#New Amiga from SGI!
77 messages in this thread
Something wonderful has happened this last week folks. Something so
wonderful that I doubt most of the major media will pick up it other than
as a token event. See how jaded I feel about the computer press?
Folks, this last week Silicon Graphics introduced what I believe to be the
high-end Amiga, high-end PC, and certainly high-end Macintosh killer if
you are into desktop video and multimedia.
This box is called the Indy. It has a 64 bit arcitecture, 100Mhz CPU
based on the MIPS processor, 8-Bit Graphics (similiar to AGA), upgradeable
to 24 bit. 16MB of RAM (uses SIMM modules), SCSI-2, Stereo In, Stereo
Out, Video In from three sources, ability to record video directly to disk
and play it back at 30fps in S-Video, NTSC, or even HDTV with an optional
board for output. (It will display these videos in real time on the
monitor w/o output video board). It comes standard with a SGI's UNIX
product with a GUI that claims that everything can be done from the GUI,
including nasty things like configuring a TCP/IP network. Bundled
software includes a sound editor, ShowMan (SCALA-like product) and other
operating system tools. The system also comes complete with a digital
video camera that can be used for network teleconferencing. The video
that is brought in can be digitally edited as well.
How much? How does $4995 List strike you? You will have to add a disk
drive, and 16MB of RAM may be a bit on the thin side. Just to give you an
idea of what a 1.2GB HP SCSI-2 (3.5 Inch) drive goes for, it is less than
$2100. SGI recmmends a minimum of 384MB.
Who is developing for this box so far? ASDG and the Blue Ribbon
Soundworks. I suspect to see more high-end Amiga software companies leap
to this machine. The one downside the SGI box has is the cost of the
current line of software. However, as this machine reaches the masses the
cost of software is likely to drop like a rock.
Amiga Developer–you don't gotta wait for AAA or Commodore to get their
collective act together. If you have a graphics product, the Indy needs
you!
Friends, SGI just got real serious.
Are we, the loyal Amiga dealer going to pick it up? Uh, will Bill Clinton
raise taxes? 🙂
Only Amiga makes it possible indeed.
-mark=
Otto Pilot Engaged..
Hi Mark. The Indy certainly does seem VERY cool. Contrary to Apple and
Commodore, SGI's stock has doubled in the past year. I believe SGI is the most
innovative, visionary and strategically coordinated computer company out there
today. They will certainly be attacked on an ongoing basis by IBM (RS/6000
etc), Sun, HP, DEC, etc., but as long as they can keep their innovative edge,
they stand to become a huge player in the years ahead. They're already bigger
than Commodore in revenue, although their installed base in tiny (less than
50,000 machines, I believe).
Regards,
Paul.
SGI has also been going through well publicized bouts of 'downsizing' woith
layoffs, pay and benefit cuts, etc. They are probably betting the company on
their ability to get into the consumer level market, after watching their reign
in high-end imaging stations eroded by pesky pcs on the low-end. Does this
sound at all like apple? will they manage to survive the transition to small
margin/large volume products?
There's been mention of this machine in some of the other forums. Very
powerful looking system for sure!
I still have a couple of questions that I haven't been able to find an
answer to… First, that's a diskless station, right? I presume that
means it boots from a network server or another SGI box. Anyway, does the
price include the OS, or is that "extra"? (I presume it's included).
Second, what kind (if any) of development system comes with it? Does it
include Motif?
Someone in another forum mentioned that the 24-bit upgrade with a disk is
something like $4000. Ouch.
If the graphics hardware engine that makes the big SGI boxes fly is in
Indy in some form, it'll take more than a quickie port to get some of the
packages we use now to take advantage of it. But when they do take
advantage of it….phew! That thing will just scream.
I can't wait to see one of those things!
Steve
Steve,
ASDG had a booth in the SGI section at NAB. I didn't have time to go in,
but I assume they were shoeing some form of ADPro/Morph plus (test or
shipping) for SGI/Unix.
wmc – via Autopilot!
If I'm not mistaken, ASDG will be at Siggraph '93 as well.
Steve
Steve,
The station is diskless at $4995, however it uses SCSI-2 drives and those
prices keeping falling. You can get an HP 1.2GB drive for right at $2k.
The OS comes standard along with many software tools built in.
I have no idea on the development system, I will find that out though.
-mark=
Otto Pilot Engaged..
It's possible to use SCSI-1 drives, though…isn't it?
The reason I asked about the software development tools is that they do
sell some systems without a development environment (no compiler…and
possibly no header files), and that might be a major expense. Sure hope
it's included. Let me know what you find out!
Steve
I would guess that you could use SCSI-1 with it, not sure about that
though.
I will let you know on the developers kit.
-mark=
Otto Pilot Engaged..
How can it be used without a disk, though? If a user adds their own hard disk,
how do they get SGI's UNIX onto it?
Inquiring minds want to know!
Paul.
Paul,
Good question, good question indeed! Perhaps you are locked into
purchasing the disk from SGI. I pray not. 🙂
I'll let you know.
-mark=
Otto Pilot Engaged..
Mark,
I'm very intereseted in this machine. If you are planning to be a dealer
in it, I wouldn't mind getting some printed material about it form you.
Just two points:
To be completely fair, what would the price of the Indy be with top-notch
modeller/animation system (Ligtwave/Imagin/Caligari-like, 24 bit of
course), able to handle CG, Switching, and digital transition duties in an
A/B Roll mode (a la Toaster with Amilink) and audio frame accurate editing
a la Studio 16/Amilink A follow V and split editing? I realize an oranges
to roganges comparison here is not possible as the Amiga will NEVER be
able to do the stuff I saw Indigo do with WaveFront at NAB (bandwidth
limit on motorola chips and Amiga bus). Which brings me to my second
point.
