Glenn9999

They're a valuable resource if you want to run stuff from that era on modern computers, or have something to play with, or something to show the kids and say "back in my day, this is what I ran". Basically, I'm thinking running them in virtual machines like Microsoft VirtualPC or VMware or VirtualBox.

I notice lately that my copy of VirtualDub just doesn't seem to cut it anymore. Most of the formats I'm finding now it simply just doesn't support. The big ones are DVD, FLV, and MP4 (I've had problems with regular MPEG too). So I thought I'd ask what people use and if they have any suggestions. I'm not looking for anything particularly fancy, but something that will get the very occasional video job done that I run into. To that end, I'm not looking to spend a bunch of money for something I would use maybe five times in the next three years. Good working converters (something that doesn't kill the picture somehow or kill the sound) that will get those three named formats (DVD, FLV, MP4) to something VirtualDub understands would be a nice substitute. Edit: And back. Edit: Also, any good solution for normalizing audio tracks on videos. So any thoughts?

Or it came with a "recovery disc". That is as worse or worser than no disk at all. Run that and you get all the garbage back again. Or the worst thing would be those "recovery files" preloaded onto the hard disk where they can be corrupted or infected with viruses or lost entirely due to a number of factors. Or the best thing of all, and what I recommend to anyone buying any computer: You better provide the ORIGINAL OS install disk, or no sale.

Which turned out to be a very wise thing. I think I figured out, mostly, what is going on. 1) IE7Pro seems to be VERY incompatible with IE8. I have all the other BHOs installed now except for that one and am running fine. Add to that the clamor on the IE7Pro forums for a new version, it seems like I'm close, there. 2) "Immunization" functions seem to be somewhat incompatible, too. I loaded the immunization functions in Spybot S&D and Spywareblaster and IE8 took about 20-30 seconds to load (to get to the point that I could browse). Removing the IE specific ones lowered that to 2-3 seconds. So my guess is that both those softwares will need to do an update. Anyway that seems to be the story.

No. Starting with IE7, downloaded and installed IE8. The described problem started. I tried a rollback point on system restore and deinstalling IE8 separate times and still no help. After I deinstalled IE8 (back to IE7), IE7 would work similarly. Yes. This would only work with IE7, but would provide a screen with no menu bar, only the address bar, and no other controls on the screen, including the main rendering area for the web pages. None to be had. The screen disappears in less than a second. Not unless IE8 is - the system was working 100% fine before it was installed, and this was the only thing that was introduced aftewards. But I do doubt IE8 is that in an intended way. I resigned myself to doing a complete re-install of my Windows XP partition now. Like I said, I ran into this problem. And then I have a couple of other reasons to do the re-install. But I would still like to know what is causing it. My going theory is that IE8 didn't take too well to the BHOs and the Add-Ons (most notably IE7Pro) that I had installed in IE7. I tried deinstalling those and still run into this problem. The first icon on my control panel options was messed up too (no caption, nothing happens when I click it). I can always try IE8 first when my install gets to that point and then try and revert to IE7 if I still have problems.

I had this problem with both IE8 RC1 and IE8 RTM. Furthermore, my IE7 is now messed up. It does the same thing in regular mode and shows with only the address bar/search bar (and nothing else) in safe-mode. Any ideas on getting things back to normal with IE6 or IE7 (I can always run the installer) without having to gut everything and start over again. And any ideas about what's causing this with IE8?

I tried this and it looks great. No problems yet.

I don't know about FCleaner, but there were certainly a lot of issues with CCleaner back in its 1.0+ days that caused some people to move to nCleaner (I tried it back then when people were talking of it). Anyway the main reason for this post was that I was going to point out the Comodo registry scanning and cleaning tool that I'm aware of and have used (and find it great, it finds more things than CCleaner and NCleaner, and seems to have more safety features on it too). But I notice on their site now that they added a disk cleaning function to it, so I'm going to have to go ahead and try that one. Here's the site for anyone that wants to try that one: http://system-cleaner.comodo.com/index.html Edit after trying the Comodo product: Their disk cleaner function needs some work (though it does catch Windows Update distribution files when the others don't). But it seems like it's a relatively new function, too. Hopefully they'll add more in future updates. It also has a reboot and rollback feature now on all the tools, which you can choose to enable or not when you perform the cleaning function. It also has a menu of system settings - I don't know how fascinating any of those are (or a good equivalent in another program), but I thought I'd point that out.

