Views
344
Replies
2
Status
Closed
In message ,
Hecate wrote:
Clear, and wrong. If you have two disks capable of 52 MB/s, on the same ribbon on an ATA100 controller, they can be accessed simultaneously at a total of about 90 MB/s. If the same two files are accessed on a single drive, in separate partitions, the total read speed will drop to about 5 MB/s, and total write speed about 25 MB/s (lazy writes save the day). Mixed access will be somewhere in-between.
On separate disks, *and* separate ribbons, the most you can hope to achieve is 104 MB/s. What is 90 MB/s closer to; 5 MB/s, or 104 MB/s?
For all intents and purposes, there is no significant bottleneck to having two hard disks on separate controllers, unless one of them is *very* stupid, or broken. CDROMs are another story; a CDROM can stop a hard disk on the same ribbon from being accessed for up to several seconds while it is spinning up.
—
<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy
Hecate wrote:
On Sun, 22 Jun 2003 03:15:50 GMT, wrote:
In message ,
Hecate wrote:
On Sat, 21 Jun 2003 03:01:28 GMT, wrote:
it may interest you to know that a myth is when something described doesn’t happen. The bottleneck in throughput clearly does, even if only to a minor extent. Now, I’m happy to agree that it is a minor extent, but it DOES still happen.
You’re changing your tune on me. Read your paragraph above that I quoted. Then read what you just wrote. 2 completely contradictory things.
You went from (paraphrased) "it’s not worth moving it to another drive if its on the same cable" to, "their is a minor loss for being on the same cable compared to a second cable".
Let me repeat it then as it seems I wasn’t clear to you – there is a minor loss from being on the same cable which makes it not worth moving to another drive on the same cable. Unless you can show major improvements from moving to another drive, and I have failed to see any on a modern, well-equipped machine, then there is little point. You will only get a measurable gain in performance from two unrelated disks. Clear enough?
Clear, and wrong. If you have two disks capable of 52 MB/s, on the same ribbon on an ATA100 controller, they can be accessed simultaneously at a total of about 90 MB/s. If the same two files are accessed on a single drive, in separate partitions, the total read speed will drop to about 5 MB/s, and total write speed about 25 MB/s (lazy writes save the day). Mixed access will be somewhere in-between.
On separate disks, *and* separate ribbons, the most you can hope to achieve is 104 MB/s. What is 90 MB/s closer to; 5 MB/s, or 104 MB/s?
For all intents and purposes, there is no significant bottleneck to having two hard disks on separate controllers, unless one of them is *very* stupid, or broken. CDROMs are another story; a CDROM can stop a hard disk on the same ribbon from being accessed for up to several seconds while it is spinning up.
—
<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy
<<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Master Retouching Hair
Learn how to rescue details, remove flyaways, add volume, and enhance the definition of hair in any photo. We break down every tool and technique in Photoshop to get picture-perfect hair, every time.
