[Avery Andrews]

HDC 3.9 seems to be misreporting the sizes of most images on my 1.5T Samsung HD154U1. e.g. the actual size as reported in Windoze explorer is 80.G (successfully restoring a 100G partition), while the reported size in HDC in 428G.

This happens with 3.9.3-5 (all the versions I happen to have), and is presumably the reason why a disk with about 500G free on it seems to be incapable of taking another image of a 100G partition. There is however one image that seems to be reported of the right size, taken off a dell 760 optiplex - no idea why that one is right and the others aren't.

[Alex]

The size reported in the dialog is the size of the original file systems, i.e. if you made an image of a 400G file system (even though only 10GB were used), HDClone would still show the image as 400G in size (this is the size you can restore the image to without having to do any modifications). If you look at the image in the explorer window, it will still only be 10GB in size (assuming you use a Smart Image).

Why the image from the Dell 760 seems to be of the correct size I do not know? Perhaps it was almost filled to the max, so there's no difference between file system size and used size?

[Avery Andrews]

OK, but, the original partition (= file system?) was only 100G (sitting on a 1T hdd), and a 100G file system won't allow its image to be written to a drive with 450G free space. So something fishy seems to me to still be going on, even if it's some kind of odd user-error. And the correctly sizing Dell partition was nowhere near full, while the 100G one was close to it.

[Alex]

If you go to the image folder, what size does the Explorer report for each image?
If you enter the image (they are only directories), you can see a file with the same name as the folder, which you can send to the support, in it all sizes are stored that are reported back to HDClone.

If you have a hex editor (or at least viewer, you don't want to destroy the image control files), you can open these files. They are, after 8 bytes of header always consisting of 8 bytes preamble containing a type and a length N and then N bytes of data.
An example:

[pre]
49 4D 47 43 01 00 00 00 IMGC, Version 1
00 00 00 00 07 00 00 00 33 39 35 2E 69 6D 67 Type 0, Length 7,
01 00 00 00 07 00 00 00 48 44 43 6C 6F 6E 65 Type 1, Length 7,
02 00 00 00 00 00 00 00 Type 2, Length 0,
04 00 00 00 07 00 00 00 08 0B DA 07 16 0C 1B Type 4, Length 7,
05 00 00 00 00 00 00 00 Type 5, Length 0,
00 10 00 00 08 00 00 00 38 36 39 46 43 33 42 35 Type 1000, Length 8,
01 10 00 00 08 00 00 00 00 00 00 00 00 00 00 00 Type 1001, Length 8,
02 10 00 00 08 00 00 00 4A 7E 00 00 00 00 00 00 Type 1002, Length 8,
03 10 00 00 08 00 00 00 3F 00 00 00 00 00 00 00 Type 1003, Length 8,
07 10 00 00 04 00 00 00 05 00 00 00 Type 1007, Length 4,
04 10 00 00 08 00 00 00 3E 00 00 00 00 00 00 00 Type 1004, Length 8,
00 20 00 00 04 00 00 00 00 00 00 00 Type 2000, Length 4,
08 10 00 00 00 00 00 00 Type 1008, Length 0,
...
[/pre]

Now pay special attention to type 1002, the 8 bytes data tell how many bytes this part of the image is large (there are as many type 1002 tags in the file as there are partitions plus one). In the example above, the value is 7E4A, which is 32330 bytes. Type 1001's value is 0, which is where this partition starts, and Type 1003's value is 3F, which is 63 in decimal notation, which means this first partition is 63 sectors in size. Type 1004's value is 3E, which is 62 in decimal, it denotes the number of the highest sector of this partition stored in the image.

By scanning the file like this, you can find out the individual sizes of the partitions in your image. You're probably better off sending the file to support@miray.de, though.

[Avery Andrews]

Using 4.02, I get the size bug off an image of a small partition (explorer-reported size 10.3 G, HDC-reported size 428G) from the 1T seagate, but not off one from a .5T WDC, both with cluster size 4. I'm just about to send off the .img file for the buggy one.

So if it's particular to images taken off certain HDD models, that would explain why the Dell image was OK.

[Avery Andrews]

A bit more info is that the wrong-reported-size images mount fine and look correct under the new driver