Re: Fast copy routine needed

From: Per Olofsson <magervalp_at_fastmail.fm>
Date: Sat, 22 Aug 2015 12:14:50 +0200
Message-Id: <1440238490.1390974.362971417.35083A03@webmail.messagingengine.com>
On Sat, Aug 22, 2015, at 10:35 AM, Ruud@Baltissen.org wrote:
> Hallo allemaal,
> 
> 
> IIRC I mentioned the fact that I lost some files of my 1541IDE and 
> 1541LPT projects. Things seems to be working now again. In fact even 
> better: I got the idea to keep the drive so I can make D64 images in 
> a faster way. The problem is: it isn't that fast at all, in fact 
> quite slow :(
> 
> The problem IMHO is reading the sectors. I read them one by one in 
> the correct order: first sector 0, 1, 2 etc. from track 1 and then 
> the same with the other tracks. IMHO it could be excelerated by 
> reading them in an other order. 

Yes, for 1541 with standard kernal IEC the default interleave is 10.

> Who has experience in reading (and writing) sectors as fast as 
> possible and is willing to share his knowledge, preferably his 
> source codes as well? 
> Note: I need the real data in the end, so it has to be GCRed first.

There are several projects in various stages of approachability.
Mafiosino's loader is afaik the fastest, it reads a full track in two
revolutions:

•
http://www.forum64.de/wbb3/board25-coder-unter-sich/board308-programmieren/board27-asm/19441-schneller-trackloader/

Graham's WarpCopy is also nice and fast and reads tracks with an
interleave of 3/4, but iirc it doesn't decode the data, just streams it
over the network, and it doesn't come with source.

lft has a great article on GCR decoding on the fly:

• http://www.linusakesson.net/programming/gcr-decoding/index.php

Other than that the best entry point is Lasse Öörni's rants on drive
coding:

• http://cadaver.homeftp.net/rants.htm

EasyTransfer is based on Lasse's code, via my ULoad M3 project:

• http://skoe.de/easyflash/doku.php?id=downloads

-- 
  Per Olofsson
  magervalp@fastmail.fm

       Message was sent through the cbm-hackers mailing list
Received on 2015-08-22 11:00:08

Archive generated by hypermail 2.2.0.