Software will always be m ore expensive on SGI since it is designed to
network right out of the box (read multi-user license assumed) and is
targetted at the established, professional users rather than the Kool,
Vidoe Yippees of NewTek. But there is a symbiotic relationship that could
be nice for both machines. I could envision several Amigas networked to
and feeding and SGI machine. The AMigas are used for storyborad, preps,
model building, scene building, maybe even lo-res approval rendering and
some off-line editting. THen the SGI is used for the online, HDTV, high
end formats and non-linear editing as well as final animation rendering an
lay-off to what have you media. What is the key ingredient that will make
this work? Developers that provide products for both platforms. And, as
you pointed out, that is already happening.
Now can CBM take advantage of this or not….
wmc – via Autopilot!
Wayne,
I pointed this out in an earlier message. Software is going to be the
pricey side of the Indy. However, if this thing hits like I think it
will, the software publishers who specialize in desktop video and
multimedia will have a major void to fill.
Things will change really quickly in the SGI market.
-mark=
Otto Pilot Engaged..
The machine sounds grand, it is indeed generating the kind of excitement that
the Amiga did when it first hit the scene.
BTW: our SGI Iris guys tell me that our SGI dealer will be in the office next
week to discuss some hardware upgrades for the Iris machines. I plan to try and
get a word with him and see if I can get some info and real pricing on a Indy.
Hmmmm….come to think of it, there's an SGI dealer around here
someplace…I better check. I'm thinking pretty hard about getting one of
those things (the Indy).
Steve
We're lining up with SGI, and hopefully in a couple of weeks will have
product literature available. I have some preliminary spec sheets on the
box now, in a word… awesome.
-mark=
Otto Pilot Engaged..
Mark:
If the machine lists for 4995.00 what do you think it's street price will
be? Did they say when the machine would be available?
// -Mike Schiller
\X/ AMIGA! On AutoPilot in sunny Fort Lauderdale
You won't see heavy discounting on the Indy except via corporate sales in
volume. This isn't exactly a commodity computer. –Redmond
For those of you interested in Indy, be advised that Desktop Video World has
been in close touch with SGI for a while. The machine ships in September, and
our next issue ships in September…
Keep your eyes peeled for that issue!
Lou Wallace Editor-In-Chief, DTVW
What are your first impressions of the machine Lou?
I swear, though I haven't seen one yet, this is the first time in YEARS
that I have actually been excited about a machine introduction. It is
almost like 1985 all over again.
-mark=
Otto Pilot Engaged..
I'll leave my comments in secret until the next issue comes out…
I do suggest you watch for the next Desktop Video World.
Lou
Mike, Mark and everybody else,
The new SGI Indy is not quite what the press releases seem to describe, and
the real 'street price' is in actuality 8 to 10 times the ballyhooed $4995!
We have been intensely evaluating the Indigo and Indigo-2 SGI machines and the
Maui Research and Technology Center has just committed to buying several
(probably 6) Indigo-2s. During this evaluation we became familiar with the Indy
specification.
The machine is very powerful – compared to a 68040 based Amiga or Mac or a
486. Its R4000 RISC CPU places it well in front of the Pentium-based machines
running Unix or Windows NT.
But at $5000, the machine can't get out of its own way. Like the early Amigas
it doesen't have anywhere near enough RAM to function. For the kind of programs
that one buys such a machine, 32 megs is not a lot and 64 to 128 is a nice
comfortable figure. And RAM for the Indy costs $90-100 per meg! Other
significant 'peripherals' such as monitors, drives, add-in boards, and the like
are very expensive.
And then there's the software. Whew! 3D modeling and animation packages start
at $8000. Right – eight _thousand_. A page layout program roughly equivalent
to Quark Express is $2900. Photoshop is $1100. Adobe Illustrator is about the
same.
To give you some perspective, my graphics prepress workstations are Macs. A
well equipped 800 Quadra with 96 megs of Ram, a 24 bit 15" monitor, about 500
megs of hard drives, running Quark, Photoshop, and Illustrator weighs in at
about $7000. It is a good performer, everything works well and I get _gobs_ of
work done on it. A similarly equipped Indy costs $30,000! And (parodoxically)
the software is not all that dependable. Bugginess in software and hardware is
inversely proportional to its cost.
Of course if you need the power of the R4000 processor, then the SGI is the
way to go. No Mac or PC can touch it in terms of sheer processing power. And
its operating system and proprietary graphic software are the best. But this
machine
[ MORE ]
Richard,
Indy comes with monitor, with color digital camera, with genlock, with
DSP, with 16M ram — at the $4995 price.
Yes, Indigo software and peripherals are extremely pricey at this point in
time. However, SGI is attempting to enter the consumer market with Indy.
It's obvious that a $5k machine will not support an $8k software package
(in any kind of quantity). I believe you'll see software/peripheral
pricing take a nose dive as more Indys are in consumer's hands. I believe
you'll also see a wealth of new companies jumping into the Indigo world
with new products are very competitive pricing.
-sja
Steve,
I agree that competition will bring the prices down. But I stand by my $30,000
figure to acheive the level of performance that I indicated.
Rich
Rich:
I would think that while your $30,000 figure is probably correct NOW, and
possibly for the near future, chances are that fairly soon it would go
down probably as low as $15,000 or lower. I have this really bad habit
which I'm trying to break, which is that I like to buy these hi tech toys
when the first come out, and I've seen many things that I've bought drop
in price almost 10 fold sometimes! While this isn't SO bad when talking
about buying something for $1400.00 and seeing it sell for $499.00 about a
year later (this actually happened to me with the 040 accelerator in my
A3000) I think I'd cry if I bought something for $30,000.00 and a year
later the same thing was selling for $5,000.00!
BTW, while talking about things that cost $30k, has the price of
imagesetters dropped much? Last time I looked into it at all, used
Linotronics were around $25k or so for the full setup (imagesetter, rip,
processor, etc). The printer that we use here was thinking of getting an
imagesetter a while ago, and they ended up getting one of the so called
1000dpi laser printers, and I'll tell you what, yes it's output is
somewhat nicer than the 300dpi output of my laser printer, but not THAT
much, and nowhere near the quality of an imagesetter.
