ðòïåëôù 


  áòèé÷ 


Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 

  óôáôøé 


  ðåòóïîáìøîïå 


  ðòïçòáííù 



ðéûéôå
ðéóøíá












     áòèé÷ :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Getting around the firewire problem was Re: filmscanners: Best film scanner, period!!!



Moreno writes:

> That's not true. How about plug and play? That's
> something that SCSI is not.

Strange--that's exactly how it was for all my SCSI devices (two scanners, a tape
drive, a disk drive, and a CD-R burner).

> And firewire, unlike SCSI, doesn't require your
> devices to be powered on at boot time.

Not a big issue for me, as I always turn everything on on those very rare
occasions when I boot, anyway.

> SCSI devices also require proper termination,
> which is one of the larger problem and support
> issues with SCSI devices.

I've never had a problem with it.

> Firewire devices do not have the cable length
> and bandwidth limitations that SCSI devices do.

SCSI has always provided all the bandwidth I need (scanners are not exactly
burning the wire with data), and cable lengths have not been a problem thus far.

> Windows 2000 has been out for more then a year
> and a half now.

When it's out for ten years, I'll think about it.

> Microsoft will officially drop NT support soon.

No, they won't.  Far too much of their corporate customer base is still running
NT, and will be for some time to come.

> Why would Nikon want to introduce new products
> that support obsolete operating systems and
> hardware?

Because it doesn't know any better.  Like I said, the camera and scanner
divisions are obviously quite separate, with the latter having learned nothing
from the former.

> Speed and cheap sound good to me.

Not when you examine the microprocessor architecture, and find out that the PPro
actually used its clock cycles more efficiently than subsequent processors based
on the simpler PII.  The PPro was much more of a native 32-bit machine, and had
a more efficient pipelining architecture, if I recall correctly.

> Back in the Pentium Pro days, Intel's profit
> margin was around 45%, not 70%.

And they still are.

> It's significantly less these days.

When it gets down to 5%, I'll worry about them.

> CPU's are very complex items to design and
> manufacture, and fabrication plants to manufacture
> the CPU's cost billions of dollars, and must
> be re-tooled every twelve to 18 months.

That is a problem for the manufacturer, not me.  And nothing requires retooling
every 18 months, except a desire on the manufacturer's part to constantly bring
newer processors to market, whether they are needed or not (the 45% margins have
to be justified somehow, I guess).

> I think a couple of hundred for a fast Intel
> or AMD CPU is a bargain.

As compared to what?

Imagine what an incredible deal the OS is, then, since it only costs $30, and
required a lot more design work than the microprocessor.

> Even a fast CPU is rarely more than 20-25% of
> the system price, not more than half as you allege.

In my case, it was half the cost of the system.  The monitor was about 25% of
the cost.

> Does it really take you two months to reconfigure
> a system?

A production system?  Yes!  Try it sometime.

> Computer technology is evolving at a much faster
> pace than your washing machine.

It is moving laterally more than it is moving forward, and much of the evolution
and especially the change is unwarranted.

> Your Pentium Pro NT box is probably still as
> fast as the day you bought it.

Yes, which is why I don't want to upgrade it.

> And it will probably run all the software and
> hardware of it's day, and do it well.

And so it does.

> It will admirably run any filmscanners of that
> era too.

Those that were sold then, yes.  But the LS-4000 is virtually identical to those
scanners, and yet it will _not_ run on my system.  There is no magically
advanced technology in the LS-4000; it's only an incremental improvement.  There
is nothing about it that requires obsoleting the previous interface.

> As long as it's doing everything you need it to do,
> there's probably not a lot of incentive for you to upgrade.

I agree, and I have no plans to upgrade.

> "Professional" computer users upgrade far more
> often than "less casual users".

No, they do not.  The largest and most critical production systems also tend to
be running the oldest hardware and software.

> I know this for a fact, as I earn a fair chunk of
> my income from "professional" computer users.

I suspect you earn your income from desktop business users, which are not the
same thing.  Production systems are different.

> In the context of this conversation, photography and
> scanning, I find this market segment to be the most likely
> to upgrade to new technology as soon as it's available.
> If they don't upgrade often, they're not likely to
> stay in business.

Not true.  The more critical a production system is to business, the less likely
it is to be even touched, much less upgraded.  I know of major multinational
companies that are still running software from 1968, because it works and
because they cannot afford to do anything that might stop it from working even
briefly (as in minutes or hours).

> Nikon did support NT at one time. Now that NT is
> no longer being sold, they've moved on to offer
> products for current operating systems &
> technology.

The mistake here is to assume that because an OS is not sold, it isn't used.
This naively assumes that everyone buying a scanner is also buying a brand-new
computer to go with it, but that isn't the case, particularly for professional
users.

> They don't seem to be having a problem selling
> their scanners.

The market is not saturated as yet.

> It's really easy to find a firewire adapter.

I'm sure it is.  Now what about the software?






 




Copyright © Lexa Software, 1996-2009.