Disk hardware emulation, was Re: Grandfather system RTE6/VM?

From: Eric Smith <eric_at_brouhaha.com>
Date: Thu Dec 11 13:52:24 2003

I wrote:
>> If you did decode the data, it would reduce the storage requirements of
>> the emulator considerably, but it would make it a lot less
>> general-purpose.

Ben wrote:
> I don't see how that would make a large difference as you only have 2x
> increase in data size. Anyhow would not information be saved on a track
> rather than sector chunk size to ram?

Without decoding the data, it is proposed to sample the data at 50 Mbps
or more.

If the data is MFM, and if the emulator decodes it, only about 5 Mbps
needs to be stored. Slightly more, actually, due to the need to
represent coding violations used for the address and data marks. In
practice, this would be most easily accomplished by storing 10 Mbps of
decoded data and clocks. This is a 5x reduction in data rate as
compared to the raw sampling.

However, given that no new 2.5- and 3.5-inch hard drives are made with
capacity lower than 40 GB, there's not really much point to trying to
save storage space.

And certainly having the emulator decode the data makes it much less
general-purpose.

> While you don't encode data,
> knowing the data format would help with removing write precom when
> syncronzing the data to the track bit stream.

Removing write precomp, if necessary, does not require any knowledge
of the data encoding. It only has to model the magnetic characteristics
of the recording channel.

Eric
Received on Thu Dec 11 2003 - 13:52:24 GMT