However, I still think that you're underestimating legacy costs, maintenance costs, and future development costs.</strong><hr></blockquote>
Not underestimating them, merely not getting the conceptual feasibility of a potential mercy-killing of SGI (with considerable benefits for AAPL) confused with the fine-print detail of the operational issues that you describe.
However, as you bring up the subject, let's discuss them:-
The costs of maintaining the legacy environment are dependant on SGI's MIPS-based technology becoming a legacy platform. In that respect, it is no different to Compaq's purchase of DEC and various others (was it Tandem or Stratus, my memory isn't what it was, and I can't be bothered to open another brower window to try and negotiate the labyrinthine "new HP" website) or HP's purchase of Apollo, or indeed SGI's acquisition of Cray.
You could complete and ship any elements of the current software release roadmap (so long as the ROI on that completion justifies the effort), and then go to maintenance releases only for a pre-defined period alongside and then web support only and then it's curtains. This is hardly a new strategy, the entire industry implements it either by degrees or completion. Try getting support for Win 98 from MS' website after 06/30.
The thing is this isn't the market Apple has been accustomed to before xServe: A 12-month warranty doesn't cut it with this class of kit - you buy on-site maintenance with specific service level agreements from the service provider who makes you comfortable and that you can afford. Maintenance is actually a profit centre, not a cost centre.
Actually, because I'm feeling expansive, how's this for a solution: Outsource the maintenance effort to an SGI specialist; or better still, spin out the maintenance assets that you acquire from the purchase of SGI as a management buy-in, or complete MBO and then sub-contract the effort to them.
[quote]<strong>To borrow the usual car industry analogy- Ford bought Jaguar a while ago, and then proceeded to pump billions of dollars into it to build new Jaguar product. It will take decades for Ford to recoup both the initial investment and the development costs of the purchase. It's very easy to argue that this ROI will occur way too slowly, and that the money would have been better spent elsewhere, say on Ford's own product line.</strong><hr></blockquote>
Actually, not quite true: Ford, GM. DaimlerChrysler, VAG (list auto conglomerates until blue) all practice platform commonality and try to use a common parts-bin where possible.
A Jaguar X-type's floorpan is actually derived from one of Ford's CDx models (can't remember which one, and would have to revert to my extensive collection of back-issues of Car to discover which) and the 2.5 V6 engine (and possibly the 2.0 unit) is/are (a) modified Ford Duratec unit. Want to know why the X-type is 4WD, becasue Jag's engineers couldn't easily achieve a rear-wheel only set-up because of the floorpan/drivetrain combination that they were obliged to use.
Jag's S-type is derived from a Lincoln floorpan, and have you ever wondered why Aston-Martin's DB7 and Jaguar's XK8 look so similar.
Don't even get me started on GM (the badge engineer's badge engineer) or VAG (the platform and parts commonality across Volkswagen, Audi, Seat and Skoda shows how it is possible to develop four different model ranges and brand images from simple (!) permutation)
When GM completes a purchase of Fiat Auto in Europe sometime towards the end of the summer, the Signum platform (which I believe is Europe only) will be used in Opel/Vauxhall's (hopefully) highly revised Vectra, Saab's replacement 9-3 and a new Alfa Romeo model (hopefully, without screwing A/R's recent renaissance on the handling front). BTW, Fiat will float the Ferrari group (i.e. including Maserati) before GM can wreak havoc with an art business that has next to nothing to do with accounting.
[quote]<strong>The same argument has been applied to GM's creation of Saturn (the billions spent could have revamped, say, Chevy, and been better spent with a faster ROI). And also the same argument has been used for GM's "roboticization" during the 80s- where they spent in the tens of billions, and could have just outright purchased Toyota for the same price, no doubt a better investment.</strong><hr></blockquote>
I seem to remember that Saturn was GM's attempt to distance itself from some of the product quality issues that were endemic throughout the empire, and an attempted response at delivering Far East-style flexible working practices and quality in a green-field company which would help GM's image in the marketplace. How succesful was it on those terms? I really don't know! (Gosh, something I don't have an opinion on - that's a Kodak moment).
As for the Toyota argument, you say that GM puchasing that producer of mainly dull, but occasionally inspiring vehicles would have been a better investment than GM re-inventing internally! Surely, you're making my case for me - no really, I'm not going to object.
[quote]<strong>For Apple to buy out SGI puts them in the same position- even assuming SGI has begun a migration to PPC (which we don't even know), the costs of migrating SGI's existing user base to PPC haven't even been estimated yet...</strong><hr></blockquote>
And why should I (or even Apple) estimate the costs of a customer's migration to a new platform. Are Apple going to give the new platform's away. Did they pay for my migratiion when they migrated from 68K to PPC? Did Data General pay for its' old 88K Aviion users to migrate to the later Pentium SHV Aviions.
Sounds to me like another profit centre. KCHING!!
[quote]<strong> and the costs of just dumping SGI's user base is larger than the cost of buying SGI and shutting it down completely- since you've now pissed off that user base who now won't become future Apple customers...</strong><hr></blockquote>
Who said anything about dumping them? You encourage them to migrate by demonstrating the increased functionality, usability and ROI potential of the new platform. If they choose not to go, they have to accept that there will come a time when Apple will no longer be responsible for direct support of legacy platforms.
Before EMC shut down Data General for good - having purchased that company purely to exploit its entry-level and mid-range SAN products (because it was cheaper than developing it itself and it eliminated a competitor that was making a loss across the board - HOLD ON, IS THAT MY ARGUMENT IN ACTION AGAIN) - I have sincere doubts that they were still providing affordable support for anything other than SHV systems. 88K, Eclipse/MV, Eclipse, and Nova were probably only supportable from third-party service providers for hardware and from no one at all on software.
[quote]<strong>Worse, assuming Apple wants to "bring SGI's customers back into the fold" (sorry- borrowing from the letter I got from Jobs when Apple bought out PowerComputing: my PowerCenter r0ck3d!), Apple will need to continue a massive investment into future SGI hardware... since the Origin will need a PPC based replacement which can offer at the least, similar performance (at similar price!).</strong><hr></blockquote>
But surely, that's exactly what I'm proposing! Build a G5 Origin architecture which is based on using one or more 4-way "nodes" (the first board can be delivered 2-way as an option and expanded in the field) so that you can deliver a solution which scales from 2 to 512 processors, with a (fair marketprice) SCI upgrade option handling the delivery of NUMA services that becomes non-optional once you decide to tie more than 32 processors/8 nodes together.
Evolve said solution so that it supports the interconnect technology of the day before you then bring out the G6 Origin.
BTW, for all we know and if The Rumour is true, SGI is/has (been) develop(ing/ed) a G5 Origin based on new interconnect technologies anyway and then the key investment is really the one that makes OS XI cope with that as a concept.
[quote]<strong>I can understand the desire to have our hypothetical PPC Origin technology "trickle down" to Apple's consumer line, but it's still a very expensive way of getting next-generation technology.</strong><hr></blockquote>
Never said that to be fair! My "trickledown" theory (how long before someone refers to this as "voodoo company strategy") referred purely software and to the fact that Apple is already using this concept with FCP/iMovie and DVDSP/iDVD.
If AAPL can abstract or abridge the technology of Alias|Wavefront or Shake/Tremor et al so that it can be delivered in multiple and appropriate versions using market/brand segmentation techniques to a variety of audiences (sometimes commercially/sometimes as a complimentary brand/market USP) it can extend the useful life of a product/technology so that the ROI on the core development is that much greater (back to my automotive analogy).
The NUMA technology can only practically exist in the server space, and the graphics/visualization technologies are definitely client space for the most part.
[quote]<strong>Not when companies like nVidia, ATi, Motorola and IBM are happy developing it on their own...</strong><hr></blockquote>
There are very few things/experiences in our industry that cannot be improved by either deploying them and/or using them in a way that never occurred to the originator of the core concept. I believe that I said there are probably few HW faculties (BTW, Programmer, that's few and would probably include IBM) that can wring every ounce of performance out of a given HW solution as well as SGI. It's the computing equivalent of the guys who turn up at Bonneville Salt Flats every year to make cars go far quicker than they really should (remember, I'm a coward).
[quote]<strong>I do like your estimated figures though, I just disagree with them representing all the possible costs.</strong><hr></blockquote>
They don't and I concede the point. They were only ever intended to provoke a headline debate, and I think that's reasonable from an intellectual perspective. Do we think that NASA/DOD/CIA goes to Congress with line item budgets or do we think that they go with abstracts?
Thanks for the argument anyway, by the time we're finished we'll have a fully debated position that Steve-O can use as a company strategy.
im just a designer, who couldnt help but throw my opinion in on this well argued thread. as an SGI customer (well, at the office anyway), the integration of apple and sgi would be great. first off irix is practically dead, sgi is supposedly going to move to linux, this from the IT guys down the hall. irix is a real beast of an operating system, an im glad to see it go. but the macs in the office just dont cut it in the hardware arena, well when compared to an origin rig. as far as xserve is concerned, apple dosnt seem to understand what "workstation" means. 3 pci slots, a half size graphics card, granted its not meant to be a "true" workstation in the first place. the upper management here is unusually mac friendly, though some depts. have PCs. if sgi got scooped by apple, it would be the catalyst for us to switch completely to macs in the next big upgrade. in my opion apple dosnt really understand the big gun markett. sgi understands, and that knowledge would shorten apples learning curve. other thing to take into account, pixar (steves other company) is dropping sgi and moving to macs(i leave that to individual interpretation). those of you making noise about ati or nvidia should be aware that when your talking about sgi's level of perfomance, Quake frame rates are irrelevent. sgi's knowledge of the ultra high-end of visualisation super computing is unparalelled, apple would gain a great deal with that IP and expeirience to integrate into arguably the best gui. is it good business, mabey not, but it would make my life a little easy around the office, not switching back an forth between machines.
