Apple's New Market / Hardware Implications
Two short-run to mid-run scenarios in Apple?s market position within the Film industry exist for Apple. Apple is either set to miserably fail in this market if it fails to deliver the goods some people are expecting from the company, or diametrically it assuredly must have something we are completely ignorant about. Let me argue why I believe this is true.
Apple's recent foray into the high-end Compositing / Film, building on Final Cut Pro's increasing number of high profile successes and an accelerated rate of adoption has positioned Apple in a peculiar middle ground. Yes, Final Cut Pro is a more powerful solution than, say, Adobe's Premiere but it still remains significantly behind feature-wise vis-a-vis other non-linear apps available today. So one question arises from this unusual circumstance: Does Apple pursue the amateur to entry-level Film markets (i.e. direct competition with Premiere et al.) or does it reach for the stars? Evidently, as seen from the glorified FCP articles found on Apple's website, our beloved computer company has chosen the latter. It is encouraging to see that Apple is actively pursuing new markets - and I would fully encourage this were I in some decision-making position - but this has direct implications on what Apple must deliver as a hardware company. Remember, FCP is a Mac-only app...
Enter Shake and Tremor, 2 highly regarded compositing tools that have been widely used on some of the "biggest" filmmaking projects in recent times. Apple, having built confidence from its FCP success seems to genuinely believe that somehow it can manage to offer a compelling solution to studios / SFX houses by in some fashion or another combining its hardware, OS and product line in a synergistic manner. This is where Apple treads from familiar ground (think "innovation" ? excluding the source acquired from Macromedia for FCP?) into completely uncharted territory. Apple has never been a company that has proven itself competitive in high-performance computing markets. Apple has always been a "desktop" computer company at best (the nomenclature used for its performance connotations). So why should we have any reasonable expectations that Apple has what it takes for it to enter a high-performance market successfully? (Several underlying basic economic implications of this move, such as the unavoidable fact that its products will face a highly nascent point on the negatively sloped EOS curve further stress its high risk, which Apple hasn?t seemingly hedged in any way). Moreover, in regards to innovation, how does Apple plan to innovate in this instance?? It has virtually no expertise to draw from in this market (high-end compositing). In all likelihood they will not push for any major GUI overhaul of the proven Shake interface. This is consistent with the FCP approach for the OS 9 to the X-native upgrade ? this is also supported by various floating rumors indicating that the Nothing Real team is and will remain for the foreseeable future virtually autonomous from Cupertino intervention). At prima fascia, the only thing Apple seems able to offer is a serious influx of resources (which it seemingly has done) but for this to be Apple?s sole strategic objective at this point seems something Microsoft would do, not Apple/Jobs.
Now, in the case of FCP, Apple has been able to deliver a workable platform. It is somewhat slower than ideal but bearably so. And they really are trying to address this specifically (IMB Cache in Ti800Mhz?) but Apple seriously lacks the computing power that studios demand. And it doesn?t lack it marginally, it is embarrassingly behind. I can picture the Nothing Real people vectorizing the hell out of every possible line of code in Shake, but with Apple?s current mobo + g4, that just won?t cut it.
It is here where I present to you my two hypotheses and hope that after reading them you will provide enlightening differences of opinion / supportive extensions of my arguments.
Scenario one: Apple fulfills its (seemingly) required cyclical flop
Apple believes that it can seriously pull this off by basically depending on altivec units and steady (albeit excruciatingly slow) mobo / cpu improvements. This seems highly unlikely, and by this I mean EXTREMLY unlikely. Apple must realize that to attract hardware developers to migrate their high performance products to the Mac (this would be the case with high-end 3D card manufacturers, for example) it must itself seem a viable platform for the people that would buy their cards (in other words, studios must perceive Apple as a strong contender for inclusion in the studios? workflow). This is a terrible position from which to enter this demanding market.
I?ve loved Apple computers since my IIc when I was 7, but some feeling inside of me doesn?t let me reject this hypothesis on the basis of its irrationality alone.
Scenario 2: Apple knows that it has a chance.