First One: Solve here. Second one: Solve here Third one: Solve here. Fourth one: This is the Knight's Tour problem. Good puzzles though. These are all well known logic problems. Edit: Got to playing. Not the solve for your fourth game, but the solve to the Easter Egg in Spybot: 1 2 3 4 5 6 7 8 9 10 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1+ 1 18 97 68 15 20 85 36 13 22 2+ 100 65 16 19 98 69 14 21 38 35 3+ 17 2 99 96 67 86 37 84 23 12 4+ 64 93 66 87 82 95 70 73 34 39 5+ 3 88 91 94 77 72 83 40 11 24 6+ 92 63 54 79 90 81 76 71 74 33 7+ 45 4 89 62 53 78 41 60 25 10 8+ 48 51 46 55 80 61 28 75 32 59 9+ 5 44 49 52 7 42 57 30 9 26 10+ 50 47 6 43 56 29 8 27 58 31

As Requested It also does the following (I added some things and consolidated some others): 1) Allocates and holds a fixed amount of memory (MB). The OP might tweak this to their use - you can figure out about how much memory you would want to hold in order to see what happens. 2) Frees the memory allocated in #1 3) Does processing against the allocated memory in #1 and returns how long it took. This takes the form of writing a character (ASCII #255) to all bytes of the memory allocated. If you set it against a significant amount of memory (greater than your physical memory), go make a sandwich because it's going to take a while (watch that HDD light light up!). 4) Does the report that was in my prior message. Hope you all find this interesting. (I had a thought too when I was playing around with this. I'm not sure how much information this would gather, but you could determine a memory sweet spot for an OS by several runs of function #1 and #3 at specified intervals and graph out the time. Perhaps memory used vs processing time per MB or some such thing. I might automate it if I have nothing else to do sometime)

Done. 1.03 released, see first post.

I tried a search and didn't find this. Is there a generally preferred set of command-line parms to run the MS Malicious Software Removal Tool version 2.0+ (890830) unattended?

I'm looking at this program now to update it. Primary of which is to add detection of a couple of routine sets of patches that have escaped the current detection routines. 1) MS MSRT 2.0 2) The newer .NET patches. (KB958481, KB958483, KB958484) Are there any others that can be suggested (i.e. routinely installed) but do not show up in Batch Patcher? Also, any other good tweaks or suggestions are welcome.

The last thing was some classic MOD and S3Ms. The thing before that was classical orchestra.

Windows ME is unsupported by Microsoft, and therefore the update list is constant. Basically, this eliminates any usefulness that something like WUD would have. If you download everything you need, then the only problem that remains is installing them each time you were to reload the OS. Something like this can be useful along those lines to not have to click through the screens for each one. http://www.msfn.org/board/Batch-Patcher-t96854.html (of course, there's always the third-party patches and things that have been put together in places like this - if you get into those, things will change a little. But something like WUD would still not be too useful in the case of 98/ME)

Black & White.

are you using a serial mouse? Yes.