First off, excellent thread. Very interesting and well thought out. Until I put in my 2¢, that is!
I think most of us Mac heads have thought about Apple buying out (or in some way aligning with) SGI over the years - the two companies have always seemed to have a complementary nature and product. However, the question I could never answer adequately is this:
why?
Why should Apple buy SGI? To solve their immediate problem of competing with Intel and AMD processors? no. To get a DDR mainboard that can compete? no. There is nothing SGI has that will help Apple solve their immediate problems. The arguments presented in the thread are good, but they don't really answer this fundamental question. In fact, the whole basis of this thread seems to be that since SGI MAY be designing one or more of their future machines using the G5, and since SGI (in the past) had good tech, Apple should buy them, because they are a potential competitor. To be sure, only the low end of the SGI line would use a G5. SGI has a wide range of machines and a large software base. The tech. and software that Apple would obtain is not going to be used in the iMac, not even the tower. This tech is way beyond the Mac, both in performance and cost. It will be very difficult to bring the cost down to a 'consumer' or even a 'pro-sumer' level. We would all love to see a Power4 in a Mac, but unless IBM develops a low-cal version, it wont happen. Likewise, a NUMA machine from SGI isn't going to happen at the pro-sumer level yet.
However, Apple has been making very serious moves into the film compositing space, and this space needs heavy duty hardware. Apple now has their first server, and SJ has said they have a lot to learn in this field. They seem to have their sites on Hollywood. But what is Apple missing? The heavy duty hardware to run these apps, and the experience selling/service high end machines. Now, if you mean Apple would buy out SGI to enhance or supplement their position, I can see this, and agree that SGI may at some point down the line be a good acquisition. But just buying SGI because they have some cool tech and MAY BE using the G5 in some of their next gen hardware...no, as much as I would like to see this happen, I can't believe this is the reason.
<strong>as far as xserve is concerned, apple dosnt seem to understand what "workstation" means. 3 pci slots, a half size graphics card, granted its not meant to be a "true" workstation in the first place.</strong><hr></blockquote>
You said it yourself, man. The Xserve is a server, not a workstation. If you want a workstation, cross your fingers and hope Dorsal M is right
[quote].....other thing to take into account, pixar (steves other company) is dropping sgi and moving to macs(i leave that to individual interpretation). /QB]<hr></blockquote>
This info speaks volumes to me I am suprised at how little notice was taken of it at the time it was announced. Pixar can't afford to follow Apple on principal, they can only move to Macs if they have a product that is on a par or superior in terms of performance. We know that this is not the case now, Apple is not superior in terms of performance, so it follows that Pixar knows that it soon will be.
<strong>is it good business, mabey not, but it would make my life a little easy around the office, not switching back an forth between machines.</strong><hr></blockquote>
I know a lot of engineers with 2 computers. One's a Unix workstation and the other is a report writing/general work computer.
Apple is definitely going after that market where people are having to currently switch machines.
Just to chime in about SGI and it's MIPS processors...
SGI is pretty committed to MIPS. The line is still being developed, and it's been stated by numerous SGI employees that they have no real plans to move away from MIPS technology.
SGI has people on staff that design the chips, and they get sent to Texas Instruments who actually fabricates them. SGI's ability to design their chips is what makes them special, and I honestly don't think they're going to move away from that anytime soon.
They're not doing great financially, I agree. I don't know what is going to happen to them, but they're 100% committed to MIPS...
<strong>Just to chime in about SGI and it's MIPS processors...
SGI is pretty committed to MIPS. The line is still being developed, and it's been stated by numerous SGI employees that they have no real plans to move away from MIPS technology.
SGI has people on staff that design the chips, and they get sent to Texas Instruments who actually fabricates them. SGI's ability to design their chips is what makes them special, and I honestly don't think they're going to move away from that anytime soon.
They're not doing great financially, I agree. I don't know what is going to happen to them, but they're 100% committed to MIPS...</strong><hr></blockquote>
Not saying anything about SGI's commitment to MIPS, but I think MIPS has been an independent company since about '98, when they had their IPO. AFAIK, they aren't part of SGI.
To borrow the usual car industry analogy- Ford bought Jaguar a while ago, and then proceeded to pump billions of dollars into it to build new Jaguar product. It will take decades for Ford to recoup both the initial investment and the development costs of the purchase. It's very easy to argue that this ROI will occur way too slowly, and that the money would have been better spent elsewhere, say on Ford's own product line.
-HOS<hr></blockquote>
Sorry, but I can't contribute too much to this thread, as I don't know much about SGI's financials. However, Ford's investment in Jaguar saved it from doom, revitalized it, and is now one of the crown jewels of PAG.
(I'm also in favor of long-term goals and investments, and not get-rich-quick buyoffs of companies, so I don't mind waiting for the ROI.)
It also gave Ford executives a better car to drive than the Mark VII, which I've always believed to be the true reason for the purchase.
I think what helped was Jaguar's brand identity. If Apple does purchase SGI, it could create an "upper" segment, much like Ford's premier auto group (PAG). I would expect that they could benefit from SGI's brand identity (if it's still positive).
Companies are always considering such opportunities. If for some reason SGI could be bought at a bargain without losing much of it's advantages, I don't see why Apple (or someone else) wouldn't consider it.
Just rambling, wondering if US can get by Mexico...
Sorry that I haven't been around to stir this now nicely boiling pot, but sleep beckoned after watching the rereun of yesterday's S. Korea/Portugal game.
For any soccer fans out there, Portugal's dismissal from the competition following on the heels of similarly lacklustre performances from France and Argentina has made my summer, and acts as proof that complacency and gamesmanship (ok, cheating ) have no place in a highly competitive competition. If only the same applied to our industry, M$ and many others would have been sent home at the end of the first round - although, some of Apple's performances would have merited cause for concern.
Right, back to my pet subject du jour!
Some of these responses are not in the order of the original questions, but the whole picture will be easier to understand.
[quote]<strong>SGI is pretty committed to MIPS. The line is still being developed, and it's been stated by numerous SGI employees that they have no real plans to move away from MIPS technology.
SGI has people on staff that design the chips, and they get sent to Texas Instruments who actually fabricates them. SGI's ability to design their chips is what makes them special, and I honestly don't think they're going to move away from that anytime soon.</strong><hr></blockquote>
I don't doubt that SGI is committed to MIPS, I remember SGI "liking the product so much, they bought (85%) the company" in a Victor Kiam sort of way. Although that said, it was probably the only way of guaranteeing MIPS future (and "independence") in the long-term.
The fact that the MIPS division is still designing is not really a surprise. But the questions are "What are they designing?" and "Why are they designing the answer to the first question?".
There are two references that could answer this question: -
There are two interesting things on this page: what it does say and what it doesn't say.
The former is illustrated by the fact that they are positioning the MIPS 64-bit line as a high-performance "system-on-a-chip" processor for portable and embedded computing. A 64-bit PalmOS or PocketPC system. Would sound a bit odd, but let's get back to that later.
The latter is illustrated by the fact that at no point on this page do they actively recommend and endorse this family of chips as a platform for standard computing environments.
"I think we're going to need a bigger boat!"
And now we get to the "Why?" question.
If you go to <a href="http://www.sgi.com/visualization/van/"; target="_blank">this</a> page at SGI, the implication is that our friends at Mountain View are seemingly tired of being tied to the office and missing out on the natural light and fresh air that their near-neighbours in Santa Clara (Palm) take for granted.
Maybe, they'd like to drive out of the office and have lunch down at Shoreline Park (nice burgers I thought) or cruise around on the pedal boats whilst doing a bit of 3D modelling (a particularly Pythonesque image, I feel).
In all seriousness, I think these two pages show some of the spirit of innovation that still exists at SGI & MIPS, which is what makes those companies special. But I would think it calls into question whether SGI are using MIPS processors out of a strategic choice or out of necessity until something "better" (I'm not dissing MIPS, hence the quotes) arrives.
We must not forget that SGI are not the only customer for MIPS: Sony, LSI, NEC and Toshiba are just some of licensees of their technology, not forgetting Texas Instruments. So sure, they still design for the future but maybe not for themselves, at least in traditional markets.
OK, next issue:
[quote]<strong>Why should Apple buy SGI? To solve their immediate problem of competing with Intel and AMD processors? no. To get a DDR mainboard that can compete? no. There is nothing SGI has that will help Apple solve their immediate problems.</strong><hr></blockquote>
And I'm not sure that's why AAPL would buy SGI anyway. If you remember, I proposed that AAPL broaden its market offering thus:
[quote]<strong>Apple ultimately delivers four hardware product segments:-
Consumer & Education: eMac, iMac, iBook, iPod (and other i-devices)
General Corporate: Power Macintosh, Power Book, xServe
Specialist Corporate: revised SGI Workstation line, running G5 and OS XI (which is 64-bit pure)
Server: revised SGI Origin line, running G5 and OS XI, but with optimised graphics sub-systems modularised and made optional. Also the SAN and other storage products
Apple would then have an integrated product range, from SoHo and K-12 through small business through to corporate and specialist vertical visualisation and scientific.</strong><hr></blockquote>
So, what I'm saying is that AAPL will carry on developing the G5 platform (BTW, my thanks to thuh Freak for pointing out that references to G5 are a placeholder for "whatever comes after G4" because we all hope/pray/know - choose your own term - that something must come soon) based on its own design and that this platform will arrive as scheduled when the RDF mothership lands in SF next January. However it will only be 32-bit, due to OS considerations.
