home *** CD-ROM | disk | FTP | other *** search
- ---------------------------------------------------------------------------
-
- Marcs Disk Cruncher 68000 V1.17 And
- Marcs Disk Cruncher 68020 V1.17 User Documentation
-
- Copyright © 1995 By Marko Pozarnik
-
- ---------------------------------------------------------------------------
- Preface
- ---------------------------------------------------------------------------
-
- Please NOTE this documentation is a kind of preview for the porgram.
- Some options described below aren't supported yet. I put this program on
- AmiNet to see your reactions. My E-Mail adress is bellow.
-
- This unregistered version of the program and files in this distribution
- are freely distributable, but are also copyright © By Marko Pozarnik. They
- may be freely distributed as long as no more than a nominal fee is charged
- to cover time and copying costs.
- Commercial usage is permitted with this sentence.
- Everything in this distribution must be kept together, in original
- unmodified form.
- The above is generally known as shareware.
-
- ---------------------------------------------------------------------------
-
- The version of the program in this distribution is not registered!
-
- ---------------------------------------------------------------------------
-
- Please contact me if you have any comments or suggestions:
-
- E-Mail: marko.pozarnik@uni-mb.si
-
- Mail: Marko Pozarnik
- Prusnikova 4
- 62000 Maribor
- Slovenia
-
- Phone: +368 (0) 62 38 303 (From 10 PM to 2 AM MET)
-
- IRC: SSMP
-
- ---------------------------------------------------------------------------
-
- I think KS 2.0 and later required. It doesn't work on KS 1.3.
-
- ---------------------------------------------------------------------------
-
- I would like to thank ZPiff for his help in using Xpk libraries. I had
- some problems and couldn't find the bug. Without him the distribution of
- this package would be in two or three months. It's although late. It
- would be distributed in early December if I hadn't had the problem. :(
-
- ---------------------------------------------------------------------------
- Contents
- ---------------------------------------------------------------------------
-
- 1. MDC compared to DMS
- 2. Installation
- 3. Usage
- 4. How to register?
- 5. To do
- 6. History
-
- ---------------------------------------------------------------------------
- 1. MDC compared to DMS
- ---------------------------------------------------------------------------
-
- Why should somebody use MDC rather than DMS?
-
- -x- MDC is faster on slower machines than DMS. On my 50 MHz 68030
- there is no difference.
-
- -*- MDC uses more different packers than DMS. There are 16 packers
- supported by Xpk at the moment and only 4 by DMS, where BEST, HEAVY1 and
- HEAVY2 use too much time for the compression. Here are the test results:
-
- Please read the complete context while the board shows that DMS is faster
- in unpacking, but this is only usable if user is sure that the disk is 100%
- O.K. In the context are some other important information too.
-
- Not so good compression with MDC acording to DMS is caused by not
- identifying empty tracks by MDC. Implementing it to MDC has the highest
- priority at the To Do things. Probably it would be better making this
- table after implementing it. I stoped making the table when I discovered
- the difference.
-
- Disk 1 Disk 2
- DICE C V3.0 Disk 1 Workbench 3.0
- Size CR PT UPT Size CR PT UPT PMemU UPMemU
-
- DMS BEST 604392 33 1:41 1:41 385728 58 2:08 0:43 615864 439012
- DMS HEAVY1 606119 33 2:19 1:43 393606 57 2:37 0:43 521952 438984
- DMS HEAVY2 604392 33 2:13 1:40 385728 58 2:39 0:44 649000 439280
- DMS NONE 902776 0 1:06 1:37 902776 0 0:55 0:42 ?????? 439280
-
- MDC BLZW.14 747236 18 0:58 0:50 656752 28 0:58 0:52 42160 40392
- MDC BLZW.28 740580 18 1:00 0:51 46016 41926
- MDC BLZW.42 744740 18 1:01 0:50 53792 46148
- MDC BLZW.57 751188 17 1:00 0:49 69200 54024
- MDC BLZW.71 757304 16 0:59 0:49 99776 70640
- MDC BLZW.85 757304 16 0:57 0:49 161216 71376
- MDC BLZW.100 757304 16 0:59 0:48 284080 70780
-
- MDC CBR0.100 727928 20 0:49 0:47 804284 11 0:50 0:47 38312 37176
-
- MDC DLTA.100 906256 0 0:47 0:47 906256 0 0:47 0:47 38312 37176
-
- MDC FAST.79 696484 23 0:52 0:46 561624 38 0:51 0:46 103848 37880
- MDC FAST.100 683740 25 1:54 0:46 508988 44 1:47 0:46 316840 37176
-
- MDC FEAL <-- Not usable with MDC, for now. CR = 0%. Password encyption.
