NEW MIO production run.

[+poobah]

It anyone is really wanting MFM drives, I have a couple full height 5.25" drive from old XT/AT machines, and a couple 1/3rd height Rodimes.

[+Larry]

Just out of interest - I don't even have an MIO yet - is this the kind of controller you're talking about?

 

I'm not an expert on compatibility of these old boards, but I'm sure that the Adaptec 4070 is an RLL controller, so that might make your drive selection more difficult.

 

Does anyone have/use a 4070? If so, what drive do you use it with?

 

BTW, if anyone wants a complete "back-in-the-day" MFM setup for an MIO (or BB), PM me -- perhaps we can work out a deal. This would be U.S. only -- shipping this overseas would cost a fortune.

 

-Larry

[MEtalGuy66]

Just out of interest - I don't even have an MIO yet - is this the kind of controller you're talking about?

 

I'm not an expert on compatibility of these old boards, but I'm sure that the Adaptec 4070 is an RLL controller, so that might make your drive selection more difficult.

 

Does anyone have/use a 4070? If so, what drive do you use it with?

 

BTW, if anyone wants a complete "back-in-the-day" MFM setup for an MIO (or BB), PM me -- perhaps we can work out a deal. This would be U.S. only -- shipping this overseas would cost a fortune.

 

-Larry

 

You can use standard MFM drives with the 4070, but it's not "guranteed" so to speak.. I have heard of people doing it..

 

The correct board to use with MFMs is the 4000A...

 

That said, those 4070s in that auction are a STEAL! Your not likely to see that again any time soon. If youve got the money, snatch them up..

Edited February 5, 2009 by MEtalGuy66

[sl0re]

Just out of interest - I don't even have an MIO yet - is this the kind of controller you're talking about?

 

I'm not an expert on compatibility of these old boards, but I'm sure that the Adaptec 4070 is an RLL controller, so that might make your drive selection more difficult.

 

Does anyone have/use a 4070? If so, what drive do you use it with?

 

BTW, if anyone wants a complete "back-in-the-day" MFM setup for an MIO (or BB), PM me -- perhaps we can work out a deal. This would be U.S. only -- shipping this overseas would cost a fortune.

 

-Larry

 

RLL and MFM are the same thing except you put more sectors per track onto a drive formatted RLL. So, you can format an MFM drive RLL, it will just fail after awhile (maybe even during the low level format or very soon after you start using it). Whereas, with an RLL drive it will also fail if formatted RLL, but it will take a bit longer . Which is why I gave up on RLL and started formatting RLL equipment with MFM specs back in the day.

 

In the end, it was mostly about 'certification'. The vendor would claim an RLL drive could handle the extra data on drive X and hence, they sold it as an RLL drive (and charged more). On the controller end, it would allow you to enter the higher numbers per track for formatting. Or translation, in the end, RLL was bullsh*t / a sort of ‘overclocking’ the original MFM spec... and it was unreliable / sucked. But, there is no reason you can’t use an RLL controller to format a drive with the original MFM specs and use it…. Other than / unless, it’s really inflexible and won’t let you use the lower numbers (never saw that though).

 

If you grew up with IDE, this stuff is something you never deal with because the controller is integrated with the drive and low level formatted at the factory. You don’t really get to pick the number of sectors per track. You just make stuff up to put in the bios to divvy up partition space; which is then translated by the drive into the actual numbers used (on the low level format)…. Yeah, back in the day when I had to walk uphill in the snow to school both ways, when you paid a tech $50 to install a hard drive for you, it was a good idea (as it was hard)... No interweb to look for help either. Had to call hardware vendors and ask for help...

[ijor]

RLL and MFM are the same thing except you put more sectors per track onto a drive formatted RLL. So, you can format an MFM drive RLL, it will just fail after awhile (maybe even during the low level format or very soon after you start using it). Whereas, with an RLL drive it will also fail if formatted RLL, but it will take a bit longer . Which is why I gave up on RLL and started formatting RLL equipment with MFM specs back in the day.

