Skip to main content
This Board Has Moved

This board has been migrated to our new platform! Check out the new home page at discussion.fool.com or click below to go directly to the new Board on the new site.

Go to the New Site
Message Font: Serif | Sans-Serif
 
No. of Recommendations: 0
Does anybody have any insight into why Arm shares just fell off a cliff? The noteworthy fact is that volume increased 10X over average. Somebody was dumping a lot of shares. I do not see any news. Is this just a high PE stock getting punished or is ominous new coming?

Thanks,
Paul
Print the post Back To Top
No. of Recommendations: 0
Kinda been wondering the same thing. Arm's biggest competitor (i think it's MIPS) announced an earnings warning. Not saying ARM is about to announce an earnings warning, but that could be why. Plus, the company is also expensive.

XaosSufer, what's your opinion on ARMHY as a company and a stock going forward (next 5-7 years)? I feel it's important to look at a company's fundamentals and business risk (risk meaning competition) and stock price. If everyone in the world knows and prices a stock like ARMHY for greatness then the potential reward is greatly reduced. That said, some argued Aol was priced for perfection in 1995 or 1997 and well...were wrong.
Print the post Back To Top
No. of Recommendations: 1

XaosSufer, what's your opinion on ARMHY as a company and a stock going forward (next 5-7 years)?


I am strictly a Gorilla Game investor. Actually, I am strictly a Gorilla Game spectator hoping to be an investor when the markets get sane again.

Arm is very close to crossing the Gorilla threshold. They have the market share lead. They have the value chain lead (I would like to see more software dependant on Arm). The only thing missing is a clear tornado. If the tornado hits and Arm can execute the transistion to Tornado tactics, I consider it very likely that they will be the embedded processor core Gorilla. I would like them to win Broadcom back from Mips, though.

Until the tornado hits and Arm is selected as the de facto standard I will not invest. I am also keeping my eyes on ARC.

Paul
Print the post Back To Top
No. of Recommendations: 2
Hi Paul:

With regards to ARM and the tornado. Look at unit shipment volumes, therein lies the indication of the tornado for IP companies like ARM, QCOM and RMBS.

Right now most tornadoes have been suspended as all the inventory glut gets worked off in the system. Strange things are happening with ARM the stock, but the company so far has been solid, we'll see if they warn or not soon, chances are that they may do so, but in the long term they seem to have a lock on the chip side of the embedded processor market even if their royalties are low, these things will be going everywhere. Who knows? In the U.S. every person might own 100 ARM processors in the next few years.

Just my two cents.

Erick

P.S. I've been getting killed with my call nights, close to zero sleep, but I've been learning a heck of a lot. It just doesn't leave much time to post. Keep up the good work everyone and keep me posted in KP. I'll be rooting for that Toronto Olympics bid, nice town, I've been a couple of times, and even saw the Blue Jays play.
Print the post Back To Top
No. of Recommendations: 0

Look at unit shipment volumes, therein lies the indication of the tornado for IP companies like ARM, QCOM and RMBS.


I disagree with this definition but I am alone in my disagreement. I think that the tornado is defined by the uptake of the end product. For example, the P4 tornado for Rambus or the 2G tornado for QCOM.


In the U.S. every person might own 100 ARM processors in the next few years.


How about disposable ARM processors (in packaging, products, paper).



I'll be rooting for that Toronto Olympics bid, nice town, I've been a couple of times, and even saw the Blue Jays play.


Thanks. Y'all come back when you need a break.

Paul
Print the post Back To Top
No. of Recommendations: 2
=====
Look at unit shipment volumes, therein lies the indication of the tornado for IP companies like ARM,
QCOM and RMBS.


I disagree with this definition but I am alone in my disagreement. I think that the tornado is defined by the
uptake of the end product. For example, the P4 tornado for Rambus or the 2G tornado for QCOM.
====

These are "integrated" products where the functionality lies in the individual components. Therefore a tornado in one of these might not mean a tornado for one of the components.

For example a P4-Rambus tornado does not mean a Rambus tornado. Intel might not support Rambus at all in the next generation.

It is the standardisation of all future end product architectures with this component as a vital feature that siginifies a tornado for this component.

In that sense, unit shipment volumes (increase thereof) are an indication of a tornado in progress. In fact, unit shipment volumes of a component for different products might be a much better indicator. Would be grateful if someone could point me towards these for ARM/RMBS.

As far as RMBS is concerned, I am not really sure why Intel would want a dominant(standard) memory supplier. Agreed, better performance, blah blah, but this is an area (motherboard) which has traditionally been Intel's stronghold in terms of interfacing components. It might be in Intel's best interests to keep two technologies in the fray. That might be the reason Intel supported Rambus in the first place. Frankly, Intel has every reason to pull the plug on RMBS in future designs just to show them their place.

Meanwhile, RMBS is playing a game of its own in the courts. If it loses, Intel has every reason to support it as a viable alternative to Micron in which case we might see wild fluctuations in the prices of both Micron and RMBS as and how Intel decides to use them in future machines.

It is this very fate RMBS is trying to avoid in the courts. Lets wait and see ...
Print the post Back To Top
No. of Recommendations: 1
Hi Paul:

I wrote:
"Look at unit shipment volumes, therein lies the indication of the tornado for IP companies like ARM, QCOM and RMBS."


You replied:
"I disagree with this definition but I am alone in my disagreement. I think that the tornado is defined by the uptake of the end product. For example, the P4 tornado for Rambus or the 2G tornado for QCOM."


I don't disagree with you, the problem is that unit shipments are much easier to quantify than end user demand. Ultimately end user demand drives unti shipments, if not then inventories rise and we get into trouble again.

I wrote:
"In the U.S. every person might own 100 ARM processors in the next few years."


You replied:
"How about disposable ARM processors (in packaging, products, paper)."


That's not so far fetched, just look at those singing greeting cards, they all have small chips. More and more products will contain chips, until one day the toilet paper will automatically tell us when our precious behinds have been wiped to perfection and I'm betting ARM™ powered paper products will be common in a few years. It could happen.

Erick

P.S. Goodnight, I have to get up at 4:30 am tomorrow.
Print the post Back To Top
No. of Recommendations: 3
It's tautological.

Geoffrey Moore defines the term tornado in term of end user uptake - not sub-component uptake. The reason is that when end user demand exceeds supply power shifts to the points of architectural control in the value chain. When this happens the companies with architectural control become Gorillas and the profits in the industry shift to them.

You may think that you have a better system. That is fine. It still does not allow you to redefine Moore's terms just because they don't fit your model.

In the the next RTW issue I will revisted the basics of Gorilla Gaming. As the posts from Erick and this post show, the language of the Gorilla Game has been applied too broadly and as a result have lost thier wonderful predictive power.

Any guesses about which company I will use as an example?

[email protected]
Print the post Back To Top
No. of Recommendations: 0
just look at those singing greeting cards, they all have small chips.

I read somewhere that one of those chips has more processing power in it than all the computers used in WWII combined.

I would be very interested to hear what competitive advantages that ARMHY has over MIPS. Is it the patents? What is it about ARMHY that makes it a possible gorilla and not MIPS?

Erik

Print the post Back To Top
No. of Recommendations: 3

========

Geoffrey Moore defines the term tornado in term of end user uptake - not sub-component uptake. The reason
is that when end user demand exceeds supply power shifts to the points of architectural control in the value
chain. When this happens the companies with architectural control become Gorillas and the profits in the
industry shift to them.

==========

Don't get mad at me boss. Just trying to discuss some things.

Who are the end users ? Persons who buy P4s -- just one end product ? Or persons who buy PCs, and Game consoles and so on ...

The value chain here is not just one product. But a whole range of products. Its like RMBS being involved in many "functional" value chains all depending upon RMBS. We do not need a tornado in all those chains for RMBS to be the dominant standard.

People are not going to buy PCs because of RMBS. But they will because of Intel. Since Intel is already the dominant gorilla in PC-microprocessors, it is Intel who must decide the memory standard. So the gorilla status for RMBS will be "conferred" by Intel because Intel are irreplaceable. And therefore, as far as RMBS is concerned, their value chain ends at Intel.