Apple knows that it has the capability of delivering the hardware that meets the job. And by knows, I mean that Apple must somehow have been assured (formally or informally, doesn?t matter) by influential market actors that it holds promise in entering the market. This would explain the willingness and ease with which Nothing Real became an Apple product group, and it would also shed light on Apple?s ability to draw even more talent, such as from the likes of Discreet (for example). It would also explain Apple?s purchase of Nothing Real as a statement in itself. If this were not the case, Apple would have to kiss goodbye any good market relations it could enjoy with the film industry. Apple would have purchased one of the best (if not the best) compositing tool available only to then drop NT / Linux support and sell it as a Apple?only solution for highly underpowered PCs. Arguing that X?s Unix roots will ease transition to the platform for the studios, and that the Apple Shake solution will be worthy of Apple "elegance" is BS. Power. Untapped, raw processing power is what these customers want. So while they are at it, Apple should be busy at work beefing up SMP scalability in X.
I believe this case to be of greatest likelihood and of significant validity as to what is rational to expect from Apple based on its recent purchasing spree. It makes economic sense too. Apple could afford significantly higher R&D budgets if it were to produce high-performance machines specifically targeted at this group of customers. It could do so because as is to be expected these machines carry much heftier profit margins than those of your typical graphic designer oriented PM G4. Apple would then be able to trickle down the benefits reaped from this expanded R&D effort to its Professional, and eventually, consumer product lines. Apple would also obviously market the hell out of its new entrenchment in the movie industry and make the wild (but funny and factual) claims it tends to make in its ads. It does seem like a highly sensible plan. The only problem is that I have a strong reservation in believing that Apple can pull off such a quantum leap from its current performance stagnation in the near future (which should be the timeframe in which it would have to begin answering questions as to why they bout Nothing Real and what are they planning to do with it).
Let me what you think.
Apple's recent foray into the high-end Compositing / Film, building on Final Cut Pro's increasing number of high profile successes and an accelerated rate of adoption has positioned Apple in a peculiar middle ground. Yes, Final Cut Pro is a more powerful solution than, say, Adobe's Premiere but it still remains significantly behind feature-wise vis-a-vis other non-linear apps available today. So one question arises from this unusual circumstance: Does Apple pursue the amateur to entry-level Film markets (i.e. direct competition with Premiere et al.) or does it reach for the stars? Evidently, as seen from the glorified FCP articles found on Apple's website, our beloved computer company has chosen the latter. It is encouraging to see that Apple is actively pursuing new markets - and I would fully encourage this were I in some decision-making position - but this has direct implications on what Apple must deliver as a hardware company. Remember, FCP is a Mac-only app...
Enter Shake and Tremor, 2 highly regarded compositing tools that have been widely used on some of the "biggest" filmmaking projects in recent times. Apple, having built confidence from its FCP success seems to genuinely believe that somehow it can manage to offer a compelling solution to studios / SFX houses by in some fashion or another combining its hardware, OS and product line in a synergistic manner. This is where Apple treads from familiar ground (think "innovation" ? excluding the source acquired from Macromedia for FCP?) into completely uncharted territory. Apple has never been a company that has proven itself competitive in high-performance computing markets. Apple has always been a "desktop" computer company at best (the nomenclature used for its performance connotations). So why should we have any reasonable expectations that Apple has what it takes for it to enter a high-performance market successfully? (Several underlying basic economic implications of this move, such as the unavoidable fact that its products will face a highly nascent point on the negatively sloped EOS curve further stress its high risk, which Apple hasn?t seemingly hedged in any way). Moreover, in regards to innovation, how does Apple plan to innovate in this instance?? It has virtually no expertise to draw from in this market (high-end compositing). In all likelihood they will not push for any major GUI overhaul of the proven Shake interface. This is consistent with the FCP approach for the OS 9 to the X-native upgrade ? this is also supported by various floating rumors indicating that the Nothing Real team is and will remain for the foreseeable future virtually autonomous from Cupertino intervention). At prima fascia, the only thing Apple seems able to offer is a serious influx of resources (which it seemingly has done) but for this to be Apple?s sole strategic objective at this point seems something Microsoft would do, not Apple/Jobs.
