wierdo124

This, too, shall pass.


y u confuz me?

__________________
Dig cars more than computers? Stop by CarForum.net!
E8400 C0 @ 4.3GHz on AIR
Quote by Syrillian: "Nono!...Kill person."
Windows 7||Fan Club The ß₤ứə Çřёώ The American Overclockers The Tux Club The Red Tide
Never Use Your Flash Drive the Same Again! Supercharge Vista!
[F@H]View Your Fahmon From Anywhere!
Member of Royal Navy GPU F@H Team

Kirmie

This still applies for one head per sector, it would just be access time per byte instead of per sector. We do not know what the size of these sectors will be on the new style of drives but the larger the sectors are, the more the advantage would go to the spinning plater method. This is because the maximum seek time per byte added would be increased by 2 per the back and forth method, but only by 1 per the spinning method. Its been a while but I think the BigO notation would be On2 for the rectangular drive and On for the spinning.

Also, I did not account for some variables. This is a comparison where the seek time from one byte to the next is equal in both methods. On a spining drive the outside sectors would spin faster so the next byte would be accessed quicker than on the inside, while the back and forth method would be the same through out. Also, the 3 head share computation I did for the back and forth method is based on an internal sector, the 2 head version would apply for an outer edge sector in that scenario.

The maximum speed a head can read will only be reachable on the outside of a disk as well, limiting the inside to lower. This does not apply to the rectangular method, but I can see that method bieng much more violent and not allowing for speeds equal to the outside of a spinning disk so those two factors will probably equal out.

In the end I still think that the spinning method would be the best, though either way if we still get faster drives I'll be happy. The head sharing per sector would also help to improve speeds though we could see either an increase in price due to extra heads, or the heads may just not fit that close together. What I can say for a fact though is that with one head per sector the seek time of each drive will be greatly influenced by sector size.

P.S. Sorry this is two days later, bachelor's party weekend.
Edit: Not my bachelor's party, a friend's out of towns. I have a year before mine since I'm not getting hitched till the fiance gets out of school.

__________________
Come join OCNs 48-hour Fold-a-Thon NOW!!

Have you ever tried simply turning off the TV, sitting down with your children, and hitting them?

From American Dad:
I touched her hand... her hand touched her boob. By the transitive property, I got some boob! Algebra's awesome!

EVE (Kirmie, Brian Fellow) | ogame.org (Univ 35, Unv 41) | Gladiatus.us (Serv 7) | NativeKingdoms (Epatlan) | DoD:S

DuckieHo

We don't know:
* head configuration.... if a sector could be read by multiple heads, a form of NCQ could optimize seeks
* what alogrithem will be implemented to optimize writes... the drive may spread writes out purposely to increase the chances that multiple heads can be used to read
* what kind of caching will be used.... the combination of the above saving to a large cache will help prevent locked seeks.

__________________
To answer most of your questions: (1) a fridge cannot cool a PC (2) 64-bit OS for over 3.4GB (3) If a PCIe card fits, it should work (4) Resolution, not screen size (5) If you have a question, it is not news (6) Report, not respond to Spam (7) Single-Rail/Non-Modular PSUs are not always better than Multi-Rail/Modular