-
- MDC HFMN.100 756408 17 1:01 1:00 46008 44200
-
- MDC HUFF.100 771432 15 1:38 0:58 67896 108488
-
- MDC IDEA <-- Not usable with MDC, for now. CR = 0%. Password encyption.
-
- MDC IMPL.10 688964 24 2:14 0:48 301496 37736
- MDC IMPL.30 631740 30 2:34 0:48 308232 37880
- MDC IMPL.50 624592 31 2:48 0:48 309032 37800
- MDC IMPL.75 620400 32 3:52 0:48 327200 37800
- MDC IMPL.98 619688 32 4:40 0:48 343648 37720
- MDC IMPL.100 619688 32 4:35 0:48 343648 37800
-
- MDC MASH.100 617308 32 2:40 0:50 234920 37800
-
- MDC NONE.100 905616 0 0:45 0:46 37688 37096
-
- MDC NUKE.100 649464 28 1:10 0:47 235008 37176
-
- MDC RAKE.25 658964 27 1:00 0:48 191928 37684
- MDC RAKE.50 658344 27 1:01 0:48 191928 37684
- MDC RAKE.75 657828 27 1:01 0:48 191928 37684
- MDC RAKE.100 657720 28 1:01 0:48 191928 37684
-
- MDC RDCN.100 673148 26 0:55 0:47 54668 37504
-
- MDC RLEN.100 721144 20 0:58 0:49 38312 37176
-
- MDC SHRI.14 633996 30 6:17 5:19 78344 40896
- MDC SHRI.28 626608 31 6:16 5:11 81416 41904
- MDC SHRI.42 624404 31 6:24 5:09 87560 43952
- MDC SHRI.56 623176 32 6:36 5:17 99848 48064
- MDC SHRI.70 622652 31 6:38 5:09 124424 56240
- MDC SHRI.84 622652 31 6:39 5:12 173576 72624
- MDC SHRI.100 622652 31 6:49 5:10 271880 105392
-
- MDC SMPL.100 772636 15 1:09 0:51 52416 37800
-
- MDC SQSH.100 666948 26 2:10 0:49 117592 37800
-
- In tests were used MDC68020 V1.16 Registered and DMS V2.01 Evaluation
- Copy.
-
- Configuration used in test:
-
- Amiga 1200
- Kickstart 3.0
- Workbench 3.1
- 3.5 '' Hard Disk Quantum 240AT (Transfer rate: 1.25 MB/s. Tested with
- SysInfo)
-
- All files were packed to hard disk.
-
- Please note: Dice C Disk 1 contains some packed files.
-
- The number after pack method (FAST, NUKE,...) means the pack mode. If
- there is only name.100, it means that there are no other modes.
-
- Size and Memory Usage are in bytes. CR is compression ration in %. PT
- is packing time, UPT is unpacking time.
-
- Memory usage is of course not precise but the results are maximal ±3 kB
- away from the precise number.
-
- Workbench 3.0 disk was unpacked with noverify but DMS didn't report error
- like MDC did without verify implemented. You can check this out by
- yourself.
-
- MDC without usage of Xpk library uses approximately 37 kB. The number by
- Memory usage is the usage with the library.
-
- At the decompression with DMS it's important to know that its
- decompression to the disk is with verify, but its verify is not better than
- MDCs' which is simplier.
-
- DMS V2.20 registered crashed once (before starting the 2nd test with
- DMS), when I wanted to pack with CMODE HEAVY1, used DMS V2.02 instead. I
- tested DMS if it's going to crash again in the same cicumstances, and it
- crashed.
-
- -*- MDC itself uses only 37 kB. DMS uses much more (it has all
- pack methods implemented) (about 600kB). MDC With the packer, which uses
- at least memory, uses 38312 bytes with 20% compression ratio and MDC with
- the packer, which uses the most of memory, uses 343648 bytes with 32%
- compression ratio.
-
- -*- If you have a DMS file which has corrupted packed tracks, there are
- nearby none possibilities to unpack it. Maybe it contains something what
- is not so important. MDC offers a possibility to unpack corrupted files
- because of it's file structure.