Now, in the case of FCP, Apple has been able to deliver a workable platform. It is somewhat slower than ideal but bearably so. And they really are trying to address this specifically (IMB Cache in Ti800Mhz?) but Apple seriously lacks the computing power that studios demand. And it doesn?t lack it marginally, it is embarrassingly behind. I can picture the Nothing Real people vectorizing the hell out of every possible line of code in Shake, but with Apple?s current mobo + g4, that just won?t cut it.
It is here where I present to you my two hypotheses and hope that after reading them you will provide enlightening differences of opinion / supportive extensions of my arguments.
Scenario one: Apple fulfills its (seemingly) required cyclical flop
Apple believes that it can seriously pull this off by basically depending on altivec units and steady (albeit excruciatingly slow) mobo / cpu improvements. This seems highly unlikely, and by this I mean EXTREMLY unlikely. Apple must realize that to attract hardware developers to migrate their high performance products to the Mac (this would be the case with high-end 3D card manufacturers, for example) it must itself seem a viable platform for the people that would buy their cards (in other words, studios must perceive Apple as a strong contender for inclusion in the studios? workflow). This is a terrible position from which to enter this demanding market.
I?ve loved Apple computers since my IIc when I was 7, but some feeling inside of me doesn?t let me reject this hypothesis on the basis of its irrationality alone.
Scenario 2: Apple knows that it has a chance.
Apple knows that it has the capability of delivering the hardware that meets the job. And by knows, I mean that Apple must somehow have been assured (formally or informally, doesn?t matter) by influential market actors that it holds promise in entering the market. This would explain the willingness and ease with which Nothing Real became an Apple product group, and it would also shed light on Apple?s ability to draw even more talent, such as from the likes of Discreet (for example). It would also explain Apple?s purchase of Nothing Real as a statement in itself. If this were not the case, Apple would have to kiss goodbye any good market relations it could enjoy with the film industry. Apple would have purchased one of the best (if not the best) compositing tool available only to then drop NT / Linux support and sell it as a Apple?only solution for highly underpowered PCs. Arguing that X?s Unix roots will ease transition to the platform for the studios, and that the Apple Shake solution will be worthy of Apple "elegance" is BS. Power. Untapped, raw processing power is what these customers want. So while they are at it, Apple should be busy at work beefing up SMP scalability in X.
I believe this case to be of greatest likelihood and of significant validity as to what is rational to expect from Apple based on its recent purchasing spree. It makes economic sense too. Apple could afford significantly higher R&D budgets if it were to produce high-performance machines specifically targeted at this group of customers. It could do so because as is to be expected these machines carry much heftier profit margins than those of your typical graphic designer oriented PM G4. Apple would then be able to trickle down the benefits reaped from this expanded R&D effort to its Professional, and eventually, consumer product lines. Apple would also obviously market the hell out of its new entrenchment in the movie industry and make the wild (but funny and factual) claims it tends to make in its ads. It does seem like a highly sensible plan. The only problem is that I have a strong reservation in believing that Apple can pull off such a quantum leap from its current performance stagnation in the near future (which should be the timeframe in which it would have to begin answering questions as to why they bout Nothing Real and what are they planning to do with it).
Let me what you think.
Comments
[ 05-03-2002: Message edited by: Nostradamus ]</p>
If you are correct about Apple's plans, then that could prove to be problematic for Apple. I have two concerns that are both interconnected.
1. For Apple to be successful in the high-performance market, they must not only produce machines with superior price/performance, but they must maintain the performance edge over time. Apple is known for putting out must have hardware, (PowerBook for example), but they are not known for keeping that hardware up-to-date with current technology (PowerBook for example). Studios are going to want to know that Apple can keep up with their performance hungry needs for the long term. High performance markets need bleeding edge all the time, not just once every three years. I fear Apple can't keep up.
