Does CSEL work on A1200 IDE?

tiffers

New member
AmiBayer
Joined
Feb 18, 2008
Posts
85
Country
Perth, Australia
Like many people before me, I've just spent the better part of my night discovering slim dvd rom drives are set to master / slave / csel in firmware (which explains the lack of jumpers on those handy adaptor boards to change the 50-pin JAE connector to 40-pin IDE)

My question is, does CSEL work properly on the A1200? I have a Teac DV-28E. Model Ver. - C93 PN: 1977067C-93 Apparently the -93 indicates CSEL.

When I have my 40GB Hitachi Travelstar drive (which is un-jumpered - set to Master) plugged into the 44-pin header of my 4-device buffered IDE board (the cheap one from AmigaKit), and the Teac connected to the secondary 3.5" IDE connector, I get the kickstart screen. (in fact I think this also happens if I have it connected to the primary 3.5" IDE connector)

Yet if I add a slave to the secondary channel, the drives all appear and work.

It's now 2.20am, and I'm not playing any longer. My question is, if I set the HDD to CSEL will this help? Or do I have to do the old ground-pin-47-to-force-master-status hack?

tiffers
 
Last edited:

Zetr0

Ya' Like it Retr0?
Joined
Nov 22, 2007
Posts
9,900
Country
UK
Region
Norfolk
@tiffers

Cable Select works with the Amiga - although somtimes with old roms and drives it can have issues with the landing zone of the secondary device -

when using CS or CSEL make sure the HDD is the first device in the chain AFAIK Optical Drives cannot configure a landing zone should a device need it.

for CS or CSEL to work properly ALL devices on the chain should be CS / Cable Select / CSEL
 

Merlin

Ministry of Retr0bright and Street Judge
VIP
Joined
Nov 24, 2007
Posts
15,597
Country
UK
Region
Manchester
@ tiffers

Most laptop CD drives are pre-set to Master at the factory, although there are some drives that have a slide switch on the back to configure them.

Your Kickstart screen is being trigged by the two devices both being set to Master and locking the IDE bus up.

You should find they work if they are put on a separate channel, since both will now be Master on a different channel.
 

tiffers

New member
AmiBayer
Joined
Feb 18, 2008
Posts
85
Country
Perth, Australia
This is the funny thing though Merlin, Both on their own channels, they disappear. When I add a CD-RW drive specifically jumpered to SL, all drives appear and work correctly.

Apparently this DVD-ROM is set to CSEL. (There's a thread on a Car audio forum somewhere discussing Teac drives having -83 or -93 at the end of their model numbers, indicating CSEL or INVERSE CSEL, and the guy came to the conclusion that his -93 drive was CSEL)

Do I need an 80-wire ribbon cable for CSEL to work correctly?

I also have 1 intriguing observation: The HDD light is permenantly on. Even if I plug the HDD directly to the 44-pin header on the A1200 motherboard, the light is still on permenantly.

I have a 1D3 motherboard, which seems to have had a hackish reset fix performed by soldering components directly to the chip.

P.S. Anyone have a Slim DVD/CD-RW Combo drive known working they'd like to part with cheaply? :roll:
 

arnljot

Dad, gamer, nerd, amigian
Joined
Feb 5, 2008
Posts
3,845
Country
Oslo
Region
Oslo
For cable select to work, you'll need a special cable.
 

Zetr0

Ya' Like it Retr0?
Joined
Nov 22, 2007
Posts
9,900
Country
UK
Region
Norfolk
For cable select to work, you'll need a special cable.

err do you mean as special as a standard 40 pin 33 ATA cable? because that will do...

yes ANY standard 40/44 pin cable (from ATA33 upto ATA133)
 

arnljot

Dad, gamer, nerd, amigian
Joined
Feb 5, 2008
Posts
3,845
Country
Oslo
Region
Oslo
To use cable select, both devices on the channel are set to the "cable select" (CS) setting, usually by a special jumper. Then, a special cable is used. This cable is very similar in most respects to the regular IDE/ATA cable, except for the CSEL signal. CSEL is carried on wire #28 of the standard IDE/ATA cable, and is grounded at the host's connector (the one that attaches to the motherboard or controller). On a cable select cable, one of the connectors (the "master connector") has pin #28 connected through to the cable, but the other (the "slave connector") has an open circuit on that pin (no connection). When both drives on the channel are set cable select, here's what happens:

  • Master: The device that is attached to the "master connector" sees the CSEL signal as grounded, because its connector has pin #28 attached to the cable, and the host's connector has that signal grounded. Seeing the "zero value" (grounded), the device sets itself to operate as master (device 0).
  • Slave: The drive that is attached to the "slave connector" does not see the CSEL signal as grounded, because its connector is not attached to the CSEL signal on the cable. Seeing this "no connection", the device configures itself as a slave (device 1).
This is what I found using a popular search engine ;-)
 

keropi

Active member
VIP
Joined
Dec 5, 2007
Posts
7,143
Country
Greece
Region
Ioannina
really? that would explain why some cables (mainly from big manufacturers) have a hole after the 1st connector (apparently pin28 is cutted)
I did not knew that :)

ide-cable-select-easy.gif
 

Zetr0

Ya' Like it Retr0?
Joined
Nov 22, 2007
Posts
9,900
Country
UK
Region
Norfolk
@arnljot

I appreciate your googlings my friend but where ever you got that info its epically wrong!