Similarly, other markets where RMBS is playing. Say Game consoles. Just because one Game console sells well does not mean the next one will standardise on RMBS being the memory standard. The processor guys will decide that. And it is their uptake which will benefit RMBS.

Print the post Back To Top
No. of Recommendations: 1

Don't get mad at me boss.


Sorry. Pet Peeve. People use the Gorilla Game language any way they choose and then blame Moore when their investments do not turn out.


Persons who buy P4s -- just one end product ? Or persons who buy PCs, and Game consoles and so on ...


Yes. Tornado happens one product category at a time.


People are not going to buy PCs because of RMBS.


So? They don't buy it for the microprocessor either. They buy it for Word, Excel, Email and Surfing.


So the gorilla status for RMBS will be "conferred" by Intel because Intel are irreplaceable. And therefore, as far as RMBS is concerned, their value chain ends at Intel.


Yes. The same way IBM conferred Gorillahood on Intel and Microsoft. Actually, IBM made Intel a King by insisting on a second source - AMD.

Question. Have you read The Gorilla Game?

Paul


Print the post Back To Top
No. of Recommendations: 1


=========
Persons who buy P4s -- just one end product ? Or persons who buy PCs, and Game consoles and so
on ...


Yes. Tornado happens one product category at a time.
=========

Yes, exactly.

=========
People are not going to buy PCs because of RMBS.


So? They don't buy it for the microprocessor either. They buy it for Word, Excel, Email and Surfing.
==========

Absolutely right. But when we have an integrated product in the mainstream, the gorilla power moves to the functional components.

======
So the gorilla status for RMBS will be "conferred" by Intel because Intel are irreplaceable. And
therefore, as far as RMBS is concerned, their value chain ends at Intel.


Yes. The same way IBM conferred Gorillahood on Intel and Microsoft. Actually, IBM made Intel a King by
insisting on a second source - AMD.
==========

Yes. Right. And then once PCs became mainstream, the power moved inside. Now Intel has the power here (and Microsoft). Why ? Because they are an enabling technology. We cannot have a win-95 without Intel underneath. But WE CAN HAVE Intel with or without Rambus, however successful a product Rambus makes because, Rambus is just another component Intel interfaces with.

Let me try again. We don't have a microprocessor layer dependant on the Memory layer because its a horizontal function. And here, Intel is the dominant gorilla and the future of all gorillas at this level of abstraction depend upon Intel because it is irreplaceable.


=========
Question. Have you read The Gorilla Game?
=========

Yes. Have you understood it ?
Print the post Back To Top
No. of Recommendations: 1
I hear this type of argument frequently. Undoubtebly Intel and Microsoft are mature, dominant Gorillas. Rambus is much younger and may or may not extend it's power beyond it's current position. However, it is still a Gorilla.



Yes. Have you understood it ?


I aksed so that I would know the level of detail to use in my answers to your questions. However, you have saved me the bother.

Good luck in your investments.

Paul

Print the post Back To Top
No. of Recommendations: 7
How about disposable ARM processors (in packaging, products, paper).

Will companies that make processors for disposable products or even simple processors used in say a alarm clock or watch really need what ARM provides?

More and more products will contain chips, until one day the toilet paper will automatically tell us when our precious behinds have been wiped to perfection and I'm betting ARM™ powered paper products will be common in a few years. It could happen.

I could see a future where chips are even more common place than they are today, but i dont feel these chips will need what ARM provides. Those chips will be so cheap and simple that it'd be a commodity to a 'T'.

Is it the patents? What is it about ARMHY that makes it a possible gorilla and not MIPS?

I'm no expert on ARMHY, but i've read that they have 5,000 patents. Although, patents can be written around. One advantage (for now) is they have superior technology. Their chips are faster, cheaper, cooler, smaller, ect..

It also seems ARMHY has developed an open-propietary system (much like Windows or Intel has with IA-64). WinCE, PalmOS, Symbian (this is a popular OS for cell phones in Europe) all support and work on ARM based processors. Palm even told Motorola that if they don't license ARM technology they'd buy processors from some one else b/c Palm was going to port its operating system to work only with ARM processors. Past palm-pilots worked with 16-bit Dragonball processors developed by Motorola i believe.

People are not going to buy PCs because of RMBS. But they will because of Intel. Since Intel is already the dominant gorilla in PC-microprocessors, it is Intel who must decide the memory standard. So the gorilla status for RMBS will be "conferred" by Intel because Intel are irreplaceable. And therefore, as far as RMBS is concerned, their value chain ends at Intel.

I feel that Intel is currently a strong king. AMD can make x86 processors and basically offers a product that works just the same. AMD and Intel compete on price, speed, and marketing.

With Intel's push into IA-64 (their 64-bit processors), software has to be rewritten for this new standard. If Intel keeps this new architecture to themselves (and if this new architecture becomes the standard) then Intel might become a true gorilla again.
Print the post Back To Top
No. of Recommendations: 0

Post of The Day!!!!!!!!!!!!!!!!

I agree 100%.

[email protected]


I feel that Intel is currently a strong king. AMD can make x86 processors and basically offers a product that works just the same. AMD and Intel compete on price, speed, and marketing.

With Intel's push into IA-64 (their 64-bit processors), software has to be rewritten for this new standard. If Intel keeps this new architecture to themselves (and if this new architecture becomes the standard) then Intel might become a true gorilla again.

Print the post Back To Top
No. of Recommendations: 1
=======

How about disposable ARM processors (in packaging, products, paper).

Will companies that make processors for disposable products or even simple processors used in say a alarm
clock or watch really need what ARM provides?

More and more products will contain chips, until one day the toilet paper will automatically tell us
when our precious behinds have been wiped to perfection and I'm betting ARM™ powered paper
products will be common in a few years. It could happen.

I could see a future where chips are even more common place than they are today, but i dont feel these chips
will need what ARM provides. Those chips will be so cheap and simple that it'd be a commodity to a 'T'

==========

Yep. I figure they will concentrate on growing markets like wireless, handheld devices. Thats where the new tornados will form.


==========
I feel that Intel is currently a strong king. AMD can make x86 processors and basically offers a product that
works just the same. AMD and Intel compete on price, speed, and marketing.

With Intel's push into IA-64 (their 64-bit processors), software has to be rewritten for this new standard. If
Intel keeps this new architecture to themselves (and if this new architecture becomes the standard) then Intel
might become a true gorilla again.
===========

Yes. It might be interesting though. If hardware-interfacing software has to be rewritten, then AMD can be big again too. Their chance at maintaing the gorilla postion would be to change the hardware architecture while keeping the os-interface same. This would force components talking to Intel to change too, to support the new Intel and keep out AMD.

Does AMD have a competing processor to the IA-64 ? But I don't think AMD has the leverage to work with software makers to change software according to their architecture.
Print the post Back To Top
No. of Recommendations: 0
Does AMD have a competing processor to the IA-64 ? But I don't think AMD has the leverage to work with software makers to change software according to their architecture.

Oh, you can bet AMD is working their little hard a$$es off coming up with something. They seem to have a pretty interesting approach. Basically, instead of starting everything over from scratch they are extending the x86 architecture to support 64 bit processing. So AMD's 64 bit processor will also be able to run all those old 32 bit applications (obviously, it would be slow at running 32-bit stuff b/c it uses an emulator)

The problem is, AMD is still supporting x86 which is many years old. Intel with IA-64 is a fresh start & probably incorporating many new technologies. Still 64-bit computing at the desktop level is probably a few years away. It'd be interesting to see how the Wintel combo does in the server space against Sun & IBM.

It seems from my very rough knowledge that a new gorilla game is developing in the 64-bit processor space. Need to do more reading, tho..
Print the post Back To Top
No. of Recommendations: 0
Someone said
Does AMD have a competing processor to the IA-64 ? But I don't think AMD has the leverage to work with software makers to change software according to their architecture.

to which scrim1 replied
The problem is, AMD is still supporting x86 which is many years old. Intel with IA-64 is a fresh start & probably incorporating many new technologies.