2. Apple runs the risk of becoming too fragmented. Every presidential hopeful knows that in order to win the primaries they have to secure the base. Apple has to secure its base and right now, I don't think they are doing a good job of that. Apple has already somewhat sacrificed the low end market by leaving them with geriatric G3 technology at prices that don't fit their offerings. They have done this while responding to the education market's needs with the powerful and affordable eMac. Some faithful budget consumers might feel a little jilted by this. Now they are apparently dumping massive amounts of resources into the high performance market. It will look bad for Apple if they produce a killer machine for this market but leave their prosumers stranded with G4 933s and aging mobo technology. Eventually the prosumers will get worthy upgrades but by the time they do, it might feel like they are only getting the crumbs that falls from a more important market's table.
I think it is poor timing for Apple to pursue other markets. They need to understand that they are first and foremost a consumer/prosumers hardware company. It seems that they have lost that focus and by pursuing other markets, that focus could be further diluted. The first thing they need to do is offer their faithful base the absolute best computing experience and value on earth. They need to spend some of that cash to release the G5, 10.2, 400MH system buses and anything else they have up their sleeves as soon as humanly possible.
Some of the worst press Apple gets comes from loyal customers on bulletin boards who feel jilted by undressed product defects and sluggish performance. And from friendly columnists in the press who are forced to write about high prices and sluggish web browsing, etc... The Mac faithful talk more about what they want from Apple that might be just around the corner, than they do about what great things they have from Apple. This is not good. Apple needs to deliver the goods to repay and renew the faith of the faithful. Only then, should Apple consider expanding to other markets. If they lose too much of their base, there won't be enough Apple to support other markets.
This is an important general observation, even in an industry that by its very nature demands a 'what's next' sensibility.
As head of Pixar, Jobs surely has a precise vision of what it will take to someday say, "We aint gonna work on Intel's, Sun's, or anyone else's render farm no more."
-----------------------------
Watching the Talent in Nashvegas
<a href="http://www.joecroker.com/flash/flashani.html"; target="_blank">http://www.joecroker.com/flash/flashani.html</a>;
I think those systems can somehow be compared, a lot more than to compare a mac to a dedicated system like a quantel editbox e.g..
from here it occurs to me that the sgi-system is a LOT more fluent in terms of responsiveness of the ui. this is a keypoint for me if it comes to work.
because of (in my eyes) the superior combination of bandwith/throughput and formidable 3d-graphics in the sgi-box i think the mac-system is definitive behind in that.
every system has to render, shure. but this is only one key to attract people to work with this or that machine.
but if i want to work fast and try different things in a short time, the system i am working on has to follow me. it can't be that fcp takes time to REDRAW its ui.
only my 5 cents....
nbd
If Apple can deliver the single and dual '32 bit' G5 to their traditional Powermac consumers then its game on.
For their 'new' markets, then quad '32 bit' G5s...and '64 bit' versions later can cater to them. Hopefully along with a 'high end' graphic card range.
The Apollo...with it's 2nd round of bumps should be driven into the consumer ranges as soon as possible.
Maybe the next half year or so will see this take place...?
At the moment...?
It's too easy to get a PC machine for a 1/3 of the price of the top Powermac and blow it out the water performance wise.
You pay BMW prices? Expect BMW performance...not a BMW chassis and Ford tyres, clutch and engine!
Like you say. I guess one supposes that Apple have something up their sleeve.
If they don't then AMD's new offerings will beat the bejeezus out them. More so. They already convincingly do!
Lemon Bon Bon
I believe Apple is acting so confidently in this high-stakes, high-performance market because they've got the hardware to run with the big boys. It's not here quite yet, but come MWNY it will be. At the very latest, the new hardware may be announced at MWSF, but I'll be surprised if it comes so late in the game.
If the G4 were all Apple had to rely on, and all they had to look forward to were incremental bumps in G4 design like the one's we've been witnessing lately, then there is NO WAY Apple could hope to become competitive in any high performance markets. It is the G5 that Apple is counting on.