 

In the end, it was mostly about 'certification'. The vendor would claim an RLL drive could handle the extra data on drive X and hence, they sold it as an RLL drive (and charged more). On the controller end, it would allow you to enter the higher numbers per track for formatting. Or translation, in the end, RLL was bullsh*t / a sort of ‘overclocking’ the original MFM spec... and it was unreliable / sucked. But, there is no reason you can’t use an RLL controller to format a drive with the original MFM specs and use it…. Other than / unless, it’s really inflexible and won’t let you use the lower numbers (never saw that though).

 

If you grew up with IDE, this stuff is something you never deal...

 

RLL is not (only) more sectors per track. It is a completely different, and much more efficient, encoding. You can consider this as being similar as MFM vs FM (double vs. single density).

 

RLL is in no way any kind of overclocking, and wasn't unreliable at all. In the same drive, both MFM and RLL controllers use exactly the same density. As a matter of fact, all the IDE and SCSI drives that followed RLL, were ... well, exactly the same RLL. They have their built-in controller, but internally, they used exactly the same RLL encoding you used in discrete RLL drives. So if RLL was unreliable, then IDE drives should be as much unreliable.

 

You can if you want, use a RLL controller and format a drive with the MFM geometry. But it would be completely useless. It would be useless because it would still use the same density and encoding, still RLL. So you woudn't improve the reliability at all.

 

The reason that some MFM drives weren't reliable under RLL, or the reason of the certification, is because RLL requires a bit more precision on the drive. Once again, it doesn't require more density (which is the fundamental aspect that affects reliability vs. capacity).

 

Seems you just had bad luck with the RLL drives/controllers you used. Most people had the opposite experience, and usually they could use RLL reliably with just about every MFM drive.

[sl0re]

You’re mostly right. Apparently there is a timing component to RLL also.

http://en.wikipedia.org/wiki/Run_Length_Limited

 

On the other hand, you can format an RLL drive on an MFM controller and vice versa. RLL works fine (in my experience better) when formatted with MFM's sectors per track setting.

 

Also, I think future tech references (IDE) are not relevant to RLL as I'm talking about RLL with 80s drive platters. They had trouble with that sector density in my opinion. They could handle MFM encoding density but failed early and often with RLL (even RLL certified drives). Of course today's drives can handle higher densities. and it was not just me, I would hear from friends howling that they had to reformat their RLL drive again... on a regular basis…

[+poobah]

Also, I think future tech references (IDE) are not relevant to RLL as I'm talking about RLL with 80s drive platters. They had trouble with that sector density in my opinion. They could handle MFM encoding density but failed early and often with RLL (even RLL certified drives).

density is irrelevant

A drive formatted RLL or MFM has the exact same number of magnetic 'bits' on the platter, it is just a question of how those bits are encoded.

RLL is more efficient, but also more complex. (I use bits here as the number of flux reversals).

 

MFM: 100 flux reversals ~= 100 data bits

RLL: 100 reversals can be as much as 200 data bits, but normally is more on the order of 150.

[+bf2k+]

You can use standard MFM drives with the 4070, but it's not "guranteed" so to speak.. I have heard of people doing it..

 

I got one of those 4070's myself.

 

I used to use MFM drives on RLL controllers (IBM-PC) a lot back in the late 80's with no problems.

[sl0re]

Also, I think future tech references (IDE) are not relevant to RLL as I'm talking about RLL with 80s drive platters. They had trouble with that sector density in my opinion. They could handle MFM encoding density but failed early and often with RLL (even RLL certified drives).

density is irrelevant

A drive formatted RLL or MFM has the exact same number of magnetic 'bits' on the platter, it is just a question of how those bits are encoded.

RLL is more efficient, but also more complex. (I use bits here as the number of flux reversals).

 

MFM: 100 flux reversals ~= 100 data bits

RLL: 100 reversals can be as much as 200 data bits, but normally is more on the order of 150.

 

Soooo, why does an RLL drive formatted on an RLL controller come out to MFM storage specs when formatted with the number of MFM sectors per track? Or an RLL drive on an MFM controller? Or an MFM drive on an RLL controller?