My theory says that what AAPL should then do 12 months later (although NAB 04 would be my target this time) is release a "PowerMac on steroids" (SGI-enhanced) workstation with more slots for both RAM and expansion, more SMP (4-way/8-way anyone), 64-bitness (OS XI) and whatever else will make a truly professional workstation sell.
As an extension to this, I also think that AAPL could/should consider segmenting their OS offering on a commercial basis. If you buy what I call either a "consumer and education" or "general corporate" Macintosh, you should be automatically licensed for 32-bit OS XI, however if you want to enable the 64-bitness of OS XI (probably more applicable to the "general corporate" market) you have to pay for an upgrade key as per QuickTime/QuickTime Pro.
Is that cute or is it just me?
[quote]<strong> In fact, the whole basis of this thread seems to be that since SGI MAY be designing one or more of their future machines using the G5, and since SGI (in the past) had good tech, Apple should buy them, because they are a potential competitor.</strong><hr></blockquote>
OK to address the basis of the thread, I have seen threads predicated on more tenous arguments than the one I'm using. Some of the threads before MWSF 02 were based on an advertising strapline ferchissake.
To address your second point, isn't that one of most often used reasons for buying a business? If The Rumour is true, than there is a tremendous opportunity to leverage synergies between two organisations who do not have overlapping product ranges, where proft & loss accounts can be improved by combining admin, accounting, support and several other areas.
HP's merger with Compaq (OK, takeover) was based on exactly the same justifications without the complementary product range argument, hence the pruning now underway.
[quote]<strong>The tech. and software that Apple would obtain is not going to be used in the iMac, not even the tower. This tech is way beyond the Mac, both in performance and cost. It will be very difficult to bring the cost down to a 'consumer' or even a 'pro-sumer' level. We would all love to see a Power4 in a Mac, but unless IBM develops a low-cal version, it wont happen. Likewise, a NUMA machine from SGI isn't going to happen at the pro-sumer level yet.</strong><hr></blockquote>
And, referring back to my reply to HOS at 19:45 yesterday as well as a section earlier in this post, I'm not saying that either.
This is a way for AAPL to become a dominant supplier of world-class workstation-derived technology all the way from Lawrence Livermore & Sandia to your study, taking in Ford, BP, DoCoMo, Stanford, Guildford Grammar School, and all points in between on the way.
It's about vision and seeing the power of bullet-proof core OS leveraged in all of those markets, on a coherent product family that can scale whether your a family going from iMac to PowerMac or a student going from iBook to TiBook or a business going from PowerMacs to PowerMac Pro and an 8-way SMP box to a 32-way box and onwards.
Business likes scalability and loves investment protection, and if gets to reduce support costs as well then you have a viable story to tell. If you can tell me how those aspirations are incompatible with what I'm proposing then I'll shut up.
[quote]<strong>To be sure, only the low end of the SGI line would use a G5.</strong><hr></blockquote>
Why? The current 64-bit element of SGI's range runs between 500 and 600Mhz on both the workstation side (Fuel and Octane2) and the Origin server space. If "G5" runs at around 1.6 - 2.0GHz and doesn't offer significantly better (I'm talking anywhere upwards of 35%) performance in a like-for-like situation, then we have definitely backed the wrong horse and someone should get a pink slip, and we should all go and price up our Clawhammer dream systems.
[quote]<strong>SGI has a wide range of machines and a large software base</strong><hr></blockquote>
Exactly and Exactly! For the latter point, imagine having it all running under OS X's successors.
[quote]<strong>If Apple does purchase SGI, it could create an "upper" segment, much like Ford's premier auto group (PAG). I would expect that they could benefit from SGI's brand identity (if it's still positive).
Companies are always considering such opportunities. If for some reason SGI could be bought at a bargain without losing much of it's advantages, I don't see why Apple (or someone else) wouldn't consider it.</strong><hr></blockquote>
GardenOfEarthlyDelights sees another facet of the argument: PAG gives Ford the ability to sell a complete range of cars to the individual and the enterprise. Hertz now supply Volvo, Jaguar and Land Rover as rental vehicles because Ford can offer them that range of product. It makes Hertz look better than Alamo if you're a status conscious user/chooser. An AAPL purchase of SGI would offer similar benefits, especially for companies tired of being robbed blind by M$.
And as GOED points out, if AAPL doesn't do it, someone else will and then we may very well land up paying a very heavy price.
as far as my complaint about the xserve being a lame excuse for a workstation, yes it is intended to be a server. but apple makes a half hearted attempt to pitch it as a workstation, go into the custom configuration (for xserve, at the apple store), they make some mention of "some users might want to use their xserve as a graphics workstation" in one of the options. all i can say is it would make a better mouse pad, as a server it might be the greatest thing since the wheel, but no itll never be a "real" workstation.
there is this little company called Grand Vitesse Systems, they repackage power macs into rackmount "psuedo" workstations, they do a fine job i might add. the cost isnt too far out of line with buying from apple, and their support is more profesional than apple (or so the IT guys tell me, i have never had to call them). now i say psuedo because its all standard apple fair under the hood, which means slow system bus, consumer grade graphics cards, and any other little problems with apple hardware that i could "nit pick", which i admit to doing. right now, apples hardware just cant cut it in the high end. sure a DP-1Ghz G4 will knock your socks off running a few p-shop filters, but this isnt what i do at work, neither does the air force (probubly sgi's favorite customer). the need for scaleable super computing, think top down solutions, is pretty big. the firm i work for could save thousands (being conservitive, im not an accountant), in misc. operating expenses. of course the downside is that apple would have us by the balls, not the worst position mind you, but not ideal.
in my opinion it will take apple a long time to be accepted in the high-end market, they have to far to go from my prospective. acquiring sgi would give them a "head start" in that direction, of course thats IF they want to go that way. its nice to dream though, think about it, a full set of applications, IRIX has the bare minimum. it would be great to run a simulation, and then crank out a findings report without walking across the office and starting up another computer. oh well whatever.
Sorry I wasn't here to post earlier- I was at a pool party all day yesterday, drinking margaritas and applying sunscreen.
MCCFR-
More additional costs which have been left out of a potential AAPL takeover of SGI include opportunity costs. Money that has to go towards the purchase and operation/shutdown of SGI is money that is taken away from Apple's warchest, Apple's R&D, and Apple's operational expenses. Manager time and talent now has to attend to a new division which at first glance doesn't appear to be in sync with Apple's "vision" of computing.
To the other posters-
Don't forget that the initial argument here for Apple buying out SGI included the following assumptions:
1) Apple is planning on delivering some sort of high performance computing (HPC), of the sort which SGI currently builds-
[quote] Steve Jobs makes a throwaway comment, which can be paraphrased as "You want performance - wait and see what we're going to release, you'll be blown away" <hr></blockquote>
2) SGI may already be looking at or working on developing high performance PPC boxes, using the G5-
[quote] SGI is alledgedly considering a move away from its' own MIPS processors to PowerPC G5. <hr></blockquote>
3) The acquisition cost of Apple buying SGI is within Apple's budget-
[quote] SGI has a market cap of under $550M, and cash/short-term investments of around $200M.
Apple has a market cap in excess of $7B, and cash/short-term investments of around $4B. <hr></blockquote>
4) Apple and SGI's strategies of video editing and graphics are very similary (synergy, to borrow a horrid word)-
[quote] Apple appears to be indulging in a fit of whatever is the corporate equivalent of retail therapy, largely based around graphics and film/video post-production. Several of the products acquired are alledgedly 64-bit capable. <hr></blockquote>
So with these assumptions in place, for Apple to acquire SGI means that Apple needs to have compelling reasons to justify doing it. My argument has been one primarily of costs- that the direct and indirect costs of Apple acquiring SGI are far higher than the value Apple gains from the acquisition. My car analogies were to illustrate examples of companies buying out other companies which failed to deliver the value that could have been gained through not purchasing.
More on this later, but as the discussion keeps moving along, don't forget the assumptions that underlie this...
<strong>Sorry I wasn't here to post earlier- I was at a pool party all day yesterday, drinking margaritas and applying sunscreen.
MCCFR-
More additional costs which have been left out of a potential AAPL takeover of SGI include opportunity costs. Money that has to go towards the purchase and operation/shutdown of SGI is money that is taken away from Apple's warchest, Apple's R&D, and Apple's operational expenses. Manager time and talent now has to attend to a new division which at first glance doesn't appear to be in sync with Apple's "vision" of computing.
To the other posters-
Don't forget that the initial argument here for Apple buying out SGI included the following assumptions:
1) Apple is planning on delivering some sort of high performance computing (HPC), of the sort which SGI currently builds-
So with these assumptions in place, for Apple to acquire SGI means that Apple needs to have compelling reasons to justify doing it. My argument has been one primarily of costs- that the direct and indirect costs of Apple acquiring SGI are far higher than the value Apple gains from the acquisition. My car analogies were to illustrate examples of companies buying out other companies which failed to deliver the value that could have been gained through not purchasing.
More on this later, but as the discussion keeps moving along, don't forget the assumptions that underlie this...
Sorry, got to run,
-HOS</strong><hr></blockquote>
HOS -
Well, your weekend sounds more relaxed than mine. But then mine involved watching England play football & cricket, neither of which are good for people of a nervous disposition. Also I rarely drink (normally cos I'm driving at some point in the day) and the UK rarely requires sunscreen.
I would agree with you that my whole concept is predicated on rumours, assumptions and possibilities of varying degrees of credibility.
