50 To 500 Step 50
                If IC = z0 Then
                    Dph20Toggle = Dph20Toggle * -1
                    Exit For
                End If
            Next z0
            Select Case Dph20Toggle
                Case 1
                    For i = 1 To 100 Step istep1
                        i2 = i2 + 1
                        Select Case i2
                            Case 1
                                SRL(i) = SRL(i) + 0.011
                            Case 2
                                SRL(i) = SRL(i) - 0.022
                            Case 3
                                SRL(i) = SRL(i) + 0.044
                        End Select
                        PX(i) = Sin(SRL(i)) * Ra(0) + H
                        PY(i) = Cos(SRL(i)) * Ra(0) + K
                        If i2 = 3 Then i2 = 0
                    Next i
                Case -1
                    For i = 1 To 100 Step istep2
                        i2 = i2 + 1
                        Select Case i2
                            Case 1
                                SRL(i) = SRL(i) + 0.011
                            Case 2
                                SRL(i) = SRL(i) - 0.022
                            Case 3
                                SRL(i) = SRL(i) + 0.044
                        End Select
                        PX(i) = -(Sin(SRL(i)) * Ra(0)) + H
                        PY(i) = -(Cos(SRL(i)) * Ra(0)) + K
                        If i2 = 3 Then i2 = 0
                    Next i
            End Select

            For i = 1 To 100
                po(i) = New Point(CType(PX(i), Integer), CType(PY(i), Integer))
            Next
            Dim P1Points As Point() = {po(1), po(2), po(3), po(4), po(5), po(6), po(7), po(8), po(9), po(10), po(11), po(12), po(13), po(14), po(15), po(16), po(17), po(18), po(19), po(20), po(21), po(22), po(23), po(24), po(25), po(26), po(27), po(28), po(29), po(30), po(31), po(32), po(33), po(34), po(35), po(36), po(37), po(38), po(39), po(40), po(41), po(42), po(43), po(44), po(45), po(46), po(47), po(48), po(49), po(50), _
                            po(51), po(52), po(53), po(54), po(55), po(56), po(57), po(58), po(59), po(60), po(61), po(62), po(63), po(64), po(65), po(66), po(67), po(68), po(69), po(70), po(71), po(72), po(73), po(74), po(75), po(76), po(77), po(78), po(79), po(80), po(81), po(82), po(83), po(84), po(85), po(86), po(87), po(88), po(89), po(90), po(91), po(92), po(93), po(94), po(95), po(96), po(97), po(98), po(99), po(100)}
            GraphicsTest.DrawPolygon(GraphicsPen, P1Points)
        End If
    End Sub

    Function SelectValue(ByVal N As Double, ByVal SelectLimit As Integer) As Integer
        Dim i As Integer ' This Function will generate numbers from 1 to SelectLimit
        If N < 1 Then N = 1
        N = Int(N)
        If N <= SelectLimit Then
            SelectValue = CInt(N)
        Else
            For i = 1 To 100000
                If N - SelectLimit * i <= SelectLimit Then
                    SelectValue = CInt(N) - SelectLimit * i
                    If SelectValue <= 0 Then SelectValue = 1
                    Exit For
                End If
            Next i
        End If
    End Function

    Function SelectValue2(ByVal N As Double, ByVal SelectLimit As Integer) As Integer
        Dim i, i2 As Integer ' This Function will generate numbers from zero to SelectLimit
        N = Int(N)
        If N <= SelectLimit Then
            SelectValue2 = CInt(N)
        Else
            For i = 1 To 100000 Step 2
                i2 = i + 1
                If N > SelectLimit * i And N <= SelectLimit * i2 Then
                    SelectValue2 = SelectLimit - (CInt(N) - SelectLimit * i)
                    Exit For
                End If
                If N > SelectLimit * i2 And N <= SelectLimit * i2 + SelectLimit Then
                    SelectValue2 = CInt(N) - SelectLimit * i2
                    Exit For
                End If
            Next i
        End If
        If SelectValue2 < 1 Then SelectValue2 = 1
    End Function

