Re: Apple Disk Server Beta. Need testers.

[Michael J. Mahon <mjmahon@aol.com>]

Egan Ford <datajerk@gmail.com> wrote:
> On Jan 5, 2:46 pm, David Schmidt <schmi...@my-deja.com> wrote:
>> On 1/5/2012 4:39 PM, Egan Ford wrote:
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>>> On Jan 5, 2:21 pm, "Michael J. Mahon"<mjma...@aol.com>  wrote:
>>>> David Schmidt wrote:
>>>>> Oh, one other thing I forgot to mention: one speedup that ADT client
>>>>> does (but not ADTPro - too lazy so far) is it starts the drive motor a
>>>>> little while before it is needed.  That avoids the spin-up delay once a
>>>>> full buffer is ready to write and lets you start writing much more
>>>>> quickly.  So, during inflate, if there's a point where you're
>>>>> "almost-done," one loop left to go or whatever, hit the motoron switch.
>> 
>>>> For reliable operation, the motor should be started about a second
>>>> before the disk is required to be up to speed.
>> 
>>>> -michael
>> 
>>>> NadaNet 3.1 for Apple II parallel computing!
>>>> Home page:  http://home.comcast.net/~mjmahon/
>> 
>>>> "The wastebasket is our most important design
>>>> tool--and it's seriously underused."
>> 
>>> What about 10 seconds before?  Is more OK?
>> 
>> Well, yeah, you could just leave the blasted thing running... but you'd
>> better be able to turn it back off again in case of failure.
> 
> 1.  Turn after download, but before inflate
> 2.  If inflate fails, turn off before error message
> 3.  If inflate ok, write starts, end of write should power down motor,
> right?
> 
> I am thinking that this will save about 5 seconds out of 3 minutes.
> It may not be worth it.

I thought that RWTS waited for several seconds after motor on if it
detected that the drive had stopped. So that would be several seconds times
the number of restarts. 

-michael - NadaNet 3.1 and AppleCrate II: http://home.comcast.net/~mjmahon