I would remind you that I'm proposing the purchase of SGI be made 50% cash, 50% paper: That ratio could move to 20/80 and on to 0/100 the worse SGI's position gets (another bad quarter or a bad analysts report) and the more desperate their investors get to find a "safe harbour" where their money does not go up in smoke. I would think if you'd bought stock at $10 and it was now worth $2, you'd grab any passing ship which offered you a viable chance of not being eaten by a passing shark.
At that point, you may only be paying $350M to $400M gross all in paper which, although it would require the reorganisation of APPL stock and would devalue the share price, would not materially affect AAPL's cash position bar the reorganisation expenses.
At the launch of xServe, SJ said "we come into this market humble", a recognition that AAPL does not have credibility of any sort in that marketplace. I do not believe that I am the only person on this thread who believes that a bulk knowledge transfer from SGI could help AAPL make inroads in both the HPC and SMP server markets.
In addition, the acquistion would help AAPL in the high-performance workstation market where AAPL needs to get some clear blue water between itself and companies which will offer (or are offering) IA-32 MP offerings, such as HP/Compaq. Read some of the other contributions from people who really do give a damn.
We may not respect the Wintel market, but it is a reality which has been eroding the Apple shoreline for too long for us not to take the possibility of pragmatic solutions to at least a practical level of due dilligence. If we slip back into the NIH (not invented here) complacency and arrogance of the past, AAPL is doomed as an independent brand, and that should be the last thing any of us want.
I've heard the nVidia buzz, and it is interesting I grant you: but I am too much of a cynic and have seen too much water go under the bridge to have faith in developing "partnerships" with companies who could disappear in the next major consumer recession or change their allegiances depending on who offers the best "peace treaty". Say what you like about SGI, the one thing it doesn't rely on are the whims of the consumer market.
More additional costs which have been left out of a potential AAPL takeover of SGI include opportunity costs. Money that has to go towards the purchase and operation/shutdown of SGI is money that is taken away from Apple's warchest, Apple's R&D, and Apple's operational expenses. Manager time and talent now has to attend to a new division which at first glance doesn't appear to be in sync with Apple's "vision" of computing.
-HOS</strong><hr></blockquote>
And this my friends is exactly why one NEVER buys a business to gut it. Remember everything that Apple does not NEED will have to be taken care of. Inventory sold at a loss to move it, buildings sold or leases ended, books closed out, contracts ended, accounting squared away, legal proceedings for accounts closings and contract endings. And this is the tip of the iceberg, details, lots of details. [edit] And how about all of those employees that Apple may not want or need?? They don't just disappear, severance pay, unemployment, help them find a new job?? Add to that the fact that these are people with families, ect. What a headache. [end edit] Easier to just offer SGI something good for MAYA, if they take it (they are stupid), all is good, if not buy the next best thing TREMOR.
when i said apple will have a hard time being excepted in the high end, mabey i should have said something like.
-apple will have some trouble breaking into the world of 500+ processor, 100 terra-flop, real-time 3D rendering super-computers.-
this is the world of sgi (which im sure your aware of), the world of custom applications and hardware, THAT high end. now i like macs, ive owned a few of em, but i dont think apple has the cred to pull that kind move off anytime soon. in seven years, i might be playing a different toon, but for now thats my take on it. i use both sgi hardware and apple hardware at work, a cluster of p-macs might cut it for some cash strapped students, but when money isnt the issue, you get the right tool for the job. to get away from the auto industrie allagories, when your child is sick with a rare disease, do you stick with your family GP, or seek the help of a specialist, heck your GP will usually refer you to a specailist. the kinds of projects that get run on sgi's hardware are paramount to the people running them. this is a difficult mentality to over come, when people want the best, expeirence counts. besides, we dont know if the board at apple really wants to get into this kind of business at all. but if they bought sgi, they wouldnt have to start at the bottom of the mountain to get to the summit. they could cut seven years of careful plodding down to say two or three, if the first apple/sgi super computer hit the bullseye, then they would have almost nothing to proove to their customers after that. in parting, if i was going to bet money, i would say apple is not going to buy sgi. just my opinion.
And this my friends is exactly why one NEVER buys a business to gut it. Remember everything that Apple does not NEED will have to be taken care of. Inventory sold at a loss to move it, buildings sold or leases ended, books closed out, contracts ended, accounting squared away, legal proceedings for accounts closings and contract endings. And this is the tip of the iceberg, details, lots of details. [edit] And how about all of those employees that Apple may not want or need?? They don't just disappear, severance pay, unemployment, help them find a new job?? Add to that the fact that these are people with families, ect. What a headache. [end edit] Easier to just offer SGI something good for MAYA, if they take it (they are stupid), all is good, if not buy the next best thing TREMOR.
If you're really sincere about your concern for the jobs of SGI employees, I would have thought you would argue for the case not against it.
Is AAPL going to continue imfringing on SGIs market? Either by developing a solution itself or by acquistion of other companies' IP by license or whatever. So the erosion still happens there. And it still happens from Sun and IBM and HP/Compaq. All companies with more money to fight a long fight than SGI. So what do you think the end result is going to be? Because SGI still being in action isn't one of them.
The solution I propose is the best option for AAPL boot-strapping itself into a market where budgets are decided by doubling the number you first fantasized, not the halving the number you put forward to your chief beancounter. Phamaceuticals, automotive, oil/gas, and high-end commercial physics like cold fusion: This is where AAPL goes under this proposal and it goes there on the fast train.
SGI's employees win by not having their work scattered to the four winds and suffering the tension of waiting for 80% of them to be given marching orders. My solution probably provides overlap in finance and a few other administrative areas and potentially SGI's manufacturing in WI - although I suspect that would stay, because AAPL would need somewhere to make the resulting product.
I regret the fact that some people and families would suffer under this scenario, but the key word is SOME. How many would survive under IBM assuming they were interested or HP/Compaq or Sun. Not nearly as many because the product overlap is too great.
And SGI's shareholders benefit because they get attached to a company with a mass market.
All of the stuff before your edit is absolutely true, that's why you reserve some cash for restructuring. I haven't been to SGI in the UK (Theale) for several years, but I tell you something it would be better for AAPL UK to move to Theale than stay in Stockley. Especially as the government is perhaps going to allow a new runway to built almost next to them. Probably cheaper in the long-run given Heathrow rents.
I covered the loss on inventory in my figures, assuming that AAPL would only be able to shift 50% of it at 50% of value. Which I thought was pretty vicious.
I'm not proposing butchering SGI, I'm proposing the surgical excision of its best technology, people and practices so that they can live to fight another day - I remember when you couldn't buy a computer without a full pre-sales programme, and lengthy proposals and planned implementation (maybe IT investment would show more ROI to business if more of that still happened), that's another thing SGI can show AAPL - how to make a proper business case for customers making an investment, and how to implement that investment in a value-added manner.
The rest of your objections are, with the greatest respect, low-level details. If it requires management time to resolve contracts, etc, etc, maybe this is where AAPL's legal, financial and property teams get to make a full-on front-and-centre contribution to helping the company grow. It's just time and money, and if due dilliegence showed it was worthwhile would you still object?
MCCFR, you have several strong arguments. Since I'm too lazy to go look up SEC filings on SGI and crunch my own numbers (I've not done any diligence at all, due or otherwise on SGI since they bought out Cray, and I had a college professor assure me that "this means supercomputers on your desktop in five years!"), so I'm definitely arguing from a weaker perspective. Especially since my position is that costs are too great for value received.
So instead of doing real work, I'll take the lazy way out and go after the assumptions- if I can topple enough of them, then the subsequent conclusions based on them will topple as well.
Assumption 1) Apple is working on a HPC product, or a product with HPC characteristics.
HPC is a very broad field, and traditionally ranges from individual vector processors (older Crays) to massively parallel machines (Beowulf clusters, ASCI White), and encompasses many things inbetween (Sun E15k). Prices are accordingly high- often starting in the $100Ks and scaling up to the millions.
There's no doubt in anyone's mind that Apple is interested in building computers which have better performance than their current machines. But a little bit better and orders of magnitude better both qualify as better than current offerings.
With prices between $800 and $4k, Macs are dirt-cheap compared to most all of SGI's offerings. In order to maintain these low prices, Apple adopts far more pedestrian, commodity parts.
Rather like the old MG or Triumph approach to performance- recycle existing parts in an exciting shell. (OK- cheap shot, but my last British car was my Dad's '52 MGTD... which doesn't go from 0-60 mph as it tops out about 55...)
Back to the argument, any HPC-like machine or technology that Apple employs (like a NUMA architecture) has to be cheap. Therefore, this must exclude many HPC characteristics which are too expensive to implement.
So if Apple is looking at HPC, it's looking at something really small-scale or cheap HPC, so we have minimal overlap with what SGI does... or charges.
Assumption 2) SGI is working on a PPC-based machine. Not provable one way or the other, and if not true, dampens a lot of the fire since this assumption is crucial to transition costs. If true though, this machine is more likely to be a million dollar server than a $1k blade. Again, we're not guaranteed that costs of this technology fall into line with what Apple can afford to charge, or even be the kind of technology that Apple's really interested in at the moment.
Assumption 3) SGI is sufficiently inexpensive that Apple can pick it up without too much sweat. MCCFR has done a very good analysis demonstrating that this is probably true. The only real hole I can knock in here is that even though it's within Apple's budget, it still represents a sizable expenditure which may be better spent elsewhere.
Assumption 4) Synergy in Apple's and SGI's video editing and 3D graphics markets.
While true to a degree, key apps like Maya already exist for MOSX, so other than control over the direction of Maya, the major benefit of Maya-on-the-Mac already exists. More importantly, the key infrastructure of generating video editing software (iMovie, Final Cut Pro) is already in place and running on both today's and yesterday's Macs. Whether software like this would run so much faster on future SGI hardware or future Apple hardware as to justify the cost of making it happen is up in the air. I don't know if high-end markets like television stations are willing to fork over the $100k for a high-end Final Cut Pro on SGI-class hardware, or whether $4k Apple hardware is sufficient to get the desired work done. The one thing I can easily state is that the demand curve clearly shows that more $4k boxes will sell than $100k boxes.
