IDE Channel

[News Groupie]

I'm confused. I have a ATA-133 IDE controller card (by Promise) and had a
200 (master) and 160 gig (slave) hard drive both on the same channel (the
card has two channel). I put the 160 hd on the second channel by itself and
made it master. I thought this would increase the speed between the drives,
but the performance is the same (4 gigs in 2 mins & 10 secs both ways). The
thinking was bandwidth was cut in half because that one cable was used to
send and recieve the data. I figured by adding a cable and using the other
channel then one would be used for sending and the other for receiving and
therefore increase speed and performance.

Why is it the same speed/performance? Why hasn't it increased?

TIA

 

[Jim]

I'm confused. I have a ATA-133 IDE controller card (by Promise) and had a
200 (master) and 160 gig (slave) hard drive both on the same channel (the
card has two channel). I put the 160 hd on the second channel by itself
and made it master. I thought this would increase the speed between the
drives, but the performance is the same (4 gigs in 2 mins & 10 secs both
ways). The thinking was bandwidth was cut in half because that one cable
was used to send and recieve the data. I figured by adding a cable and
using the other channel then one would be used for sending and the other
for receiving and therefore increase speed and performance.

Why is it the same speed/performance? Why hasn't it increased?

Perhaps the problem is that the two channels share the same processor and
the same PCI bus.
In addition, the 133 is the burst rate not the steady state rate.
You will definitely see a difference if you use disks with 8 MB on board
buffers.
Jim

 

[Larry(LJL269)]

I'm confused. I have a ATA-133 IDE controller card (by Promise) and had a
200 (master) and 160 gig (slave) hard drive both on the same channel (the
card has two channel). I put the 160 hd on the second channel by itself and
made it master. I thought this would increase the speed between the drives,
but the performance is the same (4 gigs in 2 mins & 10 secs both ways). The
thinking was bandwidth was cut in half because that one cable was used to
send and recieve the data. I figured by adding a cable and using the other
channel then one would be used for sending and the other for receiving and
therefore increase speed and performance.

Why is it the same speed/performance? Why hasn't it increased?

TIA

Aliens have kidnapped ur PC & tweaked it ! The ones with big
eyes from Roswell, & crop circles

They caused my IDE-USB2 transfers to be 6x faster than
IDE-IDE rate. They also caused no one to answer my query of
6 mos ago You should see my pics!

Damn aliens!!

Larry

Any advise given is my attempt to show appreciation for all
the excellent help I've received here but I'm no MVP so it
may only apply NUGS (Normally, Usually, Generally, Sometimes

 

[Lil' Dave]

Data movement, per your figures, is just under 31MB/sec. PCI 32 bit
standard max is 33MB/sec.
Assuming HDs that can meet additional read/write speed for throughput,
there's no purpose in using any current ide card or onboard ide controller
using anything faster than the ATA-100 standard. And, you just proved it by
maxxing out the PCI bus.

BTW This has nothing to do with XP at all.

 

[Yves Leclerc]

You first need to check to see if BOTH Promise IDE channels have a ATA-133
Ultra DMA IDE cable installed. This cable needs to be 80 wires with 40 pins
connectors.

 

[R. McCarty]

An IDE channel can only handle a single request (operation) at a
time. Whether there is a single device or a M/S.

1 Read then a Write ( Same Channel - M/S) 2 Operations (X2)
1 Read then a Write ( Primary to Secondary) 2 Operations (X2)
Same Hardware, Controller - but using different channels, time
to completion would be the same.

 

[Lil' Dave]

Separate controller operation has been shown faster on some older systems.
Slight but visible. Today's PCs work effectively either way.