I really hope Intel has learned from what IBM tried to do with the PS2 and Microchannel. This is beginning to sound eerily similar and PS2/MicroChannel was a classic example of a minor technology compared to ISA/EISA that everyone else stuck with.

DD
Print the post Back To Top
No. of Recommendations: 1

I would be very interested to hear what competitive advantages that ARMHY has over MIPS. Is it the patents? What is it about ARMHY that makes it a possible gorilla and not MIPS?


In the beginning the differentiator was that ARM used lower power, thus making it more suitable for battery power applications. Secondly ARM offered their chips in forms which made it easy to combine it with other parts so that TI (for example) could combine it with DSP technology or Intel with Flash technology.

These helped to build the value chain as it allowed ARMs licensees to get revenue themselves as well as pass some on to ARM and battery powered embedded devices are rather more of a growth area than devices where power is essentially unlimited. The value chain meant that ARM was able to develop reference designs, development environments and other things that help make it easy to use ARM and faster to bring a product to market, which meant that ARM became more popular, which meant that more developers had experience with ARM which meant they were easier to find...

Right now the competitive advantage is in the value chain and the fact that ARM works better in low power situatuations. One other reason why the value chain is so much stronger with ARM is that they were independant, MIPS was tied very much to SGI/NEC and thus other potential users felt that they would be at a disadvantage compared to NEC. With ARM no one felt that they were at a disadvantage.

DD
Print the post Back To Top
No. of Recommendations: 2
the language of the Gorilla Game has been applied too broadly and as a result have lost thier wonderful predictive power.

Amen to that! Just as we had people convinced that Gemstar had become a Gorilla before any evidence of any kind of a tornado, we now have people thinking Qualcomm is already the gorilla of 3G wireless when the product category hasn't even crossed the chasm.

Actually, I believe the predictive powers of the terminology remain as valid as ever. It's the misuse of the terms that causes people to think their predictive power is lessened.

--Mike Buckley
Print the post Back To Top
No. of Recommendations: 0
Paul,

Actually, IBM made Intel a King

What was it that you asked me not to get into over here at the Fool? Whatever the subject is, I promise to stay away from it if you promise to stay away from that debate. :)

--Mike Buckley
Print the post Back To Top
No. of Recommendations: 0

=======

Actually, I believe the predictive powers of the terminology remain as valid as ever. It's the misuse of the
terms that causes people to think their predictive power is lessened.

=======

I agree!! We can't apply the rules too broadly. Its the corner case which will break the rule.
Print the post Back To Top
No. of Recommendations: 0
and I promise to remember my promise as well as you.

Paul
Print the post Back To Top
No. of Recommendations: 0
Paul,

and I promise to remember my promise as well as you.

Great, but I'm off the hook until you tell me what I'm supposed to remember. :)

You're right that I need to read Tornado again. At a bare minimum I need to look at the pages highlighted by page points.

--Mike Buckley

Print the post Back To Top
No. of Recommendations: 10
Hi Paul:

Yes, I read all four of Moore's books. I hope I understood them! ARM is a funny gorilla.

The Gorilla Game is predicated on the Microcosm, on computers. ARM does produce processor cores but these are not restricted to computers -- they live in everything mobile and embedded. The Intel x86 basically powered the IBM PC and Windoze basically enabled the IBM PC. They are one paradigm players. ARM is inside lots of things that belong to different paradigms from PDAs to cell phones to security chips to stuff in cars, game players and whatnot.

The next difference is that when the IBM PC tornado exploded, Intel was capacity constrained but ARM will never be because they don't have fabs. Maybe some of their licensees will be capacity constrained but not ARM. In that sense, ARM is more like Microsoft than Microsoft itself, It does not even have to shrink wrap copies of ARM chips, all it has to do is to collect the royalties (easier and cheaper than shrink wrapping with no inventory problems :-)).

To return to the tornado question, there could be a PDA tornado, a G2 cell phones tornado, a G3 cell phones tornado, a security chip tornado, a car stuff tornado, a game player tornado and many whatnot tornados. From ARM's point of view, all these areas are their bowling alley pins and the tornado, for them, could be just the shipment of ARM cores in general. In my mind the big question is, how many cores does ARM ship in contrast to MIPS, Transmeta, ARC and others. The growth year over year is impressive. The market share of cell phones is impressive. Their growth relative to Transmeta and MIPS is impressive.

So why the share price contraction? Because they were valued at bubble levels and the bubble burst. As per DOW Theory, all stocks go up and down with the market. ARMHY is being brought down to reasonable levels from where it will explode once the bears go away.

Long ARMHY!

Denny
Print the post Back To Top
No. of Recommendations: 3
Hi Denny,

Thanks for sharing your thoughts, I was hoping you'd drop by after our email exchange. Just as an extension of you comment "Because they were valued at bubble levels and the bubble burst", ARM's multiple has remained much higher than most other tech companies (especially in the semiconductor market). Erik warned about this in his June article in RTW so most of us were prepared for a fall. ARM's management has been pretty cocky these past few quarters as the rest of the market faltered while they continued to hit estimates. However, since the Nokia warning last month, I think that investors are growing more concerned that ARM will no longer be able to escape the down draft. Obviously ARM would have fallen with the market regardless of the news but I think that investor fears probably exacerbated the fall.

I also agree that ARM will definitely be one of the survivors of this market as it continues to improve its competitive position and build out its value chain. Once the market starts to recover, great companies like ARM will lead the rebound.

Matt
Print the post Back To Top
No. of Recommendations: 2
<I<
The Gorilla Game is predicated on the Microcosm, on computers.


Are you suggesting that Moore's market development model does not apply in the embedded processor world? If so, what part of the model no longer works?


They are one paradigm players.


One point I think people forget is that Intel and Microsoft have been develeping the PC architecture for 20 years. It is very mature today. In the early days of the 6502 and CP/M, the industry did not seem as well defined or was the opportunity truly understood.


Intel was capacity constrained


Arm and Microsoft have capacity constraints - sales, training, intgration testing and developer support. These are very real constraints since the Gorilla gets to decide what value chain partners are integrated and in what order. This is the source of their power not manufacturing constraint. The Gorilla can say - bring me a $10M deal and I will make our product work with your product.


From ARM's point of view, all these areas are their bowling alley pins and the tornado, for them, could be just the shipment of ARM cores in general.


If unit shipments and revenue we're the only issues I might agree with this statement. However, since competitive advantage is a real issue and value chain power is a termendous source of advantage, I will disagree. The dynamic Moore talks about is when the end user market demand is so high that shipping as fast as possible is key and market share is the goal. The Gorilla has the power to decide who will win market share just by deciding who to train and support first. If you need special features in the Gorilla's product, it will cost you.


The growth year over year is impressive. The market share of cell phones is impressive. Their growth relative to Transmeta and MIPS is impressive.


Hey, I love Arm and I think it is a close to Gorilla status as any company AND they have an awesome business model. However, it is possible to misexecute the change from bowling alley strategy to tornado strategy and another technology could be choosen. The entire world of configurable computing and silicon IP is very active today. There is probably more innovation happening in this area than any other field of technology right now.

I think we agree more than disagree and I think that the future of Arm is very bright but keep your eyes on the programmable logic folks.

I enjoyed your comments.

Paul

p.s. I STILL want broadcom to switch from MIPS to ARM. Oh well I don't often get what I want.








Print the post Back To Top
No. of Recommendations: 2
Hi Paul:

>>>Are you suggesting that Moore's market development model does not apply in the embedded processor world?

No. What I'm saying is that each case is unique unto itself and that we have to be careful when we try to apply Gorilla logic to new scenarios. Moore's logic comes from selling high tech in Silicon Valley in the '80s and '90s. Crossing the Chasm is copyrighted in 1991. A decade is almost a lifetime in Silicon Valley. 20 years ago there were no "merchant" fabs. The new Moore mantra (Living on the Fault Line) is to stick to the core and to outsource the context. For Intel and ARM, fabs are context, NOT core. Today ARM does it right and Intel wrong! BTW, for RF Microdevices, the fab might be core because they deal with exotic substrates, not plain silicon.