I would say, and with a smidge of gusto - "CS requires special cable? - Special cable my arse!"


@Keropi

Those particular cables with a snip are ATA66 cables, this *snip* informs the controller that is can use ATA66 speed. otherwise the old ATA66 controllers would default to ATA33

its all about them controllers my friend, when they moved upto ATA80 speeds (and beyond) the controllers used ground inference to know the cable type (i.e. NOT ata33) so to run at optimum speed.

On some motherboards you can even overide this in the BIOS.


@thread


There is NO difference between an ATA80 / ATA100 / ATA133 cables, by using one of these cables on an ATA66 controller it will default to ATA33

an 80wire IDE cable only has extra shielding wires (every other wire) to reduce noise and signal doplar's, theres is nothing special about it at all.

For those that dont know the ATA-xx the number referes to MegaBits per second, (yes MEGABITS) so ATA100 is 12.5MegaBYTES a second (theorectical) through put -

Actual speeds will be lower, reading obviously is faster than writing but you can expect 9MB a sec read and 6MB a sec write.

---

here endeth the lesson.
 

Kin Hell

Active member
Banned
Joined
Nov 25, 2007
Posts
6,970
Country
U.K.
Region
Cornwall
Considering the level of expertise of most posters on this thread, I am so suprised the majority of people were not aware you need a specific cable when using the CS pins on a hard drive.
These cables are always 80 wire too. :shhh:

The blue header goes to the Mobo header, the Black to the Master & the Grey to the Slave.

Kin
 

tiffers

New member
AmiBayer
Joined
Feb 18, 2008
Posts
85
Country
Perth, Australia
I appreciate your googlings my friend but where ever you got that info its epically wrong!

arnljot got the info from: PC Guide

PC Guide claims to have gotten their info from: Hale Landis of ata-atapi.com

The PC Guide claims that pin 28 is grounded and this is what CS drives use to tell if they're master or slave (grounded = master, not grounded = slave)

It explains that the early ATA cables were poorly designed, using the cut wire as shown by keropi, leaving a single drive in the middle of the cable, not the end, resulting in probably signal reflection issues. The other alternative at the time was the 'Y' cable, with the mobo in the middle and master or slave at either end, making it hard to mount drives far from the mobo.

Plus it states a lot of manufacturers just didn't support it, which explains a lot about CS not working in the 40-pin ATA33 days (I had heaps of problems with it, and gave up on it myself)

Then when they bought out the 80-wire cables, they were clever enough to actually remove the 'pin-28' connector in the first connector, while leaving it in on the 2nd (and final) connector. This had the master (and possibly singles) drive at the end, as well as ensuring CS worked as designed, and it would seem all manufacturers supported this method.

I'm using a 40-wire cable which does not seem to have anything special done to it.

It still doesn't explain why the drives diesappear when I effectively have 2 masters, on per channel, but when I add a slave to the secondary channel, all 3 drives show up.

I'm going to have to try shorting pin 47 on the JAE connector to ground to see if it works.

tiffers
 

Kin Hell

Active member
Banned
Joined
Nov 25, 2007
Posts
6,970
Country
U.K.
Region
Cornwall
Just to add Tiffers:

If you use a 40 pin cable on one of the later EIDE Hard Drives, you might not be getting the transfer speeds you could from an 80 wire jobby. ;)

ATA133 Devices Mode 5
ATA100 Devices Mode 4
ATA66 Devices Mode 3
ATA33 Devices Mode 2

Imho, anything lower than Mode 4 is better off in the bin or best used as a door stop. :nod:

Kin
 

tiffers

New member
AmiBayer
Joined
Feb 18, 2008
Posts
85
Country
Perth, Australia
Just to add Tiffers:

If you use a 40 pin cable on one of the later EIDE Hard Drives, you might not be getting the transfer speeds you could from an 80 wire jobby. ;)

ATA133 Devices Mode 5
ATA100 Devices Mode 4
ATA66 Devices Mode 3
ATA33 Devices Mode 2

Imho, anything lower than Mode 4 is better off in the bin or best used as a door stop. :nod:

Kin

Would an A1200 with a simple buffered interface be able to keep up with anything over an ATA33? :D

My drive is an Hitachi Travel Star 5400RPM 40GB. Can't remember the exact model. It may be the HTS548040M9AT00. It's connected to the IDE Buffered Interface by a 44-pin 2.5" cable.