    Function SelectValue3(ByVal N As Double, ByVal SelectLimit As Integer) As Integer
        Dim i, N2, IntN As Int32 ' This Function will generate numbers from zero to SelectLimit
        If N < 1 Then N = 1
        If Double.IsNaN(N) Then
            N = 1
        End If
        IntN = CInt(N)
        For i = 0 To 100000 Step 4
            N2 = IntN - SelectLimit * i
            If N2 <= SelectLimit / 2 Then
                SelectValue3 = N2
                Exit For
            End If
            If N2 <= SelectLimit Then
                SelectValue3 = CInt(SelectLimit / 2 - (N2 - SelectLimit / 2))
                Exit For
            End If
            If N2 <= SelectLimit * 2 Then
                SelectValue3 = N2 - SelectLimit
                Exit For
            End If
            If N2 <= SelectLimit * 2.5 Then
                SelectValue3 = SelectLimit - (N2 - SelectLimit * 2)
                Exit For
            End If
            If N2 <= SelectLimit * 3 Then
                SelectValue3 = N2 - SelectLimit * 2
                Exit For
            End If
            If N2 <= SelectLimit * 4 Then
                SelectValue3 = SelectLimit - (N2 - SelectLimit * 3)
                Exit For
            End If
        Next i

    End Function

    Private Sub Form1_MouseClick(ByVal sender As Object, ByVal e As System.Windows.Forms.MouseEventArgs) Handles Me.MouseClick
        Close()
    End Sub

End Class

Module Variables
    Public GraphicsTest As System.Drawing.Graphics
    Public IC As Integer = 0
    Public a As Integer = 1
    Public Screen As Integer = 1
    Public GraphicsPen As Pen
    Public po(100) As Point
    Public PX(200), PY(200) As Double
    Public Ra(2) As Double
    Public H, K As Double
    Public i, i2, istep1, istep2 As Integer
    Public SRL(240) As Double
    Public SRLmp As Integer
    Public Dph20Toggle As Integer
    Public XLimit, YLimit As Double
    Public Const TP As Double = 6.2831853 ' Two times Pie
    Public RIC(1) As Double
    Public R As String = Chr(13) & Chr(10)
    Public PenThickness As Integer = 1
    Public TotalTimeForScreen(9) As Double
    Public StartTime As DateTime
    Public TimeAfterStart As DateTime
    Public ICincrement As Integer = 1
    Public VisualVarietyFactor As Integer
End Module

---

Use attachment in Post # 39.

I just found out about the 10000 character limit for a post because I exceeded it and then brought it down. Above post is very close to limit now. I eliminated 3 modules by making 3 functions associated with the main form. The chances that someone will take the trouble to recreate this should go up but if no one does that's ok. I'll say it looks cool visually and you can alter the value VisualVarietyFactor from 0 to 9 to get different effects and you can also play around with GraphicsPen to get different colors. I'm principally interested in getting the test results but maybe I can lure someone into taking the trouble to run it by telling them that it looks really cool to run and it does. After you get the results you could alter it for your own pleasure. Label2 needs to be big. Shouldn't be autosize.

A Vista Experience score of 1.0 is probably your problem, particularly as the lowest score is Graphics

@EntityX, you do know you can zip the project source files up and attach it to a post instead of just the files?

See Post # 39 for Graphics Test attachment.

Quote:

---

Originally Posted by EntityX

OK. Here's all the source files and the project file in one zip. If you have Visual Basic 2008 installed on your computer then just unzip the attachment.
If you're looking at the unzipped files you'll see :
Form1.Designer.vb
Form1.vb
Graphics Test
Variables.vb

Double click Graphics Test. That's the Visual Basic Project file. A solution should be created and Visual Basic should open. You can then run the program.

---

it doesn't run. you've removed too many files. zip it again + repost

ok. i've run it. interesting graphics. it runs well on my PC (vista x64 home premium, on amd phenom 8.8ghz quad core with 1gb graphics card. windows experience score = 4.8).
do you want me to time it? how?

screen 1 duration = 57842.9775 milliseconds
screen 2 duration = 9627.3135 milliseconds
screen 3 duration = 10540.341 milliseconds

Quote:

---

Originally Posted by .paul.

ok. i've run it. interesting graphics. it runs well on my PC (vista x64 home premium, on amd phenom 8.8ghz quad core with 1gb graphics card. windows experience score = 4.8).
do you want me to time it? how?