// -Mike Schiller
\X/ AMIGA! On AutoPilot in sunny Fort Lauderdale
Oh gosh, only $15,000 now I think that is a huge money saver! NOT…
-Jonathan
Mike,
I agree that the price of machines like the SGI will come down in price. I
hope you are right about the 2 to 1 ratio. But I don't think that the
machine/software combo that I priced for $30K is going to get much less than
25K any time soon.
Imagesetters seem to have gone _up_ in price. In the last year, the technology
has accelerated enormously, rendering machines only a few years old obsolete.
My old 4300P Varityper (2400 dpi, Atlas speed RIP, 12 megs of RAM, two 140 meg
hard drives) cost $60,000 just a few years ago. It's street price now is below
$6000! There are a _lot_ of older imagesetters on the used market now. They are
hopelessly outclassed by the newer machines which are faster, have PostScript
level 2, and have screening technologies which make the older machines'
(halftone color separation) output look like junk.
We are taking delivery of an Emerald RIP machine next week which is state of
the art. Fortunately, we got a killer deal on it, but it still cost dearly.
Until about 2 years ago, the conventional wisdom in purchasing this kind of
'durable equipment' was that your market had to be strong enough so that you
could pay for the equipment in 2 years and realize profit for 3 years. At the
end of that 5 year cycle, the technolgy was supposed to catch up, requiring an
upgrading of the machinery. If you could not pay for the equipment, and realize
a profit, you were in deep trouble. Today, the five year cycle is more like 3
years. That means that if a company can't pay off the cost of equipment in a
very short time, it can't stay in business.
Rich
Richard:
But correct me if I'm wrong, but a VT4300P would do the same job today
that it did a few years ago, right? Just because there are newer/better
machines available today, couldn't somebody that had a lower volume, less
money to spend buy a used VT4300P (Or L300, etc) and use it just fine, and
still have output MUCH better than any 300DPI laser? (considering that
just a few years ago 300DPI lasers were in the pricerange that used
imagesetters are now in).
// -Mike Schiller
\X/ AMIGA! On AutoPilot in sunny Fort Lauderdale
Mike,
Sure. The answer to your question is yes. But it's a qualified yes for a
number of reasons.
My 4300P is just as good right now as it was when it was new. As a matter of
fact, it's better. I've spent long hours calculating screen angles for it and I
can produce beautiful color separations which result in very high quality work
coming off the press. In fact, Varityper will not acknowlege that this machine
can produce the quality that I'm getting from it. I would assign a figure of
merit to my present color work at about 86-88 with something like Architectural
Digest or World Wild Life or National Geographic clocking in at a consistent
97-100.
That alone is the problem. I can't compete in the color market if I can't
produce the highest quality. Because my customers can _get_ that quality as the
result of a phone call and the price of a Fed Ex mailer.
Add to that the fact that a 9 x 12 full page four color job takes a good four
hours to image (150 line halftones, some graphics with blends and other
complicated elements, type matter, etc). That's half a day just to output that
one page. If there's anything at all wrong with the output (the designers
fault, a typo that got missed in proofing, an inexplicable PostScript error
message), at least another half day is shot. That's a full day gone, just for
that single page. And, during that time, _no_ other work can come off the
machine.
In other words, the 4300P is not a production machine in today's printing
environment, and it's quality, while very high simply can't compete in an arena
where every customer wants rock bottom prices and examines the printed sheet as
though his job is a Cartier jewelry ad and he has already paid me millions in
previous print jobs.
Now, if as you asked, can't these machines be pressed into service to replace
300 dpi laser printers, since used imagesetters are now at the same price point
as laser printers were when they first came out ($4000-7500), again the answer
depends on the user's ability to tolerate a number of realities.
[ MORE ]
Your experience with typesetting equipment prices and the "business case" for
return on investment is exactly the same as that for computer equipment.
Hobbyists will look at that "loaded" Indy price you quoted of $30K and say "in
my dreams only", but those who need to get real work done, may easily be able
to justify the price. While Toasters can do 3D rendering, for example, they're
very slow (even with GVP's fastest accelerator), compared to workstations. A
client will pay a lot for 3D, and the time savings an SGI box can provide can
often easily pay back the $30K price.
The $3K loaded PC's that are being bought are going to people upgrading from
older 286/386 class machines and to new users who've never been in computing.
$10K and up machines are going to those with specialized applications where the
tool is essential, just like a lathe in a machine tool shop. I know I don't
need to tell you this, but I was just trying to put the "high" price of some
machines into perspective for all here. As they say, it's all relative.
Paul.
Paul,
You're absolutely right. If I have a healthy market for the kind of work that
an SGI workstation can produce, then the price of the equipment is _almost_
irrelevant. In fact, trying to get by with less expensive solutions is the
worst decision to make, (and actually a much more expensive solution in the
end) because the lower priced machine often can't handle the work load. (as in
my messages about the speed of my old 4300P).
But, can a $30k Indy outrun 5 or more toasters each doing a segment of an
animation? Five toasters, that's about what $20 grand? What about those
SAGE devices? Aren't they in the "workstation" speed realm?
-Steve
[ continuation ]
First, an imagesetter is _not_ a laser printer, and cannot replace it as a
quick proofing device. The low resolution of the laser printer is actually an
advantage as a device to produce multiple, progressive working proofs of a
design in progress. The higher resolution of the imagesetter often results in a
job bombing that would have imaged on the lower resolution device (on which the
potential for the problem at high resolution could have been discovered). There
is also a _significant_ learning curve in aquiring the ability to operate these
machines.