Next up, some returning commentary on past running commentary...
Comments
<strong>MCCFR-
You have some very interesting figures.
However, I still think that you're underestimating legacy costs, maintenance costs, and future development costs.</strong><hr></blockquote>
Not underestimating them, merely not getting the conceptual feasibility of a potential mercy-killing of SGI (with considerable benefits for AAPL) confused with the fine-print detail of the operational issues that you describe.
However, as you bring up the subject, let's discuss them:-
The costs of maintaining the legacy environment are dependant on SGI's MIPS-based technology becoming a legacy platform. In that respect, it is no different to Compaq's purchase of DEC and various others (was it Tandem or Stratus, my memory isn't what it was, and I can't be bothered to open another brower window to try and negotiate the labyrinthine "new HP" website) or HP's purchase of Apollo, or indeed SGI's acquisition of Cray.
You could complete and ship any elements of the current software release roadmap (so long as the ROI on that completion justifies the effort), and then go to maintenance releases only for a pre-defined period alongside and then web support only and then it's curtains. This is hardly a new strategy, the entire industry implements it either by degrees or completion. Try getting support for Win 98 from MS' website after 06/30.
The thing is this isn't the market Apple has been accustomed to before xServe: A 12-month warranty doesn't cut it with this class of kit - you buy on-site maintenance with specific service level agreements from the service provider who makes you comfortable and that you can afford. Maintenance is actually a profit centre, not a cost centre.
Actually, because I'm feeling expansive, how's this for a solution: Outsource the maintenance effort to an SGI specialist; or better still, spin out the maintenance assets that you acquire from the purchase of SGI as a management buy-in, or complete MBO and then sub-contract the effort to them.
[quote]<strong>To borrow the usual car industry analogy- Ford bought Jaguar a while ago, and then proceeded to pump billions of dollars into it to build new Jaguar product. It will take decades for Ford to recoup both the initial investment and the development costs of the purchase. It's very easy to argue that this ROI will occur way too slowly, and that the money would have been better spent elsewhere, say on Ford's own product line.</strong><hr></blockquote>
Actually, not quite true: Ford, GM. DaimlerChrysler, VAG (list auto conglomerates until blue) all practice platform commonality and try to use a common parts-bin where possible.
A Jaguar X-type's floorpan is actually derived from one of Ford's CDx models (can't remember which one, and would have to revert to my extensive collection of back-issues of Car to discover which) and the 2.5 V6 engine (and possibly the 2.0 unit) is/are (a) modified Ford Duratec unit. Want to know why the X-type is 4WD, becasue Jag's engineers couldn't easily achieve a rear-wheel only set-up because of the floorpan/drivetrain combination that they were obliged to use.
Jag's S-type is derived from a Lincoln floorpan, and have you ever wondered why Aston-Martin's DB7 and Jaguar's XK8 look so similar.
Don't even get me started on GM (the badge engineer's badge engineer) or VAG (the platform and parts commonality across Volkswagen, Audi, Seat and Skoda shows how it is possible to develop four different model ranges and brand images from simple (!) permutation)
When GM completes a purchase of Fiat Auto in Europe sometime towards the end of the summer, the Signum platform (which I believe is Europe only) will be used in Opel/Vauxhall's (hopefully) highly revised Vectra, Saab's replacement 9-3 and a new Alfa Romeo model (hopefully, without screwing A/R's recent renaissance on the handling front). BTW, Fiat will float the Ferrari group (i.e. including Maserati) before GM can wreak havoc with an art business that has next to nothing to do with accounting.
[quote]<strong>The same argument has been applied to GM's creation of Saturn (the billions spent could have revamped, say, Chevy, and been better spent with a faster ROI). And also the same argument has been used for GM's "roboticization" during the 80s- where they spent in the tens of billions, and could have just outright purchased Toyota for the same price, no doubt a better investment.</strong><hr></blockquote>
I seem to remember that Saturn was GM's attempt to distance itself from some of the product quality issues that were endemic throughout the empire, and an attempted response at delivering Far East-style flexible working practices and quality in a green-field company which would help GM's image in the marketplace. How succesful was it on those terms? I really don't know! (Gosh, something I don't have an opinion on - that's a Kodak moment).
As for the Toyota argument, you say that GM puchasing that producer of mainly dull, but occasionally inspiring vehicles would have been a better investment than GM re-inventing internally! Surely, you're making my case for me - no really, I'm not going to object.
[quote]<strong>For Apple to buy out SGI puts them in the same position- even assuming SGI has begun a migration to PPC (which we don't even know), the costs of migrating SGI's existing user base to PPC haven't even been estimated yet...</strong><hr></blockquote>
And why should I (or even Apple) estimate the costs of a customer's migration to a new platform. Are Apple going to give the new platform's away. Did they pay for my migratiion when they migrated from 68K to PPC? Did Data General pay for its' old 88K Aviion users to migrate to the later Pentium SHV Aviions.
Sounds to me like another profit centre. KCHING!!
[quote]<strong> and the costs of just dumping SGI's user base is larger than the cost of buying SGI and shutting it down completely- since you've now pissed off that user base who now won't become future Apple customers...</strong><hr></blockquote>
Who said anything about dumping them? You encourage them to migrate by demonstrating the increased functionality, usability and ROI potential of the new platform. If they choose not to go, they have to accept that there will come a time when Apple will no longer be responsible for direct support of legacy platforms.
Before EMC shut down Data General for good - having purchased that company purely to exploit its entry-level and mid-range SAN products (because it was cheaper than developing it itself and it eliminated a competitor that was making a loss across the board - HOLD ON, IS THAT MY ARGUMENT IN ACTION AGAIN) - I have sincere doubts that they were still providing affordable support for anything other than SHV systems. 88K, Eclipse/MV, Eclipse, and Nova were probably only supportable from third-party service providers for hardware and from no one at all on software.
[quote]<strong>Worse, assuming Apple wants to "bring SGI's customers back into the fold" (sorry- borrowing from the letter I got from Jobs when Apple bought out PowerComputing: my PowerCenter r0ck3d!), Apple will need to continue a massive investment into future SGI hardware... since the Origin will need a PPC based replacement which can offer at the least, similar performance (at similar price!).</strong><hr></blockquote>
But surely, that's exactly what I'm proposing! Build a G5 Origin architecture which is based on using one or more 4-way "nodes" (the first board can be delivered 2-way as an option and expanded in the field) so that you can deliver a solution which scales from 2 to 512 processors, with a (fair marketprice) SCI upgrade option handling the delivery of NUMA services that becomes non-optional once you decide to tie more than 32 processors/8 nodes together.
Evolve said solution so that it supports the interconnect technology of the day before you then bring out the G6 Origin.
BTW, for all we know and if The Rumour is true, SGI is/has (been) develop(ing/ed) a G5 Origin based on new interconnect technologies anyway and then the key investment is really the one that makes OS XI cope with that as a concept.
[quote]<strong>I can understand the desire to have our hypothetical PPC Origin technology "trickle down" to Apple's consumer line, but it's still a very expensive way of getting next-generation technology.</strong><hr></blockquote>
Never said that to be fair! My "trickledown" theory (how long before someone refers to this as "voodoo company strategy") referred purely software and to the fact that Apple is already using this concept with FCP/iMovie and DVDSP/iDVD.
If AAPL can abstract or abridge the technology of Alias|Wavefront or Shake/Tremor et al so that it can be delivered in multiple and appropriate versions using market/brand segmentation techniques to a variety of audiences (sometimes commercially/sometimes as a complimentary brand/market USP) it can extend the useful life of a product/technology so that the ROI on the core development is that much greater (back to my automotive analogy).
The NUMA technology can only practically exist in the server space, and the graphics/visualization technologies are definitely client space for the most part.
[quote]<strong>Not when companies like nVidia, ATi, Motorola and IBM are happy developing it on their own...</strong><hr></blockquote>
There are very few things/experiences in our industry that cannot be improved by either deploying them and/or using them in a way that never occurred to the originator of the core concept. I believe that I said there are probably few HW faculties (BTW, Programmer, that's few and would probably include IBM) that can wring every ounce of performance out of a given HW solution as well as SGI. It's the computing equivalent of the guys who turn up at Bonneville Salt Flats every year to make cars go far quicker than they really should (remember, I'm a coward).
[quote]<strong>I do like your estimated figures though, I just disagree with them representing all the possible costs.</strong><hr></blockquote>
They don't and I concede the point. They were only ever intended to provoke a headline debate, and I think that's reasonable from an intellectual perspective. Do we think that NASA/DOD/CIA goes to Congress with line item budgets or do we think that they go with abstracts?
Thanks for the argument anyway, by the time we're finished we'll have a fully debated position that Steve-O can use as a company strategy.
[ 06-14-2002: Message edited by: Mark- Card Carrying FanaticRealist ]</p>
I think most of us Mac heads have thought about Apple buying out (or in some way aligning with) SGI over the years - the two companies have always seemed to have a complementary nature and product. However, the question I could never answer adequately is this:
why?