In my program testing with Windows XP, I have discovered that this can not be effectively done (i.e. the system configured so Total Memory = Physical Memory) Edited to clarify where I'm coming from. While you can effectively turn off the static page file (that's what the System/Advanced/Virtual Memory option does), you can not turn off the fact that the system will page parts if its memory out to diskette. In fact, if you set the static page file to zero, it will still write a large segment out to disk (482 MB on my latest test I did just now). Memory Usage Report ------------------- Available Physical Memory: 295.852 MB. Total Physical Memory: 511.531 MB. Available Page File: 316.777 MB. Total Page File: 482.879 MB. Available Virtual Memory: 2033.074 MB. Total Virtual Memory: 2047.875 MB. Memory Load is: 42%. This is from a little program I have that I will share if requested. It exposes the memory reporting API call and reports it. The terminology is different (Microsoft's API documentation is different than the terms in the UI). As you can see, physical memory is much different than virtual memory. Yet physical memory + page file does not equal virtual memory. For my app (and I have done it), I can allocate 2 GB of memory and Windows will happily comply with it - and will happily slow down to a crawl once processing is done on that memory, among other problems. Indeed, allocating 2GB will create a low memory situation (which Windows reports in the taskbar). But it's just not too practical. Edit again: Another fun one I discovered when I really started playing around by allocating huge sections of memory and monitoring the page file (more than I did before). Virtual memory as set in the control panel is only a suggested starting point more than anything definite. Which means as I was referring to before in this post, Windows will swap memory, whatever it wants to. Windows will also keep as much of the physical memory as free as it can. When I allocate the maximum memory possible, a significant portion of physical memory is still free. I observe that when the page file hits the maximum, then Windows will start producing error messages, not load apps at all, not load apps entirely (I only got one tab in Task Manager once). Even more, there really aren't any guarantees on what kind of memory you can really allocate at any one time, or how much is available. The status of memory in Windows is very fluid. So the API above really doesn't even give you a good starting point, whatsoever. Programmatically speaking, you have to allocate the memory, error check and hope for the best. Really, you have to write your program to prepare for anything. If you go for a static portion of memory, it may succeed or not. You may have no virtual memory left for other things, 20MB, or 1.5GB. You just never know. You dive in and check whether you got a good result or not. (Anyway I hope that was educational, or at least entertaining to some)

It would probably be very useful, given the nature of some of the responses if the OP could clarify him/herself. But the way I read this is that the OP is wanting to test an app in a low-memory available/high-memory used situation. More or less to see how the app might handle not having a lot of memory to run with. As the OP mentioned, there are many programs out there that will "soak" a certain amount of CPU as to test how an app will run with slower CPUs. But no really good solution for memory, because of what I mentioned in my first response. Of course, burn-in of memory is an easy task, and many have pointed out good solutions (I like Memtest86). But for what I'm reading, I don't think that's what the OP is looking for.

Maybe I am not understanding what you are saying, but how would this accomplish a low-memory situation, without hitting the page file? If you are speaking of the host OS, you would still have the potential of other applications going to page file (even assuming Vmware stops that for its VMs). If you're speaking the guest OS, the Windows installed in a VM will still go to page file, no matter how small the memory. I mentioned Application Verifier because I think I saw the option in it when I was playing with it (though I'm not 100% sure of it). If it does it, the only way it could possibly is to redirect the API calls to fool the program into thinking that there is much less memory to work with than there really is.

