|  Post Your Message | |
| |
| | Previous Messages | Posted By
MIK
on 2003-01-30
 |  Marcus site.
From the past more than often it would be down.. Server problems, but I haven't been to his site in a long while so I don't know 
You could try asking on Lemon 64 forum.
Sure thats ok by me. By all means Marko is welcome to download ours and work on them for Funet if he wishes. The saving that was made on Tank Attack was cool I have to say. Even more cool is they are intresed in the TAP format at last 
|
|
Posted By
Csabo
on 2003-01-30
| Re: Csabo.
No problem. I have no reason to touch the existing files on our server. The 'other website' for clean TAPs already exists, it's Funet.fi. Marko had no problem with PRG files (I've been slowly uploading all the missing games), but he wanted the TAPs to be smaller. After seeing Tank Attack zipped to 6K (which loads perfectly, and after loading the memory is a byte exact match to ours), he had me thinking.
BTW, Marcus' MTAP site seems to be down. Does anyone know any good links to more info on the TAP format?
|
|
Posted By
MIK
on 2003-01-30
| Csabo.
Cheers mate, had me worryed for a second there.
|
|
Posted By
Csabo
on 2003-01-29
| Re: ..
Don't worry mate, I wouldn't.
|
|
Posted By
MIK
on 2003-01-29
| ..
Please do not clean-up or replace any of the 100+++ I have made that are here on this server Csabo!!
By all means only clean-up the ones you have downloaded for your own use.
I don't fully understand the in's and out's of cleaning, but do know there are different ways to go about it even if you think your way is correct. I've seen alot of talk on the C64 side about it and really the last thing you want to do is replace the original you were working on. One mistake, one error, untested 100% ect, ect means you would lose that original if you understand. I know Marcus author of MTAP isn't keen on cleaning, but hes the one that made this all possible so I rather we kept to his way of doing things in general....
Those TAP's I've spent many, many, many hours on, even lost original cassettes in some cases just trying to pull of the impossible, and with those I have lost the rewards are sat here.
If you want to make another TAP (Clean) web site then please do, but don't replace the ones we have here. Cheers.
|
|
Posted By
JamesC
on 2003-01-29
| TAP Cleaning
Csabo, please be sure that you are up to date on Microsoft Windows Washer(TM) and Microsoft Demagnetizer(TM). Just please do NOT use Microsoft Teflon(TM) otherwise any patches will not stick.
|
|
Posted By
Csabo
on 2003-01-29
| TAP Cleaning
While talking with Marko Mäkelä, the TAP issue came up. He explained to me how the TAP format works (I'd like to send him my thanks once again). He has 'cleaned up' our Tank Attack image for exampe. Ours was 41K zipped and his is only 6K zipped (big difference).
Today I sat down to write a small utility for cleaning up (quantitizing) TAP files. It seems to work so far. I have a 450K multi-loader TAP image, for example (from the nice hungarian fellow who sent me tons of stuff, his identity is soon to be revealed). The original compresses to ~140K. My first cleaned-up version is ~72K zipped.
I just want to make sure that I'm doing this right, and I will release this new utility for the public complete with source code and everything.
Okay, so here's what I know so far about the TAP format: 20 bytes header (left alone, irrelevant for this util), then data. The data is pulse widths (non zero bytes), plus pauses. Pause is a zero, then 3 more bytes, right? So my util leaves all those alone as well always (4 bytes in sequence). The cleaning up is basically the following: if we have 20,000 byte 13's and 21,000 byte 14's (just an example), they are "the same", that is it doesn't make a difference which one we use. So changing all of them to one value means the file loads just the same, but it compresses a lot better (whatever pattern the changing of 13's and 14's would give is gone). So how far can we take this?
Marko wrote to me: "There seem to be two major variants of the TAP format for the plus/4. Normally, the pulse widths are measured from a falling edge of a pulse to the falling edge of the following pulse. The Zip*.tap files use a more detailed format where the time is measured for each change of the CASS READ line.
The standard loader uses three different pulse widths, while most fastloaders seem to use only two different pulse widths."
Well, this is what I know so far. Any feedback? (Ati, get back home from your visit already.)
|
|
| |
Copyright © Plus/4 World Team, 2001-2024. Support Plus/4 World on Patreon |