Second, no hard copy actually comes out of an imagesetter. It exposes
photographic media (film or paper) which then has to be sent through a
processor. A decent rapid access chemistry processor or a litho chemistry
processor runs about $7000. Although these machines are also available used.
The true cost of a page of output is close to $5 in quantity.
Third, the imagesetter is a very complex machine that depends on the
simultaneous functioning of fantastically accurate components. Its electronic
innards are the equivalent of a mini computer. Every board inside is a
proprietary thing that bears no similarity with plug in boards on an Amiga or
Mac other than the fact that both have a lot of chips on them. Only the
manufacturer can or will repair a malfunctioning board. Go try to get a circuit
diagram! The optical elements consist of a laser beam and a spinning octogonal
mirror that are aligned to an accuracy measured in fractions of microns. The
film drive mechanism has to drive a 12 inch wide roll of film past the laser
output port at about half an inch a second so smoothly that there is not the
slightest distortion down to a measurement of less than 1/2400 th of an inch.
If one has obtained one of these machines for the 'street price' of $4000,
that means that it has come from a private sale, not from the manufacturer.
(Official factory reconditioned price of a 4300P is about $26,000). That of
course means, no
[ MORE ]
[ continuation ]
warranty. If repairs are needed (very likely in any piece of equipment more
than a few years old) it is very possible that the cost of repairs will exceed
the purchase price.
So . . . if any one wants to get a used imagesetter, and dabble with it in a
non-productivity environment . . . have at it! 8)
Rich
You are correct, Steve. Indy sets a price point that many (for example) Amiga
third parties would be happy to follow. Amiga software brought over to the Indy
and sold at reasonable prices might fair pretty well. Gee, maybe one could
actually port AMIGADOS itself! There's a novel concept.
Anyway: a relatively low cost platform will inspire lower priced good software
if properly marketed. –Redmond
Redmond:
I don't think a ported AmigaDOS would go over too well. I don't know about
the SGI market, but if it's anything like the PC market, AmigaDOS would
never make it as it's too good, and too simple to setup. After all, if you
make an OS that's so simple to setup that almost anyone can do it, where
are all the consultants that make their living setting up computers for
other people gonna go?
It seems that Unix is gaining lots of users on PC systems, and now
NextStep has been ported to the 486, and it looks like the next generation
of machines will be able to run either windows or the Mac finder
interchangably, so who knows, maybe there IS a market for AmigaDOS in the
PC world. It sure would be cool to be able to run MS-DOS, AmigaDOS, Unix,
MacOS (does the Mac OS actually have a name?), all on the same machine,
all in a native mode (isn't this possible using the PowerPC?).
I think the next few years will be some very interesting ones for sure.
10 years from now: "Can you believe they actually used to use K to measure
the amount of RAM in their computers?"
// -Mike Schiller
\X/ AMIGA! On AutoPilot in sunny Fort Lauderdale
Mike: the only "official" name I've ever seen was Mac System 6.x or 7.x. Apple
doesn't call it MACOS. Maybe it means something off-color in a foreign
language?
😎 –Redmond
AmigaDOs isn't needed forIndy. It comes with Inigo Magic, a very nice
multitasking interface that effectively hides Unix…
Lou: if you had AmigaDOS on the Indy you would not only have a GOOD operating
system, but you would also have a SMALL operating system. Then the basic Indy
would be immediately useful at the 5000 bux level. AmigaDOS runs in a few megs
and a modest amount of HD space. Unix on the other hand…
The GUI ain't the issue. –Redmond
Maybe that's why companies like ASDG have (wisely) ported their MorphPlus to
SGI, and are charging significantly more for it than they do on the Amiga.
Looks like a good business opportunity.
Paul: economics can't be ignored in business (if you want to be in business
that is). No matter how loyal we are to the Amiga, one cannot ignore the fact
that Commodore management has utterly failed to make a market within which
third parties can grow. The installed base of the Amiga is about 1/3 that of
the Mac yet the third party marketplace is hardly one-thirtieth as lucrative.
If the Amiga had been vigorously marketed (directly) by Commodore in its
natural niche in the USA (the "prosumer" video market) there would be at least
500,000 machines in the USA in the hands of people with enough money to really
support a vigorous third party aftermarket. Once that was established, all the
other good elements of a computer market would have come together and we would
easily be rivaling the installed base of the Mac *in North America*. Commodore
had a clear shot at this market *UNOPPOSED* for about five years. They did
nothing with it except nibble at it. Scattershot ads. No real support. They
dropped the ball right on their own foot –and it was a bowling ball! This is
the result of "quarterly report" management and a complete failure of vision of
upper management.
Commodore engineering and sofware literally worked MIRACLES with a literal
handful of people to create what I believe to be the best small operating
systems and machine architecture in the world to date. Commodore management
took these diamonds and hid them in the fog of their "thinking" all these
years. Sad ain't it? And if CBM croaks, it won't be the upper management that
gets hurt. They'll already have made their getaway money long before the
sheriff puts the lock on the door. The people who'll get hurt are the people
who actually created the Amiga and AmigaDOS we know today. Plus we third
parties and all of us users and lovers of the Amiga.
I long for some intelligent company to buy the Amiga and do right by it. Maybe
if we all concentrate on that thought, we can bend the engine of reality by
sheer will, hmm? –Redmond
No arguments there Redmond, (Unfortunately.)
Bob Comer — Cruising the nets on AutoPilot from Cheyenne, Wyoming
Paul (and all),
I've had an Amiga since the very early 1000's. I've wondered why alot of
times. Not why I own the machine but, why there aren't more people who do!
It seems as though I'm seeing here alot of what I've seen with the banking
industry. NOT, to compare the two directly in these times but, one
particular thing that has got on my nerves after hiring on at Ranier Bank
and loving it, then getting bought by Security Pacific and hating it then
learning to enjoy it, then being bought by Bank of America and being
transferred to Seafirst. I'm only saying all this and using all the big
names 'cause it happened! And all the people at the tops of these
organizations had thier 'GOLDEN PARACHUTES' worth millions and millions.
