Amiga.org
Amiga computer related discussion => Amiga Hardware Issues and discussion => Topic started by: Wain on September 18, 2003, 06:08:04 PM
-
I read somewhere (don't remember where) that the A1200 was pushed into mass production with a few manageable bugs still in the system, the most major one being a lot of CPU taxation on the internal IDE bus.
I was wondering if anyone knows if this issue was ever resolved, and how much more CPU usage are we talking about compared to IDE generated via a different controller(which has an already much higher rate of CPU taxation than SCSI).
-
It isn't really a bug per se, more like a cheap design descision..
I haven't used the internal ide in a long time, however, so I can't give numbers. :-/
-
I read somehwere (don't remember where) that the A1200 was pushed into mass production with a few manageable bugs still in the system, the most major one being a lot of CPU taxation on the internal IDE bus.
Non-DMA ide is like this, no way to get over it. This is not a bug, but a feature.
I was wondering if anyone knows if this issue was ever resolved
No it was never fixed.
With A600/A1200/A4000 IDE, all data goes thru 16-bit register. CPU need to push/pull each word one at a time. This is slow and CPU demanding.
-
It isn't really a bug per se, more like a cheap design descision..
I agree that something like that wouldn't really be a bug, just the thing I read made it sound like the engineering department really wanted a little more time to make the whole thing flow better, and were unhappy with having to go to production with it like that.
-
Okay, so it's unbuffered non-DMA IDE (eeewww), got it.
that's all I needed to know...
Thanks.
-
@ Wain
the IDE interface on A600/1200/4000 works *only* in PIO Mode 0, its maximum transfer rate would be 3.3 MB/s but most likely 2.3 MB/s is what you get...
This is why there are some "cards" that plug on a particular chip (don't remember its name) in the A1200 and provides UDMA modes
/ipino
-
@ ipino
if you are speaking about the PowerFlyer cards, they aren't UDMA but E-IDE with PioMode4 and eats more CPU power, than internal IDE. (anyway, are very useful and fast)
Ciao
-
This is why there are some "cards" that plug on a particular chip (don't remember its name) in the A1200 and provides UDMA modes
I was under the impression due to recent conversations on forums here that UDMA was not possible with any hardware add-ons, end of story...
I'll try and dig out the thread.
-
What about UDMA capable IDE drives attached to an accelerator card's SCSI adaptor via a bridge?
Does that still work in DMA mode?
If so, where can I get the cabling and bridge? :lol:
-
What about UDMA capable IDE drives attached to an accelerator card's SCSI adaptor via a bridge?
Works fine. I have two SCSI2 AEC-7720U (http://www.acard.com/eng/product/scside/aec-7720u.html) SCSIDE modules in my A1200 connected to BlizzPPC SCSI.
Does that still work in DMA mode?
Yes, of course.
If so, where can I get the cabling and bridge?
ACARD SCSIDE (http://www.acard.com/eng/product/scside.html). At least KDH (http://www.kdh-amiga-shop.com/) is selling them.
No special cables are required, normal SCSI cables will do.
This is no longer A1200 IDE bus, however. So it's abit OT.
-
I was just reflecting on mikeymikes remark.
Piru wrote:
No special cables are required, normal SCSI cables will do.
I was actually thinking about that awkward connector on the card itself :-)
This is no longer A1200 IDE bus, however. So it's abit OT.
Granted, but I was reflecting on mikeymike's post about no UDMA IDE solutions for A1200 you see...