Why should Apple buy SGI? To solve their immediate problem of competing with Intel and AMD processors? no. To get a DDR mainboard that can compete? no. There is nothing SGI has that will help Apple solve their immediate problems. The arguments presented in the thread are good, but they don't really answer this fundamental question. In fact, the whole basis of this thread seems to be that since SGI MAY be designing one or more of their future machines using the G5, and since SGI (in the past) had good tech, Apple should buy them, because they are a potential competitor. To be sure, only the low end of the SGI line would use a G5. SGI has a wide range of machines and a large software base. The tech. and software that Apple would obtain is not going to be used in the iMac, not even the tower. This tech is way beyond the Mac, both in performance and cost. It will be very difficult to bring the cost down to a 'consumer' or even a 'pro-sumer' level. We would all love to see a Power4 in a Mac, but unless IBM develops a low-cal version, it wont happen. Likewise, a NUMA machine from SGI isn't going to happen at the pro-sumer level yet.
However, Apple has been making very serious moves into the film compositing space, and this space needs heavy duty hardware. Apple now has their first server, and SJ has said they have a lot to learn in this field. They seem to have their sites on Hollywood. But what is Apple missing? The heavy duty hardware to run these apps, and the experience selling/service high end machines. Now, if you mean Apple would buy out SGI to enhance or supplement their position, I can see this, and agree that SGI may at some point down the line be a good acquisition. But just buying SGI because they have some cool tech and MAY BE using the G5 in some of their next gen hardware...no, as much as I would like to see this happen, I can't believe this is the reason.
Finis, sorry so long...
<strong>as far as xserve is concerned, apple dosnt seem to understand what "workstation" means. 3 pci slots, a half size graphics card, granted its not meant to be a "true" workstation in the first place.</strong><hr></blockquote>
You said it yourself, man. The Xserve is a server, not a workstation. If you want a workstation, cross your fingers and hope Dorsal M is right
This info speaks volumes to me I am suprised at how little notice was taken of it at the time it was announced. Pixar can't afford to follow Apple on principal, they can only move to Macs if they have a product that is on a par or superior in terms of performance. We know that this is not the case now, Apple is not superior in terms of performance, so it follows that Pixar knows that it soon will be.
Roll on MWNY.
<strong>is it good business, mabey not, but it would make my life a little easy around the office, not switching back an forth between machines.</strong><hr></blockquote>
I know a lot of engineers with 2 computers. One's a Unix workstation and the other is a report writing/general work computer.
Apple is definitely going after that market where people are having to currently switch machines.
SGI is pretty committed to MIPS. The line is still being developed, and it's been stated by numerous SGI employees that they have no real plans to move away from MIPS technology.
SGI has people on staff that design the chips, and they get sent to Texas Instruments who actually fabricates them. SGI's ability to design their chips is what makes them special, and I honestly don't think they're going to move away from that anytime soon.
They're not doing great financially, I agree. I don't know what is going to happen to them, but they're 100% committed to MIPS...
<strong>Just to chime in about SGI and it's MIPS processors...
SGI is pretty committed to MIPS. The line is still being developed, and it's been stated by numerous SGI employees that they have no real plans to move away from MIPS technology.
SGI has people on staff that design the chips, and they get sent to Texas Instruments who actually fabricates them. SGI's ability to design their chips is what makes them special, and I honestly don't think they're going to move away from that anytime soon.
They're not doing great financially, I agree. I don't know what is going to happen to them, but they're 100% committed to MIPS...</strong><hr></blockquote>
Not saying anything about SGI's commitment to MIPS, but I think MIPS has been an independent company since about '98, when they had their IPO. AFAIK, they aren't part of SGI.
To borrow the usual car industry analogy- Ford bought Jaguar a while ago, and then proceeded to pump billions of dollars into it to build new Jaguar product. It will take decades for Ford to recoup both the initial investment and the development costs of the purchase. It's very easy to argue that this ROI will occur way too slowly, and that the money would have been better spent elsewhere, say on Ford's own product line.
-HOS<hr></blockquote>
Sorry, but I can't contribute too much to this thread, as I don't know much about SGI's financials. However, Ford's investment in Jaguar saved it from doom, revitalized it, and is now one of the crown jewels of PAG.
(I'm also in favor of long-term goals and investments, and not get-rich-quick buyoffs of companies, so I don't mind waiting for the ROI.)
It also gave Ford executives a better car to drive than the Mark VII, which I've always believed to be the true reason for the purchase.
I think what helped was Jaguar's brand identity. If Apple does purchase SGI, it could create an "upper" segment, much like Ford's premier auto group (PAG). I would expect that they could benefit from SGI's brand identity (if it's still positive).
Companies are always considering such opportunities. If for some reason SGI could be bought at a bargain without losing much of it's advantages, I don't see why Apple (or someone else) wouldn't consider it.
Just rambling, wondering if US can get by Mexico...
For any soccer fans out there, Portugal's dismissal from the competition following on the heels of similarly lacklustre performances from France and Argentina has made my summer, and acts as proof that complacency and gamesmanship (ok, cheating ) have no place in a highly competitive competition. If only the same applied to our industry, M$ and many others would have been sent home at the end of the first round - although, some of Apple's performances would have merited cause for concern.
Right, back to my pet subject du jour!
Some of these responses are not in the order of the original questions, but the whole picture will be easier to understand.
[quote]<strong>SGI is pretty committed to MIPS. The line is still being developed, and it's been stated by numerous SGI employees that they have no real plans to move away from MIPS technology.
SGI has people on staff that design the chips, and they get sent to Texas Instruments who actually fabricates them. SGI's ability to design their chips is what makes them special, and I honestly don't think they're going to move away from that anytime soon.</strong><hr></blockquote>
I don't doubt that SGI is committed to MIPS, I remember SGI "liking the product so much, they bought (85%) the company" in a Victor Kiam sort of way. Although that said, it was probably the only way of guaranteeing MIPS future (and "independence") in the long-term.
The fact that the MIPS division is still designing is not really a surprise. But the questions are "What are they designing?" and "Why are they designing the answer to the first question?".
There are two references that could answer this question: -
Firstly, the "What?" question can be addressed by <a href="http://www.mips.com/products/s2p2.html"; target="_blank">this</a> page at mips.com.
There are two interesting things on this page: what it does say and what it doesn't say.
The former is illustrated by the fact that they are positioning the MIPS 64-bit line as a high-performance "system-on-a-chip" processor for portable and embedded computing. A 64-bit PalmOS or PocketPC system. Would sound a bit odd, but let's get back to that later.
The latter is illustrated by the fact that at no point on this page do they actively recommend and endorse this family of chips as a platform for standard computing environments.
"I think we're going to need a bigger boat!"
And now we get to the "Why?" question.
If you go to <a href="http://www.sgi.com/visualization/van/"; target="_blank">this</a> page at SGI, the implication is that our friends at Mountain View are seemingly tired of being tied to the office and missing out on the natural light and fresh air that their near-neighbours in Santa Clara (Palm) take for granted.
Maybe, they'd like to drive out of the office and have lunch down at Shoreline Park (nice burgers I thought) or cruise around on the pedal boats whilst doing a bit of 3D modelling (a particularly Pythonesque image, I feel).
In all seriousness, I think these two pages show some of the spirit of innovation that still exists at SGI & MIPS, which is what makes those companies special. But I would think it calls into question whether SGI are using MIPS processors out of a strategic choice or out of necessity until something "better" (I'm not dissing MIPS, hence the quotes) arrives.
We must not forget that SGI are not the only customer for MIPS: Sony, LSI, NEC and Toshiba are just some of licensees of their technology, not forgetting Texas Instruments. So sure, they still design for the future but maybe not for themselves, at least in traditional markets.
OK, next issue:
[quote]<strong>Why should Apple buy SGI? To solve their immediate problem of competing with Intel and AMD processors? no. To get a DDR mainboard that can compete? no. There is nothing SGI has that will help Apple solve their immediate problems.</strong><hr></blockquote>
And I'm not sure that's why AAPL would buy SGI anyway. If you remember, I proposed that AAPL broaden its market offering thus:
[quote]<strong>Apple ultimately delivers four hardware product segments:-
Consumer & Education: eMac, iMac, iBook, iPod (and other i-devices)
General Corporate: Power Macintosh, Power Book, xServe
Specialist Corporate: revised SGI Workstation line, running G5 and OS XI (which is 64-bit pure)
Server: revised SGI Origin line, running G5 and OS XI, but with optimised graphics sub-systems modularised and made optional. Also the SAN and other storage products
Apple would then have an integrated product range, from SoHo and K-12 through small business through to corporate and specialist vertical visualisation and scientific.</strong><hr></blockquote>
So, what I'm saying is that AAPL will carry on developing the G5 platform (BTW, my thanks to thuh Freak for pointing out that references to G5 are a placeholder for "whatever comes after G4" because we all hope/pray/know - choose your own term - that something must come soon) based on its own design and that this platform will arrive as scheduled when the RDF mothership lands in SF next January. However it will only be 32-bit, due to OS considerations.
My theory says that what AAPL should then do 12 months later (although NAB 04 would be my target this time) is release a "PowerMac on steroids" (SGI-enhanced) workstation with more slots for both RAM and expansion, more SMP (4-way/8-way anyone), 64-bitness (OS XI) and whatever else will make a truly professional workstation sell.
As an extension to this, I also think that AAPL could/should consider segmenting their OS offering on a commercial basis. If you buy what I call either a "consumer and education" or "general corporate" Macintosh, you should be automatically licensed for 32-bit OS XI, however if you want to enable the 64-bitness of OS XI (probably more applicable to the "general corporate" market) you have to pay for an upgrade key as per QuickTime/QuickTime Pro.
Is that cute or is it just me?
[quote]<strong> In fact, the whole basis of this thread seems to be that since SGI MAY be designing one or more of their future machines using the G5, and since SGI (in the past) had good tech, Apple should buy them, because they are a potential competitor.</strong><hr></blockquote>
OK to address the basis of the thread, I have seen threads predicated on more tenous arguments than the one I'm using. Some of the threads before MWSF 02 were based on an advertising strapline ferchissake.