This drive seems to be rated for:

100MB/sec Ultra DMA mode-5
16.6MB/sec PIO mode-4

I doubt the A1200 would handle these speeds though. Am I wrong?
 

arnljot

Dad, gamer, nerd, amigian
Joined
Feb 5, 2008
Posts
3,845
Country
Oslo
Region
Oslo
Considering the level of expertise of most posters on this thread, I am so suprised the majority of people were not aware you need a specific cable when using the CS pins on a hard drive.

For cable select to work, you'll need a special cable.

Oh, I'm sorry. A spesific cable, not a special one. My mistake :cool: :D
 

Zetr0

Ya' Like it Retr0?
Joined
Nov 22, 2007
Posts
9,900
Country
UK
Region
Norfolk
Firstly -

You can use Cable Select on ANY IDE (IDC) cable, the ONLY time you would need a "specific cable" is if the Devices you use do not allow/support Cable Select.

Those devices were phased out by the mid 90's, any HDD or optical drive made past 1996 is very unlikely to have this problem.

It is the DEVICE that sets its mode NOT the cable, (well not since the late 80's and very early 90's). The data is very out of date, mis-leading and inacurate.

SOME Harddisk devices will ignore any device thats second in the chain (this can also happen when set from MA/SA as well) this is a device and controller issue, the controller inquires the 2nd unit and the first unit in the chain intercepts the signal but wont pass it on.

For you problems you need to review these pics-

main.php


main.php


this will enable your device for CS, the IDE on the A1200 will support this, it now comes down to if your HDD will intercept the signal properly.

ATA 33 is 33MBits which is (33/8 = MBytes) [4.1MB per sec] thats twice as fast as the Miggy native IDE with an 060 pushing it
 

Kin Hell

Active member
Banned
Joined
Nov 25, 2007
Posts
6,970
Country
U.K.
Region
Cornwall
@ tiffers

Different HD's will give you different results on the Miggy & @ best, the default header will shove anything between 2 & 4 Mb/s depending on file system, size of partition & hard drive type. It all starts to go a bit pear shaped with drives over 4GB & there is also a CPU hit for it as well. ;)

Kin
 

tiffers

New member
AmiBayer
Joined
Feb 18, 2008
Posts
85
Country
Perth, Australia
It is the DEVICE that sets its mode NOT the cable, (well not since the late 80's and very early 90's). The data is very out of date, mis-leading and inacurate.

So, how do the devices do this? One would have thought 'cable select' would mean the 'cable' determined the designation.

The grounded pin method sounds plausbile. The only other way I could think of would be a 'race' condition, where the one closer to the board would reply fater than the next one along the cable.

Are you sure they don't use grounded pin 28 anymore?

If only I had:

a) a bunch of old and new drives (which all supposedly supported CS)
2) a sampling of old cables (including those cut cables)
iii) enough time on my hands to do conclusive, process of elimination tests

I'd go ahead and work it all out.

tiffers

tiffers
 

arnljot

Dad, gamer, nerd, amigian
Joined
Feb 5, 2008
Posts
3,845
Country
Oslo
Region
Oslo
@Zetr0

The wikipedia article on IDE/PATA/ATA also mentions #28 and CS. Is there something we're missing?
 

Zetr0

Ya' Like it Retr0?
Joined
Nov 22, 2007
Posts
9,900
Country
UK
Region
Norfolk
okay...

when you set the device to CS, it (the device) grounds pin 28 on the IDE chain

does that help?
 

Zetr0

Ya' Like it Retr0?
Joined
Nov 22, 2007
Posts
9,900
Country
UK
Region
Norfolk
@Tiffers

you will need to configure your IDE harddisk, as this MUST be cable select (CS) for CSEL to work for two devices on the chain.

This is the most common for 2.5" (IBM/Hitachi Travel Star)
addjum.gif


although some models might be a little different so check here first

worse case scenario you could try to modify the cable by by cutting wire ( pin 28 ) between the HDD and CDROM, however as the drive (unit) is not configured internally as CS, it may block the attempts of the controller to inquire the second device in the chain.

The best you can do, is to configure your HDD as CS and your Optical ROM as CS, remember not all components play nice, as not all standards are standard lol.

It is worth noting that while *some* SFF optical drives are firmware fashed as Master, most, of the few that are firmware flashed would be flashed as Cable Select.
 
Top Bottom