If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Rate Thread | Display Modes |
#46
|
|||
|
|||
3.5 floppy
Mark Lloyd wrote:
On 11/15/2017 08:59 PM, Paul wrote: Ant wrote: IIRC, 3.5" disks could do 2.88 MB but I think it used compression to do that? Apparently, they were just a doubling of density. http://www.pcguide.com/ref/fdd/formatKB2880-c.html And they were a flop commercially. Paul Also, disk compression claimed a doubling of capacity, but except in very unusual circumstances was more like 1.2 times than 2. Hey, we don't want to give away any secrets now. Andy thinks a compressor doubles, so we'll just leave it at that. ******* If you try: dd.exe if=/dev/random of=samplefile.bin bs=1024 count=1000 and try and compress that, it shouldn't compress at all. So it won't even be 1.2 times. Incompressible data is ... not compressible. What the compressor does in such cases, is inserts its header in front of the file, and the header will declare "payload not compressed" as leaving it uncompressed is the best you can do. The compressor works at it, but notices the compression ratio is no damn good, so it punts and just slaps a header on the file, declaring defeat. It's like when your car hits a brick wall, and the "check engine" light comes on :-) I tried to use 7Z Ultra on this, and the output is *bigger* than the original file. Directory of H:\ 11/16/2017 04:45 PM 1,024,000 samplefile.bin 11/16/2017 04:46 PM 1,024,196 samplefile.bin.7z You can see the impact of making that test file larger, and how much the header scheme adds. This tells you the compressor likely works in "chunks" of some particular size. Directory of H:\ 11/16/2017 05:17 PM 1,024,000,000 samplefile.bin 11/16/2017 05:26 PM 1,024,054,451 samplefile.bin.7z Whereas if you make a second file like this dd.exe if=/dev/zero of=zerofile.bin bs=1024 count=1000 that file will compress like gang-busters. The ratio is 2917x. I could store a zillion of these "easy" files on my compressing floppy. Directory of H:\ 11/16/2017 04:47 PM 1,024,000 zerofile.bin 11/16/2017 04:48 PM 351 zerofile.bin.7z Real life is somewhere between those two extremes. ******* In the year 2017, many pieces of media may have attempted compression on their own. Some PDFs for example, might not compress all that well. Microsoft Word has a default encryptor (encrypts with the default password, so it isn't actually password protected), and any time you encrypt a file, the random data patterns involved don't compress. This is one reason, if you use GPG (GNU Privacy Guard) and prepare a file, it compresses the file first, before it encrypts it. If the order was reversed, the file would be huge. Really, the results can be quite variable. Compression results factor in the nature of what you're compressing, as well as the "class" of compressor you are using. A compressor that "runs fast", generally doesn't compress all that well. Paul |
Ads |
#47
|
|||
|
|||
3.5 floppy
Andy wrote:
That is correct Ant Yay, I can remember correctly again. :P -- Quote of the Week: "I go out of my way to avoid stepping on ants." --Terry McGovern, daughter of Senator George and Eleanor McGovern, subject of the book "Terry by her father" Note: A fixed width font (Courier, Monospace, etc.) is required to see this signature correctly. /\___/\ Ant(Dude) @ http://antfarm.ma.cx (Personal Web Site) / /\ /\ \ Ant's Quality Foraged Links: http://aqfl.net | |o o| | \ _ / Please nuke ANT if replying by e-mail privately. If credit- ( ) ing, then please kindly use Ant nickname and AQFL URL/link. |
Thread Tools | |
Display Modes | Rate This Thread |
|
|