| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 28.1 | `low' is a relative term | HYEND::BLYONS | VAXcluster SASE | Thu Jan 11 1990 10:36 | 16 |
| > Are customers price sensitive or are they
> more interested in the technology no matter what the cost?
At some level, isn't everyone cost sensitive :-)
We need to qualify the term `low cost'... low with respect to NI
or CI interconnects. I can't picture DEC charging $30K+ for it
like we do for CI adaptors. NI interfaces are 1/3 or less (I think)
that amount and that is probably where the FDDI price band will fall.
Mainly because the market thinks of the FDDI as a replacement for
the NI, not a cost reduction for the CI.
I agree with you that the FDDI market should be driven mostly by
price and only a little by its mechanical security type features.
Bob L.
|
| 28.2 | What I expect: FDDI cost equal to CI cost | CVG::PETTENGILL | mulp | Thu Jan 11 1990 15:52 | 8 |
| I expect the FDDI `system' cost to be in the CI `system' cost range. In other
words, by the time you total up the cost of adapters, cable concentrators,
CI hubs, cable, etc, the interconnect cost for 4 CPUs connected by FDDI will
be about the same as 4 CPUs connected by CI.
This is my expectation based on the complexity of the components and based on
my assumption that we will go higher margin on CI and lower margin on FDDI.
DO NOT TAKE THIS AS FACT!!! I DO NOT HAVE ANY SPECIFIC KNOWLEDGE OF THE PRICING.
|
| 28.3 | I hope DFON $ is NO indication | MAMTS3::PDORNAN | Patrick Dornan, NWSS 8-339-7169 | Thu Jan 11 1990 17:45 | 7 |
| If pricing for DFON products is any indication, we are going to
be extremely expensive. A small example: The BN24E-01 Fiber Optic
Jumper Cable (ST/ST) is $130. I can get a 50 ft cable from AT&T
for $75. I really can't wait to see how our FDDI products compare
with the competitions'.
Patrick
|
| 28.4 | > Ethernet, < CI | DENVER::CASE | SWS DNT DVO | Tue Jan 16 1990 13:59 | 16 |
| My customers are expecting cost to be higher than Ethernet, but that
competition and a large nuber of interconnect products from multiple
vendors to quickly drive the cost down. For bridges and routers, a
cost in the $15-$25K range would be expected. For lower-end SAS
controllers (e.g. SCSI, VME bus), perhaps $3-10K. For higher-end SAS
and DAS controllers, $5-20K.
We're already seeing a wave of competitive products based on the AMD
Supernet(-1) chip set at these prices; so I hope we won't have to do
extensive "justify the high price by additional capabilities" selling
in this space. FDDI connections will become commodity items much
faster than Ethernet and Token Ring did.
No special knowledge of our marketing directions here, just a synopsis
of what I'm hearing from customers.
|
| 28.5 | Adapters aren't the only expense | CVG::PETTENGILL | mulp | Wed Jan 17 1990 18:56 | 13 |
| re:.4
Do your customers understand about cable concentrators? Are they including
the cable concentrator cost in the $3K low end cost? On a per port basis,
the component count for the concentrator is basically the same as for a low end
adapter.
The next generation of `low end' CI adapters will have less than half the
component count on a per port basis than equivalent FDDI; in fact that might
already be true for TODAY's CI adapters. The major cost for CI adapters is
related to the transport protocol support that is included in every adapter.
If a low end FDDI adapter gave you Ethernet adapter performance, can you
justify FDDI?
|
| 28.6 | | KONING::KONING | NI1D @FN42eq | Fri Jan 19 1990 13:33 | 5 |
| No, but I don't expect anyone would be foolish enough to plan an adapter that
would offer merely Ethernet performance. All the ones I know of are factors
of N better than that.
paul
|
| 28.7 | Look at the 10/100 bridge | DC101::BLASINGAME | Craig @EKO, GSG/DCC, DTN-339-7245 | Fri Feb 09 1990 16:23 | 8 |
| I'm currently trying to price an FDDI to interconnect 35+ Ethernets.
Based on that exercise, the preliminary pricing for the bridge given at
NETU, and what the "competition" showed at COMMNET; I would say that
the fact that we will not be able to connect multiple (say at least 4
ethernets to the same bridge) is going to be a problem.
In case you are curious, the DEC pricing for a network of this size, without
any fiber plant cost, could be over 1M$ - THAT'S just CONs and BRs!
|
| 28.8 | Don't spec a Ferrari if a VW will do | KONING::KONING | NI1D @FN42eq | Fri Feb 09 1990 18:55 | 7 |
| I suppose that's true if you have 35 10/100 bridges. But that would be
overkill unless every Ethernet is near full load. If not, then the right
answer would be to build a two-level structure: some 10/10 bridges onto
a smaller number of Ethernet backbones, then 10/100 bridges from those
backbones to FDDI.
paul
|
| 28.9 | You may not sell an Italian on the VW | ARCHON::BLASINGAME | Craig @EKO, GSG/DCC, DTN-339-7245 | Thu May 17 1990 15:28 | 6 |
| I would agree that the use of 10/10 bridges seems the answer here.
However, my customer (Uncle Sam of course) looks on FDDI as a
"backbone" for connecting unrelated departmental LANs. Many of these
departments have not spoken to one another in years, let alone thought
about the consequences of having someone elses traffic flow across
THEIR ethernet LAN.
|