View Single Post
Old 08 July 2011, 14:07   #18
IFW's Avatar
Join Date: Jan 2003
Location: ...
Age: 44
Posts: 1,838
Originally Posted by alexh View Post
Out of curiosity.

Disks which are written back using Kryoflux are distinct from ones written using a trace machine when looking at them with the analyser?
KryoFlux software has a greater freedom of using processing power than Trace could have used at the time.

On a Trace machine variable density tracks had to be written in multiple-passes - you could set the bitcell width only for an entire track for one pass.
KF can write anything in a single pass, greatly improving reliability and data consistency.
Right now the only bottleneck of the system is the type of drive attached and the media being used.

KF uses various modulation techniques to achieve a near identical copy of the IPF image - Trace did not have this capability, so virtually every single disk written was slightly different, especially as drives went out of alignment gradually over the course of usage.

The KF host software automatically and continuously adjusts the cells, modulation and pre-compensation to the current drive speed AND the target system's drive speed to make the image written as optimal as is possible - Trace couldn't do either of that, it required recalibration of the drives after a while or adjusting the FreeForm script describing how to write the data...

As long as your disk and drive is good enough (I know... not exactly a scientific description) - each disk written with KF is practically identical, they differ only in a few microseconds range - which is pretty good considering the variad nature of the drive signals, speed wobble etc.
Actually, it has become much better than I ever thought it could be.

We also use some magic pre-compensation, which was not always correctly set for duplicated disks at the time.

See the attached pictures for the effect, which is very clean bands making it easy for the controllers to read the data correctly, even at the very dense inner tracks of a disk using non-standard (very high) bit densities.

Note, many original disks wouldn't look as "good" as the picture with the scattered cell bands shown here on the left hand side.
Attached Thumbnails
Click image for larger version

Name:	noprecscatter.PNG
Views:	399
Size:	83.1 KB
ID:	29039   Click image for larger version

Name:	noprechist.PNG
Views:	292
Size:	60.4 KB
ID:	29040   Click image for larger version

Name:	precscatter.PNG
Views:	279
Size:	75.2 KB
ID:	29041   Click image for larger version

Name:	prechist.PNG
Views:	311
Size:	58.2 KB
ID:	29042  

Last edited by IFW; 08 July 2011 at 14:27.
IFW is offline  
Page generated in 0.05709 seconds with 10 queries