Welcome, Guest. Please login or register.

Author Topic: IDE PCI controller card with Mediator- anyone tried?  (Read 2637 times)

Description:

0 Members and 1 Guest are viewing this topic.

Offline Karlos

  • Sockologist
  • Global Moderator
  • Hero Member
  • *****
  • Join Date: Nov 2002
  • Posts: 16882
  • Country: gb
  • Thanked: 6 times
    • Show all replies
Re: IDE PCI controller card with Mediator- anyone tried?
« on: December 01, 2003, 01:44:40 AM »
Digital data stored on a grammophone would be faster than the A1200's bare IDE ;-)
int p; // A
 

Offline Karlos

  • Sockologist
  • Global Moderator
  • Hero Member
  • *****
  • Join Date: Nov 2002
  • Posts: 16882
  • Country: gb
  • Thanked: 6 times
    • Show all replies
Re: IDE PCI controller card with Mediator- anyone tried?
« Reply #1 on: December 01, 2003, 01:52:19 AM »
I've no IDEa...

/sorry
int p; // A
 

Offline Karlos

  • Sockologist
  • Global Moderator
  • Hero Member
  • *****
  • Join Date: Nov 2002
  • Posts: 16882
  • Country: gb
  • Thanked: 6 times
    • Show all replies
Re: IDE PCI controller card with Mediator- anyone tried?
« Reply #2 on: December 01, 2003, 03:21:36 AM »
Theres this whole odd thing going on with mediator and drivers and stuff right now...

Maybe the prometheus? IIRC it supports open PCI?

Patrik would know...

Thinking about it, he knows waaay too much....

/me looks around with theatrical slyness..

Seriously though, I don't doubt its possible to get better transfer rates over PCI than the 1200 IDE but it would likely be very CPU hogging all the same.

If you want low CPU overhead DMA transfers on A1200, SCSI is still the best way to go (and add a few IDE drives with a bridge) for now, IMHO. 10Mb/s sustained transfer is feasable if phase 5 SCSI controllers live up to their specs.
int p; // A
 

Offline Karlos

  • Sockologist
  • Global Moderator
  • Hero Member
  • *****
  • Join Date: Nov 2002
  • Posts: 16882
  • Country: gb
  • Thanked: 6 times
    • Show all replies
Re: IDE PCI controller card with Mediator- anyone tried?
« Reply #3 on: December 01, 2003, 03:52:16 AM »
Or you could split a driver in two. A generic pciide.device that serves as the standard AmigaOS interface.
This in turn would delegate the low level control to a chipset/card specific subdriver. A bit like the AHI design...
int p; // A