Re: Windows and black boxes
Posted by
mariss92705
on 2002-01-25 18:58:45 UTC
Hello Art and others,
Indeed the "devil is in the details". The details in this instance is
a balance scale where a "black box" is on side and a PC is on the
other side. The ideal is where a balance is struck. Here's how I am
looking at it:
(1) No Black Box:
Pros:
No hardware is always cheaper.
Maximum flexibility; no hardware constraints and formats.
Upgrades, improvements, etc. Can be done via download.
Complete internal control of product, not externally dependent.
No hardware related support necessary.
Cons:
Low step pulse frequency.
Non-linear and poor step pulse granularity.
Step pulse "jitter", i.e. poor spectral purity.
Step pulse timing "donkey work" burden placed on the PC.
Adding more axies makes the above even worse.
(2) Dumb Black Box generates Step pulses only:
Pros:
Even and very fine step pulse frequency resolution.
No granularity or jitter problems.
Reduced burden on the PC.
High step pulse frequencies easily available.
Additional axies are less of a burden.
Maximum flexibility because PC generates all motion routines.
Cons:
Hardware costs more than $0.
Though less, the PC burden is still significant.
Still have to handle the nuts and bolts of motion routines.
High communications bandwidth. Parallel port necessary.
(3) Smart Black Box takes care of motion routines:
Pros:
Need to only supply coordinates, lin or circ data, etc.
Much less real-time computational burden on PC.
Low bandwidth serial ASCII string communications.
Costs the same as the "Dumb Black Box"
Cons:
Inflexible "canned" motion control routines.
(4) Genius Black Box does everything! No PC needed!
Pros:/Cons:
This gets back to and becomes option (1) in a curiously circular way.
It is only meant only to point out there is a very tricky balance
between (3) and (4). That is where the "detail devil" is.
Seriously, in my opinion option (1) will forever have a problem with
the "Cons:" mentioned above. There is no hardware mechanism in a PC
to generate evenly distributed granularity for step pulses. This
requires external hardware to do it natively.
By offering to make the whole thing open with software and hardware,
I feel there will be people who use their ingenuity to find where the
proper balance is.
I will build the hardware cheaper than anyone else, which is what I
expect to get out of this, should it fly.
Mariss
Indeed the "devil is in the details". The details in this instance is
a balance scale where a "black box" is on side and a PC is on the
other side. The ideal is where a balance is struck. Here's how I am
looking at it:
(1) No Black Box:
Pros:
No hardware is always cheaper.
Maximum flexibility; no hardware constraints and formats.
Upgrades, improvements, etc. Can be done via download.
Complete internal control of product, not externally dependent.
No hardware related support necessary.
Cons:
Low step pulse frequency.
Non-linear and poor step pulse granularity.
Step pulse "jitter", i.e. poor spectral purity.
Step pulse timing "donkey work" burden placed on the PC.
Adding more axies makes the above even worse.
(2) Dumb Black Box generates Step pulses only:
Pros:
Even and very fine step pulse frequency resolution.
No granularity or jitter problems.
Reduced burden on the PC.
High step pulse frequencies easily available.
Additional axies are less of a burden.
Maximum flexibility because PC generates all motion routines.
Cons:
Hardware costs more than $0.
Though less, the PC burden is still significant.
Still have to handle the nuts and bolts of motion routines.
High communications bandwidth. Parallel port necessary.
(3) Smart Black Box takes care of motion routines:
Pros:
Need to only supply coordinates, lin or circ data, etc.
Much less real-time computational burden on PC.
Low bandwidth serial ASCII string communications.
Costs the same as the "Dumb Black Box"
Cons:
Inflexible "canned" motion control routines.
(4) Genius Black Box does everything! No PC needed!
Pros:/Cons:
This gets back to and becomes option (1) in a curiously circular way.
It is only meant only to point out there is a very tricky balance
between (3) and (4). That is where the "detail devil" is.
Seriously, in my opinion option (1) will forever have a problem with
the "Cons:" mentioned above. There is no hardware mechanism in a PC
to generate evenly distributed granularity for step pulses. This
requires external hardware to do it natively.
By offering to make the whole thing open with software and hardware,
I feel there will be people who use their ingenuity to find where the
proper balance is.
I will build the hardware cheaper than anyone else, which is what I
expect to get out of this, should it fly.
Mariss
--- In CAD_CAM_EDM_DRO@y..., "Art Fenerty" <fenerty@h...> wrote:
> Hi All:
>
> I just got back from vacation and have followed this thread with
great
> interest. The "black box" threads have been running for the last
year or
> more, but I notice the discussion getting more and more detail
oriented as
> it evolves over time. As the saying goes "the devil is in the
details"
> Art
> Master5 Software
> http://users.andara.com/~fenerty/master.htm
Discussion Thread
ron ginger
2002-01-25 04:48:51 UTC
Windows and black boxes
ccs@m...
2002-01-25 05:48:19 UTC
Re: [CAD_CAM_EDM_DRO] Windows and black boxes
mariss92705
2002-01-25 08:34:44 UTC
Re: Windows and black boxes
Bert Pirson
2002-01-25 10:32:08 UTC
Re: [CAD_CAM_EDM_DRO] Windows and black boxes
CL
2002-01-25 11:04:06 UTC
Re: [CAD_CAM_EDM_DRO] Windows and black boxes
j.guenther
2002-01-25 11:41:50 UTC
RE: [CAD_CAM_EDM_DRO] Windows and black boxes
hllrsr@c...
2002-01-25 14:13:36 UTC
Re: [CAD_CAM_EDM_DRO] Re: Windows and black boxes
Carol & Jerry Jankura
2002-01-25 14:18:44 UTC
RE: [CAD_CAM_EDM_DRO] Windows and black boxes
mariss92705
2002-01-25 14:38:59 UTC
Re: Windows and black boxes
stevenson_engineers
2002-01-25 15:11:10 UTC
Re: Windows and black boxes
Art Fenerty
2002-01-25 17:13:00 UTC
Re: [CAD_CAM_EDM_DRO] Re: Windows and black boxes
mariss92705
2002-01-25 18:58:45 UTC
Re: Windows and black boxes
Art Fenerty
2002-01-25 19:15:56 UTC
Re: [CAD_CAM_EDM_DRO] Re: Windows and black boxes
Russell Shaw
2002-01-25 20:02:45 UTC
Re: [CAD_CAM_EDM_DRO] Re: Windows and black boxes
Jon Elson
2002-01-25 22:30:07 UTC
Re: [CAD_CAM_EDM_DRO] Windows and black boxes
hllrsr@c...
2002-01-25 22:55:39 UTC
Re: [CAD_CAM_EDM_DRO] Re: Windows and black boxes
Alan Marconett KM6VV
2002-01-26 10:37:37 UTC
Re: [CAD_CAM_EDM_DRO] Windows and black boxes
ballendo
2002-01-27 05:07:42 UTC
Re: Windows and black boxes
ballendo
2002-01-27 05:30:42 UTC
Concatenate vs. CP vs. CVV was Re: Windows and black boxes