Some more code I thought I'd share. This code repeatedly cycles colors on the whole monitor at the refresh rate that is returned by the XP API. Again something I'm not sure might be too useful from a utility standpoint (I do see sites that mention doing this to LCD monitors to try and unstick pixels). At best, it is a good burn-in test for a monitor. Demonstrates in Delphi: 1) Basic threading. 2) Key and mouse controls. 3) Sending and dealing with messages (interprocess communication). The app won't start at the true refresh rate for Windows 98/ME (the API doesn't support it, I think you have to use DirectX on that platform), but will start at a 60hz refresh rate, which hopefully should be sufficient to start from in adjusting that. unit lcdunit; interface uses Windows, Messages, SysUtils, Classes, Graphics, Controls, Forms, Dialogs, StdCtrls, ExtCtrls; type TForm1 = class(TForm) Memo1: TMemo; TimeIndicator: TLabel; Timer1: TTimer; procedure FormCreate(Sender: TObject); procedure FormKeyDown(Sender: TObject; var Key: Word; Shift: TShiftState); procedure FormDblClick(Sender: TObject); procedure ScreenThread; procedure Terminate; procedure FormClick(Sender: TObject); procedure Timer1Timer(Sender: TObject); private public { Public declarations } protected procedure UpdateScreen(var WinMsg:TMessage); message WM_USER+1; end; const adjust: Longint = 5; var Form1: TForm1; start_freq: Longint; freq: Longint; Processing: Boolean; screen_id: DWord; Screen_Delay: Integer; implementation {$R *.DFM} procedure TForm1.FormCreate(Sender: TObject); var DevMode: TDevMode; begin { set form size to screen size } Form1.Width := Screen.Width; Form1.Height := Screen.Height; { bring window to foreground } SetForegroundWindow(Handle); SetActiveWindow(Application.Handle); ShowCursor(false); { try to start screen frequency to be same as vertical refresh of monitor } EnumDisplaySettings(nil, 0, DevMode); if DevMode.dmDisplayFrequency = 0 then start_freq := 16 { ms for 60hz } else start_freq := (1000 div DevMode.dmDisplayFrequency); Processing := false; end; procedure TForm1.Terminate; { terminates all threads and then the application } begin Processing := false; ShowCursor(True); WaitForSingleObject(screen_id, INFINITE); Application.Terminate; end; procedure TForm1.FormKeyDown(Sender: TObject; var Key: Word; Shift: TShiftState); begin if key in [vk_up, vk_right] then begin if freq < 100 then begin inc(freq, adjust); TimeIndicator.Caption := IntToStr(freq) + ' ms'; end; TimeIndicator.Visible := true; Timer1.Enabled := true; end else if key in [vk_down, vk_left] then begin if freq-adjust > 0 then begin dec(freq, adjust); TimeIndicator.Caption := IntToStr(freq) + ' ms'; end; TimeIndicator.Visible := true; Timer1.Enabled := true; end else Terminate; end; procedure TForm1.UpdateScreen(var WinMsg:TMessage); begin Form1.Color := WinMsg.LParam; TimeIndicator.Color := Form1.Color; end; procedure TForm1.ScreenThread; { process screen. Keep changing screen from Red -> Green -> Blue } const ColorList: array[1..3] of TColor = (clRed, clGreen, clBlue); var colortype: integer; begin sleep(1000); { to let the user see a pure black screen } while Processing do for colortype := 1 to 3 do begin PostMessage(Form1.Handle, WM_USER+1, 0, Integer(ColorList[colortype])); sleep(freq); end; EndThread(0); end; procedure TForm1.FormDblClick(Sender: TObject); begin Memo1.Visible := false; processing := true; freq := start_freq; TimeIndicator.Caption := IntToStr(freq) + ' ms'; TimeIndicator.Left := Screen.Width - TimeIndicator.Width - 50; TimeIndicator.Top := Screen.Height - TimeIndicator.Height - 20; BeginThread(nil, 0, Addr(TForm1.ScreenThread), nil, 0, screen_id); end; procedure TForm1.FormClick(Sender: TObject); { single click, we do not want to respond if we have not started processing via double-click } begin if Processing then Terminate; end; procedure TForm1.Timer1Timer(Sender: TObject); { timer is triggered to make the TimeIndicator disappear after 2 seconds } begin TimeIndicator.Visible := false; Timer1.Enabled := false; end; end.

You can't really do that on a system with virtual memory, as Windows has. In essence, you have to be able to shut down all memory paging or virtual memory, and then fill the remaining physical memory. Windows won't let you get away with this. While it is possible to fill up all the memory in usage (I've done it), the problem there is that in forcing the system to use its page file extensively that you make the system run about as fast as a pig on its back in slop. In essence, you run into the same problem that the system would have if you were to provide it a small amount of memory (say 64MB for a XP system) - extensive use of the page file slows a system down considerably. A thought though: You might try Application Verifier and see what comes of that. Though, it probably won't get you what you're looking for either...

My mouse still doesn't work when it loads into setup. Pretty much the same as the last time I tried it (v6 or 5 or thereabouts). Pretty much makes it a non-starter for me.

Okay, I would have no way to know - in this case, it probably wouldn't work if it doesn't put things into that registry path. I might experiment with WUA some someday, but I really didn't mean to produce anything *too* usable - just provide a programming example of a few things from something I wanted to play around and do (and people tend to have questions on how to do). As the other poster said, I do find it useful for XP in tracking patches (the update files only write *some* of the patch files there) in some way. YMMV, but hopefully it will work out if someone does want it for that. But bugs do tend to irritate me, so I thought I'd try to track this one down and fix it.

An attempt to fix. I got it working in my Windows ME VM now (it had the same error), so I took that check out. Can update the source in the first post once I know it's better. http://rsgp0g.bay.livefilestore.com/y1pMMb...CK.ZIP?download If it's what I think it is, it's a good lesson that Windows isn't all that standard from version to version. It's definitely a lesson that it's good to have as many versions as possible to test on. As for me, I only have XP and ME to test on... Hopefully this will fix it.