Chuck, So are you saying that with a CC IV you don't need the DBIT adapter, for writing, and just need the 34pin to 50pin converter? -Ali

Chuck, I've seen the wiring info in the manual and I just looked through the PDF version of the manual I have. I could not find any mention of TG43 in my quick look through. Do you by any chance know where the info is? Thanks. -Ali

On Wed, 4 Feb 2015, Chuck Guzis wrote: Oversimplified: On inner tracks on a constant speed drive, the data is closer together. Therefore, the reliability is lower. (Some variable speed drives will change speed and write less data on inner tracks) One effort to help a little is "write precompensation". On inner tracks, if adjacent flux transitions that would end up real close together are written slightly out of position, slightly further apart, it can help reduce the problem of flux transitions being too close together. As part of doing that, there is a signal available called "TG43" ("track number greater than 43") to help decide when to enable precomp. It is NOT needed for reading, but is sometimes used to enable precomp for WRITING.

Well, Fred, not exactly--you're correct in that it's used for innter tracks, but RWC/TG43 is a signal INPUT to the drive. The function is to reduce write current by about 20% to improve the readability of tracks 44-76. It's not on all drives--the Shugardt SA800, for example, doesn't have such a line. The issue is one of "bit shifting". When a transition is written close one already present, the tendency will be to "push" the already written transition away from the one being written. This can lead to "bit crowding". This effect can be seen if one has a controller, such as the Catweasel, which records a histogram of time between transitions. Outer tracks tend to show nice clean frequency peaks with relatively empty "valleys" between them--that is, transition times fall in nice, neat "bins" with little overflow adjacent ones. Go to the inner tracks and things are a lot more "fuzzy" statistically. RWC/TG43 was an attempt on early controllers to mitigate this by lowering write current, particularly on FM-encoded disks. Sometime along With MFM, the idea of "precompensation" came along. This is done in the controller, not the drive. The idea is that instead of spewing bits out of a shift register in the controller straight to the drive, one can run them through an extension of the shift register, so one can determine whether to advance or delay slightly the transition being recorded, based on what has been written and what will be written adjacent to the current bit. This will cause the transitions upon reading to be closer to optimal. You can still use the RWC/TG43 facility on the drive, but the effect is far less pronounced when the controller-side precompensation takes place. I hope this wasn't too confusing. It can be a bit puzzling. --Chuck On 02/04/2015 12:45 PM, Fred Cisin wrote:

Is TG43 available on both the internal 34-pin and the external DC(DB)-37 connector on the CompatiCard? Some old posts suggest that it's only output externally, but the CC manual seems to treat both internal and external connections to an 8" drive equally valid. m ----- Original Message ----- From: "Chuck Guzis" <cclist at sydex.com> To: <General at classiccmp.org>; "Discussion at classiccmp.org:On-Topic and Off-Topic Posts" <cctalk at classiccmp.org> Sent: Wednesday, February 04, 2015 3:32 PM Subject: Re: ISA ( PC ) 8" floppy disk CONTROLLER ???