On 10/22/2010 3:29 AM, R.S. wrote:
Well... Since we talk about FBA drives, then reliationship to 3390 track size is simply irrelevant. I don't know what google searching results you've seen, but IMHO no consideration applies to FBA, especially because most OS'es does not address single sector, they use clusters of sectors, 4k is typical siz, but it's variable.
I didn't say that 3390 size has any relevance, only that the same considerations apply. Whatever physical medium is used for the FBA disk, each platter will hold multiple records. Due to (cheap) electronics, the gaps are of significant size, so a design with larger blocks will get more data per track. It would be possibly to design a drive specifically for FBA use, with better speed control, to shrink the gaps, but at horrendous expense. Since the manufacturers profit from selling DASD, there is limited incentive to reduce their income.
I first ran into this in the seventies, when my boss brought in a DEC-system 20 (at the time we had a 360/65 and two 360/165s). It used 512 byte blocks on something that looked like 3330 stand-alone drives. Had they used 1024, they would have gotten 10% more data per track.
BTW: blocksize 512 uses approx. 44% of the 3390 track, it's much more than 1%.
I tracked that down to a stupid omission in the code. Gerhard Postpischil Bradford, VT ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