>>>In the early days of the 6502 and CP/M...

6502 early days? In the early days I programmed vacuum tube based computers (IBM 650 :-))

>>>Arm and Microsoft have capacity constraints...

During the Tornado, the demand is so great that you can hardly cope with it. Moore says to ignore the customer and just ship, ship, ship and let the devil take care of the hindmost (it's market share, stupid! :-). When Intel's fabs are running at 100% on three shifts, that's the max they can produce. If a customer is left waiting, he can go to AMD or some other fab. Arm and Microsoft never have that kind of constrain. "Training, integration testing and developer support" can wait until the Tornado slows down and it becomes imperative later on.

>>>This is the source of their power not manufacturing constraint.

The Gorilla power comes from "Open Proprietary Architecture with High Switching Costs." Once you are locked in, you are locked in.

>>>The Gorilla has the power to decide who will win market share just by deciding who to train and support first. If you need special features in the Gorilla's product, it will cost you.

Sorry, Paul, but here you are missing the boat. The market share that the Gorilla worries about is his own market share, not that of the various members of the value chain. The only function of the value chain, from the Gorilla's point of view, is to hand off to them the less productive aspects of building the "whole product." To use Intel as an example. First they only built the chips and the value chain put them on motherboards and put the motherboards in pretty boxes. When Intel felt stronger, it started assembling the motherboards themselves and the value chain could only install these motherboards in the pretty boxes. A big chunk of added value was shifted from the value chain to the Gorilla. Compaq complained bitterly but had no choice but to accept. That is Gorilla Power!

>>>but keep your eyes on the programmable logic folks.

Nick Tredennick of Dynamic Silicon, designer of the one of my favorite chips, the 680x0, agrees with you on that! He thinks that ARC is a real threat to Arm.

>>> I enjoyed your comments.

Same here! I was looking for a forum where I could have intelligent chats with knowledgeable people about investing in technology. I think I have found them right here! Thanks!
Print the post Back To Top
No. of Recommendations: 0
Hi Matt:

I'm very glad to be here. The idea of putting the hot links in the PDF document is brilliant! I'm working on a list of Gorilla candidates and I will eventually present it here. It will have some candidates that are missing on Paul's list. I'll post the list as soon as I can put my thoughts on "paper."

Regards.

Denny
Print the post Back To Top
No. of Recommendations: 0
I disagree with almost everything you said with regards to Gorilla Gaming / TALC marketing. My take - I think you are refering to the bullet point outcomes, I was refering to the process that leads to Gorilla power. Since I am writing an article for the August RTW you will have to wait until then before getting a shot at my thinking on the subject.


Sorry, Paul, but here you are missing the boat.


Why so harsh?


He thinks that ARC is a real threat to Arm


and Clayton Christensen thinks ARC is disrupting the entire microprocessor industry.


Paul

Print the post Back To Top
No. of Recommendations: 1
>>>The Gorilla has the power to decide who will win market share just by deciding who to train and support first. If you need special features in the Gorilla's product, it will cost you.

Sorry, Paul, but here you are missing the boat. The market share that the Gorilla worries about is his own market share, not that of the various members of the value chain. The only function of the value chain, from the Gorilla's point of view, is to hand off to them the less productive aspects of building the "whole product." To use Intel as an example. First they only built the chips and the value chain put them on motherboards and put the motherboards in pretty boxes. When Intel felt stronger, it started assembling the motherboards themselves and the value chain could only install these motherboards in the pretty boxes. A big chunk of added value was shifted from the value chain to the Gorilla. Compaq complained bitterly but had no choice but to accept. That is Gorilla Power!



Sorry for jumping into the middle but I just wanted to mention something about the above quote. I usually don't like to make assumptions regarding what points people are trying to make but since Paul and myself are often on the same wavelength I'll give it a shot (Paul, please correct me if I'm wrong). I think that Paul's statement of how the Gorilla's power over a value chain allows it to control the destiny of the other members is just meant to demonstrate true Gorilla Power, which is exactly what you (Denny) are alluding to with the Compaq example. In essence you're both coming to the same conclusions from different angles. But then again, I could be wrong :^)

I'm very glad to be here. The idea of putting the hot links in the PDF document is brilliant! I'm working on a list of Gorilla candidates and I will eventually present it here. It will have some candidates that are missing on Paul's list. I'll post the list as soon as I can put my thoughts on "paper."

Sounds great, I'm looking forward to it! Make sure you come up with a cute name for the list like Paul's XaosZoo Index.


Matt
Print the post Back To Top
No. of Recommendations: 3
I think we agree more than disagree and I think that the future of Arm is very bright but keep your eyes on the programmable logic folks.

If the ASIC-integration trends in end markets transition to adoption of programmable logic cores then ARM needs to make sure some programmable logic vendor has their cores for integration (I haven't kept up on this so they may already have this type of arrangement). They could still thrive in this environment, I think (unless you meant something else?).

My take is that the hardware for platforms trends toward a commodity as an industry develops, in that the designs are just building blocks on a printed circuit board. The value-added is in the overall system design, in the silicon (designed by suppliers) and in the software.

Networking and cell phone markets may not be there yet but the forces of designing, maintaining and supporting these designs will almost certainly drive them in this direction (network processors, for example). In this situation you want to minimize the cost of the hardware and the effort needed to develop it, and put your effort into where the value is added -- which is typically the software/system design. Programmable logic is usually more expensive (recurring cost) than a custom solution.

I don't see programmable logic supplanting ARM-type cores in this environment, although there will always be those new emerging markets where the system manufacturers need to develop new, custom solutions themselves. But you're right, programmable logic is compelling to use for a large number of reasons and is a potential threat.
Print the post Back To Top
No. of Recommendations: 0
Matt

You are right on, as usual.

Paul
Print the post Back To Top
No. of Recommendations: 1

If the ASIC-integration trends in end markets transition to adoption of programmable logic cores then ARM needs to make sure some programmable logic vendor has their cores for integration (I haven't kept up on this so they may already have this type of arrangement). They could still thrive in this environment, I think (unless you meant something else?).

...

I don't see programmable logic supplanting ARM-type cores in this environment, although there will always be those new emerging markets where the system manufacturers need to develop new, custom solutions themselves. But you're right, programmable logic is compelling to use for a large number of reasons and is a potential threat.


The real question:


"Is Programmable Logic a Disruptive Technology?


I am being paid to answer that question so I can't voice an opinion but I do know that Clayton Christensen was engaged by Applied Materials to answer this very question and his answer is yes.

I think it is a very interesting question to ask and the journey is a bit of a 'through the looking glass' experience.

Paul


Paul


Print the post Back To Top
No. of Recommendations: 1
The real question:

"Is Programmable Logic a Disruptive Technology?

...but I do know that Clayton Christensen was engaged by Applied Materials to answer this very question and his answer is yes.


I liken programmable logic to the introduction of the microprocessor. You have a more complex, more expensive device -- but the uses for that device are endless, compared to an existing solution. You can do one design and then make that design do a bunch of wonderful things.

Say for the sake of argument that programmable logic is disruptive (I'm not arguing that it isn't). The question is whether or not value chains will develop in markets that push out existing solutions (processors, custom ASICs, etc.) and drive the performance and cost of implementing PL solutions down, which will then potentially fuel the demand in other markets. (Oh, and by the way, these solutions are on the architecture and uses the tools that are proprietary to the PL supplier, right?!) As I've said, I know these markets exist but I haven't seen the cost/performance become competitive with alternative answers. But I have been extremely impressed with the level of integration and features that are available -- so there is progress in the required direction.

Print the post Back To Top
No. of Recommendations: 0


Say for the sake of argument that programmable logic is disruptive (I'm not arguing that it isn't). The question is whether or not value chains will develop in markets that push out existing solutions (processors, custom ASICs, etc.) and drive the performance and cost of implementing PL solutions down, which will then potentially fuel the demand in other markets. (Oh, and by the way, these solutions are on the architecture and uses the tools that are proprietary to the PL supplier, right?!) As I've said, I know these markets exist but I haven't seen the cost/performance become competitive with alternative answers. But I have been extremely impressed with the level of integration and features that are available -- so there is progress in the required direction.