-
- -*- According to this that MDC isn't finished yet, I can't tell more
- about the disadvantages and advantages. Please register and/or wait for
- the next version of MDC. I will get a turbo board tomorrow with 8 MB of
- RAM, so there won't be excuses for not continuing developing it.
-
- ---------------------------------------------------------------------------
- 2. Installation
- ---------------------------------------------------------------------------
-
- You should have Xpk package installed (it's on AmiNet in util/pack
- directory named xpk25usr.lha or later), then the only thing to do is to
- copy the version of MDC that works with your processor (68000 is for all,
- 68020 is for 68020, 68030, 68040 and 68060):
-
- Copy MDC680X0 C:MDC
-
- Where X is one of 2 or 0
-
- That's it. If you want to test it you don't have to read the doc to the
- end, but PLEASE NOTE (!) these facts for this unregistered version:
-
- -*- It doesn't tell you to insert the disk to unpack or pack to. It
- starts to do it so be careful (I am used to it, but there are some people
- who could have problems with it.
-
- ---------------------------------------------------------------------------
- 3. Usage
- ---------------------------------------------------------------------------
-
- MDC From/A,To/K,Q=Quiet/S,L=Low/N/K,H=High/N/K,M=Mode/K/N,P=Packer/K,
- Test/S,Info/S,Extract/S,Repack/S,Join/S,Beep/S,WFK=WaitForKey/S,
- Verify/S,Corrupt/S,MDCOnly/S,Prefs/K
-
- Some examples:
-
- 1. MDC DF0: To RAM:Test.MDC 0 0 M 100 P Nuke
- 2. MDC DF0: To RAM:Test.MDC Low 0 High 0 Packer Nuke Mode 100
- 3. MDC RAM:Test.MDC Test
- 4. MDC RAM:Test.MDC To DF1: Quiet
-
- Examples 1 and 2 are the same. Both are packing track 0 of disk in DF0:
- to file RAM:Test.MDC with packer Nuke and mode 100.
- Example 3 tests archive RAM:Test.MDC if it is corrupt or not.
- Example 4 unpakcs archive RAM:Test.MDC to DF1: Nothing is printed to the
- output.
-
- From/A - It can be file or DFx:. It can be file only if To/K is DFx: or
- one of following options is used: Test/S, Info/S, Split/S, Repack/S,
- Join/s. And it can be DFx: if To/K is file or none of before written
- options is used. If this argument is DFx:, MDC will pack to To/K. If this
- argument is file, MDC will unpack it from From/K to To/K.
-
- To/A - See the context at From/K. It can be file if From/A is DFx:.
-
- Q=Quiet/S - This argument causes MDC not to print anything to the output.
- The only things printed to the output are errors. Options Info/S, Test/S
- and WaitForKey/S disable this option.
-
- L=Low/N/K - It can be a number between 0 and 79 (by now). It stands for
- the low track to be packed. There is no affection on unpacking by now.
-
- H=High/N/K - It can be a number between 0 and 79 and has to be greater or
- equal than L=Low/N/K. If only one of H=High/N/K or L=Low/N/K is given, MDC
- uses the default value for the other.
-
- M=Mode/K - It can be a number between 0 and 100. It represents pack
- mode. It has no affection on unpacking.
-
- P=Packer/K - It can be a packer of the Xpk package except IDEA and FEAL.
- It represents packer method. It has no affection on unpacking.
-
- Test/S - It tests the archive. If there is only one error it stops and
- reports it. If there is everything O.K. it prints that file is O.K.
-
- Info/S - Displays a short information about the archive.
-
- Extract/S - it extracts some packed tracks from the archive. From/A is
- the name of the archive, To/K is the name of the new, extracted archive and
- low and high are the first and the last number of the tracks to extract.
-
- Repack/S - The same as extract, only that Packer and Mode have affection
- to it and it repacks the archive to new one and not only extracts.
-
- Join/S - It joins extracted archives. It doesn't matter if there are
- some tracks not available. It only matters if the disk with this unpacked
- archive will function like it should.
-
- Beep/S - It causes the program to beep after finishing his job, since
- there are some people who watch TV during the long operations and want to
- know if the processing is finished without to have a look to the monitor or
- drive LEDs.