[ijor]

Soooo, why does an RLL drive formatted on an RLL controller come out to MFM storage specs when formatted with the number of MFM sectors per track?

 

Because in that case you are using only a portion of the track on the disk surface. Or in other words, the number of raw unformatted data bits is still the same, only the number of formatted bits is smaller.

 

It would be the same if in a 1050 you format a disk using enhanced density at 18 sectors per track, instead of 26 (the 1050 firmware doesn't let you, but you could if you want with some 1050 enhancements). Density is the same, encoding is the same, just some big unused area somewhere on the track.

 

Or to take the idea even further, you can (again, if you want) take an RLL or MFM drive and format it with just one single sector per track.

Edited February 6, 2009 by ijor

[_The Doctor__]

Run Length Limited- The variables of run length and run limit are the combination here, It considers groups of several bits instead of encoding one bit at a time. It mixes clock and data flux reversals to allow for a more dense packing of encoded data, thus improving efficiency. The two parameters that define RLL are the run length and the run limit. A "run" refers to a sequence of spaces in the output data stream without flux reversals. The run length is the minimum spacing between flux reversals, and the run limit is the maximum spacing between the reversals. The amount of time between reversals cannot be too large or the read head can get out of sync and lose track of which bit is where, which is why some drives failed rll certification. They simply can't handle the timing involved.

 

The RLL system often included numbers to identify the schemes involved for example 1,7 RLL or RLL 1,7.

I was able to format some mfm drives using the RLL 1,7 encoding scheme and enjoyed having the extra storage for free. I was only able to use certified RLL drives with the 2,7 scheme. I found this when running Bulletin Board systems and getting every ounce of storage for database and video streams when hard drives were Very Expensive Real Estate. Following that rule I save some money and never lost my video or data. You mileage may vary.

Edited February 6, 2009 by _The Doctor__

[sl0re]

Run Length Limited- The variables of run length and run limit are the combination here, It considers groups of several bits instead of encoding one bit at a time. It mixes clock and data flux reversals to allow for a more dense packing of encoded data, thus improving efficiency. The two parameters that define RLL are the run length and the run limit. A "run" refers to a sequence of spaces in the output data stream without flux reversals. The run length is the minimum spacing between flux reversals, and the run limit is the maximum spacing between the reversals. The amount of time between reversals cannot be too large or the read head can get out of sync and lose track of which bit is where, which is why some drives failed rll certification. They simply can't handle the timing involved.

 

The RLL system often included numbers to identify the schemes involved for example 1,7 RLL or RLL 1,7.

I was able to format some mfm drives using the RLL 1,7 encoding scheme and enjoyed having the extra storage for free. I was only able to use certified RLL drives with the 2,7 scheme. I found this when running Bulletin Board systems and getting every ounce of storage for database and video streams when hard drives were Very Expensive Real Estate. Following that rule I save some money and never lost my video or data. You mileage may vary.

Edited February 6, 2009 by sl0re

[_The Doctor__]

Maybe these will help,

http://en.wikipedia.org/wiki/Run_Length_Limited

 

It took some time but I found this one which is more like what I was trying to say.... the picture of the data is off but close enough to get the idea

 

http://www.storagereview.com/guide/dataRLL.html

[DamageX]

I finally got the updated binary onto a chip and tried some 50-pin SCSI drives. One was a 4.2GB IBM drive which acted like it was working but after copying files to it the directory would come back garbled. The others... 250MB Quantum, 4GB Seagate, and a 'mystery' Seagate could not be accessed at all. Does parity need to be disabled? What else? My cable includes a terminator.

[_The Doctor__]

disable parity on drives, make sure on drive termination packs are out if using cable termination.

[+poobah]

Run Length Limited- The variables of run length and run limit are the combination here, It considers groups of several bits instead of encoding one bit at a time. It mixes clock and data flux reversals to allow for a more dense packing of encoded data, thus improving efficiency. The two parameters that define RLL are the run length and the run limit. A "run" refers to a sequence of spaces in the output data stream without flux reversals. The run length is the minimum spacing between flux reversals, and the run limit is the maximum spacing between the reversals. The amount of time between reversals cannot be too large or the read head can get out of sync and lose track of which bit is where, which is why some drives failed rll certification. They simply can't handle the timing involved.