Go ask Alice!

<-;

Paul
Print the post Back To Top
No. of Recommendations: 1
Paul:

>>>I disagree with almost everything you said with regards to Gorilla Gaming / TALC marketing.

Your privilege, I'm sure!

>>>Sorry, Paul, but here you are missing the boat.
>>>Why so harsh?

No offense meant. I am by nature not Politically Correct. Is PCness a requirement here?

In my 40 year professional life I have always encouraged no-holds-barred discussions. By this I don't mean that I like or encourage personal offense. What I do mean is that people should speak their mind in the understanding that what is under debate is the issue and not the people involved. When I say "Sorry, Paul, but here you are missing the boat" what I mean is "In my opinion, Moore said X and you are interpreting it as Y." If I'm not allowed to say that here, I don't want to take part in the discussions.

Regards from,
Denny Schlesinger
Caracas - Venezuela
[email protected]
Print the post Back To Top
No. of Recommendations: 0
For Intel and ARM, fabs are context, NOT core. Today ARM does it right and Intel wrong! BTW, for RF Microdevices, the fab might be core because they deal with exotic substrates, not plain silicon.

I think you are wrong about Intel. Given the complexity of the stuff they make fabs are CORE because they get their huge cost benefit from being able to use the latest fab technology before anyone else. Of course they have to get the bugs out but thats the point. Intel have an institutional body of knowledge about how to make successful fabs which others simply don't have.


Nick Tredennick of Dynamic Silicon, designer of the one of my favorite chips, the 680x0, agrees with you on that! He thinks that ARC is a real threat to Arm.

Take a look at the ARM board on TMF UK - lots of people have strong opinions about ARC and its future viability

DD
Print the post Back To Top
No. of Recommendations: 1

No offense meant. I am by nature not Politically Correct. Is PCness a requirement here?


I am not offended. I prefer arugments be about companies, markets and technology. When arguments get personal people tend to take positions and thoughtful discussion stops.


I have always encouraged no-holds-barred discussions.


Amen. Read my posts. I am sure you find many no-holds-barred discussions. I am not interested in anything else.


"Sorry, Paul, but here you are missing the boat" and
"In my opinion, Moore said X and you are interpreting it as Y."


These are not the same statements. I cannot learn for or refute the first. I may learn from or refute the second.


If I'm not allowed to say that here, I don't want to take part in the discussions.


Always your choice but this statement is not consistent with your stated preference for no-holds-barred discussion. Hopefully you will continue to contribute. I am particularily interested in the Gorilla candidates you mentioned to Matt.

Paul





Print the post Back To Top
No. of Recommendations: 0

Given the complexity of the stuff they make fabs are CORE.


I am going from memory but I remember reading somewhere recently that Intel would be in production in 2005 with 450mm wafers and 0.05 micron geometry with SOI interconnects. Cost of building the fab - $5B - if they can get the lithography to work.

Paul
Print the post Back To Top
No. of Recommendations: 0

Take a look at the ARM board on TMF UK - lots of people have strong opinions about ARC and its future viability


DD - could you cross post a couple of the relevant links when you have a moment?

Thanks,
Paul
Print the post Back To Top
No. of Recommendations: 3
Matt:

I think Paul and I are talking about different things. When Paul says "the Gorilla gets to decide..." then the company he is talking about is already a crowned Gorilla. As I understood his position on Arm, he thinks that Arm is not yet a Gorilla and that is why I am talking about the Tornado that might crown Arm as Gorilla. For me, the point under debate was: "What do you do in a Tornado?" My answer: "You ship."

Since people here are worried, and rightly so, that Moore's words are being twisted out of context, allow me to quote from Inside the Tornado:

===================
"The Significance of the Tornado

"In sum, the tornado is a hugely significant time. It is also a bit confusing for the marketing department. They are used to defining their value to the company as creating demand, but in a tornado there is no need for this service. So what should the marketing strategy be now?

"To put it as succinctly as possible -- just ship!

"Don't segment. Don't customize. Don't commit to any special projects. Just ship. It is like a sardine run. You don't bait hooks, you just stick in your bucket, pull them out, and go back for more.
===================

As you can see, there is no mention of anything but ship, ship, ship and do it once again and do it till you drop. Your future depends on what you manage to ship. If you win, you are the Gorilla. If you don't win you are either chimp, monkey or broke.

How does this apply to Arm? The reason why I mentioned that the concepts behind the Gorilla Game are from the Microcosm and that they are over ten years old, is precisely because it is difficult to apply them directly to new situations, in this case to a fabless chipmaker that is inside not just PCs but inside everything mobile and embedded. There was no such thing (as far as I know) before 1991 when Moore wrote the Chasm. For Intel, the x86 was a single paradigm product and all PCs could tornado at the same time. That is not likely to happen for Arm because it covers so many diverse markets which will each tornado it its own good time. As a matter of fact, some of the tornados need to be sequential, for example G2 and G3 cell phones.

Since there won't be one tornado but many, I think it's right to talk about core shipments in general as a measure of Gorillaness. But then, it's only my opinion.

Regards from,
Denny Schlesinger
Caracas - Venezuela
[email protected]
Print the post Back To Top
No. of Recommendations: 2

DD - could you cross post a couple of the relevant links when you have a moment?


The thread ARC Warning - starts at http://boards.fool.co.uk/Message.asp?mid=6657682&sort=threaded

Also not completely OT for ARC but interesting is this recent post by Scollag
http://boards.fool.co.uk/Message.asp?mid=6656844

Also the posts around http://boards.fool.co.uk/Message.asp?mid=6624221
(Why ARC is not ARM)

Search for posts by troglodytes, ptoboley, simoan, scollag and license.

I'll summarise one reason why ARC/programmable logic isn't a threat:- ARM gives you very programmable logic very efficiently itself. And because it enforces standardization you get consistency and portability.
An ARM7 thumb core is minute, low power, memory efficient and probably already a reference desing doing 50% of what you want.

Consistency/portability is a huge bonus as you well know - think back to how it was IBM PC compatibles that won not computers running MS-DOS

DD
Print the post Back To Top
No. of Recommendations: 1
Paul:

>>>These are not the same statements.

Understood, sorry.

Denny Schlesinger
Caracas - Venezuela
[email protected]
Print the post Back To Top
No. of Recommendations: 0
Another interesting thread on the ARM UK board

http://boards.fool.co.uk/Message.asp?mid=6663690&sort=threaded

DD
Print the post Back To Top
No. of Recommendations: 0
DD:

Thanks for the links. Very interesting. The arguments have the feel of disruptive technology type arguments. However, I don't think that the embedded processor market is being overserved today. Configurable computing / PLD is a trend worth watching but it's time is not yet come.

Paul
Print the post Back To Top
No. of Recommendations: 0
Denny,

As an innocent bystander, (okay, so maybe I'm a guilty bystander), when I read your and Paul's posts in your recent discussion I get the impression that there is very, very little disagreement about anything substantive. When seeing online discussions such as the one you're having, I imagine that if the two of you were able to sit at the corner table away from the crowd sipping wine, latte, or whatever for an hour or so, you'd find that the advantages of talking face to face would whet out any apparent disagreement seen in your posts.

I think it's right to talk about core shipments in general as a measure of Gorillaness.

I do to, but I wouldn't overly emphasize them. Shipments are a quantifiable symptom. They are to a large extent (though not entirely) the result of Gorilla-like characteristics rather than being characteristics that are the driving factors of the company's long-term success. To that extent, when I pull up my chair to your corner table, I sit on Paul's side of the table.

--Mike Buckley
Print the post Back To Top
No. of Recommendations: 1

when I read your and Paul's posts in your recent discussion I get the impression that there is very, very little disagreement about anything substantive.


I say potato, he say potato. Uhmm, it works better as a song. I do agree that Denny and I agree.


Shipments are a quantifiable symptom. They are to a large extent (though not entirely) the result of Gorilla-like characteristics rather than being characteristics that are the driving factors of the company's long-term success.