---

With a quad core and all that power its still only a 4.8? :confused:
I have a dual core and average memory and I get a 5.5 minimum index score :D

I'll test the code when I get home tonight. ;)

Thanks very much .paul. I appreciate the results. You have very serious power on your computer yet my XP screen durations are lower. When the pen thickness is 1 the screen duration is still about 6 times slower in comparison to when the pen thickness is 2 just like on my computer when Vista is being used.

From XP Home on my computer:
Screen 1 duration = 7,781 milliseconds or 7.781 seconds - Pen Thickness = 1
Screen 2 duration = 7,796 milliseconds or 7.796 seconds - Pen thickness = 2
Screen 3 duration = 7,796 milliseconds or 7.796 seconds - Pen thickness = 8

From Vista Business on my wimp computer results are as follows:
Screen 1 duration = 119,062 milliseconds or 119 seconds - Pen Thickness = 1
Screen 2 duration = 20,234 milliseconds or 20.23 seconds - Pen thickness = 2
Screen 3 duration = 21,609 milliseconds or 21.6 seconds - Pen thickness = 8

From Vista x64 Home Premium on .paul's super power PC
screen 1 duration = 57842 milliseconds or 57.84 seconds - Pen Thickness = 1
screen 2 duration = 9627 milliseconds or 9.627 seconds - Pen thickness = 2
screen 3 duration = 10540 milliseconds or 10.54 seconds - Pen thickness = 8

It's pretty close to a 2 : 1 ratio between our computers as far as durations go for running the screens when comparing results from Vista.

There may be some kind of glitch in Visual Basic 2008 in it's compatibility with Vista or some glitch in Vista in it's use of the .NET Framework 3.5 and Visual Basic 2008 that creates the unusually slow drawing when the pen thickness is set to 1.

.paul I'm giving you a positive rating for your post. Anyone who runs my test program from Vista and gives me the results I'll give a positive rating at this time up to 10 different results from 10 different members. That should be enough.

.paul was there something that you did to eliminate the original problem you had after unzipping.

If you look at the comments in the code it will tell you how to get some variations in the visual effect. You can alter the value of VisualVarietyFactor. It's set to 0 but you can set it to any value 0 to 9 to get different results. Look at second line down under where you see Sub Timer1_Tick. After you run the test and get the results you could set the pen thickness to 2 which will give you 6 times better speed from the first screen. You can play around with the values for GraphicsPen to get different results for colors.

Code snippet from line 76 of Form1.vb

Code:

---

GraphicsPen = New Pen(Color.FromArgb(SelectValue3(IC * 2, 255), 160 - SelectValue3(IC * 2, 150), SelectValue(IC * 2, 50) * 4), PenThickness)

---

The RGB color component values shouldn't go above 255 so for the functions SelectValue, SelectValue2 and SelectValue3 the second value inside the parentheses should be 255 or less unless your playing around with the result some how. You could replace PenThickness(see above code snippet and scroll right) with 2 and delete label4 if you already have the test results and just want to play around with the graphics.

Also if you run the test and get the screen durations could you also give me the subscores for Graphics and Gaming Graphics. See post #7. It tells you how to get to the screen that gives you the Windows Experience Index score and subscores for the different types of performance.

]I didnt read the entire thread but are you comparing times when tested in the compiled exe state or in the IDE as source code?

Quote:

---

Originally Posted by RobDog888

[color=navy]With a quad core and all that power its still only a 4.8? :confused:
I have a dual core and average memory and I get a 5.5 minimum index score :D

---

yeah well it would be off the scale but the 1gb graphics lets it down:afrog:

I've run the test both ways. When using XP there's just about zero difference in the test which was surprising to me. When I used Vista and compared the times running from IDE or from the program installed and running outside Visual Studio the numbers were very close. IDE numbers were just slightly slower, that is screen durations were just slightly longer.

I just did another comparison. These are results from Vista Business only.