Once the company went under or was sold they were set for life! It doesn't
seem to me that they could care about anyone at my level (or shareholders,
or customer base or the original and existing designers/employees, in
Amiga's case). Although, it seems that Commodore is a bit smaller in
scale, hey, if I had my 2 or 3 Mill what the heck?
Just 'cause, ignore, if you will.
Robb
Redmond,
I have one slight argument with your points. I do not find it
suprising at all that the Amiga market is but 1/3rd that of the Mac while
the third party is hardly one-thirtieth as lucrative. If any industry
exhibits significant economies of scale, it is the computer market.
Nothing we can do, nor Commodore in a very real sense, will change those
facts. Only if the installed base increases significantly, and I do not
mean by 50% but on the close order of 300%, will significant economies of
scale occur.
The barriers to entry are high and I do not think that C= has the
management to pull off a marketing coup. They seem to have a significant
presence in the video market, at least to judge from the number of Amigas
I've spotted in video studios in news reports, but this is a very small
market niche. Given the state of the market, C= doesn't have a choice
chance in hell of penetrating the consumer market unless it is willing to
play with the big boys with their significantly tighter profit margins.
They can't do that, period. If they try, they will be in bankruptcy
proceedings so fast it'd make your head spin as they can't afford the
advertising budget required. As evidence, I offer the campaign by Intel.
There is no way C= could afford a comparable campaign, and this is in
comparison to a chip company, which is but a small part of C=, not to a
full-fledged OEM.
In comparison to an OEM, C= also fails the test. There is no way they
can afford an advertising budget similiar to Apple or IBM, period (again).
Face it, we are stuck with a niche machine with a evangelistic following
that will never break into the "mainstream." Wishing won't make a
difference nor will advertising by a company that can not afford the
effort.
As for myself, I'm going to sit here enjoying a five year old machine
that is easily comparable to a 20 MHz '486 and, for now, cheaper.
Brian
Brian,
I like your post and agree.
Let me add one thing though. Let us just suppose that Commoodre through
some miracle could afford the ad campaign. What would be the point of the
campaign if there were not distribution points where a customer could see
this wonderful machine?
There are simply not enough distribution points for the Amiga now to make
it worth the expense.
-mark=
Otto Pilot Engaged..
You are absolutely correct. Economies of scale have made the PC industry what
it is. This is most evident in the case of hardware development (which of
course, GVP is keenly involved in). Hardware development is EXPENSIVE.
Contributing factors are board prototype runs, testing equipment, regulatory
issues (FCC, UL, etc.), and the requirement of thousands of components.
Compare this with software! Anyway, we at GVP are somewhat lucky in being the
biggest Amiga vendor, but we could probably bring SIGNIFICANTLY better prices
to the market if it were SIGNIFICANTLY bigger.
I don't foresee this situation changing, however. We're not that
psyched about doing add-ons for the game machine!
Paul.
Brian: I agree with some of your points. But I can tell you as someone who has
been engaged in advertising and marketing for a long time (25yrs) that CBM
could have EASILY afforded a "signficant" print campaign targeted at the
"prosumer" (not CONsumer) video market. They wasted almost 15 million on a
pointless television campaign that got them almost zero sales (they sold 55,000
Amiga 500's the year they did "tv"). For the same amount of money they could
have bought several YEARS of targeted print and direct mail selling and gotten
a huge US installed base of folk who were actually using their machines above
the hobbyist level. This would have caused a flood of software and peripherals
to be created which would have further enriched the installed base. Sure
Commondore has done "ok" in the video market –and they didn't even try! Think
of how well they could have done with even reasonably competent marketing?
There are 20+ million video camera owners in the USA. About 15% of them have
"high end" vcr's and cameras. Every one of those folk are candidates for an
Amiga with the right software bundle (and a simple IR interface). With a decent
effort and a few million a year in real advertising and marketing, CBM could
have cemented the Amiga into that prosumer market segment as the general
purpose editing/digitizing/titling/effects machine. A hunderd grand a month in
print ads in selected video publications (repeatedly, not just once or twice)
and a half million to a million in direct mail annually. It would have taken a
few years to "stick" but it would have stuck and paid off bigtime. Commodore
was/is unwilling to think about more than one-quarter at a time in their
"marketing". When they did spend significant money, they did exactly the WRONG
thing –they tried to buy the mass market with a lame campaign that cost them
too much and did nothing except confuse people. –Redmond
Redmond,
This forum has seen a lot of messages about CBM's unforgivable management of
this marvelous machine called the Amiga. Your message is about as accurate and
relevant as any I have seen here on the subject. The Amiga, sadly is a
non-entity in the business world in any framework that one wishes to place it.
It is a minor curiosity with an interesting video add in board (Toaster),
nothing more and this situation will never change.
Still mourning the NeXT machine's passing, I think the Amiga will follow.
Rich
While it will never be a busines machine, I don't think it's just a minor
curosity. The Indy is a really nice machine, but I doubt it will make
changes in regard to the Amiga, I expect it to effect the Quadra market
more. The picture isn't good, but I haven't written off the Amiga just
yet.
Richard -> via the Amiga 1000 AutoPilot
Richard,
I've re-read my message. And I retract some of it. The Amiga will probably
_not_ suffer the fate of the NeXT machine. It will remain in existence for a
long time. Look at the C-64! CBM will be able to maintain a niche market as it
always has. Unfortunately, and sadly, the CBM-Amiga will never become any more
significant in the computing world than it now is.
Rich
Rich,
No it's highly doubtful the Amiga will ever break into the mainstream. I
am wondering how CBM and the Amiga will fit in after the chip and OS
changes that are on the horizon.