Well said Mike. For the audience, this is one of the moments of agreement between --MB and [email protected] that has taken a couple of months to reach. As usual, Mike put the puck in the net and I get to take credit. Life is good.


To that extent, when I pull up my chair to your corner table, I sit on Paul's side of the table.


I appreciate your joining our conversation but you realize we have been at this all night and you are buying the first round. I am sure DD can suggest a nice little spot in his home town.

[email protected]



Print the post Back To Top
No. of Recommendations: 0
For the audience, this is one of the moments of agreement between --MB and [email protected] that has taken a couple of months to reach.

Really? I'd be willing to bet that it wouldn't have taken more than ten minutes at the corner table. The first round is on me, but be careful about all the subsequent rounds; my wine is a lot more expensive than your expresso.

--Mike Buckley
Print the post Back To Top
No. of Recommendations: 0
my wine is a lot more expensive than your expresso.

It's also more expensive than your espresso.

--Mike Buckley
Print the post Back To Top
No. of Recommendations: 0

I appreciate your joining our conversation but you realize we have been at this all night and you are buying the first round.

I think you need to go to bed. I'be given up trying to guess which timezone your in - you've bene responding to all my posts which indicates Europe but you appear to contine to do more when I'm safely tucked up in bed as well!!!


I am sure DD can suggest a nice little spot in his home town.


Indeed. I suggest Morrisons (just off the Rue d'Antibes - http://www.mapblast.com/myblast/map.mb?CMD=GEO&AD2=rue+d%27antibes&AD4=FRA&AD2_street=rue+d%27antibes&AD3=cannes&x=49&y=15)

DD
Print the post Back To Top
No. of Recommendations: 0
DD:

>>>I think you are wrong about Intel. Given the complexity of the stuff they make fabs are CORE because they get their huge cost benefit from being able to use the latest fab technology before anyone else.

You are in good company with that opinion, George Gilder agrees with you. IMO, Intel's business is (or should be) the IP part of it and they should outsource the actual making of the chips. No reason why they cannot have the outsourced fab do exactly what they do in their own fabs. The real experts in the field are companies like Applied Materials and KLA-Tencor.

Intel not only has their own fabs for making the chips, they also assemble the chips on motherboards, having taken that business away from their customers, from their own value chain. But how good was that for Intel shareholders? Have a look at this interesting comparison:

DELL, the king, outperforms INTC, the gorilla.
http://cchart.yimg.com/z?s=intc&c=dell,cpq&a=v&p=s&t=5y&l=on&z=m&q=l

and this one
http://cchart.yimg.com/z?s=intc&c=emc,aol,msft&a=v&p=s&t=5y&l=on&z=m&q=l

IMO, had Intel stuck to the core, as Moore suggests, their stock price performance would have been better. Of course, we will never know!

>>>Take a look at the ARM board on TMF UK - lots of people have strong opinions about ARC and its future viability

Biased, I'm sure, but interesting. Thanks!

Denny Schlesinger
Caracas - Venezuela
[email protected]
Print the post Back To Top
No. of Recommendations: 2
Mike and Paul:

>>>Shipments are a quantifiable symptom.

OK, so we are splitting hairs. (BTW, make mine... well, that depends in which country this table sits at: in Vienna, coffee; in Lisbon, house wine; in Casablanca, arak...)

But let me tell you an interesting opinion a local doctor has. "Until you show symptoms that bother you, you are not sick." We agree that the underlying conditions for gorillahood exist. The only question left is whether there has been or not a tornado. Intel sells to box makers who sell to the public. If all box makers acted like DELL, there would be an almost one-to-one correlation between customer uptake and Intel shipments. With Arm, the chain is different and longer. Arm does not have fabs. The fabs might or might not belong to the "box" makers. This means that there are one, two, or maybe three extra steps in getting the Arm core to the consuming public. When the consuming public buys widget X with an Arm core inside, I can't tell if that is Arm's tornado or someone else's. Maybe there is Pink Hoola Hoop tornado going on that just happens to have Arm inside but it's the Pink part that's the rage. This uncertainty leads me to look for more reliable symptoms! The symptom I particularly like, besides core shipments, is the long list of Arm associates, partners and customers.

So, where are this drinks?

Denny Schlesinger
Caracas - Venezuela
[email protected]
Print the post Back To Top
No. of Recommendations: 1
No reason why [Intel} cannot have the outsourced fab do exactly what they do in their own fabs. The real experts in the field are companies like Applied Materials and KLA-Tencor.

Especially in the earlier days, the amount of cash Intel had (has) that the fabs didn't have is one reason.

Intel not only has their own fabs for making the chips, they also assemble the chips on motherboards, having taken that business away from their customers, from their own value chain.

For me, the crux of the matter is whether or not Intel's in-house fabrication was (is) a vital component of their susatinable, competitive advantage. If the answer is "yes," the fab is part of Intel's so-called core. If the answer is "no," it's not part of the core.

--Mike Buckley

Print the post Back To Top
No. of Recommendations: 2
Denny,

What is arak? I travelled all over Morocco years ago and have never heard of it. I need to know whether or not I can afford to join you at the corner table. :)

But let me tell you an interesting opinion a local doctor has. "Until you show symptoms that bother you, you are not sick."

Fire the doctor. Millions of people with life-threatening high blood poressure and cancer have no bothersome symptoms. They wouldn't no they were sick based on that criterion.

Maybe there is Pink Hoola Hoop tornado going on that just happens to have Arm inside but it's the Pink part that's the rage. This uncertainty leads me to look for more reliable symptoms!

I completely understand your point (it's a good one) and I comisserate with the problem of identifying the symptoms of a Gorilla. But I wouldn't call the issue splitting hairs mostly because I think it's dangerous to use anything other than revenue as the quantitative symptom. In other words, quantitative symptoms other than revenue can be misleading because if they might impact the top line in a way befitting of a Gorilla.

Let's use Gemstar as an example, mostly because I'm familiar enough with its business to discuss it. A lot of people want to use the number of set-top boxes conatining their interactive program guide as the indicator of a tornado. If I remember correctly there were fewer than 2 million of them at the beginning of last year. By the end of March of this year there are already more than 12 million. Based on that historical information, some people want to proclaim a tornado and thus, Gemstar, as the Gorilla. However, looking at the revenue Gemstar recieves from that part of their business, it would be very, very difficult to justify calling the tornado.

Looking forward, there are expected to be 20 million of them by the end of the year and Wit-Soundview estimated in May that there will be 40 million of them by the end of next year. The tornado will most likely blow soon, but I don't think it's there yet. And if there's no tornado, by definition there's no Gorilla. Therefore, counting set-top boxes instead of revenue is very misleading.

Similarly, we have to be very careful about the metrics we use and how much emphasis we give to them in identifying the relative strength of Gorillas and whether or not they really have become a Gorilla.

--Mike Buckley
Print the post Back To Top
No. of Recommendations: 0
Sheesh! Do I get a second try at one of those sentences?

In other words, quantitative symptoms ...

Change that sentence to "In other words, quantitative other than revenue can be misleading because they might not impact the top line in a way befitting of a Gorilla."

Sorry about that.

--Mike Buckley

Print the post Back To Top
No. of Recommendations: 1
Mike:

Arak is a strong aniseed flavored drink called anis in Spain or pastis in France.

Gorilla hunting is very difficult because as the world moves on the gorilla terrain changes. What was fairly easy to do for Intel is more difficult for Arm because the business model is different. As much as I like to stick to the original definition of the wonderful words that Moore coined (he used to be an English professor before becoming a marketing consultant) at times you have to play it by ear. After all, the final purpose of the game is to make money in the market.

Denny Schlesinger
Caracas - Venezuela
[email protected]
Print the post Back To Top
No. of Recommendations: 0
Another ARM concern:

With MSFT, MSFT was able to leverage the OS into applications and move into to dominate other sectors like servers; INTC just kept taking over more and more of the PC board utilizing its gorilla position to do so. It then took this experience and moved into servers et al; Cisco, simialr story. BEAS, assuming it is a gorilla and I am beginning to have my doubts given Schwab's use of both IBM and BEAS' web-servers and the tightening of marketshare, can leverage its product to expand outwards.