Screen 1 duration from IDE = 117,703 milliseconds or 117.703 seconds
Screen 1 duration from installation = 117,390 milliseconds or 117.39 seconds

Screen 2 duration from IDE = 20,109 milliseconds or 20.109 seconds
Screen 2 duration from installation = 20,062 milliseconds or 20.062 seconds

Screen 3 duration from IDE = 21,656 milliseconds or 21.656 seconds
Screen 3 duration from installation = 22,281 milliseconds or 22.281 seconds


I tried it again and got similar results. Note how the screen 3 duration is higher from the installation but when I did the tests again the screen 3 durations were the same from IDE and installation.

In the above testing Visual Basic 2008 was targeting .NET Framework 3.5. I was wondering if I targeted .NET Framework 3.0 or 2.0, would it make a difference. It didn't. I tried both 3.0 and 2.0 from IDE and 2.0 from installation and all the results pretty much stayed the same. You still see the times about 6 times slower when the pen thickness is 1 in comparison to when the pen thickness is 2. Microsoft still has some debugging to do of something. Vista probably since the problem doesn't exist when you run Graphics Test from XP.

I was gonna try it for you but its missing the resx and other files so I cant rebuild the exe. Sorry but I want to do a fresh compile of the exe.

Don't worry about it. I reported the problem as a bug at the Microsoft Connect site and just got an email that said that they were able to reproduce the problem I reported. I sent them a zip attachment of the installation program for Graphics Test.

Quote:

---

We were able to reproduce the issue you are seeing. We are escalating this bug to the product unit who works on that specific feature area. The product team will review this issue and make a decision on whether they will fix it or not for the next release

---

The one result I got from .paul.'s computer is really good enough. He has enormous power on that computer. I didn't even know they had 8.8 GHz computers out there. And now I have confirmation from Microsoft Connect too.

Hi, I've got some interesting results for you (at least I think they are.)
They are somewhat near your XP results.

These are my computer specifications:
Asus P5K (Intel) Motherboard
Intel Quad Core Q6700 @ 2.66GHz
4GB DDR2 RAM
Asus/nVidia GeForce 8800GTS 512MB
Windows Vista Ultimate 32 with Aero fully enabled.
Index Score: 5,8
Screen 1 Duration: 28082ms (it is weird that this takes so long!)
Screen 2 Duration: 7790ms
Screen 3 Duration: 7781ms

So it's only the first that takes longer.
By the way, I've never heard of a quad core processor with 8.8Ghz o_O must be giant overclocking or either it might be a 2.2Ghz. It's just a slight difference :P

Quote:

---

Originally Posted by RobDog888

With a quad core and all that power its still only a 4.8? :confused:
I have a dual core and average memory and I get a 5.5 minimum index score :D

I'll test the code when I get home tonight. ;)

---

Well the problem with the scores is that they only state the minimum score. So he may have 4 scores in the 5+ region (e.g. CPU, HDD and the two graphics scores) but score only 4.8 in the memory (a friend of mine has x64 Vista with 1GB of memory and his memory score is 4.3, says that x64 requires about twice as much memory as x86).

Quote:

---

Originally Posted by EntityX

The one result I got from .paul.'s computer is really good enough. He has enormous power on that computer. I didn't even know they had 8.8 GHz computers out there. And now I have confirmation from Microsoft Connect too.

---

Actually, there are no commercially available 8.8 GHz CPUs. Last time I checked, the highest clock speeds were in the 5 GHz region. I believe that what .paul. meant was that he has quad core (4 processors) at 2.2 GHz each.

Btw, I also couldn't do a fresh compile of the exe, and I'd really like to see the results. I'm running Win Vista Business, dual core 2.4 GHz, 2 GB RAM. Could you pls upload the entire project, along with the .resx files and everything?

Thanks for the extra results. I've got to run now but later today I'll post again with a better attachment that has all the goods for my Graphics Test app since I see that there is some interest in it. You could start a fresh project and copy and paste from above post. Place 4 labels. Timer1 should be enabled and interval = 10.

Actually, if you download the attachments from #13, and place them in a project, it works excellent.

Now, I've done some additional testing for you: I added some logging code to the project: It logs the time needed to perform each drawing action. I ran it:

I've attached the file. It is kind of big (80kb :P) as it logged the values for each screen.