Richard -> via the Amiga 1000 AutoPilot
Of course software for UPmarket systems like the Indigo has to be more
expensive than software for cheaper (and thus, widely-spread) systems like
the Amiga or Mac or PC. There's 100 Miliion PCs out there (oh, well…),
10 Million Apples and 5 Million Amigas and maybe just as many Ataris.
Spread the cost to develop a software for a system among the number of
sales it's likely to get and take into consideration the price of the
system itself (people with 15,000$ systems are usually willing to pay more
than people with 600$ systems) and there you are…
Funny that the prices for PC 'standard software' are still so high – oh,
well, as long as people are actually paying the asked prices, who cares?
(Superbase 4 Pro for the Amiga (english version) sells for about 175$ in
Germany, the German version for about 235$, and the PC version for about
765$ which is 3 to 4 times the price of the Amiga version, and the PC
version has SOLD over one million(!) copies there.)
// Chris coming in from Germany
\X/ on AutoPilot 1.051……….
I recently bought Superbase 2.0 for Windows at CompUSA for only $149. That was
as an upgrade to my Amiga version. The competitive upgrade was $199. Although
PC software still posts a very high list price, you can find amazing values out
there that make Amiga prices look inflated! –Rick Rodriguez
I think you'll find that Superbase is now having MAJOR problems against
Microsoft's Access for Windows, Microsoft's FoxPro for Windows, and Borland's
Paradox for Windows. Lotus is about to jump in the pool too, having recently
purchased the company that makes the award winning Approach database for
Windows.
PC software is also pretty cheap here in the U.S. Many of the big name
packages can be had at their "upgrade" prices without any proof of owning a
prior version (they might ask if you've HEARD of the package before <g>!). I
paid $84 for Access, and a friend got Lotus AmiPro (killer wordprocessor) and
Lotus Improv spreadsheet for about $90 each.
I don't know what kind of cause and effect relationship there is due to the
MASSIVE amounts of piracy that goes on in Europe (far worse than the U.S,
according to SPA).
Paul.
Paul,
I don't know why, but here in Europe (or at least in Germany) we don't get
the special (competitive) 'upgrade' ;v) offers that you do in U.S. or at
least not at your prices. That MAY be to do with the german regulations
regarding 'dumping prices', I don't know. Maybe that's why so much
software is pirated over here. In US, the companies think, ok, we sell
this new software cheap but we'll have the customers by the b*lls for
updates/upgrades later. I think that's where the software companies make
their money nowadays. Once you have a software package you rather stick
with it than change except the other is REALLY much better AND you get it
at a low price (= reason for 'competitive upgrades'). If its data
compatible to your old software the better…
// Chris coming in from Germany
\X/ on AutoPilot 1.051……….
Steve,
Absolutely my take as well. In fact, I suspect that with this machine
introduction that many of the top notch Amiga developers will leap to this
platform if they product graphics, or desktop-video related products.
I am not sure SGI realizes what they have done.
-mark=
Otto Pilot Engaged..
Indy doesn't stand a snowball's chance in hell. The consumer computer
market ends at $2300, not $5000. Software will never drop more than 12% in
price from todays level. 128 megs of RAM is required.
Only a few thousand Indy's will sell, average configuration will be
$18,000 for a starters kit.
The big Software companies (SGI guys) are damn good at what they do and
they know of what they speak.
This comes from them.
In two years you'll see Indy just getting warmed up with needed support
and developement. This will absolutely not be a storm of excitement. Just
barely a trickle of interest at it's highest.
You'll see, probably closer than you wished.
–Paul < Motorola Inside >
Paul,
First off, I don't think SGI has built a 'consumer' home computer, nor
have they built a killer box for 90% of business.
What they have done though, is build a desktop video machine that clearly
destroys the high-end Amiga and Macintosh market. You may not realize
this, but a nicely layed out toaster system is usually in the 10k range.
I have customers who WANT more capability than what the Amiga can deliver.
I intend to continue to carry Amiga and Video Toaster as they are quite
viable products. I see the Indy as my high-end solution to users who need
more than what the current Video Toaster has to offer. I also see the
Indy as a system that preserves system value and as a box that will allow
a reseller to make a profit, unlike the toaster business today.
My excitement lies in the fact that SGI has dropped into the high-end of
our desktop video market and basically have dared the other hardware
manufacturers to beat them. Tell me, if Commodore was to build an AAA
system and it listed for $4995, would you buy SGI or Amiga if the SGI were
only a few thousand more. Most of my corporate buyers would NOT give the
Amiga a second look, nor would you if your business was based on a need to
produce video. Not that the Amiga would not produce video, but given the
fact that most of our corporate customers either already have an Amiga, or
generally have budgetted $20k toward a video solution, the Indy will shine
in their eyes as they all know who and what SGI is.
I think you need to understand something, something that apparently you do
not see, and perhaps even SGI itself doesn't see. The mere fact that SGI
has a entry-level box in the under $5k mark has redefined the playing
field. I am exited as I have not seen a company attempt to redefine the
market since Commodore did it in 1985 with the Amiga. The difference here
is that SGI has the potential to literally dominate the market due to
their already established good name, and the fact that their dealer
network is designed to allow resellers to make money. This is not the
case when an Amiga dealer works with NewTek.
Let me address your software concerns. It is my belief that this box has
the potential to DRAW a large group of graphics oriented developers who
ARE NOT, in general, making money in the Amiga market to the Indy. These
companies will fill the void that the big name software companies on SGI
boxes are not willing to fill. ASDG and the Blue Ribbon Soundworks are
already developing for it. If the Indy sells like I think it might, there
will be a lot of money to be made in the less than $1k software range.
Many will be willing to stoop over and pick up those dollars.
Perhaps this makes my feelings clear on this. It is not the Amiga killer,
today, a year from now–I guarentee it is from a desktop video perspective
anyway.
-mark=
Otto Pilot Engaged..
I agree totally! Heck, PC people percieve the Amiga as to expensive right
now, no way are they going to buy an Indy.