 

The RLL system often included numbers to identify the schemes involved for example 1,7 RLL or RLL 1,7.

I was able to format some mfm drives using the RLL 1,7 encoding scheme and enjoyed having the extra storage for free. I was only able to use certified RLL drives with the 2,7 scheme. I found this when running Bulletin Board systems and getting every ounce of storage for database and video streams when hard drives were Very Expensive Real Estate. Following that rule I save some money and never lost my video or data. You mileage may vary.

 

MFM is actually a special case of RLL

(I believe RLL 1,3)

[+bf2k+]

Success! (somewhat)

Using the new MIO 1.4b2 firmware, I got a Seagate ST32430N (2.1gb) drive to work. This Seagate drive was actually manufactured for Apple. I didn't do a LL format on it... just the HDFMTDIR. Made 2 65535 sector partitions. Was able to save the MIO cfg (option 4) to sector 0 of the HD.

 

This is the first drive I have gotten this far with... fingers are crossed! I am copying files to it now...

 

Using SDX 4.42 and COPY /R A: C: the copy appears to go fine until some distance into it, the HD flashes once and everything stops.

 

Any ideas?

[MEtalGuy66]

Ok.. I need updated mailing adresses for:

 

Dan Olson

Pontus Oldberg

Rob Patton

 

either PM me on here, or email me..

[_The Doctor__]

Success! (somewhat)

Using the new MIO 1.4b2 firmware, I got a Seagate ST32430N (2.1gb) drive to work. This Seagate drive was actually manufactured for Apple. I didn't do a LL format on it... just the HDFMTDIR. Made 2 65535 sector partitions. Was able to save the MIO cfg (option 4) to sector 0 of the HD.

 

This is the first drive I have gotten this far with... fingers are crossed! I am copying files to it now...

 

Using SDX 4.42 and COPY /R A: C: the copy appears to go fine until some distance into it, the HD flashes once and everything stops.

 

Any ideas?

 

wow! apple used to have a wire or two soldered on those to modify them and make them not speak with non apple hardware/software. cool you got it to work without tinkering on the hard drives pcb.

[+bf2k+]

Success! (somewhat)

Using the new MIO 1.4b2 firmware, I got a Seagate ST32430N (2.1gb) drive to work. This Seagate drive was actually manufactured for Apple. I didn't do a LL format on it... just the HDFMTDIR. Made 2 65535 sector partitions. Was able to save the MIO cfg (option 4) to sector 0 of the HD.

 

This is the first drive I have gotten this far with... fingers are crossed! I am copying files to it now...

 

Using SDX 4.42 and COPY /R A: C: the copy appears to go fine until some distance into it, the HD flashes once and everything stops.

 

Any ideas?

 

wow! apple used to have a wire or two soldered on those to modify them and make them not speak with non apple hardware/software. cool you got it to work without tinkering on the hard drives pcb.

 

Maybe that's what the problem is, but I cannot get files to copy to it. Sometimes the SDX COPY routine dies, sometimes it aborts with a error 135 (file is read only?!??? - weird error). Hardback will not even start copying to it. My own sector copy program (very slow) will get anywhere from 25-50% into the copy before it dies (1 flash on the drive and the MIO activity LED stays on solid). It usually copies enough

for me to boot the drive and it boots very fast - definitely faster than my MIO 1.1-Adaptec-St225 setup at booting. When I say booting I mean SpartaDOS 3.2g (not X).

[drac030]

error 135 (file is read only?!??? - weird error)

 

Except the situations, where it makes sense this error sometimes appears when the DOS gets some mess instead of valid data when reading file and/or directories.

[mellis]

wow! apple used to have a wire or two soldered on those to modify them and make them not speak with non apple hardware/software. cool you got it to work without tinkering on the hard drives pcb.