You can notice that from certain IC's the time required is increased at the first screen. I dont know if it is much use to you though, but it didnt take much time so I just did it.

(System specifications are in previous post)

@EntityX, Thanks. I know we can run the included exe but for safty and following the Forums rules it really shouldnt have been included. ;)

RobDog are you referring to Andicioz's post just before your last post. I haven't posted any executables on this site since you told me about the policy.

Andicioz thanks for your work. I ran your test. It illustrates clearly how the frame rate changes during different portions of screen 1. When the drawing area diameter is small the frames run by rather quickly but when the drawing area diameter goes up the frame rate(IC for Image Count) slows way down. On my computer when using Vista certain portions of screen 1 will run at only around 1 frame per second. Andicioz you should know that RobDog(and probably any other VBForums moderator) doesn't want us posting executables. I did myself before and they were unposted.

I'm attaching a zip of the VB files. No executables. If you have Visual Basic 2008 on your computer unzip and click on Graphics Test solution or Graphics Test project file(inside folder Graphics Test) and Visual Basic should open and you should be able to run it. Go to Post # 39 for attachment.

I said before that anyone who runs the test and gives me the results I'd give their post a positive rate. I withdraw that statment now because I have all the results I need. So only run it if you want to run it. I might give your post a positive rating if I feel you've contributed something worthwhile but no promises. Here and there when I'm really helped out I do rate the posts.

If you run the test and have some results that are significantly different from what you've seen so far then that would be a reason to post the results. Screen 1 is very slow. Screens 2 and 3 are a lot faster. This is what everyone is seeing in Vista.

Andicioz I see you just attached a text file. So I'm not sure what RobDog is referring to.

To add a note, in case anyone is wondering if the just released Visual Basic 2008 with SP1 makes any difference in the test, it doesn't. I've already installed it and tested and no difference. I installed Visual Basic 2008 Express with SP1. I doubt the Standard version with SP1 would make any difference but if someone finds that it does they can let us know. Microsoft Connect team would have tried it and told me I would think. See post # 30.

The attachment in Post # 36 has executables in the bin and obj folders but I didn't know about them when I made the attachment. It was an unintentional post of executables.

Here's an attachment that has no executables. After you double click the Graphics Test project file inside folder Graphics Test the obj and bin folders will be created and they will have executables in them then so it doesn't make that much difference but I can say that this zip file doesn't have any exe files in it as far as I am aware.

No problems here.

Vista x64 Ultimate
AMD 9500 Quad Core
4gb RAM
1x ATI Radeon 3850 HD (512 mb)

syntaxeater what were the durations for the 3 screens. You say no problems here. I'm wondering exactly what you mean. So far everyone that is running Vista sees the duration for Screen 1 to be about 6 times longer than Screen 2 and 3. You need to let all 3 screens run full length and then the times for the screens will be displayed.

If anyone runs this test in Vista and the Screen 1 duration is much like the duration of Screen 2 and 3 then that is news and I would like to hear about it. Maybe there's some special thing that can be done to get Screen 1 to run at the same speed as 2 and 3. Maybe some particular hardware setup or maybe some particular something. I'm guessing that some fix in either Vista or Visual Basic 2008 has to occur.

Ohh ok.

Screen 1:
Fast until IC 40 then slows.
Fast again at IC 150.
Slows at IS 230
Fast again at IC 330
Slow at IC 400 until end
Screen 2:
Fast throughout
Screen 3:
Fast throughout
Screen 4:
Freezes at IC 1

Ok. Thanks for that. Again we see the same pattern. If anyone out there runs this test and the Screen 1 duration is close to the Screen 2 and 3 durations then let me know. I doubt this is going to happen and I think that this thread is going to fade into the background now which is just fine.

Quote:

---

Originally Posted by EntityX

The attachment in Post # 36 has executables in the bin and obj folders but I didn't know about them when I made the attachment. It was an unintentional post of executables.

Here's an attachment that has no executables. After you double click the Graphics Test project file inside folder Graphics Test the obj and bin folders will be created and they will have executables in them then so it doesn't make that much difference but I can say the this zip file doesn't have any exe files in it as far as I am aware.