I'd be interested, but even I couldn't justify spending the required cash
to get a capable machine.
Bob Comer — Cruising the nets on AutoPilot from Cheyenne, Wyoming
I certainly don't think Indy will set the world on fire, so I basically agree
with you. I find Indy to remind me a lot of the first NeXT machine. Cool
stuff, but not that many people need it or can afford it in a usable
configuration.
I believe that video capture / compression etc., will only really take off when
it is seamlessly integrated with corporate networks, because then "video
chatting" will be a reality and save a TON of "hall walking".
Paul.
Paul, Indy is not a 'consumer' machine. It is a workstation that offers
features of very high end systems at prices unheard of before. It will compete
with Quadras, pentium and maybe even 040 Toasters. It is not a home machine.
Lou
Don't limit yourself to the seller's definition, Lou. If the right software and
peripherals come along, the Indy is cheap enough to sell as a pure "end user"
machine. Many video "pro-sumers" would be interested in the capabilities of the
Indy (for example). –Redmond
I think SGI know's EXACTLY what they've done. At NAB, they had a very big
booth in the multimedia hall, right near all those Toaster-related vendors,
among others, and they were giving out gobs of developer application forms as
far as I could tell.
Rich:
Yes, I do think the Indy will end up costing much more than the 5k base
price, oh well, I don't need it THAT badly. I did order NextStep 3.1 for
486 today tho. They currently have a special deal on where you can get
NS3.1 for $299.00 for the complete system, both user & development
systems. It comes on 2 CDROMs, and a 3.5" boot disk, and I'll be replacing
the Interactive Unix on my PC with it. The only 'gotcha' to the deal is
that it's NOT upgradable, when NS3.2 comes out, and you wanna get it, you
gotta pay full price. Of course there's nothing stopping you from
continuing on with NS3.1 as long as you want.
I think it's funny that now that the NeXT machine isn't a threat to the
Amiga anymore, I don't mind looking at it's OS as a replacement for the
character based Unix I was using (NextStep is basically an X windows like
GUI on top of BSD Unix), especially considering that with their special
deal the entire NextStep 3.1 OS is cheaper than I'd have to pay for any
version of X for my PC (either from Interactive [now Sun], or 3rd party).
As for the pricing of apps for the Indy (or any other computer), I think
it's absolutely DISGUSTING how the software companies RAPE the users of
non MS-DOS machines. I looked into a couple of software packages for my
Unix 486 machine, and was shocked at the prices! While I realise there are
costs to porting software, I can't believe that something that sells for
$500.00 on a PC (already overpriced if you ask me!) should sell for double
or more on a Unix machine?
As I'm gonna have to get a larger HD than I currently have in my PC to run
NextStep, I'm probably going to have a small (around 50M or so) DOS
partition, and for the first time in a LONG time have a machine that can
bootup into real MS-DOS. It's gotten to the point that even with DOS being
brain dead, some of the games available on it look really nice.
// -Mike Schiller
\X/ AMIGA! On AutoPilot in sunny Fort Lauderdale
Mike,
Quite thinking for a moment as a consumer. Think a moment as a
manufactuer. Price is set at what the market will pay. The market
clearly pays for the software on UNIX machines, and thus, the companies
exist.
Price is set based on market-size, the amount of support necessary and
finally what the precieved profit will be. The UNIX market on the PC is
not a large one, and thus the very market is designed for the vertical
marketeers.
I really wish the consumer in general would get off of this 'software
should be less kick'. It sounds as if many think, not necessarily you
Mike, but many think that software companies should be more of a social
program than a profitable company.
The market demands what happens to software, not really based on consumer
outcry, but rather on actual cold cash in the coffers. If the coffer gets
a bit empty, the price will drop. If the coffer remains full, the cries
will never be heard, nor should they.
It is my thinking that the Indy will draw a lot of the high-end, top notch
Amiga developers who are looking for someplace to go. You'll get the
software in the range that is clearly less expensive than the current line
up of software products.
You'll be amazed, things can change in the blink of an eye.
-mark=
Otto Pilot Engaged..
Mike,
Thanks for the information about NextStep. $299 for a 486 machine, huh? I've
got to take a look at that.
The sticker shock for software for Unix-based machines is heart-stopping!
Rich
I don't get it… you move from a more popular market to a more obscure market
(PC to Unix) and expect the software prices to be the same. For sake of
argument, let's say the development expenses (marketing, sales, programming)
are exactly the same for those two markets. (They're not.) Consider the actual
and potential sales penetration for each market. Divide the total cost (let's
call it 100K per employee per year, in a small company with five people) by the
total sales. You still think it's "rape and pillage"?
Richard,
I seem to remember a certain machine that cost $1995 with only _one_
floppy drive, 256K RAM (expandable to 512K), and a questionable, at best,
paint program. You had to pay _significantly_ more for an extremely buggy
compiler and a word processing program, also buggy, as was the
documentation. The machine, my '85 Amiga 1000.
Frankly, anytime you are on the 'bleeding edge' you should expect to
pay a price, but I have serious questions on your evaluation of the
"minimum" configuration, again harkening back to my experience with
another introductory machine. I do page layout here on a 2.5 MB machine
feeding a laser printer, an impossibility to some. 96 MB??? Sheesh! I
think you will find that this machine will enter a particular niche which
will attract quality developers that are not satisfied with the current
status quo on pricing.
Just my two cents.
Brian
Brian,
I don't need 96 megs for page layout. But I _do_ need that much RAM and more
to handle the image manipulation that I do of the scanned images that I place
in those page layouts.
For electronic pre-press work, 256 megabytes will soon become the 'standard'
work station RAM configuration.
Rich
[ continuation ]
is not the Mac killer, or the PC killer that SGI's ads would like the world to
believe. Other forces are killing the Mac. But SGI's star is certainly in an
ascending mode lately.