Where does ARMHY leverage its core to expand outward and collect more and more value?

Tinker
Print the post Back To Top
No. of Recommendations: 2

Where does ARMHY leverage its core to expand outward and collect more and more value?


Here is one -

ARM Holdings has licensed Sun Microsystems' Java technology, a formalization of an existing effort to spread Java to cell phones and other small mobile devices.
ARM isn't the first chip company to sign a Java license. Zucotto is working on the J2ME project; Zucotto is a licensee and has passed Sun's Java compatibility test for J2ME, the company said.

But whereas Zucotto is a start-up, ARM already has a huge presence in the cell phone market. The company licenses its chip designs, called "cores," to other companies that add to them and take care of manufacturing them.

ARM licensed the Java technology because its customers desire the technology, said ARM General Manager Reynette Au. NTT DoCoMo sells about 60,000 Java phones a day, and cell phone giant Nokia expects to sell 100 million Java-enabled cell phones by the end of 2003.

Paul

http://news.cnet.com/news/0-1003-200-6375791.html?pt.fool..nefeed
Print the post Back To Top
No. of Recommendations: 1
A retrospective view

http://boards.fool.co.uk/Message.asp?mid=6666261

And yes I suppose the same could happen in reverse

DD
Print the post Back To Top
No. of Recommendations: 0
A fairly good articulation of the bearish case for ARM - http://boards.fool.co.uk/Message.asp?mid=6670015

Also worth reading the posts following

DD
Print the post Back To Top
No. of Recommendations: 0
DD,

Given your networking expertise what do you think of Micromuse? I've followed them since the middle of 1999. These guys seem to dominate their market to the extent (and please gorilla gamers withhold your sighing because this is a true rarity) that they have become known as the "industry standard" with what looks like inhouse solutions being the only major alternatives.

I know MUSE has their own "Center of Excellence" type of program going where firms as diverse as Cisco to never heard of software companies collaborate to leverage the Micromuse solutions.

What do you think if these guys as having a lock on the installed base and ability to leverage this going forward?

I ask because Micromuse, although not cheap in a "value" sense is now cheap in a growth sense, and I sense it ready to pop, assuming it doesn't lay an egg when it reports earnings on Wednesday.

Tinker
Print the post Back To Top
No. of Recommendations: 0

Given your networking expertise what do you think of Micromuse? I've followed them since the middle of 1999. These guys seem to dominate their market to the extent (and please gorilla gamers withhold your sighing because this is a true rarity) that they have become known as the "industry standard" with what looks like inhouse solutions being the only major alternatives.


About the only thing I can sat about Micromuse is that they are a network management company. Very few network management companies (if any) have been successful long term because fundamentally you have to keep on adding support for new things while maintaining compatibility with old things and people won't pay enough to let this be done at a profit.

I know MUSE has their own "Center of Excellence" type of program going where firms as diverse as Cisco to never heard of software companies collaborate to leverage the Micromuse solutions.

If this means that micromuse have managed to outsource the development of new capabilities and support of new thingys then the fundamental problem I mentioned above doesn't apply.

Sounds worth a little more investigation

DD
Print the post Back To Top
No. of Recommendations: 2
Tinker,

I finally remembered what worried me about MUSE. Their dependance on SUNW. It's been quoted that 70% of MUSE customers have deployed NetCool in Sun environments.

Given that Microsoft are making headway into Linux marketshare, not to mention Intel too in server solutions, I think MUSE could be threatened here. The slowing growth for servers in general leaves much to be desired. Many companies are now daisy-chaining Intel servers together to provide the redundancy of Sun servers for a fraction of the cost.

Also, I think BEAS's middleware solution could eventually make MUSE's software effectively worthless?

Steve
Print the post Back To Top
No. of Recommendations: 1
Thanks Steve,

It did raise a flag when I saw MUSE trumpeting their joint deal with SUN only for me later to find out that SUN is practically the company in terms of distribution. However, like BEAS, a slow down in Sun servers does not necessarily indicate a slowdown in applications sold for those servers, although of course its better when they are selling like hot cakes (hmmmmm, hot cakes...dang, forgot dinner again:( )

In regard to BEAS and MSFT however, I don't see how that would affect MUSE. What MUSE does is very complicated. I can see it maybe running on top of BEAS platform but BEAS is not going to emulate anything close to what MUSE does in the network. It also is not a place that you'll see MSFT moving into. So in regard MUSE would need to make its software cross platform to avoid any of these issues.

It is the equivalent of I2 software running on top of BEAS or NT. Neither company is gonna replace I2's supply chain solutions.

But I only bought a small portion of MUSE today in my IRA account. An account I just started two months ago for tax purposes for 2000. So playing with a smaller risk there. I did, however, buy a lot more OPWV today. The timing may turn out to be wrong in the short-term as we seem to be in the lull before 2.5g starts to roll-out in Europe. However, even Motorola sounded positive about phone sales going forward. So I am excited about that purchase and will have to dig deeper to really find out why. But it just feels good. If it feels good just do it! {okay, I'm not from the 60s generation but sometimes it just feels good}.

Tinker
Print the post Back To Top
No. of Recommendations: 0
Tinker,

Thanks for the response. It cleared up a lot in my mind. Sorry about dinner, try a Hot Pocket. :)

I'm not particularly up to speed on MUSE, but I will do some research on it and let you know what I think. With regards to BEAS and MSFT - thank you for the clarification.

It still bothers me about the SUN server issue, the whole DSL CLEC thing was downright nasty. Now I'm not saying Sun will go out of business, far from it, but selling into a slowing market, where margins could potentially drop through the floor due to intense competition is not good. Sun is not in a good place right now. However, MUSE could do fine here, I'm just not sure how they will benefit if the server market collapses.

Not pretty:

http://search.ebay.com/search/search.dll?MfcISAPICommand=GetResult&pb=&ht=1&st=2&query=Sun+Server&SortProperty=MetaEndSort

Also, where is the rebound in server buying? Look at EXDS and MFNX - both are nearly bankrupt. I do not even want to think about how many Sun servers those guys bought. The web hosting industry in general is on the ropes, and you can bet those guys bought millions and millions of dollars worth of servers.

MUSE may be a great company, and I will look into it more deeply, but Tinker - their customers...eh?

Looking forward to your thoughts,
Steve
Print the post Back To Top
No. of Recommendations: 2
MUSE may be a great company, and I will look into it more deeply, but Tinker - their customers...eh?

Muse's customers are basically all of the telecommunication companies in the world, all of the networking companies in the world, and a good portion of the major corporations in the world.

MUSE grows as the IP network grows. It is said by analyst's that VoIP is another great growth area for MUSE given the QoS issues involved. Wireless build-out is another such area.

Given the not-so-subtle slowdown in telecom there certainly is some much deserved caution in regard to MUSE. So far they have continued growing even into the slowdown. Argument A is that this will continue as networks scramble more and more to make shift solutions in this period of capital spending slowdown and compete on things like QoS. Muse's product in the network is essential to this strategy. It is also essential to the integrating of networks, like if AOL cable properties were to buy AT&T broadband and needed to hook them together.

This said, it is hard to predict exactly what will happen. The telecom slowdown is so vast that even the worse case scenarios have been breached.

But I'm playing the fact that I think the worries are over done and the stock will pop nicely in regard. If I'm wrong MUSE is still as well positioned as any other company out there to profit over the coming years from the network build-outs, integrations, expansions, wireless, VoIP, etc.

Anyway, my reasons for buying MUSE both short and long-term. I was trying to find out if anyone had any information about what sort of gorilla lock MUSE might have in the network. It seems to me it might be a bit like Mercury (which does QoS for Internet sites) because once Mercury's products are in the network and running switching costs become very high. I am assuming this is the case with MUSE in telecom networks but I don't really know beyond this comparison.