 

Apple OEM drives were not crippled.

 

I've used old Apple OEM SCSI drives on all kinds of hardware, and I've never had any trouble using them with other computers. Specifically, I have used 40MB, 80MB, 160MB, and 500MB Apple OEM drives (made by Connor and Quantum) on an Atari ST, a NeXT 040 Cube, and PCs with Adaptec SCSI card.

 

The way that the drivers were customized is that Apple had the OEM include a custom ROM chip on the drive. In the System 7 days, the included "HD SC Setup" utility would check for the Apple ROM, and if it was missing, would refuse to format or install a driver on the drive. In those days, you could use a third party, non-Apple OEM drive if you used a third party formatter and driver. This "check for custom ROM" nonsense ended around the System 7.5-7.6 era.

[+bf2k+]

I seem to have struck out completely with the 1.4b2 firmware. I have ordered a new SCSI 50-pin cable to try the Seagate Hawk one more time. Keeping my fingers crossed....

[_The Doctor__]

The ancient apple drives refused to work with my Black Box, MIO, Falcon030, and my old 520ste(1meg) so I took a trip to double click computers and the fellow there helped me by pointing out the wires at that time, this was around 1990's or so and since this was the only apple dealer around I took his word. He helped me remove one, move another, and one we had not luck with as it only wanted to talk with the apples for whatever reason. We swapped the pcb and the last one was fine. So Apple must have experimented with making it so Apple was looking to limit who and what could be used. Must have been a known issue for the service tecch to know pretty much what was needed for each drive. He even advised me about the need to lubricate the one drive as it was susceptable to what he termed stiction, and a year later I was doing just that. Lubing the spindle bearing on the Apple HD. seems the lube used by the manufacturer would dry out and cause a sticky spindle not starting, hard to start or wrong rpm. After the lube all was fine again. Kudo's to double click for having the answers back in the '90's... the drives were old back then as well.

 

wow! apple used to have a wire or two soldered on those to modify them and make them not speak with non apple hardware/software. cool you got it to work without tinkering on the hard drives pcb.

 

Apple OEM drives were not crippled.

 

I've used old Apple OEM SCSI drives on all kinds of hardware, and I've never had any trouble using them with other computers. Specifically, I have used 40MB, 80MB, 160MB, and 500MB Apple OEM drives (made by Connor and Quantum) on an Atari ST, a NeXT 040 Cube, and PCs with Adaptec SCSI card.

 

The way that the drivers were customized is that Apple had the OEM include a custom ROM chip on the drive. In the System 7 days, the included "HD SC Setup" utility would check for the Apple ROM, and if it was missing, would refuse to format or install a driver on the drive. In those days, you could use a third party, non-Apple OEM drive if you used a third party formatter and driver. This "check for custom ROM" nonsense ended around the System 7.5-7.6 era.

 

 

Edited March 4, 2009 by _The Doctor__

[mellis]

The ancient apple drives refused to work with my Black Box, MIO, Falcon030, and my old 520ste(1meg) so I took a trip to double click computers and the fellow there helped me by pointing out the wires at that time, this was around 1990's or so and since this was the only apple dealer around I took his word. He helped me remove one, move another, and one we had not luck with as it only wanted to talk with the apples for whatever reason. We swapped the pcb and the last one was fine. So Apple must have experimented with making it so Apple was looking to limit who and what could be used. Must have been a known issue for the service tecch to know pretty much what was needed for each drive. He even advised me about the need to lubricate the one drive as it was susceptable to what he termed stiction, and a year later I was doing just that. Lubing the spindle bearing on the Apple HD. seems the lube used by the manufacturer would dry out and cause a sticky spindle not starting, hard to start or wrong rpm. After the lube all was fine again. Kudo's to double click for having the answers back in the '90's... the drives were old back then as well.

 

Were the problematic drives made by Rodime, by any chance? Those drives were the only brand of Apple OEM drives I've not personally owned. The Connor and Quantum OEM drives (circa 1992-1996) always worked well for me.

 

Cheers!