---

Yea that was what I was referring to :) You really dont need to nclude the debug/release folders as they will be generated again when you open the solution.

@syntaxeater, w00t! Mine was faster :D
Attachment 66052

Thanks for the information everyone. .paul., RobDog, Andicioz, and syntaxeater I have all your information for screen durations. syntaxeater could I get your Windows Experience Index also and if you could give me all your performance subscores that would be neat to.

Control Panel > System and Maintenance > Performance Information and Tools. That screen will give you the full set of performance subscores on your computer. .paul., RobDog and Andioz and anyone else who wants to perform the test it would be cool if you could give me that info too but if you don't no big deal. If you just give me Graphics subscore and Gaming graphics subscore that's good enough. There are 5 subscores given. Processor, Memory, Graphics, Gaming Graphics and Hard Disk.

So far Andicioz has lowest Screen 1 duration(28.02 seconds) and highest Windows Experience Index score(5.8) so it looks like the score is a good indicator of graphics performance. I have longest Screen 1 duration(119 seconds) and lowest Windows Experience Index score( 1.0).

Hey again.

Processor: 5,9 (Intel Quad Core Q6700 @ 2,66 GHz.)
Memory: 5,8 (Corsair DDR2 4GB)
Graphics: 5,9 (nVidia GeForce 8800GTS 512MB)
Graphics for videogames: 5,9
HDD: 5,8 (Maxtor 80GB, Cheap :p)

It seems that the working of your application is indeed hardware-dependant in Windows Vista. I hope you can find a software-solution or a valueable reply from Microsoft. If you need any more assistance, I'm always around.

Good luck.

I'll post mine tonight but I know my lowest index is 5.5 and thats a memory category.

When I think about it I don't know if I really need all those different scores. If someone has a Windows Experience Index of 5.5 then that means that none of the subscores will be below 5.5. So I know that RobDog all your subscores are 5.5 or higher. And for Andicioz all his subscores are 5.8 or higher since his Experience Index is 5.8. From Post # 47 I see that Andicioz just has 2 scores above 5.8 and they're both 5.9 so the range of scores is very narrow.

If you want to give all the subscores to me fine or not it's fine. If anyone else does the test(use attachment in Post # 39) and gives me screen duration results at least give me your Windows Experience Index Score. See Post # 46 paragraph 2. And if I don't get any more test results that's OK.

.paul. has a Windows Experience Index of 4.8 so it might be interesting to know where the weak link in his system is that is bringing the score down. His screen 1 duration is 57.84 seconds which is about twice the duration for RobDog and Andicioz.

the weak point is my graphics card. all the other scores are 5.8 or 5.9

I just learned that at this time the highest Windows Experience Index score that you can have is 5.9. That being the case I really wouldn't need all the subscores especially for those computers that have a high rating. I know for instance that all of RobDog's subscores are from 5.5 up to a maximum of 5.9.

from http://windowsvistablog.com/blogs/wi...es/458117.aspx

Quote:

---

Base scores of 6.0 and higher are not defined yet. They will be defined when the time comes and new innovations in hardware allow new capabilities. From past experience, it is expected this will happen at a rate of once every 12-18 months.

---

Yes, the index is the lowest score.
In my case its the memory but I have two decently good graphics cards which probably are helping to get the faster times.

Quote:

---

Originally Posted by obi1kenobi

(a friend of mine has x64 Vista with 1GB of memory and his memory score is 4.3, says that x64 requires about twice as much memory as x86).

---

I run both and they take approximately the same memory. However the x64 install is nearly twice as large on the hard drive. Perhaps that's what he was referring to. It has hundreds of megs of compatibility files for 32-bit software. Not only that but .net is twice as large as it contains both 32 and 64 bit code. You can sink nearly a Gigabyte just in the .net files in x64.

Back in October I received an email from Microsoft Connect about the issue I reported to them wherein the graphics runs slower when the pen thickness is set to 1 in comparison to higher pen thicknesses.

Quote:

---

Greetings from Microsoft Connect!

This notification was generated for the following Feedback item:

http://connect.microsoft.com/VisualS...dbackID=361354

which you submitted at the Microsoft Connect site (http://connect.microsoft.com).