Rich
To be perfectly fair Richard, a 24 bit Indy with harddisk (340 MB) will cost
$9500. This includes the 15 inch monitor, 16 MB ram, camera abd a fair amoount
of software.
But you are right that SGI peroripheals cost a lot more then those on other
platforms. However, Indy – if it is the 'volume' machine SGI hopes, should
change that to a more normal cost structure.
Lou
Lou,
I stand by my estimate of $30,000 to have a useable Indy work station. I have
no doubt that SGI is serious about penetrating the 'low end' market. But the
uses that SGI envisions for this machine _require_ RAM in the _hundreds_ of
megabytes. Graphics processing demands resources that are orders of magnitudes
greater than what most people are used to using.
The $30,000 figure comes from a long discussion that I had with our dealer. We
compared the Indy to my 'middle of the road' pre-press setup (also vulgarly
known in the industry as DTP). To acheive the same level of functionality as my
Quadra 800 set up, the Indy _easily_ costs as much as I've indicated. Those are
_his_ figures – and believe me, he is a motivated salesman. He _wants_ to sell
me an Indy. But he's also honest and realistic. He knows my operation well and
has no desire to mislead.
There is no denying that getting the Indy outfitted as I've described gives it
the ability to outperform the Mac in many other ways. To try to so equip the
Mac would be even _more_ expensive, and probably not possible at all.
Rich
Richard,
I must be confused because I find your statement "[g]raphics processing
demands resources that are orders of magnitudes greater than most people
are used to using" curious, at best. I'm running only a 9 MB Amiga here
and I have yet to run out of memory on ADPro or any of the other graphics
programs I have. This in 24 bit mode. Now, I can picture a few reasons
for my confusion, so I thought I'd enumerate them.
One, I'm not reaching the high-end. What _are_ you doing with all that
RAM? Am I missing something here?
Two, the machines, the Mac and the Indy, require many times more
resources to accomplish a transformation. That doesn't quite make sense,
especially if a transformation is involved. Shouldn't the machine
dedicate itself to this particular task?
Three, as a matter of convienence, are you keeping massive amounts of
images in memory at the same time? Why is this significant?
As I stated, I must be missing something here if the minimum
configuration is as you state.
Brian
Brian,
The answer is #1, you are not reaching the high-end. 🙂
Try the following: load a 300 dpi 5×7 or 8×10 (inches) image into ADPro
(24-bit color) and see what happens to your 9 megs. 🙂 That is what is
necessary for color DTP. If you do DTP professionally, it will be a common
occurance. (Heck, try an 8×10 in 8-bit grey!)
Also…many renderings in 3D (using Lightwave, Aladdin or Real 3D 1.3 or 2.0)
suck up a LOT of RAM. 18 Mb makes Lightwave comfortable; 12 for Aladdin; 24+
Mb for Real 2.0 (more if you want to do massive collision detection within a
particle system — with transparency, etc.) I've seen a person with 32 Mb of
RAM crash Real 3D 2.0 for lack of RAM. 🙂 (Gigamem is a nice solution.)
Nine is fine for most normal business use (WP, Database, Spreadsheets, games,
etc.) but not for professional graphics. It is borderline low.
-Chas
Brian,
My business is electronic pre-press. We prepare images for reproduction on
printing presses. (We also are a printing house, but that's another story). If
you look at a magazine like Vogue, or Architectural Digest, or National
Geographic you will see Four Color Process printing at its best. Pictures in
those magazines are four-color halftones which have a 'resolution' of 150 to
200 lines per inch. That is, if you examine the fine dot pattern that makes up
the pictures, you will count 150 dots to the inch in a 150 lpi picture.
Those pictures result from original continuous-tone transparencies that have
been scanned on a drum scanner. In general, the scanner must produce an
electronic file containg a scan frequency which is twice the lpi figure of the
desired final halftone. That is, if a halftone of 150 lpi is needed, the
scanner must scan the original at 300 dots per inch (pixels per inch).
Each scanner pixel contains one byte of data. That byte is able to specify one
of 256 levels of gray at the position in the original that it represents. The
file size for such a sampling of data is given by the following formula:
horizontal scan resolution x vertical scan resolution x area scanned.
One square inch scanned at 300 dpi therefore results in a file size of 300 x
300 x 1 or 90,000 bytes. a full page scan (8 x 10) is 300 x 300 x 80 or 7.2
megs. For a full color scan (24 bit RGB), that figure becomes 300 x 300 x 80 x
3 or 21.6 megs.
So, for an 8 x 10 full color image at 300 dpi, I have to have _at least_ 22
megs of RAM for the image _alone_, to say nothing of the operating system or
the image manipulation program. To handle such an image, that is, to allow the
manipulations that one performs on such an image, the image manipulation
program requires about five times as much RAM (or virtual memory) as the image
itself. For the 22 meg image, over 100 megabytes of RAM is a _must_. Virtual
memory (hard drive) is hopelessly slow for this kind of work.
[ MORE ]
[ continuation ]
Now add these realities to the game. The RGB image needs to be converted to
CMYK. That makes the 22 meg image 29 megs. Five times 29 megs is 144 megs of
needed system ram! And that's just for one 8 x 10 image at 150 lpi. Magazines
are going to 175 and 200 line halftones almost routinely now. An 8 x 10, 175
line halftone in full color results in a file of 40 megs.
Want to try the math for a double page spread, full bleed 11 x 17 at 200 lpi
(400 dpi)? The file itself is over 120 megabytes.
See why so much RAM is needed for this kind of work?
8)
Rich
I expect the street price to be around $4995. The difference in the SGI
market versus the Amiga market is that in the SGI market, you, the dealer,
are expected to make a profit. 🙂
Honestly I don't know, mostly it will depend on the cost of the system. I
already know that it will not be made available for mail order, something
I particularlly like.
-mark=
Otto Pilot Engaged..