Tinker

Print the post Back To Top
No. of Recommendations: 0

How would Openwave fare in this scenario? I feel that this is probably the most likely one given all the problems that people are having rolling out a true 3G wireless world.

All in all, a really awesome article. Thanks to imuafool for the orginal link.

http://www.techreview.com/magazine/jun01/print_version/knorr.html

Looking toward the end of the decade, you may end up using 2.5G wireless for convenient cell-phone calls and Web access while traipsing around town, then cut over to a fixed-wireless network when you step into the coffee shop, subway station or meeting room, perhaps using a software-radio Web phone that switches between voice and data as needed. Flip open your 2.5G CellMate while walking down Main Street to call home, then convert to data mode to download a shopping list after your spouse tells you about a sudden party you didn't know you were hosting. When you step into Mammoth Grocery, CellMate switches over to the store's fixed-wireless network so you can quickly check Online Wine to see which vintage will complement dinner. The store's map appears, leading you to the wine aisle. You point CellMate at the checkout's infrared scanner to debit your bank account. And that indoor/outdoor hybrid system, rather than the grand vision of "3G," might be what the future really looks like for broadband wireless.


I think everyone will agree that the first generation of WAP was a dismal failure. Openwave and Co. have acknowlaged this recently and has begun work on second generation WAP.

http://www.thestandard.com/article/0,1902,27353,00.html

This will push us towards the scenario in the the first link I believe. I just don't think high speed data services through the cell phone will be successful.

It's the videophone all over again, part of the reason we own a phone is so we don't have to talk to the person face to face, or even see the person - so if we are having a bad hair day, nobody notices. :)

Just my 2cents,
Steve
Print the post Back To Top
No. of Recommendations: 2
Yep, nobody knows for sure how it will go. It is consistent with my theory going forward on the network that voice and data, instead of becoming one and th same, may actually be separated out. The cost of doing voice is greater than the cost of doing just data due to QoS requirements. Except for the most particular of markets most people do not need voice quality data on the Internet. Removing the need for this millisecond or less QoS that voice requires can dramatically reduce the cost for data networks and enable data networks to be created in much more efficient architectural designs. Such as the stuff players like Cogent and its ilk are doing. Combining voice and data bring on necessities like SONET.

But, I must say, however, that 802.11a and b do have some short-comings that are material. As an example, they require direct line of sight. Therefore, no data in the automobile whether your pulled to the side of the road or moving. No data while walking the dog. And, due to the increased complexity in a phone, it will create more equipment expenses. These equipment expenses will not be subsidized by the communication provider because he will have no incentive to give away free phones with 802.11 capability because it competes with him. And finally, 100,000s, even millions of small businesses would have to all invest in 802.11 technology to make it a ubiquitous experience and most will find little to no profit potential in doing so.

Law of economics: If they build it people will come, only works if first they build it. They won't build it unless they can make money from the people that are coming. 802.11a or b may provide extra money for some businesses but for the vast majority it is just an expense. And even for those businesses that might profit from it, a 2.5G or 3G network would be more convenient to use once launched to do the things like ordering your next drink at a restaurant utilizing the phone instead of having to purchase, carry with you, pull out, and then set-up your extraneous 802.11 data device when you could just pull out the phone quickly and hit a few buttons.

Another phone advantage, always on. How many of us even keep our PDAs always on? We don't. But an always on phone creates many advantages for keeping up to date with information. Things as simple as the school forwarding an SMS to your cell phone informing you that Johnny has not been to school today, or is in the doctor's office, to mission critical things like the nuclear plant has had an unexpected coolant link please return from lunch ASAP. A device that is not always on defeats much of the utility of this functioning.

Given these things, if the 2.5g and then 3g networks are enabled the marginal value of 802.11 as a mobile data source loses much of its appeal. As such, given the higher equipment costs, the lack of ubiquity, the redundancy, the lessor utility, etc., if the telecoms put it out 2.5 and 3G will remove the need for 802.11 as a mobile communication mechanism.

802.11, Bluetooth, et al., will still have a place in fixed networking in the home and office however, particularly in regard to removing all of those unsightly wires. And Bluetooth will be incorporate into wireless phones, no doubt.

As for the phone as a viewing device, I highly agree. As of today I have not seen a decent phone. However, this is just a matter of design. Do you want to carry around 2 or 3 pieces of equipment everywhere you go? I don't. And I know I will be carrying around my phone - I have to.

It gets back to a marketing theory I was pushing with eBooks last summer. If we pushed them into the enterprise market as wireless digital document management devices, and thus coerced employees in having to utilize and haul them around, then how long before they got sick of carrying their phone, PDA and eBook (digital document management device)? Not very long, and as such, the PDA could be engulfed by the eBook and a window of extra revenues open up. The same thing between the PDA and phone today, except either way it goes the wireless technology to enable both is virtually the same.

But just some thoughts on the future of wireless.

Tinker
Print the post Back To Top
No. of Recommendations: 1
Steve,

Just read the article after posting. This is exactly the vision we were proposing last summer. Where possible utilize the existing 802.11 as the wireless experience.

Again though, the problem is which device do you use? The cell phone is ubiquitous, tiny, cheap, and always on. The laptop is clumsy, expensive, heavy, not ubiquitous, and not always on. So if not the cell phone its not the laptop. The PDA? Yes, but again, darn expensive. I can't see the articles vision becoming the ubiquitous solution. However, I can indeed see it serving many large niches. There are a lot of compelling mrkets out there for such use and we looked at many of them.

I just don't think that streaming media in a fixed wireless environment is what people want. That is what television is all about.

Again, just thinking about all this from a marketing and product adoption perspective. But indeed fixed wireless does have many compelling uses and will be available many places because of this utility. However, I don't think it replaces the more ubiquitous and more convenient data access for carriers through 2.5G and 3G and I don't think it cannibalizes 2.5g or 3g either. Rather, it will be used for specific computing purposes in different environments for applications far removed from what people will use their phones for.

Tinker
Streaming graphical media is not one of the killer apps of wireless data IMHO anymore than it is one of the killer apps of the wired Net, other than the very practical utility of voice streaming media and accompanying charts and diagrams.

So when thinking wireless net, think simple, think practical, think what is the utility. Bells and whistles are a long way off and not something most people desire to pay for on a phone like, PDA like, wireless device.

Print the post Back To Top
No. of Recommendations: 1
Tinker:

It is consistent with my theory going forward on the network that voice and data, instead of becoming one and the same, may actually be separated out. The cost of doing voice is greater than the cost of doing just data due to QoS requirements. Except for the most particular of markets most people do not need voice quality data on the Internet. Removing the need for this millisecond or less QoS that voice requires can dramatically reduce the cost for data networks and enable data networks to be created in much more efficient architectural designs. Such as the stuff players like Cogent and its ilk are doing. Combining voice and data bring on necessities like SONET.

Disagree. The original intent of the Internet and TCP/IP was survivability, DARPA wanted a system that could survive atom bomb attacks. That was achieved by allowing packets to wend their way through whatever was left standing of the net. When VoIP was invented, QoS did become a problem because packets had to arrive in proper sequence and timeframe, something not envisioned in the original packet switching. But a funny thing happened on the way to the third millennium, a highly reliable all or mostly optical transport system was put in place, a system that does not drop packets. Once packet switching is married to lambda transport, the QoS thing will be a dim memory.

So yes, QoS is a short or medium term problem that will simply disappear long term.

Comments?

Denny
Print the post Back To Top
No. of Recommendations: 0
p.s. I STILL want broadcom to switch from MIPS to ARM. Oh well I don't often get what I want.

It may be a while.

http://www.mips.com/pressReleases/071701.html

MOUNTAIN VIEW, Calif., July 17, 2001 - MIPS Technologies, Inc. (Nasdaq: MIPS, MIPSB), a leading provider of industry-standard processor architectures and cores for digital consumer and network applications, today announced that Broadcom Corporation (Nasdaq: BRCM), the leading provider of integrated circuits enabling broadband communications, has taken a license for several additional MIPS® products, including the new MIPS64™ 5Kf™ core.

The momentum with MIPS in BRCM chips is gathering.
Print the post Back To Top