Microsoft QuickC 2.51 Professional (5.25) Missing 7 Disk images

CraigAB69

Hi,

There are only 3 disk images in the 7 zip file.
I looked at the Packing list on the Disk01 and it list that this software is a 10 disk set.
Is the original poster able to re-upload it with the correct disk set please.

Regards,
Craig

SomeGuy

Thanks for pointing that out. It is very, very crufty. I have marked it as needing a redump.

I think that all of the files are there, but they are supposed to be on 10 *360K* floppy disk images, not 1.2mb images. Clearly the files were just stuffed in to the disk images. The date stamps are also borked.

CraigAB69

Thanks for looking at that.

There is also "Microsoft QuickC 2.50a (3.5)" that has combined the 360k disks to 720k images.
I am not a fan of such things, I personally like to have copies of the originals.

There is something odd going on with file qc25\tutorial\learn.com it crashes VirtualBox 6.1

EDIT: Actually it looks like it has something to do with VirtualBox. I'm using 6.1, not sure that means anything, but when I run learn.com I get a Guru Meditation.

I'll give it a try from another install.

EDIT: I installed Microsoft QuickC 2.50a (5.25-360k) in to 86Box, with no mods to the autoexec and config, and it works straight away.
Set the Drive to 360K - No problems.
Also works on real hardware (P3, BX2000...)

Cheers

SomeGuy

Were you able to get "Microsoft QuickC 2.50a (3.5)" to install? When I try that it refuses to recognize the "learning" disk. I can't quite tell why.

The 360k version seems to install fine.

Both of those are crufty, and should be redumped, but they appear complete.

Learn.com on a number of Microsoft products is known to be buggy in such a way that it does not like some emulators. Of course, if Virtualbox itself crashes then there is a serious bug in Virtualbox.

SomeGuy

Ok, I figured it out, the file qcc.ov$ on the setup disk of "Microsoft QuickC 2.50a (3.5)" is corrupt, and setup will get confused after that.

Also, the 5.25 360k version is actually 2.50, not 2.50a.

ibmpc5150

In particular, many of the materials uploaded to WinWorld a long time ago are still disk images that have not been properly dumped from the original media, and some are uploaded without checking the operation.
In order to prevent such confusion, I hope WinWorld to upload data that has been verified to work properly from the original diskette.

tarlabnor

I hope WinWorld to upload data that has been verified to work properly from the original diskette

In a perfect world - yes, would be awesome to have all apps dumped from original disks. But in a real world it's better to have a crufty, but complete and working copy of a program than to have nothing at all. At least unless someone finds a better copy.

SomeGuy

Still, I encourage everyone to examine and test all archives, and if there are any issues like this please post them.

Unfortunately, "crufty" can also mean corrupt or incomplete in some way - and if no one notices, then it can be worse than having none, as nobody thinks they need to re-dump.

Even if a title is already present on Winworld, providing additional disk dumps can provide extra assurance that a dump is correct. Many archives on Winworld are missing genuine label scans, box art, and manual scans too. Older archives like these QuickC versions probably originated on BetaArchive, and are likely to be crufty, corrupt, or incomplete.

tarlabnor

Ok, I figured it out, the file qcc.ov$ on the setup disk of "Microsoft QuickC 2.50a (3.5)" is corrupt

It contains a 'bad sector' (512 bytes) at the address 0x23600 - spaces instead of compressed data. This file can be replaced with the one from 5.25" copy, they are identical.

and if no one notices, then it can be worse than having none, as nobody thinks they need to re-dump

That is true. Therefore one could add to such releases notes like "non-original, tested" (i.e. to the 5.25" 360k copy in this case), "non-original, untested" etc.

CraigAB69

SomeGuy said:
Learn.com on a number of Microsoft products is known to be buggy in such a way that it does not like some emulators. Of course, if Virtualbox itself crashes then there is a serious bug in Virtualbox.

It seems to be fine on 86Box and real hardware, so all good.
I started a thread on the VirtualBox forum about it and mentioned it worked on 86Box....
It seems that the DOS template and the Win 3.1 template are different in some way. I don't understand that, Win 3.1 sit's on DOS, as I understand it, like a DOS extender. Happy to be wrong, as I just started the Undocumented DOS book.
And that the "Video Memory" setting in not just for the VM, not sure about that at the moment how that works.
I increased it to 32MB and it seemed to improve video a bit in the VM. Which I thought was odd as video cards never had more than about 512kb or RAM in 1990's.
So, I obviously that needs a little more investigation.

tarlabnor said:
It contains a 'bad sector' (512 bytes) at the address 0x23600 - spaces instead of compressed data. This file can be replaced with the one from 5.25" copy, they are identical.

Very Interesting, may I ask how you looked at the img?
Testing the img via a VM? Or mounting the img and looking at it with a hex editor?

SomeGuy

Ok, I've added a better copy of 2.51, it is on 3.5" but it is the same thing so no point keeping the messed up 5.25" copy around.

I've also repaired and cleaned up the 2.50 versions and re-uploaded them.

> Or mounting the img and looking at it with a hex editor?

Yea, a big block of 512 identical bytes sitting in the middle of a compressed file stands out as an obvious error. In this case the file was supposed to be the same as on the other 2.50 copy, so easy to fix.

BTW, the only difference between 2.50 and 2.50a is the QC.EXE.

Also, Microsoft apparently backdated the files on 2.51 so they have the same dates as 2.50, yet 2.50a has a later date. Annoying. I'm guessing 2.51 was actually released either very late in 1990 or early 1991. Can't tell from the files.