-
- WFK=WaitForKey/S - It is for the people, who want to start operations
- direct after typing the command and pressing return, like me. And not
- being bored with the request: Do this and that and pres whatever.
-
- Verify/S - I think it's a little bit useless, but there are people who
- want to be sure. The only thing what it is doing more than without it, is
- that, when reading from disk, it is reading twice to different memory
- locations and comparing them. Similiar to this is writing to disk.
-
- Corrupt/S - It unpacks corrupted archives as best as possible. It
- unpacks files which were packed from track 0 to 39 in one archive and
- packed from track 40 to 79 (for example) to the other archive and then
- joined with AOS join command. But I think the last possibility won't have
- much sense after implementing the Join option.
-
- MDCOnly/S - affects only searching files with a pattern. If it's set,
- MDC will search only #?.MDC files. Otherwise it will get all files no
- matter if they are with MDC extension or not.
-
- Prefs/K - it has an file, for argument, which contains the settings for
- the program.
-
- ---------------------------------------------------------------------------
- 4. How To Register?
- ---------------------------------------------------------------------------
-
- The registration costs 10 Deutsche Mark, 4 Pounds Sterling, 8 US Dollars,
- 10 Austraian dollars, 70 Oesterreichische Schilling, 800 Slovenskih
- Tolarjev or 8 Schweizerische Franken. That are all currenties which are
- accepted. Any other won't be accepted and won't be sent back and you won't
- become a registered user.
-
- I think it's worth it, not now, but as though I am going to do the planed
- things.
-
- You can send me the money via mail (not E-Mail :) ). All registered
- users will get the newest version of MDC sent via E-Mail or on from you
- sent disk.
-
- ---------------------------------------------------------------------------
- 5. To Do
- ---------------------------------------------------------------------------
-
- -*- Make him not to pack the empty tracks. Serious problem! I didn't
- realize it, before making the complete test with DMS and DMC in MDC
- compared to DMS.
-
- -*- Verification. It makes a simple verification but it doesn't compare
- the data read from disk or written to it.
-
- -*- It still not unpack corrupt archives. Please see the Usage for
- further explainations.
-
- -*- Repack isn't done yet. Please see the Usage for further
- explainations.
-
- -*- Extract isn't done yet. Please see the Usage for further
- explainations.
-
- -*- Join isn't done yet. Please see the Usage for further explainations.
-
- -*- WaitForKey isn't done yet. Not much work to do.
-
- -*- Beep isn't done yet. Not much work to do.
-
- -*- Info isn't done yet. Not much work to do.
-
- -*- Prefs isn't done yet. I haven't planed it yet. It's only an idea at
- the moment.
-
- -*- Searching archives with wildcard isn't done yet. For testing or
- maybe for unpacking too, would be great when the user could write MDC #?
- test.
-
- -*- MDCOnly isn't done yet. There won't be much work to do after doing
- the previous To Do.
-
- -*- It doesn't check if the to file is existant or not. It overwrites
- it, if it exists and creates new if it doesn't.
-
- -*- If the disk is full, to which a disk is packed it reports only that
- there is error but not which. Not much work to do this.
-
- -*- Controlling of CTRL C. There is no possibility to break the program
- (exept reset or turning off the computer). It worked fine, but I removed
- it (don't ask me why). Will be implemented in the next released version.
-
- -*- Better output on the end of processing. There is an ugly output at
- the moment. I am about to make a new one which would look prettier:
-
- Estimate time: 1:41
- Tracks: 60
- Empty tracks: 20
- Size: 647236 bytes
- Compression ratio: 37%
-
- -*- Password encryption if it is needed. I haven't used it till now.
-
- -*- I could modify the track header produced by Xpk. There is some data,
- which is always the same. That would make MDC archives shorter (I know how
- to eliminate about 11 long words per track, minimal >= 80*11*4 = 3520 bytes
- shorter archive). But that would make new archives uncompatible with the
- old. But compatibility won't take that much memory usages (6 LW).
-
- If you think there is something that could be done in addition to the
- above, tell me about it. All suggestions are welcome.
-
- ---------------------------------------------------------------------------
- 6. History
- ---------------------------------------------------------------------------
-
- I don't have enough time to write it. Besides I don't even know if it's
- worth doing it, because since 68030 machines will rule this world there
- won't be need for a program like mine it is.
-
- ---------------------------------------------------------------------------
-