Thanks for your feedback on the .NET Framework! Unfortunately the issue that you have reported is caused by GDI+, a component external to Windows Forms and as such cannot be fixed by our team. We have informed the responsible team about the issue, however they do not use this forum to track issues and hence we are closing it. If you urgently need a fix, please escalate this issue through Microsoft technical support channel in your region. They could help you with the Hot-Fix/QFE process.

---

I got a new computer a couple of weeks back that has a Windows Experience Index of 4.6.
Processor = 5.6
Memory = 5.9
Graphics = 4.6
Gaming Graphics = 5.3
Hard Disk = 5.9

Even though my experience index is only 4.6 note my times running the test from Vista.
Screen 1 = 10.73 sec
Screen 2 = 7.8 sec
Screen 3 = 7.78 sec

Someone might want to run the test again. Andicioz had a time of 28 seconds for Screen 1 duration and his Experience Index is 5.8.

Maybe some fix was made in Vista perhaps; some recent update possibly. The SP1 for Visual Studio was out a while back and I know that shouldn't make a differnce because I installed it and it didn't change anything. Maybe some other Microsoft Update for the .NET Framework or Visual Studio changed something.

My screen 1 duration went from 119 seconds with my last computer to 10.73 seconds with this computer but as I said above I'm thinking something other than just the change of hardware may have made part of the difference. Screen 1 is still slower than screens 2 and 3 but the difference is much less.

I noticed that if you download from post # 39 you should unzip using WinZip. I opened the files with Windows and there were problems that way. With my new computer I didn't have Graphics Test so I had to download from this thread.

Vista SP-1 was out back in April but maybe some didnt have it installed

The previous results we were getting in this thread were from August and I know I had all the latest updates and service packs then and I would think you (RobDog) would have all the latest so perhaps something happened since August that fixed this problem but it would be interesting to see someone who ran the test before besides myself run the test again and see if the times changed.

My results for screen 1 running from Vista are more than 17 seconds less than the previous fastest result that was from Andicioz who has a computer with a 5.8 Experience Index and the Experience Index on my computer is only 4.6. Maybe RobDog or Andicioz or someone who ran the test before could try it again and see if your times came down considerably.

Back in August what we saw for people using Vista was that everyone's time for Screen 1 (pen thickness = 1) was about 6 times greater than the duration for Screen 2 (pen thickness = 2) and Screen 3 (pen thickness = 8). The times I got yesterday broke away from the pattern completely for the first time.

Typically when the pen thickness is lower the drawing will take place quicker.

Still somewhat the same. I'm running Vista Ultimate 32bit with SP1 installed (and I didn't have SP1 installed back then) but it doesn't really make a difference it seems. Also have the same specs as before.

Very curious. I wonder why my times are so much lower than everyone else's. My time for Screen 1 is 10.7 seconds versus your Screen 1 time of 28.3 seconds and you previously had the fastest Screen 1 duration and the highest Experience Index. (I should have had the results display in seconds)

Doesn't make much sense. The Experience Index was predicting performance pretty well before for this test but my latest test result breaks away from everything that went before. I noticed turning Aero off didn't make a difference in the test for me.

I have a Intel Core 2 Duo E8400 @ 3.00 GHz with 4 GB Ram. I have a NVIDIA GeForce 8500 GT graphics card according to the device manager on my computer but it's actually an XFX graphics card that I'm guessing uses NVIDIA chips. Maybe XFX is a part of NVIDIA.

Either there's some recent update that I have that makes a difference in this test that Andicioz doesn't have or there's something in the hardware on my computer that makes quite a difference.

Hello again (sorry for the late replies, I don't visit this forum that often anymore and pretty busy).

Yeah, it is pretty weird. I have:

Asus P5K Motherboard
Intel Core 2 Quad Q6700 @ 2.66GHz
4GB DDR2 RAM 800 MHz
Asus EN8800GTS (Which is actually a nVidia 8800 GTS)

All newest drivers are installed.
As for OS I have Windows Vista Ultimate SP1 x86 and Visual Basic .NET 2008 SP1 Express Edition.
Your processor is clocked higher, so you might get better results due to non-multicore support of the application (I have no idea).