To address your second point, isn't that one of most often used reasons for buying a business? If The Rumour is true, than there is a tremendous opportunity to leverage synergies between two organisations who do not have overlapping product ranges, where proft & loss accounts can be improved by combining admin, accounting, support and several other areas.
HP's merger with Compaq (OK, takeover) was based on exactly the same justifications without the complementary product range argument, hence the pruning now underway.
[quote]<strong>The tech. and software that Apple would obtain is not going to be used in the iMac, not even the tower. This tech is way beyond the Mac, both in performance and cost. It will be very difficult to bring the cost down to a 'consumer' or even a 'pro-sumer' level. We would all love to see a Power4 in a Mac, but unless IBM develops a low-cal version, it wont happen. Likewise, a NUMA machine from SGI isn't going to happen at the pro-sumer level yet.</strong><hr></blockquote>
And, referring back to my reply to HOS at 19:45 yesterday as well as a section earlier in this post, I'm not saying that either.
This is a way for AAPL to become a dominant supplier of world-class workstation-derived technology all the way from Lawrence Livermore & Sandia to your study, taking in Ford, BP, DoCoMo, Stanford, Guildford Grammar School, and all points in between on the way.
It's about vision and seeing the power of bullet-proof core OS leveraged in all of those markets, on a coherent product family that can scale whether your a family going from iMac to PowerMac or a student going from iBook to TiBook or a business going from PowerMacs to PowerMac Pro and an 8-way SMP box to a 32-way box and onwards.
Business likes scalability and loves investment protection, and if gets to reduce support costs as well then you have a viable story to tell. If you can tell me how those aspirations are incompatible with what I'm proposing then I'll shut up.
[quote]<strong>To be sure, only the low end of the SGI line would use a G5.</strong><hr></blockquote>
Why? The current 64-bit element of SGI's range runs between 500 and 600Mhz on both the workstation side (Fuel and Octane2) and the Origin server space. If "G5" runs at around 1.6 - 2.0GHz and doesn't offer significantly better (I'm talking anywhere upwards of 35%) performance in a like-for-like situation, then we have definitely backed the wrong horse and someone should get a pink slip, and we should all go and price up our Clawhammer dream systems.
[quote]<strong>SGI has a wide range of machines and a large software base</strong><hr></blockquote>
Exactly and Exactly! For the latter point, imagine having it all running under OS X's successors.
[quote]<strong>If Apple does purchase SGI, it could create an "upper" segment, much like Ford's premier auto group (PAG). I would expect that they could benefit from SGI's brand identity (if it's still positive).
Companies are always considering such opportunities. If for some reason SGI could be bought at a bargain without losing much of it's advantages, I don't see why Apple (or someone else) wouldn't consider it.</strong><hr></blockquote>
GardenOfEarthlyDelights sees another facet of the argument: PAG gives Ford the ability to sell a complete range of cars to the individual and the enterprise. Hertz now supply Volvo, Jaguar and Land Rover as rental vehicles because Ford can offer them that range of product. It makes Hertz look better than Alamo if you're a status conscious user/chooser. An AAPL purchase of SGI would offer similar benefits, especially for companies tired of being robbed blind by M$.
And as GOED points out, if AAPL doesn't do it, someone else will and then we may very well land up paying a very heavy price.
[ 06-15-2002: Message edited by: Mark- Card Carrying FanaticRealist ]</p>
Remember A|w folks? They make a little app called Maya...
And we have follow-up rumors that SGI said, "Uh, no."
So, I could see Stev-o looking at the Apple board of directors and saying "Phuck them then, let's buy the entire flipping corporation!"
This way, Apple gets what they want (Maya & it's software engineers), and a healthy bonus to boot.
That would be all of Mark's hardware suppositions listed above... Which I wholeheartedly agree with...
I am just saying, I believe the MAIN reason that Apple would buy SGI is simply for Maya...
And then Pixar licenses PRRenderMan to Apple for inclusion into Maya as it's new default renderer...!
Thank you, I'll be here all week!
;^p
:cool: Maya for Mac OS X :cool:
[ 06-15-2002: Message edited by: MacRonin ]</p>
there is this little company called Grand Vitesse Systems, they repackage power macs into rackmount "psuedo" workstations, they do a fine job i might add. the cost isnt too far out of line with buying from apple, and their support is more profesional than apple (or so the IT guys tell me, i have never had to call them). now i say psuedo because its all standard apple fair under the hood, which means slow system bus, consumer grade graphics cards, and any other little problems with apple hardware that i could "nit pick", which i admit to doing. right now, apples hardware just cant cut it in the high end. sure a DP-1Ghz G4 will knock your socks off running a few p-shop filters, but this isnt what i do at work, neither does the air force (probubly sgi's favorite customer). the need for scaleable super computing, think top down solutions, is pretty big. the firm i work for could save thousands (being conservitive, im not an accountant), in misc. operating expenses. of course the downside is that apple would have us by the balls, not the worst position mind you, but not ideal.
in my opinion it will take apple a long time to be accepted in the high-end market, they have to far to go from my prospective. acquiring sgi would give them a "head start" in that direction, of course thats IF they want to go that way. its nice to dream though, think about it, a full set of applications, IRIX has the bare minimum. it would be great to run a simulation, and then crank out a findings report without walking across the office and starting up another computer. oh well whatever.
MCCFR-
More additional costs which have been left out of a potential AAPL takeover of SGI include opportunity costs. Money that has to go towards the purchase and operation/shutdown of SGI is money that is taken away from Apple's warchest, Apple's R&D, and Apple's operational expenses. Manager time and talent now has to attend to a new division which at first glance doesn't appear to be in sync with Apple's "vision" of computing.
To the other posters-
Don't forget that the initial argument here for Apple buying out SGI included the following assumptions:
1) Apple is planning on delivering some sort of high performance computing (HPC), of the sort which SGI currently builds-
[quote] Steve Jobs makes a throwaway comment, which can be paraphrased as "You want performance - wait and see what we're going to release, you'll be blown away" <hr></blockquote>
2) SGI may already be looking at or working on developing high performance PPC boxes, using the G5-
[quote] SGI is alledgedly considering a move away from its' own MIPS processors to PowerPC G5. <hr></blockquote>
3) The acquisition cost of Apple buying SGI is within Apple's budget-
[quote] SGI has a market cap of under $550M, and cash/short-term investments of around $200M.
Apple has a market cap in excess of $7B, and cash/short-term investments of around $4B. <hr></blockquote>
4) Apple and SGI's strategies of video editing and graphics are very similary (synergy, to borrow a horrid word)-
[quote] Apple appears to be indulging in a fit of whatever is the corporate equivalent of retail therapy, largely based around graphics and film/video post-production. Several of the products acquired are alledgedly 64-bit capable. <hr></blockquote>
So with these assumptions in place, for Apple to acquire SGI means that Apple needs to have compelling reasons to justify doing it. My argument has been one primarily of costs- that the direct and indirect costs of Apple acquiring SGI are far higher than the value Apple gains from the acquisition. My car analogies were to illustrate examples of companies buying out other companies which failed to deliver the value that could have been gained through not purchasing.
More on this later, but as the discussion keeps moving along, don't forget the assumptions that underlie this...
Sorry, got to run,
-HOS
<strong>Sorry I wasn't here to post earlier- I was at a pool party all day yesterday, drinking margaritas and applying sunscreen.
MCCFR-
More additional costs which have been left out of a potential AAPL takeover of SGI include opportunity costs. Money that has to go towards the purchase and operation/shutdown of SGI is money that is taken away from Apple's warchest, Apple's R&D, and Apple's operational expenses. Manager time and talent now has to attend to a new division which at first glance doesn't appear to be in sync with Apple's "vision" of computing.
To the other posters-
Don't forget that the initial argument here for Apple buying out SGI included the following assumptions:
1) Apple is planning on delivering some sort of high performance computing (HPC), of the sort which SGI currently builds-
So with these assumptions in place, for Apple to acquire SGI means that Apple needs to have compelling reasons to justify doing it. My argument has been one primarily of costs- that the direct and indirect costs of Apple acquiring SGI are far higher than the value Apple gains from the acquisition. My car analogies were to illustrate examples of companies buying out other companies which failed to deliver the value that could have been gained through not purchasing.
More on this later, but as the discussion keeps moving along, don't forget the assumptions that underlie this...
Sorry, got to run,
-HOS</strong><hr></blockquote>
HOS -
Well, your weekend sounds more relaxed than mine. But then mine involved watching England play football & cricket, neither of which are good for people of a nervous disposition. Also I rarely drink (normally cos I'm driving at some point in the day) and the UK rarely requires sunscreen.
I would agree with you that my whole concept is predicated on rumours, assumptions and possibilities of varying degrees of credibility.
I would remind you that I'm proposing the purchase of SGI be made 50% cash, 50% paper: That ratio could move to 20/80 and on to 0/100 the worse SGI's position gets (another bad quarter or a bad analysts report) and the more desperate their investors get to find a "safe harbour" where their money does not go up in smoke. I would think if you'd bought stock at $10 and it was now worth $2, you'd grab any passing ship which offered you a viable chance of not being eaten by a passing shark.
At that point, you may only be paying $350M to $400M gross all in paper which, although it would require the reorganisation of APPL stock and would devalue the share price, would not materially affect AAPL's cash position bar the reorganisation expenses.
At the launch of xServe, SJ said "we come into this market humble", a recognition that AAPL does not have credibility of any sort in that marketplace. I do not believe that I am the only person on this thread who believes that a bulk knowledge transfer from SGI could help AAPL make inroads in both the HPC and SMP server markets.
