Master Vs. Slave for my Cd bruner and my CD-Rom

[gomcse]

Hi folks,

If I want to hook up my Cd rom and my Cd burner on the same IDE channel by using one IDE cable.

Should I set my Cd burner as master and Cd-Rom as slave ??

Please kindly advise.

[AllGamer]

i did the same

and i have set my CDRW as Master
DVD as Slave.

just a small Paranoia of mine, thinking Master should always has the most timing during access by the IDE bus :D

[Quandary]

Yeah; burner master, CD slave. I do not see why you would want such a configuration, though - a CD burner will read CDs just as well as a plain ROM drive, and only one drive can be being read from (or writen to) at any given time.

And just in case you have any silly ideas in your head, do NOT do on-the-fly copies in this configuration.

- Q

[AllGamer]

huh ?!?! :confused:

it works perfectly for me

and that's Precisely WHY i set it up to have both CD drives on the same IDE channel

so i can multitask.

do stuff meanwhile burning without affecting the other IDE channel

if you had it the other way around

then your Copies usually ends up bad, if you are doing other stuff that needs HDD access.

but having both on the same channel, it remains unaffected. :D

so CD 2 CD on the Fly works good this way.

at least it does for me. :t

[Quandary]

Um, have you looked at the IDE spec? Only one drive can be accessing the channel at any one time.

So, what?

In an OTF copy, here is what happens (with 2 channels, both master)

Source drive has the channel the whole time and streams data off of the disk

Destination drive hooks up to where the source drive is writing to, and reads all of that data - also using the whole channel the whole time.

Both drives work in /parellel/ and everything is OK.


On one channel, this is what happens:

source drive locks the IDE channel and starts writing. The destination drive starts eating up it's internal cache because no more data is coming in. If the dest. runs out of data, a buffer underrun occurs. If you have burn proof, this results in linking blocks, which cause the copy to be imperfect.

The source drive uses up its write-out time, and releases the bus to the destination drive. The dest drive starts reading the data that the source drive read out. The source drive has to pause while this is happening.


The result? You are not working in parellel. Just like a single processor computer does not (can not) perform parellel tasks, your single channel drive is not performing parrellel data transfers. The result? You end up with the source idling and the destination starving. If you like a whole helluva lot of link blocks on your CD copies, then fine. The setup should, however, be an absolute last resort.

- Q

[AllGamer]

so what, my 80min CDs recording still finishes at or under 3 min, with no error, that's all i care about.

:D

and the best part is that i can still run other stuff, without worrying for corrupted copies :p

[Quandary]

Well, then you don't need to have them on the same chanel, in that case.

-Q

[AllGamer]

if i have them on separate channel then Buffer under runs happens when i run other Apps or Surf the net.

why do you think i switched in the first place? :rolleyes:

[Quandary]

Okay, pay CLOSE attention...

The drives are under the SAME stress as they would be if you had active disks on the channel. Having two active CDs on the channel is a Bad Thing. Don't take my word for it; hook both of the buggers up as masters. Hey, look, you burn just fine, but *now* you feel what the slaved CD-ROM has to go through when it has an active drive on top of it. This will also demonstrate why you never slave a HD to an ATAPI device.

You would be able to have the following set up (your scenario):

IDE0
Mas - HD
Sla - HD

IDE1
Mas - CD-R
Sla - CD-ROM


and it would be just as bad as this:

IDE0
Mas - HD (OS/boot disk)
Sla - CD-ROM

IDE1
Mas - CD-R
sla - HD (data)

The real solution? Get an IDE card. IDE was never meant for simultaneous device access; that's why SCSI is around. Most new mobos have four IDE headers anyway, so your optimal set up would be each device on its own channel.

-Q

An afterthought:
You could have
IDE0
HD (defragging)
CD-R
IDE1
HD (defragging)
CD-ROM

And still not burn a coaster. It's this wonderful little thing called burn-proof. You'll end up with a whole bunch of linking blocks (ick) but it is almost impossible to burn a coaster. If you burn a CD in 3 min, you should have burn proof, too.

[AllGamer]

you know a weird irony?

when i was using my SCSI CDRW and SCSI HDD i was getting more coaster than my current pure IDE + RAID IDE setup.

i gave the old SCSI system to my sis
since she doesn't Multitask like i do.

[Quandary]

DUH. They don't have buffer underrun prevention (burn-proof). That is the only thing that is saving your *** in the current setup, and it is leaving nasty little link blocks all over your CD.

- Q

[thxmanu]

gomcse- works the same either way.

[AllGamer]

Originally posted by Quandary
DUH. They don't have buffer underrun prevention (burn-proof). That is the only thing that is saving your *** in the current setup, and it is leaving nasty little link blocks all over your CD.

- Q

I use Nero
and Burn prof if not enable
no check mark in the Box
Nero does not recomend it
it gives you the warning the first time when you go try to check them

:t :D

so i know i'm not using Burn proff

i'll say it's that spacious 8 meg of cache that is saving my CDs

the old SCSI burner only had 2 meg 8x scsi CDRW

[RamonGTP]

I know you're not suppose to do on-the-fly burning when both devices are on the same IDE channel... But I do it anyway and haven't had any issues with it. I try not to do it when i'm burning at speeds above 12x though.

-Ramon

[AllGamer]

as far as the 8x, 12x , 24x, 40x
i don't think there's much difference since 8x to 12x
cuz now regardless of what speed you set it to above 12x it still records that fast. or i meant at the same speed as 12x just with maybe a slight less time, hardly noticeable
8x was like 10 to 15 min
12x was like 5 to 6 min
24 and above has remained around the 3 min mark, a tiny bit less, a tiny bit more

so to my best guest if we ever reach 100x it'll still be around the 2 min to 3 min mark.

it'll be **** cool if a CDR or RW can instanly record in a sec or 2 sec like with HDD

but it's just a dream
it takes at least 1 sec to get the Lazer head moving
another sec at least for it to spin up
and then another sec for it to scan the empty disk, to choose a starting point.

that's 3 to 4 sec already.
so even at our best current technology it'll stick take 3 to 4 sec of delay

:D ahhh ... nice dream.... :x

[mattj2012]

:D you know it really doesnt matter how you have them set to master or slave it will work all the same.. all that is doing is allowing you to see both devices when you set one to master and one to slave... and those drives will multitask with no problem.. there really isnt a set way when it comes to CD-r or CD-rom from what ive seen unless you have a particular way of doing it for some special reason.
For IDE hard disk, this does matter because of the boot sequence so be carefull which one you set as master and slave. Of course having a CD rom and A CD-RW is redundant and a waste of space.. you could be using a DVD player there and copy DVD's and copy them to VCD formats. hee hee, not saying that i would do that or that anyone else should do that.. anywhooo, im outa here.. have fun and you dont have to prove a point to anyone except yourself.. unless your vain like that..:p

[mcjamesk]

I'm guessing that "gomcse" got what he or she needed out of this thread.....huh