It all fits. The recent migration of consumer lines to the G4, the stagnation of the Powermac line with the same pitiful motherboard for so long, the recent low-key revision of the Powermacs...Apple is on the verge of introducing a radically new Powermac line, one that will change the face of computing as we know it. Get ready to see the iMac bumped to a GHz G4 at MWNY, and for the iBook to migrate to the G4. The current processor lineup represents Apple's future low end.
I suspect the new Powermacs will have a new motherboard that lays to rests any worries about bandwidth, and GHz will be impressive to say the least. The high end will be around 1.4 or 1.6 GHz, dual CPU. It will rock.
If I'm wrong then may God save Apple, because they are otherwise doomed.
I rally myself when I think of the days when Apple had the best bar none PC's on the market for any price. When they had "desktop rockets."
My theory:
Apple realized sometime during the early life of the G4 that it wasnt going to be the end all solution. I think at that time Apple and Steve went to a new plan: START OVER. I think Apple has an entirely different processor than what anybody is even guessing will be released.
W.W.D.C. or M.W.N.Y. will see the release of this new processor. I can see the entire Apple line with G4's while the powermac gets its new megachip. It makes some sense. The G4 is a pretty fast and efficient chip by most standards, nothing to scoff at. Perfectly reasonable in the consumer and portable lines. The powermac will be priced slightly higher than it is now and may even retain one model at a sub 2k pricepoint while all the new ones become REAL machines, even if it means REAL MONEY. Lets say from $2600 to $6,000.
[ 05-03-2002: Message edited by: fellow722 ]</p>
There might be room for uber-configs on the top-top end -- like a 4 proc machine with huge fast data caches and a pro-level OpenGL card -- but they must keep a couple of 2000K and under machines, or they just won't sell enough powermacs.
I love it when The Dawg gets his bark up! Bellow, Dawg! Hound it!
And Fellow722 dropped a big bone on the deck, too:
[quote] I can see Apple releasing its next gen processor VERY soon. A processor that REDEFINES computing. I wouldnt put it past them to release a product so staggering that it left the likes of Intel and AMD so far behind, as in GIGAHERTZ behind. <hr></blockquote>
I just hope you fellers are right. These are bold dreams. Let's wake up to them soon.
C'mon Apple.
-------------------------------
Sniffing for Talent in Nashvegas
<a href="http://www.littlevinnie.com"; target="_blank">http://www.littlevinnie.com</a>;
Why not take advantage of some of the G4 advantages and make a render farm on a card?
Take 4 low power G4's with about 256 to 512MB of Ram each and slap them on a PCI-X card. Span a huge heatsink over the whole affair and give a strong fan. In the re-designed PM case, you make a special PCI slot on the end of the board with it substantial power and it's own vent.
You sell PM's in two flavors. The regular SP and DP configs OR the regular configs plus a 2000-2500 optional multipurpose render card. It doesn't draw the GUI, or load the OS or programs, it only crunches numbers as instructed by your favorite computationally intensive 3-d renderer, simulation, or science app.
Even now, take your average SOI G4 800. $125 x 4 = 500 bucks. plus RAM, heat sink, and fan, I think you could certainly supply a 4 proc card for 2500. Get .13u G4's and that's even better. Space efficient too. You'd need at least three machines to get 6 procs to work crunching your data in a farm. This way, it costs, but you can get it all in one Apple machine.
3000 for a DP powermac + 2500 for a specially designed low power quad G4 render card and you get a number crunching monster for about $5500. Worth it, if that's where you make your living!
The other good thing about it is it could be kinda standard in that it takes advantage of PPC developments as they happen. When a faster PPC part comes along, Apple can start outfitting the render cards with newer low power chips to always keep the whole package fresh.
Just have to build a case with a good strong PS and adequate cooling. Loud, yeah maybe, but who gives a shit if you've got 6 PPC's working for you in there?