In addition, the acquistion would help AAPL in the high-performance workstation market where AAPL needs to get some clear blue water between itself and companies which will offer (or are offering) IA-32 MP offerings, such as HP/Compaq. Read some of the other contributions from people who really do give a damn.
We may not respect the Wintel market, but it is a reality which has been eroding the Apple shoreline for too long for us not to take the possibility of pragmatic solutions to at least a practical level of due dilligence. If we slip back into the NIH (not invented here) complacency and arrogance of the past, AAPL is doomed as an independent brand, and that should be the last thing any of us want.
I've heard the nVidia buzz, and it is interesting I grant you: but I am too much of a cynic and have seen too much water go under the bridge to have faith in developing "partnerships" with companies who could disappear in the next major consumer recession or change their allegiances depending on who offers the best "peace treaty". Say what you like about SGI, the one thing it doesn't rely on are the whims of the consumer market.
<strong>
More additional costs which have been left out of a potential AAPL takeover of SGI include opportunity costs. Money that has to go towards the purchase and operation/shutdown of SGI is money that is taken away from Apple's warchest, Apple's R&D, and Apple's operational expenses. Manager time and talent now has to attend to a new division which at first glance doesn't appear to be in sync with Apple's "vision" of computing.
-HOS</strong><hr></blockquote>
And this my friends is exactly why one NEVER buys a business to gut it. Remember everything that Apple does not NEED will have to be taken care of. Inventory sold at a loss to move it, buildings sold or leases ended, books closed out, contracts ended, accounting squared away, legal proceedings for accounts closings and contract endings. And this is the tip of the iceberg, details, lots of details. [edit] And how about all of those employees that Apple may not want or need?? They don't just disappear, severance pay, unemployment, help them find a new job?? Add to that the fact that these are people with families, ect. What a headache. [end edit] Easier to just offer SGI something good for MAYA, if they take it (they are stupid), all is good, if not buy the next best thing TREMOR.
End of one story, beginning of another.
[ 06-16-2002: Message edited by: Brendon ]</p>
when i said apple will have a hard time being excepted in the high end, mabey i should have said something like.
-apple will have some trouble breaking into the world of 500+ processor, 100 terra-flop, real-time 3D rendering super-computers.-
this is the world of sgi (which im sure your aware of), the world of custom applications and hardware, THAT high end. now i like macs, ive owned a few of em, but i dont think apple has the cred to pull that kind move off anytime soon. in seven years, i might be playing a different toon, but for now thats my take on it. i use both sgi hardware and apple hardware at work, a cluster of p-macs might cut it for some cash strapped students, but when money isnt the issue, you get the right tool for the job. to get away from the auto industrie allagories, when your child is sick with a rare disease, do you stick with your family GP, or seek the help of a specialist, heck your GP will usually refer you to a specailist. the kinds of projects that get run on sgi's hardware are paramount to the people running them. this is a difficult mentality to over come, when people want the best, expeirence counts. besides, we dont know if the board at apple really wants to get into this kind of business at all. but if they bought sgi, they wouldnt have to start at the bottom of the mountain to get to the summit. they could cut seven years of careful plodding down to say two or three, if the first apple/sgi super computer hit the bullseye, then they would have almost nothing to proove to their customers after that. in parting, if i was going to bet money, i would say apple is not going to buy sgi. just my opinion.
<strong>
And this my friends is exactly why one NEVER buys a business to gut it. Remember everything that Apple does not NEED will have to be taken care of. Inventory sold at a loss to move it, buildings sold or leases ended, books closed out, contracts ended, accounting squared away, legal proceedings for accounts closings and contract endings. And this is the tip of the iceberg, details, lots of details. [edit] And how about all of those employees that Apple may not want or need?? They don't just disappear, severance pay, unemployment, help them find a new job?? Add to that the fact that these are people with families, ect. What a headache. [end edit] Easier to just offer SGI something good for MAYA, if they take it (they are stupid), all is good, if not buy the next best thing TREMOR.
End of one story, beginning of another.
[ 06-16-2002: Message edited by: Brendon ]</strong><hr></blockquote>
If you're really sincere about your concern for the jobs of SGI employees, I would have thought you would argue for the case not against it.
Is AAPL going to continue imfringing on SGIs market? Either by developing a solution itself or by acquistion of other companies' IP by license or whatever. So the erosion still happens there. And it still happens from Sun and IBM and HP/Compaq. All companies with more money to fight a long fight than SGI. So what do you think the end result is going to be? Because SGI still being in action isn't one of them.
The solution I propose is the best option for AAPL boot-strapping itself into a market where budgets are decided by doubling the number you first fantasized, not the halving the number you put forward to your chief beancounter. Phamaceuticals, automotive, oil/gas, and high-end commercial physics like cold fusion: This is where AAPL goes under this proposal and it goes there on the fast train.
SGI's employees win by not having their work scattered to the four winds and suffering the tension of waiting for 80% of them to be given marching orders. My solution probably provides overlap in finance and a few other administrative areas and potentially SGI's manufacturing in WI - although I suspect that would stay, because AAPL would need somewhere to make the resulting product.
I regret the fact that some people and families would suffer under this scenario, but the key word is SOME. How many would survive under IBM assuming they were interested or HP/Compaq or Sun. Not nearly as many because the product overlap is too great.
And SGI's shareholders benefit because they get attached to a company with a mass market.
All of the stuff before your edit is absolutely true, that's why you reserve some cash for restructuring. I haven't been to SGI in the UK (Theale) for several years, but I tell you something it would be better for AAPL UK to move to Theale than stay in Stockley. Especially as the government is perhaps going to allow a new runway to built almost next to them. Probably cheaper in the long-run given Heathrow rents.
I covered the loss on inventory in my figures, assuming that AAPL would only be able to shift 50% of it at 50% of value. Which I thought was pretty vicious.
I'm not proposing butchering SGI, I'm proposing the surgical excision of its best technology, people and practices so that they can live to fight another day - I remember when you couldn't buy a computer without a full pre-sales programme, and lengthy proposals and planned implementation (maybe IT investment would show more ROI to business if more of that still happened), that's another thing SGI can show AAPL - how to make a proper business case for customers making an investment, and how to implement that investment in a value-added manner.
The rest of your objections are, with the greatest respect, low-level details. If it requires management time to resolve contracts, etc, etc, maybe this is where AAPL's legal, financial and property teams get to make a full-on front-and-centre contribution to helping the company grow. It's just time and money, and if due dilliegence showed it was worthwhile would you still object?
MCCFR, you have several strong arguments. Since I'm too lazy to go look up SEC filings on SGI and crunch my own numbers (I've not done any diligence at all, due or otherwise on SGI since they bought out Cray, and I had a college professor assure me that "this means supercomputers on your desktop in five years!"), so I'm definitely arguing from a weaker perspective. Especially since my position is that costs are too great for value received.
So instead of doing real work, I'll take the lazy way out and go after the assumptions- if I can topple enough of them, then the subsequent conclusions based on them will topple as well.
Assumption 1) Apple is working on a HPC product, or a product with HPC characteristics.
HPC is a very broad field, and traditionally ranges from individual vector processors (older Crays) to massively parallel machines (Beowulf clusters, ASCI White), and encompasses many things inbetween (Sun E15k). Prices are accordingly high- often starting in the $100Ks and scaling up to the millions.
There's no doubt in anyone's mind that Apple is interested in building computers which have better performance than their current machines. But a little bit better and orders of magnitude better both qualify as better than current offerings.
With prices between $800 and $4k, Macs are dirt-cheap compared to most all of SGI's offerings. In order to maintain these low prices, Apple adopts far more pedestrian, commodity parts.
Rather like the old MG or Triumph approach to performance- recycle existing parts in an exciting shell. (OK- cheap shot, but my last British car was my Dad's '52 MGTD... which doesn't go from 0-60 mph as it tops out about 55...)
Back to the argument, any HPC-like machine or technology that Apple employs (like a NUMA architecture) has to be cheap. Therefore, this must exclude many HPC characteristics which are too expensive to implement.
So if Apple is looking at HPC, it's looking at something really small-scale or cheap HPC, so we have minimal overlap with what SGI does... or charges.
Assumption 2) SGI is working on a PPC-based machine. Not provable one way or the other, and if not true, dampens a lot of the fire since this assumption is crucial to transition costs. If true though, this machine is more likely to be a million dollar server than a $1k blade. Again, we're not guaranteed that costs of this technology fall into line with what Apple can afford to charge, or even be the kind of technology that Apple's really interested in at the moment.
Assumption 3) SGI is sufficiently inexpensive that Apple can pick it up without too much sweat. MCCFR has done a very good analysis demonstrating that this is probably true. The only real hole I can knock in here is that even though it's within Apple's budget, it still represents a sizable expenditure which may be better spent elsewhere.
Assumption 4) Synergy in Apple's and SGI's video editing and 3D graphics markets.
While true to a degree, key apps like Maya already exist for MOSX, so other than control over the direction of Maya, the major benefit of Maya-on-the-Mac already exists. More importantly, the key infrastructure of generating video editing software (iMovie, Final Cut Pro) is already in place and running on both today's and yesterday's Macs. Whether software like this would run so much faster on future SGI hardware or future Apple hardware as to justify the cost of making it happen is up in the air. I don't know if high-end markets like television stations are willing to fork over the $100k for a high-end Final Cut Pro on SGI-class hardware, or whether $4k Apple hardware is sufficient to get the desired work done. The one thing I can easily state is that the demand curve clearly shows that more $4k boxes will sell than $100k boxes.
Next up, some returning commentary on past running commentary...
-HOS