Nostradamus:
[quote] Firewire 2, Quicktime 6 and OS 10.2 will play key roles for Apple's entry into this market. None of them are released yet, but come this summer, I think we will have them. The Cinema HD is a definite sign that Apple is seriously interested in pursuing the film industry.<hr></blockquote>
You say that these technologies are representative of Apple's explicit efforts in attempteing to better serve the Film market. I would agree with a statement that would argue that these technologies may have some usefulness in the Film market, but they are hardly indicative of any serious commitment. These technologies represent functionality that is 100% aligned with Apple's current niche markets / strategies (i.e "creative professionals," digital hub strategy, etc)and thus are "hedged" efforts that to any hardcore mac loyalist would not represent a strategic departure by Apple in any real way.
I do agree with you that it is very difficult to realistically picture Apple as increasing hardware performance as much as is really required for usability by these customers.
Mac Voyer
[quote] Studios are going to want to know that Apple can keep up with their performance hungry needs for the long term. High performance markets need bleeding edge all the time, not just once every three years. I fear Apple can't keep up <hr></blockquote>]
I couldn't agree more. Apple has not proven itself as a company capabale of keeping at pace with the hardware market. Actually, not even the software market (look at how long it took them to fundamentally improve their OS, after failing repeatedly with Rhapsody, (2nd Gen) Gershwin, etc...). This failure rate I may say was quite comparable, at times even inferior, to Microsoft's development efforts. Now, XP is poised to overhaul its filesystem to a database-akin structure...which is extremely more powerful than it may sound. Sure, Apple has hired the engineer behind BeOS' BFS, but Apple really needs to make a public commitment to significant inovations if we are to expecti it to not remain on its laurels as it did throughout System 6 to 9.
[quote] The Mac faithful talk more about what they want from Apple that might be just around the corner, than they do about what great things they have from Apple. This is not good. Apple needs to deliver the goods to repay and renew the faith of the faithful. <hr></blockquote>
i couldn't disagree with you more. All my claims are based on factual, historical experience of Apple's performance. If any noe way were to be deemed soundest as a manner in which one evaluates a company, I would say looking at its history record would be it. This is the most objective manner by which we can forecast future behavior / perfromance.
This basis on Apple's track record dispenses all the religious undertone in your remark. And as to the claim that "this is not good," this is actually the BEST thing we can do as Apple suporters / advocates. At no point in time have I un-objectively trashed Apple with no merit, but instead I have only offered constructive criticism which, along with the others who offer it, allows Apple to better understand its current standing and perception vis-a-vis its customers.
Syabritic:
[quote] This is an important general observation, even in an industry that by its very nature demands a 'what's next' sensibility. <hr></blockquote>
I see nothing wrong with a company having a demanding customer base. Would a complacent customer base provide the incentive structure for Apple to innovate? Would a complacent customer base be conducive to market growth for Apple?
Basics in economic theory would explain why this is true, and moreover why it is more generally a market reality.
nbd:
I don't quote you because I would quote your entire post..which seems pointless.
I would just like to thank you for such an informative post. What we do need on these boards is better informed discussion that leads to greater understanding of Apple's problems.
I myself have experience with certain sgi boxes (onyx (very limited) and flint/flame) although my use was not compositing (more scitech stuff,: protein modeling)
I couldn't agree more that Apple is unacceptably behind on the performance curve.
Lemon Bon Bon:
[quote]The G5, whatever the form it takes becomes more and more imperative to apple's plans. <hr></blockquote>
Well put. It is somewhat paradoxical that Apple, a company that has been understood by some as primarily a software company (in many ways I would agree with this) is becoming prisoner of its "revitalized" software strategy because of its inferior hardware.
There is a ray of hope that a new processor (whichever that is) is coming relatively soon, with the iMac and eMac already sporting G4s at speeds so near the professional line.
However, if the "QuickSilver 2002 Mobo" on eBay is any indication, we will just be seeing more of the same old Apple insignificant upgrades.
Junkyard Dawg:
I agree with you, as I have aforementioned in this post, that Apple's current pro hardware lineup seems posied for a major hardware revision. But I can easily see Apple trying to oversell a good upgrade as a best upgrade, as it currently does with its pr machine that seems permantly stuck on overdrive.
(Note, sometimes this strategy has worked, as in the case with the iPod in which most people became disappointed at the face value features of the player, but over time the industry has repeatedly called it the "absolute best mp3 player"...)
If Apple choses this strategy (delivering a good upgrade, and then running specialized tests to show that "performance" on a Mac is better than on Intel/AMD (by lack of a qualifier "performance" as Apple calls it seems to reference "overall perfromance" which is not true at all in these tests)
The fundamental problem with this isn't Apple's core markets, as they have come to accept this and live with it. It is, however, a critical problem with the Film industry. These people, because of their extremely demandind computing needs, literally don't even have the time to deal with this nonsense.
Matsu:
[quote] 2600 to 6000 = no sales to anyone but very high-end production houses. No edu, no science, no consumers, and no servers.<hr></blockquote>
I would highly disagree with this. I would believe that in fact this would increase sales to some of the markets you deem as negatively affected. Mobo improvments are always welcome addition to the server market, and this market does not obey any pricing schedule. Servers run from very cheap, to very very expensive, price points at which even $10k seems insignifcant (think $150k). Space also exists for such hardware in .edu. Actually, it creates space. Why is Apple makeing the eMac? Because this is what the market wants for mundane scholastic use. But schools also have unix infrastructure (aka Sun and IRIX labs, for example, which are not only Sun / IRIXZ because they are unix, but because they are high-performance labs). Apple's X has awarded Apple a unique opportunity to penetrate the REAL places where computer science is taught (students do not learn cs in mac or wintel labs...) but it desperatly needs the hardware necessary to push it through the door.
Matsu:
[quote] Why not take advantage of some of the G4 advantages and make a render farm on a card? <hr></blockquote>
Why intorduce new inefficencies by builiding on existing (but surmountable) ones? We could make the arguement that Apple could do this with G4's, G3 machines, even, say, have a renderfarm of 6100s...
You cannot expect that just multiplying the number of machines will solve all problems. This approach alone also introduces added costs to the end-users. The fact remains that the problem rests on Apple's side of the court.
Your proc card idea is another common band-aid many advocate.
I only critically evaluate Apple so strongly because I expect the BEST from them. And I do so because I truly believe that they are the only ones capable of deliviering what I would ever consider BEST. But I do spend so much talking about them because I fundamentall believe that the better equiped we are as a demanding customer base, the more "insanely great" products we will see, and a better market standing Apple will enjoy.
None of my comments above are meant to be "personal attacks" in any way. I just beleive that it is important to keep a serious conversation about Apple's future as focused and objective as possible.
By the way, Apple still remains very vague on <a href="http://maccentral.macworld.com/news/0205/03.shake.php"; target="_blank">Shake</a>[
I love to hear what other people think so let me know what you think.
My theory:
Apple realized sometime during the early life of the G4 that it wasnt going to be the end all solution. I think at that time Apple and Steve went to a new plan: START OVER. I think Apple has an entirely different processor than what anybody is even guessing will be released.
[ 05-03-2002: Message edited by: fellow722 ][/QB]<hr></blockquote>
That would explain the stories that Moto has stopped work on the "G5"/8500. If it 's something totally different, then maybe some insider type folks could poke around for other areas of intense chip development.
Indeed. THAT is THE question.
The smoke signals say something(!) is coming. But exactly what?
Shake certainly gives a tantalising clue.
I can't believe Apple thinks that people are going to pay almost £3,000 smackers for a dual 1.2 or 1.4 gig G4 processor with its lame single fpu unit.
It's so laughably out of date compared to the Wintel/WinAMD boxes out there. Even with a mobo bump.
It's going to take a significant shift in Apple's Power(!)Mac strategy to be a mover and 'shake(r)'(groan) in the film industry...
If Apple 'cancelled' the 8500 G5 then why? Apple have recently moved to more G4s. Are they forgoing the 'performance edge' for mhz?
Will they incorporate the G5 architectural perks in favor of ramping the G4's clock speed?
Doesn't make sense to me...
Unless Apple have something significantly better behind it?
Many rumours have whispered of the G5...why cancel it so close to release?
'It' is still on the PPC road map...
I'm frustrated as hell with the whole processor thing. I can't wait to see it resolved.
Lemon Bon Bon
[ 05-04-2002: Message edited by: Lemon Bon Bon ]</p>
Whatever you give them, they'll put it in front of a render-farm. Right now the render farm comes from a slew of cheap boxes or rack-mounts. Little farms (1-6 machines) can add quite a boost. I know of people using one and two DP Athlon boxes as render farms at home! Nutters, die hards really, but you can imagine that businesses with a real interest would certainly cough up the cash.
The Powermac still needs to get the fastest hardware possible (and soon) this would be a way to offer that much more power in a flexible solution that creates sales for Apple and not the AMD/intel Box vendor du jour. Do we need a G5 or something like it? Yes sir, and soon too. But when we get it, there's no reason not to give people as many as can be crammed into one box.
If you don't want the render farm on a card, you don't buy it, you just buy the fastest G5 ddr PowerMac you can afford knowing that it will run faster than anything else available from anyone. I want this for Apple as well. But if you're one of those people whose going to take that machine and make the front end for a little cluster of boxs (likely not macs) wouldn't you consider a solution that let you have one box on your desk instead of 5?
I 100% agree, that such a farm-card should not be used as a band-aid. The box itself must be competitively priced and powerful in it's own right. The card is an extra targeted at render-farm sales, not workstation sales. I think it brings a very Apple like solution (integration and space efficiency) that only multiplies it's advantages as your needs grow.
For example:
Say I'm a geology prof who has the budget for a small cluster. If I buy 24 machines at 1500 dollars a piece, that's $36,000. Forget networking costs for now, let's just keep it simple. I'll also need a lot of power to run them, but I don't pay that, so forget that too. I will need another room to hold this little cluster cause they'll squeeze me out of my office/lab if I keep them all in there. (I've actually seen this, some profs labs become so rammed with boxes/equipment that you can barely move) I like mac, I'm spending university money so hey, I treat myself good and put a nice powermac on my desk, with the cluster in the room next door.
But If I bought 4 or 5 DP PowerMacs with render cards, I'd have 24-30 procs for $22-27,500 (at 5500 each). Instead of putting them all in another room, I can put them under a table, or in a custom cart int he corner of my office/lab and have them right there where no one but me can tinker with them. I save a room worth of space, and I can get through models/simulations and calcs FAST. Were talking the difference between a computation that takes 2 days versus one that takes 90 minutes. There are dept/people that will pay good money to do that, especially to do it without giving up the use of a whole room.
In the first case, Apple sells 1 machine, and a lucky vendor sells 2 dozen boxes. In the second case Apple sells 4 or 5 very expensive machines, and a community that really matters (ie, they'll spend big money often) walks away impressed.
however, in <a href="http://forums.appleinsider.com/cgi-bin/ultimatebb.cgi?ubb=get_topic&f=1&t=001608&p=3"; target="_blank">this</a> thread (is this a hoax?) a remark from bodhi caught my eye, because a "friend" told him a story that matches the story i heard.
now i'm really eager to find out if it's true... <img src="graemlins/surprised.gif" border="0" alt="[Surprised]" />
I couldn't agree more....
-----------------------------
Enjoying the Nashvegas Talent Grind
<a href="http://www.jencohen.com/"; target="_blank">http://www.jencohen.com/</a>;
Our next server purchase will most likely be a couple blade server racks for our web servers coupled to a high-end Xenon database cluster on the back-end.
To a company like ours MHZ DOES NOT MATTER! What matters is how fast will the server run our application. The new Xenon processors based off the P4 are NOT faster than our current setup...that's why we haven't upgraded to them yet!
This having been said, "IF" Apple releases servers they MUST have the following options:
Redundant power supplies, Redundant cooling, hot swap HD's, the support for Intel NICs with port trunking, etc.
Would our company switch away from Dell to Apple IF they came out with quality server hardware? Yes, if pricing was similar and performance was similar.
Also, the argument that Univerities will not spend $10,000+ on a server cluster is crazy. Good universities do it every year...try $100,000+ for a high end cluster with many nodes.